Home

Online Help

1. FID Connect C2BatteryC Online Help Status Disconnected C Show details Connection lt gt 458 5 Status Status Load TX Unit Name Disconnected 62 42 52 C2BatteryC Disconnected a Buss 5 1 Introduction amp Content Overview 2 Lower Screen C2 MMI Overview 3 Message Windows 4 Card deck panel 5 Graphical display Unit Name Properties Opens the Filter Property dialog for the selected filter This is also accessible when double clicking a filter in the list 6 Tool bar IC 7 Status Bar 8 Menu C2 a 9 Menu Item Situation Picture 10 Menu Item Mission 11 Menu Item Order 12 Menu Item Communication 9 12 1 Control Panel 912 2 Control Panel 12 2 1 Control Panel Settings 12 2 2 Control Panel New Filt E 12 3 L11B Control Panel 12 4 Asterix Control Panel Online Help The implementation of user manuals into a radar application from Saab Microwave Systems 9 12 5 Unit Register 12 6 2 Planning Mode 13 Menu Item Tools 14 Menu Item Window
2. fal Air Track List X TN Ste Sro Info TON distr Sensor TN 4 IAD Platform ThreatLev Strength Liveness Emergency 0116 0003 Remote Incoming 000 Helicop No 1 Alive 0117 0004 Local 000 1 Dying 0118 0003 Common Common 1000 1 Alive 0119 0003 Local 000 1 Alive 0110 0002 Remote Incoming 000 d Identity of track Possible values are 1 Alive S 0111 0004 Local Outgoing 000 d Pending 1 Alive 0112 0002 Common Common _ 000 Unknown 1 Alive 0113 0003 Local Outgoing 000 Assumed friend 1 Alive 0114 0003 Remote Incoming 000 4 1 Alive 0115 0001 Local 000 Friend 1 Alive 0129 0001 Local 000 Neutral 1 Dying 0127 0003 Local 000 Suspect 1 Alive 0128 0003 Common Common _ 000 Hostile 1 Alive 0121 0002 Common Common _ 000 1 Alive 0122 0004 Local Outgoing 000 1 Alive 0120 0004 Local 000 Missile Medium 1 1 Alive Local Eloctranic os 1 1 il D gt Add air tracks manually 0 30 Add Remove Clear Figure 13 13 Pop up help in ejected mode The menu item windows The other part the context sensitive help are the windows which are accessed from the menus The context sensitive help in these windows works the same way as the ejected windows previous mentioned The only difference between them is that there are normally a lot of more components and therefore more Pop ups The TDN Control Panel that was used in the last usability test contains a
3. 12 2 Basics This dialog show information regarding the collected from unit register The tables have cq information of the defined filters The operate simultaneously Procedures Connecting to unit Choose the unit you wish to connect to from transmit in full or limited power FTD or LT Setting up or editing a filter In order to reduce the amount of outgoing Filter dialog Click here to get to the New TDI Interface context from top down 18 TDN Control Panel TONI Master of Science Thesis in the Programme Interaction Design Sebastian Dreyer Chalmers University of Technology University of Gothenburg Department of Computer Science and Engineering G teborg Sweden November 2009 The Author grants to Chalmers University of Technology and University of Gothenburg the non exclusive right to publish the Work electronically and in a non commercial purpose make it accessible on the Internet The Author warrants that he she is the author to the Work and warrants that the Work does not contain text pictures or other material that violates copyright law The Author shall when transferring the rights of the Work to a third party for example a publisher or a company acknowledge the third party about this agreement If the Author has signed a copyright agreement with a third party regarding the Work the Author warrants hereby that he she has obtained any necessary permission from t
4. 33 RNA 33 e DA CTIVIEWER 2 33 8 6 FIREFOX PORTABLE 33 86 1 Testing 33 8 7 CHOOSING ONLINE HELP PLATFORM rr sr sr sr 34 9 DESIGNING THE INITIAL PROTOTYPE 35 9 1 BUILDING THE PROTOTYPE 35 9 2 TESTING THE 36 9 2 1 Feedback what WAS 8004 36 9 2 2 Feedback on what was bad ccccccccccccssssssceccccccsssseccecccecsussseccecccessusssecceccceususesecsssessuusaessesceeees 36 9 2 3 Miscellaneous feedback 36 9 3 FOCUS OF THE NEXT PROTOTYPE 37 10 DESIGNING THE SECOND PROTOTYPE cssscccssssscesssscccssssccscssscccessscccessseccscssscecesscecessnee 38 10 CREATING THE HELP CONTENT 38 10 2 BUILDING THE SECOND PROTOTYPE ccccscceesesesesseeseseseseseseseseseseseseseseseseseseseseseseseseseseseeeseeesees 39 10 3 TESTING THE SECOND PROTOTYPE 39 10 3 1 what was goods esiri 39 10 3 2 Feedback on wha
5. Thomson 2009 Raynald Thomson Ltd What Is Software Prototyping http www reynardthomson com what is prototyping html 2009 UPA 2005 Usability Professionals Association UPA Heuristic Evaluation http www usabilitybok org methods p275 2005 UsabilityNet 2006 Usability Net Design Guidelines http www usabilitynet org tools designguidelines htm 2006 68 Appendix Evaluation of Usability Test 1 Positiva noteringar Klassificering pers Generellt mycket h g acceptans p det totala konceptet 3 Att f upp Popup f nster med ett musklick Interaktion 3 Att f kort snabb hj lp med vissa komponenter Effektivitet 3 Att kunna f hj lp via klick p label och inte bara p aktuell Effektivitet 3 komponent tillfredst llelse Att n popup med mittenknappen Interaktion 1 Bra att f nsterplacering sker rakt under mark ren F nsterplacering 1 Negativa noteringar Inte g r att f rstora minska Popup F nsterhantering 2 Inte g r att flytta Popup kan ibland d lja GUD F nsterhantering 1 Att inte kunna anv nda h germusknapp f r att ppna Popup Interaktion 1 Scrollbar har l gre acceptans n st rre storlek p popup F nsterhantering 1 F nsterplaceringen gjorde ibland att labels skymdes F nsterplacering 1 Att inte markerat objekt highlightas i F1 dialog n r den Effektivitet 1 ppnas Tillfredst llelse Ovriga kommentarer Om popuperna kan flyt
6. 2 19 decision but have not really understood first then manual Read first act later Seeking understanding Needed hint Not until breifing Very narrow Mean value time 18 5 min Standard deviation 6 36396103 Observations 4 Want to drag and drop directly into the Sit display Found L11B s chapter about filters uses it instead Annoying that the Online manual disappears every time you click on something in the user interface Tries to open Pop up in coord field does not try to do the same thing on the label Right reasoning with Point still unsure what really means 5 Filter inside outside hard to understand Point feels superior to platform ID in GUI Did not use the Online Help to understand it Thinks that it should say what is needed for the options in the group box to work It is not apparent why you can t set geographical boundaries Not clearly stated in the manual Hardly uses Pop ups Does not think that New is associated with filter should say New Filter instead 6 Hardly used the Online Help 100 reasoning from reading labels and trial and error Started reading the online manual when he noticed Point but did not understand if it should be checked or not Thinks that Pop ups seems to be a good idea if you are uncertain of what a component means Although there are not Pop ups for all components nor do they always help An interfa
7. 30 8 Searching for Online Help System Platforms One of the first challenges with the thesis was to determine which platform to use for the Online Help System Researching the subject gave the answer that there were a multitude of such platform on the market This was done by first using Wikipedia to see what system was available there were a few but only JavaHelp and Oracle Help were of any real interest since they were Java based After this some time were put on using Google search to find more examples of Online Help platforms there were a lot of hits but most were either outdated or not Java based The only real option that was found was Eclipse Platform Help System EPHS There was also some research done within Saab which resulted in two possible options a solution using Firefox Portable and another one using ActiViewer In order to decide which of these were most suitable each platform was tested and evaluated The description of the platforms can be read in chapter 4 5 8 1 Building a test GUI In order to test the help platforms some sort of test Graphical User Interface GUT was needed Because the underlying code was not important it was decided to use NetBeans to create the GUI In NetBeans you can simply drag and drop different components on a surface and the code for the resulting GUI is auto generated It all worked out great until the GUI was to be imported into Eclipse The problem was that there was no way to import the pr
8. DOKUMENTATION STRIC MUSEIPLATS OPPEN UNCLASSIFIED FHL OPPEN UNCLASSIFIED SekrL APPROVED VASS 98 APPROVED DATE 2009 05 18 Dokumentation STRIC museiplats 1 Dokumentationsbeskrivning 2 Sakerhetsinstruktion 3 Visningsregler 4 Kanda restriktioner fel problemomraden 5 Systembeskrivning 5 1 Beskrivning 5 2 Operat rsplatsen 5 3 Presentations och inmatningsutrustning 5 3 1 Presentationsutrustning 5 3 1 1 Beskrivning 5 3 1 2 Huvudindikator 5 3 1 3 Tabl indikator 5 3 2 Inmatningsutrustning 5 4 Presentation 5 5 Uppspelningsfunktion av civila radarstationsinformation 5 6 Kortfattad beskrivning av m lf ljningsfuktionen 5 7 vriga funktioner 6 Handhavande sy mp 09001 pm305231 xml 3 ssk 2009 05 19 10 24 RELEASE Approved Ala r ttigheter f rbeh lls Reproduktion kopiering eller utel mnande till tredje man i n gon som helst form medges icke utan skriftligt samtycke fr n All rights strictly reserved Reproduction or issue to third parties in any form whatever is not permitted without written authority from the proprietors Figure 2 5 Online Help with Firefox Portable 11 3 Theory Since this is master thesis in interaction design it means that Usability and User Centered Design is a main part of the thesis There are also a number of methods that can be used in an interaction design process whereas some of them w
9. number of iterations and the improvement in usability As can be seen in the figure the curve will go upwards after each iteration early in the development after a while though the curve will begin to smooth out and eventually it will plateau Ideally each iteration will be better than the last one but this is not always the case sometimes the changes made in the design will actually make the interface worse than before The curve in Figure 4 2 should therefore be seen as an ideal example of how iterative design should work Neilsen 1993 The main advantages with an iterative design process over a more classic one like the waterfall model are numerous for example 1 Serious misunderstandings and usability problems are discovered early in the project when there is still time to do something about them 2 It makes it possible for as well as encourages users to give feedback 3 Finding the most serious issues early forces the development team to focus on and can thus ignore at least temporary other lesser issues 4 Doing continuous iterative testing enables a more objective view of the project s status 5 The development team workload is spread out more evenly throughout the development cycle 6 The team will quickly get used to the iterative design process and thus continuously improving it 7 It is easier to show the stakeholders that the development is going forward Kruchten 2000 4 4 Software Prototyping Software p
10. 2 52 14 DISCUSSION 56 141 THE FINISHED PROTOTYPE 57 142 THE DEVELOPMENT PROCESS 58 5 60 TA Z WRITING THE REPORT sb ense E E 60 14 5 THE RESULT FROM THE THIRD USABILITY 61 14 6 FUTURE 64 IS CONCLUSION tosvoss osts os vssssovssv vesssossseos os voosisssssosbsvsooonsosssss saoss usssssnsssbovesssv senn ns nsesss ss rsessssdosessnrs 65 15 1 CONCLUDING REMARKS ON THE THIRD USABILITY 65 16 REFERENCES 67 APPENDIX EVALUATION OF USABILITY TEST 1 69 APPENDIX EVALUATION OF USABILITY TEST 2 71 APPENDIX USABILITY TEST ASSIGNMEN T csssccsssssscssssccccssssccesssscccssscccsssscssesssccssssnees 73 APPENDIX D OBSERVATION PROTOCOL sscccssssscssssscccssssccccssscccscssscccssssccscsscecsesssscccssnees 74 APPENDIX QUESTIONNALURE cccssssssssssscscssssccscssscccssssscccssssccccssscccssssccscsssaccscssccssesssscccsssees 75 APPENDIX F RESULTS FROM USABILITY TEST 3 sccssssscscssssccesssscccssssccccsscccsesssccsessnees 77 1 Introduction This report will describe
11. except that he could not test the prototypes himself since they were not allowed to leave the building Technically he could have come to SMW and test it here but since he could look at screenshots of the prototypes and get a first hand explanation of the how it worked he thought that it were enough All this also ties into the fact that the end users could not be used to test the prototype since they are military personal from another country Instead people within SMW were used to test the prototype since they had the necessary experience to do this Also since this thesis is not meant for commercial use as well as having a limited budget this seemed like a feasible solution 59 The supervision of this thesis been bit different At SMW one of the senior programmer were appointed as supervisor for this thesis He have been a great help with everything programming and Eclipse related but the one that have put down most time helping out with this thesis is the Human Factor Engineer HFE that work with interaction design and usability at SMW Considering his background and the focus of this thesis his expertise has been most helpful in the development of the prototypes and usability tests He sort of became the unofficial supervisor of this thesis Having two supervisors had its pros and cons it was easier to get help with different aspects of the thesis both in the area of interaction design and in programming but mostly interaction
12. 1 1 2 3 mean val stand dev 5 6 7 10 X Numerical Question 2 1 10 N 77 2 X 8 3 X 6 mean val 7 3333333 stand dev 1 1547005 Question3 1 2 3 4 5 6 7 8 9 10 1 x 2 X 3 X mean val X stand dev Question 4 1 2 3 4 5 6 7 8 1 2 3 mean val stand dev Question5 1 2 3 4 5 6 7 8 1 9 2 X 7 3 X 6 mean val X 7 3333333 stand dev 1 5275252 Question6 1 2 3 4 5 6 7 8 9 10 1 X 2 X 3 X mean val X stand dev OJN 10 gt lt gt lt X lt O 10 COIN Participant thoughts What kind of help did you prefer All six participants preferred the Online Help over the traditional manuals What do you think are the biggest advantages with the kind of help you prefer Since all user answered Online Help on the previous question that will be the preferred help type It s more detailed It is easier to see in which order you have to do stuff There is a search function available Quicker explanation of buttons and fields Easier to find what you look for Direct access to information about the component function that you are currently using Other thought People should be educated in the use of manual and then practise with the Online
13. 2 6 5 These slowdowns are caused by the single humongous HTML topic file that the help content consisted of Usually in a help content each topic and subtopic should have their own HTML file so splitting the file would also make the help content seem more like the real deal It was decided that splitting each subtopic would be too cumbersome so for the moment only the main topics or chapters would have their own HTML file Unfortunately RoboHelp had no function that made it possible to automatically split the file neither was there any function to do this while converting the Word document into RoboHelp format So this had to be done manually with copy paste The topic IDs had to be updated so that they referenced to the new topic files After this a new set of JavaHelp files were generated and viewed in RoboHelp The slowdowns were completely gone Interestingly some of the new HTML files were pretty large but this did not cause any slowdown whatsoever It seems like the size of the files have to pass a magic border of some sort before they start to slow down the JavaHelp system 38 10 2 Building Second Prototype The first step was to implement the new help content into the application There was a small problem the Pop ups could no longer be used since the helpfile used for the new online manual was no longer the same as the one the Pop ups were in The options were to either un JAR the help content file and manually add the
14. G l Dir cted DESIGN sioe eiee eee eaa teva bind pie ease Osea 14 A METHODS 19 A USABILITY 19 A D HEURISTIG EVALUATION 19 AS ITERATIVE 20 A A SOPEWARE PROTOTYPING 22 AS DESIGN GUIDELINES 23 5 PLANNIN G E E EEEE 24 SF TIMEPLAN 24 5 2 PREPARATION AAAA EA EAE sd AAEE AA A AEREE IE EEA EE PEIE AEREE AEAEE E AEE 24 5 3 LO e L0 a ONKO A d 2 EAAS NAE A EEA S E E EEA A EAA AE 25 5 4 FIRS T PROTOTYPE 25 5 5 SECOND PROTOTYPE 25 5 6 THIRD AND FINAL PROTOTYPE 26 6 MAKING HCI GUIDELINES FOR ONLINE sssscssssscscsssccssssccccssssccscssscccsssscccesssseces 27 6 1 DESIGN GUIDELINES FOR ONLINE HELP 27 7 CREATING AMOCK UP 0ss ss v es sssssoo ss soss s ocss sp s usdssvsnsoosps sssssns nsov sssovesssr ssonisssosos ss ssonssssnsssorssnre 29 8 SEARCHING FOR ONLINE HELP SYSTEM PLATFORM sssccssssssccssseccsssssccessscccssscecoes 31 8 T BUIEDINGA TEST GUL 31 8 2 ECLIPSE PLATFORM HELP SYSTEM 32 ek RO ca 32 OF JA VAHELP 32 SDT OSUNG 32
15. Help in order to get a feeling of what kind of functions that exist and then quickly be able get hold of details through the Online Help It is a bit annoying that the Online Help disappear every time you click on something in the interface You want to be able to have the Online Help open beside the interface 78 while working Instead there is lot of clicking every time you want to open Online Help again The Online Help works very well when needing information about a certain field i e short brief information But it the user manual feels like a better choice if you want to read longer sections to understand how a function works No matter how good the Online Help is it is still important that the interface is as intuitive as possible so that you won t need it Results from the observation protocols Traditional manuals Time Tester min Questions 1 2 3 4 5 6 7 1 2 3 1 22 3 xX X 2 16 2 X gt Ro x 3 13 1 X X gt 6 17 Wrong values After hint 1 half 2 half Mean value time 17 min Standard deviation 4 58257569 Observations 1 2 3 Took a while to find New TDN Filter does not read to get help Want to have drag function in Sit display Trial and error all the way Does not even read the labels Have problems to create geographic boundaries because of EW strobes Want to have warn
16. SMW develops their software in The reason it is important to use the already existing documentation to generate the content of the Online Help is because SMW wants to avoid as much extra work as possible since they have limited time and resources to put on this As long as the traditional manual is requested by the customers it is important to generate it from the same source as the Online Help One complication possibility is that the department that handles documentation at SMW today develops an IETM Interactive Electronic Technical Manual based the system ActiViewer and the material is therefore adapted to ActiViewer Optimally SMW wants to use the same material to create the traditional manual IETM and Online Help 1 2 Constraints As stated above the point of this thesis is not to implement a complete working help system but rather explain how such a system would work and how to implement it The prototype will have working context sensitive help for a number of windows those that is deemed necessary in order to demonstrate the help system as well as online manual in the product It will not have help for all windows and the online manual will feature only part of the user manual The importance is to show how to use suitable tools to create help system files from the source material 1 3 Saab Microwave Systems Saab Microwave Systems is a leading supplier of Radar Systems that are used for military purposes around the world They hav
17. and to make sure that the code for this thesis was safely backed up 9 1 Building the Prototype With this done it was time to start on the prototype In order to get more familiar with the code the first thing that was implemented was a Help menu in the interface with the same options as in the mock up mentioned earlier see chapter 7 To start things of a help class called MMIHelp was created to handle the creation of a HelpSet variable that is linked to the helpset file as well as a HelpBroker variable that is created from the HelpSet variable and handles all the JavaHelp operations This class was written as a Singelton class so that only one instance of it can exist at a time or else the JavaHelp would start to malfunction To make things easy in the beginning the same help files that were used on the test GUI was used here as well The menu item Help Content under the Help menu was then set to open the online manual when pressed But the online manual should also be available when pressing F1 in a window or side panel The starting point of the Online Help implementation was at a side panel called Weapon Engagement Status the reason for choosing this panel was that it contained a table as well as a few checkboxes and labels and therefore a perfect place to start testing the Pop up windows But before that the panel was linked to a page in the online manual with the enableHelpkKey function so that that specific page was opened when the
18. become the most popular of usability inspection methods The main goal is to find usability problems in the design so that they can be attended as a part of an iterative design process Heuristic Evaluations are done by a few numbers of evaluators that examines the interface to see if it coincides with recognized usability principles so called heuristics 19 It is not recommended that heuristic evaluation is done single person because it is not possible for one person to find all usability problems in an interface People are different and different people find different usability problems Nielsen 2005 Because of this it is possible to find most usability problem in an interface with just a few participants Jakob Nielsen the creator of the heuristic evaluation recommends the use of three to five evaluators using more people than that doesn t give much additional information The heuristic evaluation is done by letting the each evaluator inspect the interface alone Not until all of the evaluators are finished are they allowed to talk to each other This in order to make sure that each evaluation is independent and unbiased The result can be written down in a report by each evaluator or can be verbally done to an observer that is present during each evaluation The latter is recommended because it is easier to organize one set of personal notes than a set of reports written by others The observer can also assist the evaluator i
19. examples where appropriate Support searching with the help of indexes table of content glossaries and free search Include lists with known and common error messages Include FAQ with clear answers Lastly there are some pointers of how the user support system should be designed gt gt gt Online Help should not be seen as an add on to the product Online Help should ideally be designed together with the rest of the system There are several ways the user can access the help system commands buttons functions that can be turned on or off or a separate application Using keyboard or mouse buttons are suitable for context sensitive help The way the help system is shown is also important an application that is run in windowed mode should show the Online Help in a separate window while a fullscreen application should have it in splitscreen ToolTips are useful to show short bits of information to the user The documentation and looks of the Online Help should be designed the same way as the interface in the application by taking into account the ability of the user and the user demands 28 7 Creating In order to get a clear picture of what kind of Online Help functionality that was going to be implemented as well as how it would look and work it was decided that the easiest way of doing this was to do a mock up of the interface with the Online Help functionality implemented This was done by taking scr
20. file and then delete the old one or close the document and then rename it there is no Rename menu command This is the natural way for a computer to work but not how a normal person does things Warning Failed to notify library Figure 3 2 A non helpful warning window Cooper 2007 The main problem is that the digital industry does not have a good understanding of what makes users happy Sure they have information on what kind of market they are in and what jobs they have how much money they make and spend and what they like to buy But this does not really tell anything about what makes them happy or how they are going to use a product There is also conflicting interest between the users and the ones that build the products the programmers Programmers often have to choose between making the coding easy for themselves and making the product easy to use It is not hard to figure out which of these tracks they are going to take It is never a good idea to let the people who are building the product also design it But the real issue here is the lack of any process to create successful products Lacking may not be the right word there are different processes used by both by the engineering and marketing departments but they are focused on other things than making products that meets the users needs and desires Cooper 2007 Designing Digital Products It is important to realize that creating a digital product is not the same thing as c
21. is in this example 14 Figure 13 1 The Help Content window from the prototype 13 2 1 The Toolbar The toolbar contains six button icons that can be seen more clearly in Figure 13 2 Butcon one and two handles navigation back and forward and the third is the home butcon like a web browser The fourth is Print next is Page Setup and the last one is Add to favorites which puts a bookmark of the current topic in the favorites tab Figure 13 2 The Toolbar 13 2 2 The Navigation Panel This panel contains five tabs the first is the Table of Content TOC tab followed by Index Search Glossary and Favorites tab Table of Content tab The TOC tab as the name says contains the table of content of the online manual It has a hierarchical structure just like a file tree in windows explorer see Figure 13 3 Each item is mapped to a topic file all the items on the same branch is linked to the same topic file but to different places 48 1 Introduction amp Content Overview 2 Lower Screen C2 MMI Overview 3 Message Windows 4 Card deck panel 4 1 Upper deck Overview 4 2 Upper deck Air Track List J 4 3 Upper deck Associations 4 4 Upper deck Presentation Settings FE 4 5 Upper deck Map Control Panel 9 4 6 Upper deck Weapon Engagement Stat 747 Unner derk Mission Control Figure 13 3 The Table of Content tab Index tab An index
22. med online manual F1 support f r att anv nda den och Pop up hj lp i hela applikationen Det tredje testet l t testarna l sa en uppgift mha antingen pappersmanualerna eller OH f r att se om det var en f rdel att anv nda den ena hj lp typen ut ver den andra Slutsatsen r att det r m jligt att billigt och relativt enkelt implementera OH i en redan f rdig produkt Dock s r JavaHelp l ngt fr n en perfekt l sning och kommer att beh va arbetas om f r att kunna anv ndas kommersiellt Det r ocks m jligt att skapa en online manual fr n redan existerande material som Framemaker och Word dokument genom att anv nda programmet RoboHelp Preface This thesis is the last step in my education in Interaction Design at Chalmers University of Technology and the IT University in Gothenburg The thesis took place at Saab Microwave Systems SMW in Gothenburg during May October 2009 I would like to thank my supervisor tutor Ulf Dahlgren at SMW for giving the technical support I needed to do my prototypes Gustav Klock that as the resident Human Factor Engineer at SMW help me a lot with interaction design related issues and acted as an unofficial second supervisor mentor it was great to have someone to talk ideas with Rafael Aramburo my boss for helping me to get this thesis Also like to thank everyone else at SMW for acting as test users or otherwise helping me with this thesis and making me feel welcome I would like to than
23. never implemented ADA APT BOC CANTCO CDO CEH Data mile DLP FCO Air Raid Warning Figure 13 6 The Glossary tab Favorites tab The tab will be empty in the beginning but when the user clicks on the Add Favorites butcon in the toolbar a bookmark will be added for the currently opened topic in the favorites tab see Figure 13 7 Clicking on it later on will immediately open that topic in the topic window 50 58 Menu Item Situation Picture 3 Card Deck Panel Figure 13 7 13 2 3 The Topic Window The topic window shows the HTML topic files that are linked to the TOC and Index see the right portion of Figure 13 1 13 3 Online Help in the Prototype The main part of the Online Help in the prototype apart from the help content is the context sensitive help Context sensitive help is as the name states help that is given differently depending on the context There are two main context sensitive help types in the prototype F1 key help support that opens the help content on the particular topic that handles that window panel The other one is Pop up help that are shown for a number of components in the interface buttons text fields checkboxes etc Aside from that there is also another minor but important part of the Online Help that have not been mentioned yet and that is the Help menu 13 3 1 The Help Menu The help menu as in most application is the last menu in the applica
24. new prototype 11 1 Building the Final Prototype The work was started by updating the help content with the help of RoboHelp an index and glossary was created The index could be created automatically with a Wizard in RoboHelp while the glossary needed to be created manually All the topics were checked and all references to other topics were changed into actual links The most tedious part was to implement Pop ups in the rest of the application As mentioned earlier each unique Pop up has a unique topic HTML file the files were saved in hierarchical way All in all there were around 180 topic HTML files and 30 folders Creating all these HTML files were not that difficult just tiresome Something that was a bit difficult was implementing Pop ups in the table of the Air Track List panel That table was custom made by SMW and did not work the same way as the JTable used in the rest of the application It took a while to figure out how to link the topic IDs to the column headers and the problem was solved Around this time the bug that made the Pop ups appear in the left corner started to make itself known again This happened in the Air Track List in window mode as well this window has a table with over 30 columns and the window therefore has a scrollbar The bug happens when Pop ups are shown in the columns that are normally hidden by the scrollbar This seemed strange at first but it soon made sense because the Pop ups in the tables are set
25. reduced Some of the test participant preferred to us the right mouse button instead together with What 15 this menu It was thus decided to implement this additional way of showing Pop ups as a complement to the other one giving the user options of using whichever mouse button that felt the most comfortable As a last touch most of the windows in the application was liked to their respective topic in the online manual as well as creating a new menu item in the Help menu called Acronyms and Abbreviations and linked that to the similar topic in the online manual 10 3 Testing the Second Prototype The test was of similar structure as last time Three new participants were asked to play around in the interface mainly the Mission Control and Weapon Engagement Status panel window The goal was to see if the problems with the window management had been solved and if there were any new problems had appeared or not been noticed last time The result was put together in tables following the same template as last time and can be seen in Appendix B The result can be summarized as followed 10 3 1 Feedback on what was good Interaction 39 last time test users liked the simplicity and showed high acceptance towards the use of Online Help Window placement and management This time the test users hardly had anything bad to say about the window placement and management None of the users noticed that there were two
26. the air traffic which is shown in a very large table If the F1 key is pressed the help content will open at the topic explaining this panel see Figure 13 10 Clicking with the middle mouse button on one of the table column headers will show a Pop up window placed under and to the left of the column header explaining that particular column see Figure 13 11 As mentioned earlier the table is very large too large to be studied in the panel Because of this it is possible to eject the panel onto the screen in a windowed mode This is done by clicking the eject button in the upper left corner of the panel see Figure 13 12 TN Air Track List 011 Associations 011 Presentation Settings 011 Control Panel Weapon Engagement Status 011 Mission Control 011 Attentions 011 Conflicts 011 Orders In 011 Out 1129 ANNA I ntal Figure 13 9 Dropdown Menu 52 2008 08 11 13 15 18 Upper deck Air Track List 1 Infraduetian Content Overview a The alr track list show all avallable air tracks or jam strobes found in the graphical display The card deck panel offers 1 1 General window amp button notation limited space for visual inspection However a full viewing of the air track information can be seen when pressing the FS 1 2 The Operator Workstation Eject button The air track list has the abiity to provide an enormous amount of information which can be fully configured 7 4 3 Man Machine Interfac
27. there was problem downloading a trial version of the latest version from Adobe for some reason Importing a FrameMaker document may give a better result because both FrameMaker and RoboHelp are Adobe products The rest of the chapter will discuss the prototype the development process if this thesis will be useful for SMW problems that made the authoring of this report problematic the test results from the last usability test and future work on the prototype 56 14 1 The Finished Prototype Of course the prototype is far from perfect it is a prototype after all The main problem is the use of JavaHelp to implement the Online Help JavaHelp does not seem to be a fully developed product yet and probably never will be since there have been no major changes in it since 2004 The bugs and shortcomings are many window management is one of the greater ones For example the Pop up windows are created inside the window which they are shown this has the unfortunate side effect that the Pop up windows cannot be appear outside that window This is not a problem as long as the window is of considerable size but in small windows there may be problems to show Pop ups without either of them covering the whole window or making them so small that scrollbars would be needed to show all the text Understandably neither of these solutions are good ones so it might be better to not use Pop ups at all in small windows Another window management problems is t
28. to open right under the first column this works fine on those tables without scrollbar The problem is that when the first column disappear out of sight while scrolling right the Pop up can no longer be shown at those coordinates and therefore appear in the left corner of the window Fixing this was easy the Pop up location was just set to the x coordinate of the mouse pointer Doing this also makes the interaction seem more intuitive when the Pop up window appears right where the user clicks on the middle mouse button After realizing this all Pop up locations were set to show at the x coordinate of the mouse pointer This was not the only place that the bug happened Most of the windows in the application were pretty slim and thus made it hard for the Pop ups to fit under some of the components also this problem only became worse when the Pop ups were set to open at the mouse pointer To fix this problem a number of things were done Some windows were made wider other windows that had very few components had Pop ups that said that the user should check the help content instead but mostly the locations of the Pop ups were restricted to certain positions for example if a certain Pop up window s location passed a certain coordinate its position became fixed 11 1 1 Adapting the prototype to the next usability test The following usability test would let the test users do a task where he or she would have to create a filter in the Tactical Data
29. tt identiteter 5 Inf ra geografisk begr nsning 6 S tta h jdbegr nsning 7 Filtrera r tt sida av begr nsningen Intressanta kommentarer reflektioner fr n anv ndaren under testets g ng Observat rens spontana k nsla av skillnad vid anv ndning av andra hj lptypen Hur l ses uppgiften sett ur Sekventiellt med infoh mtning allt eftersom trial and error perspektivet p vilket s tt anv ndaren nyttjar tillg nglig hj lp Laser f rst igenom allt f r f rst else och agerar d refter Hybrid l ser delar f r f rst else agerar och g r samma sak om igen 74 Du har f tt i uppgift att anv nda pappersmanual __ onlinemanual Appendix E Questionnaire 1 Hur upplevde du tids tg ngen f r att hitta r tt information Mkt tidskr vande Mkt tidseffektivt 1 2 3 4 5 6 7 8 9 10 2 upplevde du kvaliteten informationen som s dan Intets gande Mkt givande 1 2 3 4 5 6 7 8 9 10 3 Till vilken grad upplevde du att du fick hj lp att n ditt m l Mkt liten grad Mkt h g grad 1 2 3 4 5 6 7 8 9 10 4 l tt r hj lpen att l ra sig att anv nda ett effektivt s tt Mkt sv rt Mkt l tt 1 2 3 4 5 6 7 8 9 10 5 I vilken grad upplever du att hj lpen ger information vilka steg som n dv ndiga att genomf ra och varf r de r viktiga att genomf ra f r att n ett visst m l Mkt liten grad Mkt h g
30. was possible to implement Online Help but due the limited time and budget it was not feasible to put extra resources to do this at that time 5 2 5 Other applications with Online Help There are of course other applications by different companies that have Online Help Mostly though they are limited to just implementing help content online manual in their applications This include popular applications like Microsoft Word or Eclipse which has a online version of their help content which demands an internet connection to use There are very few applications that have context sensitive help Pop up help for example No such applications were found and studied during this thesis 2 6 The Available Online Help Platforms There are a number of available platforms on the market that allows implementation of Online Help in the product This thesis will only focus on those that are Java based since that is the programming language used in the application There are three platforms available freely on the market and two more that are Saab solutions These are Eclipse Platform Help System EPHS JavaHelp Oracle Help and within Saab ActiViewer and a Firefox Portable solution 2 6 1 Eclipse Platform Help System In Eclipse it is possible to create help content Online Help to a project in the form of a plug in This is easily done by creating a new project and selecting Plug in Project instead of Java Project after filling out some sta
31. 6 Se ovan 4 70 Appendix Evaluation of Usability Test 2 Positiva noteringar Klassificering pers Generellt mycket h g acceptans p det totala konceptet 3 Att f upp Popup f nster med ett musklick Interaktion 3 Att f kort snabb hj lp med vissa komponenter Effektivitet 3 Att n popup med mittenknappen Interaktion 2 Att n popup med h gerknappen Interaktion 1 L gger inte m rke till de olika storlekarna F nsterhantering 3 Bra att f nsterplacering sker under komponent F nsterplacering 1 Negativa noteringar Inte g r att flytta Popup F nsterhantering 1 Texten st mmer inte Rubriker beh vs ndras Manual 2 Manual inte tillr ckligt utf rlig beh ver mer hj lp n vad vissa knappar mm betyder Manual 1 Ser inte po ngen med What is this menyn K nns lite konstigt med bara ett menyalternativ Interaktion 2 Vissa komponenter ska inte beh va Popup hj lp ska vara sj lvf rklarande Andam lsenlighet 1 Ingen knapp Windows standard Interaktion 2 Tillfredst llelse Fokus p 1 fungerar inte alltid Interaktion 3 Effektivitet Ingen hj lp f r m l p kartan Interaktion 1 Tillfredst llelse Ingen hj lp i menyer Interaktion 1 vriga kommentarer Beh vs fels kningsschema varf r fungerar det inte som det ska etc ndam lsenlighet 1 Wizards kanske r en l sning Tryck p F2 f r att f fokus pa Popu
32. Content The help content window is divided into three sections a toolbar a navigation panel with several tabs and a topic window all of which are shown in Figure 13 1 below 47 B Online Help 5789 lt 855 1 Introduction amp Content Overview Introduction amp Content Overview D 2 Lower Screen C2 MMI Overview This manual refers to the Saab Giraffe AMB System Generation III The way of reading and this document follow a certain flow and notation in order to reduce search time and increase intuitive understanding of its contents Below follows a brief layout of the notations and ways of 3 Message Windows R describing the contents of this document 4 Card deck panel 5 Graphical display x 6 Tool bar G New TDN Filter 7 Status Bar 8 Menu C2 ae Suppress source 9 Menu Item Situation Picture 10 Menu Item Mission oat 11 Menu Order Faered object types 12 Menu Item Communication vi Point 13 Menu Item Tools Airtrack and EW Jam Strobe 14 Menu Item Window 1510 gt Hostile Suspect 16 Symbols Shapes amp Areas gt Neutral v Assumed friend 17 Popup Windows gt Friend Unknown D 18 Acronyms and Abbreviations r Pending Filtered volume Fig 1 Dialog example In the example above the main chapter name is presented as 9 Menu Item Communication represents an item in the menu bar and its sub menu or subchapter
33. F1 key was pressed Implementing the Pop up help was not that complicated either at least not for the checkboxes First of all the enableHelp function was used to link topic IDs to the checkboxes and a mouseListener was added to those same component The responding function was programmed so that when the middle mouse button was clicked on such a component a new Pop up window would be created that showed the topic that corresponded to that component The Pop up window was closed and destroyed as soon as something else outside the Pop up was clicked This was easily done for the checkboxes but it was not as easy to implement in the table headers The problem here is that the table headers in a JTable lies under a single component that can be gotten by using the function getTableHeader on the JTable the table headers themselves have no components of their own The solution to this problem was to make some changes to TableModel of the JTable The project code contains a class called ColumnDefinition that contains information about the columns in a table like max and minimum size among others This class was edited so that the topic ID was included as well for each column In order this to actually work it has to be included into the TableModel1 somehow fortunately the TableColumn class that the TableModel uses contains a variable called identifier that is not used for anything This variable is of the type object the most basic type in Java and
34. ISO 9241 11 1998 E 1998 03 15 http www idemployee id tue nl g w m rauterberg lecturenotes ISO924 1 part11 pdf JavaHelp 2004 Sun Microsystems Inc JavaHelp 2 0 System User s Guide http java sun com javase technologies desktop javahelp download_binary html 2004 12 Kruchten 2000 Kruchten P From Waterfall to Iterative Development A Challenging Transition for Project Managers The Rational Edge 2000 http www ibm com developerworks rational library content RationalEdge dec00 Fro mWaterfalltolterativeDevelopmentDec00 pdf 2004 12 Mcclendon et al 1999 Mcclendon C M Regot L Akers G What is Prototyping http www umsl edu sauterv analysis prototyping proto html 1999 05 26 Nielsen 1993 Nielsen J Iterative User Interface Design http www useit com traditionals iterative_design 1993 11 Nielsen 2005 Nielsen J How to Conduct a Heuristic Evaluation http www useit com traditionals heuristic heuristic_evaluation html 2005 Oracle 2009 Oracle Oracle Fusion Middleware Developer s Guide for Oracle Help 11g Release I 5 0 http download oracle com docs cd E12839_0O1 doc 1111 e14149 toc htm 2009 Saab 2007 Saab AB Operations http www saabgroup com en AboutSaab Organisation SaabMicrowaveSystems operat ions htm 2007 06 01 67 Saab 2006 Saab Product Offerings http www saabgroup com en AboutSaab Organisation SaabMicrowaveSystems produ ct_offerings htm 2006 09 11
35. Net TDN Control Panel window To make this test as successful as possible extra time and effort were put into the TDN Control Panel to make the Online Help as helpful as possible The help content topic regarding 41 this window and its sub windows was rewritten so that control and operation chapters were integrated together in the traditional version these are two separate manuals Extra care was also made to make sure that there was Pop ups on all important components Before the test the whole prototype were looked over it was then something strange was noticed A number of windows that had F1 context sensitive help were no longer working i e pushing F1 resulted in an exception from the application it did not crash just sending error messages whenever F1 was pressed Those windows had worked perfectly just a few days earlier and no change had been done in them since so the origin of this exception was a mystery and still is the only conclusion that have been drawn is that it has something to do with the helpset but what is unclear Fortunately this did not happen to the TDN Control Panel window and the usability test was thus unaffected 11 2 The Third Usability Test As mentioned in the previous chapter the third usability test would let the users do an assignment The assignment was given in paper form that the testers had to read before the test and is shown in Appendix C It focuses on making a TDN filter that removes airp
36. Pop up topic files edit the map file and then make it into a JAR file again or just make them into separate helpsets The latter one seemed the best choice so the MMIHelp class was edited to create two different helpset references one for the online manual and one for the Pop up help Next the code was changed so that F1 help used the online manual helpset and the components with Pop up help used the Pop up helpset As mentioned in chapter 9 3 the issues with the positioning of the Pop up windows have to be solved There is a function in the Popup class that makes it possible to set the position of the Pop up window The position was set to appear a bit under the component To solve the problem with the scrollbars in the Pop up windows a second Pop up window size was defined in the helpset file which was twice as large as the original This new size was used on Pop ups those texts did not fit in the original size A strange bug showed itself while opening Pop ups in Mission Control and Weapon Engagement Status in windowed mode The nature of the bug is that some of the Pop ups appear in the left corner of the window distorted instead of under the components While this bug was mystery at first it was soon discovered that if Pop up did not fit inside the window it was shown in the left corner instead It seems like the Pop up windows can only be shown inside the window they are created in To temporary solve this problem the Pop up window size was
37. a Linux environment Each tool platform that fulfills these criteria will be further researched and tested to see how it works out in practice Time would also be taken to look within Saab to see if anyone else has done any work research on Online Help A meeting with SMW s documentation department DP K will also be held sometime within the first few weeks to see how the customer documentation is made and see if there is a way to use any of this to create Online Help 5 3 Mock up prototype After doing some research and getting a clear picture of how the Online Help would look and work a mock up prototype would be made to illustrate this and later be shown to the thesis supervisor to get feedback This feedback could then be used in a second mock up or perhaps in the actual prototype In this stage a choice of what tool platform to use to create the Online Help must be done as well as acquire the source code to the software that the Online Help was going to be implemented in A few days would probably be needed to study the software code in order to figure out how to implement the Online Help Followed by researching and deciding what type of usability test that could be used to evaluate the prototypes 5 4 First prototype When that was done it was time to start working on a first prototype that would likely be very simple just testing out the help system on the product to see how simple hard it would be to implement and give a feeling o
38. a thesis project done at company Saab Microwave Systems SMW located in Gothenburg Sweden Saab Microwave Systems develops sensor systems that are used for military purposes around the world In order to use these sensor systems different kind of software is also developed The software is often custom order and are only developed and sold to specific customers The software is then delivered to the customer together with thorough documentation describing how the software works but the actual software has no kind of help system in it only traditional manuals are available The main reason for this is that it would take to long and cost too much to implement considering the limited customer base Another lesser reason is that the people that are going to use the software have been trained to use these systems and thus not need a help system Recently this has started to change and a need for some sort of digital help other than the traditional manual The foremost reason to this is to improve the operator s operation environment by giving him quick access to the operations and controls manuals in the form of Online Help This is one step to in the future stop shipping and making the part of the complete documentation that is classed as the user manual although as a first step it is important to be able to generate the Online Help and traditional manual from the same source and offer this to the customers The introduction of an Online Hel
39. andard navigators are table of content index and full text search Other information is what title shown in the top of the help content window as well as presentation settings i e the size of the window if there is a toolbar or not and if so what it will contain like print and home The second important file is the map file mentioned above This file is used to associate topic IDs with the URL or path name of the HTML topic files Lastly there is the Table 6 of Contents TOC file that describes TOC navigator and layout of The format of the TOC file is based on the World Wide Web Consortium W3C Extended Markup Language XML The helpset and map files also uses the same format Besides these three there are other optional files that can be included these are the index file glossary navigation file and favorites navigation file All the help files can be compressed into a JAR Java Archive file it is not necessary to unJAR the files to run the help system It is also possible to merge two or more helpsets This is useful when installing different modules for a platform each module has its own helpset that will be merged with something called the master helpset probably the helpset of the main application or similar There are two ways of merging helpsets one is static merge which is done by specifying the helpsets in the XML code of the master helpset s helpset file The other is done by writing code in a Java pr
40. as several years ago so RoboHelp must have improved during this time After a quick Google search it was found out that the latest version was RoboHelp 8 and was now owned by Adobe A trial version was downloaded to test its capabilities to create help content RoboHelp supports a number of different help content types JavaHelp is only one of them There is also an option that lets the user import FrameMaker and Word documents that will be transformed into RoboHelp s file format XPJ among others This seemed much easier than trying to create everything from scratch The traditional version of the manual was in Word format so it was imported into RoboHelp The result was one big HTML topic file all the headings had been converted into topic IDs and all the pictures had been put into a separate folder and linked into the topic file The next step was to generate the JavaHelp files to see how it would look After configuring the JavaHelp generator the JavaHelp files were generated into a JAR file It was then possible to view the generated JavaHelp directly in RoboHelp doing so showed that there were serious slowdowns with the JavaHelp loading the topic file took forever To make sure that it was not just RoboHelp that were malfunctioning the JavaHelp files were tested in the application as well The slowdowns were even worse the topic page never loaded at all It seems like the problems that Saab Systems had with JavaHelp was still there see chapter
41. at help type differently than the first one Mostly it seemed like the traditional manual was read more carefully than the Online Help no matter which group the testers were from originally There seems to be some rooted conception that if the information is read from the screen it is not as important as something written on paper but this is just a theory It could also be at least those that started with Online Help that because they were instructed to talk about what the differences were between the two types they felt obliged to spend extra time with the traditional manual to have something to say This applies especially for the participant that did not use the Online Help to complete the assignment but studied the traditional manual for several minutes So it is not certain they would have used the traditional manual that way if it had been used from the start 62 There were some important differences with the two manual types other than that one were written on paper and the other was online The operations manual was not as thorough as the one used in the Online Help and the part describing the creation of filters was not in the right chapter This made it harder for those using the traditional manuals to complete the test then again the Online Help is supposed to make it easier for the users to find the information they are looking for by combining the control and operations manuals into a single manual There were also other parts of t
42. can therefore be set any type in this case a topic ID Setting it to a topic ID is done at the same time as the configuration of the table is done 35 This itself is not enough to make window work though topic must extracted somehow When a MouseEvent is triggered the position of the mouse pointer is recorded by getting the JTableHeader and the TableColumnModel it is possible to get the index of the column by using the x coordinate of that position The index can then be used to get the actual column and thus its topic ID After this the same thing was done to the Mission Control Panel Instead of using the old test topics in the Pop ups new HTML topic files were made with material from the original manual this because the test files did not really say anything 9 2 Testing the Prototype With a finished prototype it was time to put it through a usability test Three people were chosen without any special criteria to do the test The test itself was unstructured it did not follow any special protocol instead the testers got to play with the system for 15 20 minutes while answering a few questions The testers were told to speak out loud any thoughts they had The result was put together into a couple of tables which can be seen in Appendix A It can be summarized into positive negative and miscellaneous parts 9 2 1 Feedback on what was good Interaction The interaction was simple with a high accep
43. ce should be intuitive enough to not need help Differences between Online Help traditional manuals 4 Hard to see the relations between the operations and controls manuals this can be because they are written by two different people Takes longer to look something up in the table of contents Generally takes a longer time 5 Easier to see that New is associated with filter Thinks that some parts are explained more clearly are not sure if the same was explained in the online manual or not though Feels more at ease with the traditional manuals and read it more thoroughly than the online manual 6 Does not find filter easily in the Op manual Because it is placed under L11B you do not connect that it is the same thing Control manual gives enough help if you are uncertain Checks the papaer manual more thoroughly probably becasue he was told to talk about the differences between the two manual types Previous experiences background 4 No experience with HMI about TDN Works with stuff at a deeper level 5 Worked with UndeE HMI No TDN filter 6 System tester Nothing about TDN 80
44. cle Help for Java Developer s Guide help in a web environment D e Oracle Help for the Web Developer s Guide Throughoutthis Oracle Help Guide Oracle Help is used when the comments apply to both Oracle Help for S Licensing Support and Copyright Java and Oracle Help for the Web OHJ is used when the comments apply only to Oracle Help for Java OHV is used when the comments apply only to Oracle Help for the Web Figure 2 2 Oracle Help Content Window Docked Oracle 2009 Oracle Help like JavaHelp also supports merging of helpsets This is done in runtime in Oracle Help making it possible for multiple authors to create multiple helpsets that can be merged seamlessly without having to rework the system The table of contents from the different helpsets will be put atop of one another except nodes with the same text and topic ID these are combined into one as long as there are no conflicts Help Navigator E 3 He p Topic Window DAR File File View Help Tools oracle Guide 2 Ch EEIE Contents Oracle Help Overview 4 b Oracle Help Overview Previous Next E Introduction to Oracle Help El Aboutthe OHJ User Interface Introduction to Oracle Help El OHJ Topic Windows ERO The Oracle Help technologies provide the means for The OHJ Navigator Window developing and displaying HTML based help systems B Merged Helpsets for Java applications and for web applications Authors can create a single h
45. contains a list of key items in alphabetical order like Available IFF followed by Base Defence Zone and so on see Figure 13 4 It works just like an index in a book except when a key word is clicked the user will immediately be moved to the topic containing that key word Auto Acquisition Auto M4 Auto Target Acquisition Automatic Activates Available Available as Base Defence Zone Base Defense Zone Popup Windows Race Nefence 7 tem Situation Figure 13 4 The Index tab Search tab In the search tab the user can search the online manual to find whatever he or she is looking for the word filter for example would give the result shown in Figure 13 5 Clicking on one of the resulting items will open the topic that that word is written on that specific place with the word highlighted 49 Find filter 57 Menu Item Communication 24 Popup Windows 3 Menu C2 6 Symbols Shapes amp Areas 3 Card Deck Panel 2 Introduction amp Content Overview 1 Menu Item Situation Picture Figure 13 5 The Search tab Glossary tab Acronyms and abbreviations are shortly explained in the glossary tab see Figure 13 6 There were supposed to be functionality that made it possible to connect these abbreviations in the topics to the glossary so if the user clicked on an abbreviation the glossary would automatically open and show that specific abbreviation but this was
46. d into a number of XML files that ActiViewer uses to build the The IETM looks and works similar as the normal help content see Figure 2 4 ActiViewer 2 01 DEMO AECMA BAX Open Help About Tree Index Search Demo IETP He Message options Status options functional description for user SST 2 ey Another functional description for user Typical functional description for technican Date To From Pri Message itno Une Operator System i 0507362 OER 5 002 LOCALIZATION REPORT 00001 70019 Voice Not sent Typical procedure for user 0507362 OER 5 001 CALL FOR FIRE 00001 0018 Voice Not sent Typical procedure for technican Typical fault isolation for technican Typical circuit diagram as pdf Typical spare part module Operators Manual Introduction Abbreviations Precautions System Overview General Function General amp amp 2 tia Tracking Beam Control RESTRICTED Design Technical Data Mechanical Design Message Log Functional Description Controls All Messages are listed in the Message log It is possible to select certain Operation messages or Message formats by use of Filter criterias If a Message is high Preventive Maintenance lighted it is possible to open a popup menu By double clicking on a Message Operator Training the operators can open the Message The Message log has fields indicating Appe
47. design This led to that the official supervisor supervised less and less of the thesis and thus was not fully aware of what was going on In the end this did not really matter since the unofficial supervisor knew what was going on The contact with the supervisor from Chalmers was mostly to make sure that the thesis went in the right direction and give guidance on what to do Since some of the things in the thesis were classified he could not give any detailed guidance on the prototype or coding Most of the tutoring was on how this report should be written 14 3 Usefulness One of the major results from this thesis is that a new project is in planning This project will look into the possibility to use context sensitive help mainly Pop ups in similar systems like the one used in this thesis This may sound like a repetition of this thesis but it is more focused on using Pop ups in tables to explain what the different column headers means Today meetings are held regularly discussing what the text in the column headers should say if an abbreviation should be used and if the users would understand it If a Pop up window was just a mouse click away this would simplify things an abbreviation could easily be used if a user could find out its meaning just by clicking on it and read the Pop up There are also long term plans of introducing online manuals help content into their systems but this would take too many resources out of their current projec
48. different sizes on the Pop up windows until it was pointed to them And one of the testers even said that he liked that the Pop ups appeared right under the components 10 3 2 Feedback on what was bad Interaction There were a lot of complaints on the interaction part this time although not on anything serious Two of the users did not see the point with the What is this menu they thought it was strange that there were only one menu option plus it was seen as an unnecessary step to have to do an extra mouse click to open the Pop up This can be related to the Iterative Design chapter chapter 4 3 that states it is not always that the each iteration will be an improvement Then again this was only one of the new features implemented in this iteration There was also some complaints that there was button in the windows that could be used instead of pressing F1 Other than that there were just some minor things like no help on the map or in the menus Efficiency In some cases during the test the help content did not open when the F1 key was pressed the window or panel had to be re opened to get this to work again The problem occurred after the test users had clicked around in the interface it seems like the focus is moved from the window to the components and thus the F1 help no longer work Manual Help Content Because a real version of the help content was used this time it was a given that there would be a lo
49. e MM amp ka emmy up window accessed through right click at the table header The configuration pop up 2 Lower Screen C2 MMI Overview 3 Message Windows E 3 4 Errors 5 3 2 Wamings 4 Card deck panel 5 4 1 Upper deck Overview D 4 2 4 Upper deck Track List ejected vie 4 3 Upper deck Associatons 63 4 10 Upper deck Orders in 4 11 Upper deck Orders out Fig 9 Air track list 7 1 Len side and middie part 7 2 Right Side 9 1 Properties 3 2 C2 Presertaton Semings Button Description TS 8 3 Enter Leare Simulation Mode 2 4 Bustin test Add alr tracks manually Provides ability to manually manage alr tracks Enables the Source Info Information regarding what unit that has identified the air track Possible values are Common Remote Local Ex Figure 13 11 Pop up window showing help Figure 13 12 Eject button 1171 nnna Once ejected the window can be moved and resized as any other window The table can now be studied in more detail There are around thirty columns in the table and it thus necessary to use the scrollbar to see it all The Pop ups works the same way as before except that it is shown under to the right of the column header instead of to the left of it see Figure 13 13 53
50. e any valuable information to the result The result from the test showed that it took around 17 minutes per participant using the manual to complete the assignment and 8 5 for those using the Online Help i e twice as fast as manual group More participants would have been needed to make sure that this was really the case though Still this provides a clear indication of the Online Help effectiveness compared to the traditional manuals The Online Help group completed 17 out of 23 subtasks and the other group 19 out of 23 The only difference is that those using the Online Help completed the subtasks faster than those that did not This means that the use of traditional manuals or Online Help did not particularly affect how many subtasks that were completed Based on this result it is clear that no matter how good an Online Help is it can not replace a poorly designed interface Each user differs from each other and therefore uses the help available differently if at all To create a help system that works for every one is therefore not possible but it is possible to create a system that makes it easier for those that actually uses it Both of the help types need to be reworked and complemented to fully help the users understand the interface and how it works All the users seemed pleased with the Online Help and preferred to use it over using the old traditional manuals at least this was what they stated afterwards The scores given in th
51. e help buttons for example instead of using F1 but since these were not used in this thesis it is no real reason to describe them here 2 6 3 Oracle Help Oracle Help is very similar to JavaHelp which is only natural since they have used JavaHelp as a base for their own help system First of all the help content window has a more modern look to it see Figure 2 2 and it is possible to undock the topic window from the navigation window to get two different windows see Figure 2 3 The benefit of this is not clear Oracle 2009 amp Help Navigator File View Tools Contents i oracle Help Overview Oracle Help Overview 4 Previous Next B About the OHJ User Interface Introduction to Oracle Help Topic Windows The Oracle Help technologies provide the means for The OHJ Navigator Window developing and displaying HTML based help systems Merged Helpsets for Java applications and for web applications Authors 8 can create single help system that be displayed 2 Other OHJ Features without modification both in a Java environment using SEJ About the OHwW User Interface Oracle Help for Java OHJ and in a web environment i using Oracle Help for the web OHVY Alternatively you gt e Oracle Help elle Formats can just use OHJ if you only need help in a Java Authoring Oracle Help Systems environment or you can just use OHW if you only need gt Ora
52. e done this for over 50 years and are considered the world leading competence center for microwave and antenna technology So far over 3000 sensor systems have been delivered to more than 30 countries Saab 2007 These systems are made to function in a variety of conditions and environments no matter if it s in the air on land or at sea Examples of these systems are ERIEVE air sea GIRAFFE AMB land and Sea GIRAFFE AMB sea Saab 2006 Special software is needed to use these Radar Systems The software also differs from system to system The system this thesis will focus on is the GIRAFFE AMD To this system there are two software products that work together one Sensor Graphical User Interface SGUI that handles and show the sensor input from the local connected GAMB and one Tactical GUI that show the sensor input from the local connected GAMB and all other C2 Command amp Control Units see Figure 1 1 and are used to give orders and other tactical functions Only the latter one is included in this thesis Unit Unit Link to L11 for example Figure 1 1 A C2 System 1 4 Outline The disposition of the rest of the report is as follows A background chapter describing how everything was before this thesis started i e the software that the Online Help was going to be implemented on the existing documentation practices that are currently being used as well as those that are about to be
53. e immediate area looks like This is shown on the upper screen of the operator workstation see Figure 2 1 The TGUI takes input from the local GAMB as well as input from all other C2 Units connected to the Tactical Network These inputs are then combined and shown in a graphical display window in the TGUI which is placed in the lower middle screen of the workstation The operator now has a full overview of the whole tactical grid and can plan and give orders on a much greater scale than before Figure 2 1 The Operator Workstation 2 2 The documentation As things are right now the documentation for the SMW software is not written until the end of the development cycle This because that the documentation department DP K are not usually part of the development team and are thus not contacted until the end of the development The amount of work the DP K currently is having also affects how quickly they can start working on a specific project Making the documentation at those stages have disadvantages such as the people writing the documentation usually have not worked with the system before and thus must spend a lot of time figuring out the system and interface talking to the programmers and even then the manual probably will not be entirely correct This as well as writing manuals from scratch is very cumbersome and also take into consideration that this is done under a limited time period This more times than not results in manuals t
54. e information if needed without having all their consumers download updates instead they can just change the version of the help they have on their server The downside is that if the customers lose their internet connection for some reason they can no longer use the help content The Online Help in this thesis could have been made online but there were no real reason for doing it there is just one version of the help content plus that the operators that are going to use it usually sits in a C2 unit somewhere in the middle of nowhere without any internet connection so keeping it locally together with the prototype made sense Changing the name from Online Help to something else because of this does not make sense since Online Help is what these sort of systems are called by people no matter if it is online or not Changing the name would just bring confusion Because this is a thesis done at a company that produce products for military use there are certain parts of the thesis that is classified for people outside SMW This would in most cases been a problem but fortunately the application that the Online Help was going to be implemented in was not entirely classified How the interface looked and worked was unclassified but the actual code and specs of the system were as well as whom the system was meant for Since the supervisor at Chalmers did not need to see the code or know the specs just see the interface this did not pose such a great issue
55. e questionnaire can be tied to how long it took for the participants to complete the assignment those using Online Help had higher scores and low completion times compared to those that used the traditional manual that got low scores and had high completion time The conclusion that can be drawn from this is that the Online Help is effective in regards to helping the user find and give the information they need to do their work All in all it can be concluded that more participants would have been needed in the final test to get a more conclusive result 66 16 References Cooper 2007 Cooper A Reinmann Cronin D 2007 About Face 3 The Essential of Interaction Design Indianapolis Indiana Wiley Publishing Inc 3 26 70 71 560 563 Dix et al 2004 Dix A Finlay J Abowd G D Beale 2004 Human Computer Interaction Third Edition Madrid Spain Pearson Prentice Hall p 396 414 Ottersten et al 2008 Ottersten I Berndtsson J 2008 Anv ndbarhet i praktiken China Studendtlitteratur p 14 16 ISO 13407 1999 SS EN ISO 13407 1999 Human centered design processes for interactive systems Stockholm SIS F rlag AB Eclipse 2009 Ecipse Fundation Eclipse Galileo Documentation http help eclipse org galileo index jsp 2009 ISO 9241 11 INTERNATION STANDARDS ISO 9241 11 First Edition Ergonomic requirements for office work with visual display terminals VTDs Part 11 Guidance on usability Ref
56. eated defining the framework for the products behavior visual design and possibly physical form interaction framework is then synthesized by using two methodological tools together with the context scenarios These two are sets of interaction design principles and interaction design patterns The first one is used to help determine suitable system behavior in different context and the other one contains solutions to previous analyzed problems These patterns are not set in stone but continue to evolve as new contexts arise and thus provide help with proven design knowledge when approaching difficult problems The interaction framework is then translated into design elements by using interaction design principles and organized into design sketches and behavior descriptions This will result in an interaction framework definition that provides logical and formal structure for details to come in the Refinement phase The interaction framework is also used by the interface designers to produce several suggestions for a visual framework also called visual language strategy The Refinement phase is similar to the previous phase but focuses more on details and implementation Interaction designers focus on task coherence and validation scenarios while visual designers define the style of the interface such as sizes icons and other visual elements All this results in a detailed design document delivered in traditional or interactive media The S
57. ed create a complete and robust index as possible There should be a menu item in the Help menu that handles shortcuts for keyboard and functions There should also be a menu option that gives an overview of different functions in the application things like scope effect power upside downside and why the function should be used ToolTips should not be underestimated it is easier for the user to point at something he wants help with and gets a small box quickly describing it than have to open the help content that takes up a lot of space and try to look for it Wizards should be avoided if possible because they often have the tendency to interrogate the user and also asking obscure questions A user that does not know what an IP address is will not get any help from a wizard asking for it Intelligent agents if used take a lot of work Use Clippy Microsoft Office little annoying assistant as an example of how an agent should not be Dix 2004 also has a few suggestions about how Online Help should work First there are some general requirements that have to be fulfilled gt gt gt gt The user should always be able to access the Online Help while using the product Online Help should always give correct information i e if the product is updated the help system have to be updated as well to show correct information Online Help should also be complete i e there should not any sections of the product that is not b
58. editing tool like Adobe Photoshop The main purpose of this kind of prototype is 22 to create look and feel of system although it will be lacking any kind of real functionality and operation flow It is actually a bit misguided to call it a true prototype it is really just a visual mock up The visual prototypes are usually done from the designers viewpoint rather than that of a business or software expert Interactive prototypes are more useful than the previous types this also means that they require a more time and resources than the others to create The purpose is to model the system design more faithfully and make actually interaction possible i e navigation use of real web controls or even mock data processing Thomson 2009 The advantages of using prototyping are many it reduced development time and thus reduced development costs Users are involved at an earlier stage giving the developers quantifiable user feedback to work with Since users knows what to expect it facilitates system implementation and results in higher user satisfaction This process also exposes the developers to future system enhancements Where there are advantages there are also disadvantages Using prototypes can lead to insufficient analysis i e the developers become distracted from focusing on a limited prototype instead of properly analyzing the complete system The users may expect the performance of the complete system to be the same as t
59. eenshots of the current interface and then modifying them in Adobe Photoshop This was done before any real background check of possible Online Help system platform had been done Because of this it was not known during this time what was possible to do with the different platforms and therefore did not influence the making of the mock up The mock up was meant to show how ideally the Online Help would look and work The choice of Online Help functionality that was implemented into the mock up was a Help menu in the menu bar and a combination of ToolTips and Pop up windows The Help menu contained four items Help Content Keyboard Shortcuts Symbol Legend and About see Figure 6 1 The ToolTip was used show what certain abbreviations meant explaining what kind of values that were expected in certain field and also what some of the table headers meant while Pop ups were used to explain certain labels checkboxes buttons text fields table headers combo boxes etc The main difference between ToolTips and Pop ups are that ToolTips are shown when the mouse pointer hovers over certain components a short while and are usually very short see Figure 6 2 for an example Pop ups on the other hand are shown when the user activate them in this case by right clicking on it to show a What is this menu that he has to left click on to show the Pop up see Figure 6 3 It closes when the user clicks on something else and often contains a lot more tex
60. elp system that be displayed Other OHJ Features without modification both in a Java environment using El About the OHW User Interface Oracle Help for Java OHJ and in web environment using Oracle Help for the Web OHW Alternatively you T Oracle Help File Formats can just use if you only need help a Java gt Authoring Oracle Help Systems environment or you can just use OHV if you only need help in a web environment D e Oracle Help for Java Developer s Guide gt Oracle Help for the Web Developer s Guide S Licensing Support and Copyright Java and Oracle Help for the Web OHJ is used when the comments apply only to Oracle Help for Java OHW is used when the comments apply only ta Oracle Help for the Web Throughout this Oracle Help Guide Oracle Help is used when the comments apply to both Oracle Help for Figure 2 3 Oracle Help Content Window Undocked Oracle 2009 The help files are the same JavaHelp there is the helpset TOC keyword index and search index file the new one that is not in JavaHelp is the link file The helpset file is more or less the same as JavaHelp except that presentation is called wintype instead and there is a new option called link that have to do with the link file mentioned A few internal differences and some new options otherwise it is the same No change in the map TOC or index file The link file is an XML file that defines lin
61. en as an international standard nevertheless but as mentioned above there can be other ways to explain describe Usability Usability is also defined as a quality attribute in interactive products meaning that products with high usability fulfill the customers and target groups groups of users that have similar expectations and intensions of using the product purposes To create a usable product three important aspects have to be taken into consideration e The human system general and specific skills that the users have e The context in which the product is going to be used The product has to be adjusted to physical psychological social and organizational contexts that it is going to be used in e The benefit that is expected from the product Both the one that uses as well as the one that provides the interactive product expect to benefit from it The provider may expect economical gain while the user expects efficiency and or satisfaction 12 Usability is not something objectively observable nor does it have internal product attributes like a color or function Usability is something that emerges from product use Quality in use is a concept that puts the focus on the use of the product instead of the actual product This makes it clear that usability is a measure of quality and relies on the context in which it is used Ottersten et al 2008 Ottersten amp Berndtsson 2008 also gives their own view on ISO 9241 11 They t
62. en people and the test were supposed to be held the beginning of the following week it was a really fortunate that there were six that agreed to take part of the test All this resulted in that the third usability test was held two weeks later than originally planned There was also plans of using two versions of the application in the test one with Online Help and one without But it was not until the day before it was realized that this was not possible At the time of the fist prototype a of the current version of the application was made so that it could be used to implement the Online Help The development of the application did not stop because of this of course so at the time of the third usability test it had changed so much so that it was no longer suitable to use in the test Neither was there an unchanged copy of the version with the Online Help that could be have been used instead Since SMW used a version control system that backs everything up it was in theory possible to go back and recreate a copy of the version used in the Online Help But this would take too much time and effort to do however so in the end the version with the Online Help was used by both groups in the test even those that used the traditional manual Could this had been planned better probably but considering that the thesis supervisors at Chalmers and SMW had their vacation during this time and since their expertise were needed to do the plan
63. enough there does not seem to be any way of implementing Pop ups in the application with Oracle Help there is support for Pop ups but only in the help content i e it is possible to define a link that then clicked on will open the referenced page in a Pop up inside the topic window There is a boolean needPopupHelp that can be set to true in the addComponent function but this will only result in that the topic window will open not the navigation window and it need to be closed manually So calling it Pop up help is a bit misleading Considering that Pop ups is an important part of this thesis just making help content and show help for each window with F1 is not enough It is important that the user also can get immediate help with components those purpose may be unclear or those meaning has been forgotten This can easily be done with Pop up windows but since Oracle Help do not support these within the actual application its uses in this thesis are limited 8 5 ActiViewer Before doing any testing it was decided to find more information about the product so the distributer s homepage was visited but no information about ActiViewer whatsoever was found A simple Google search was also done but resulted in no relevant hits After checking with DP K it seems like the distributer is no longer selling ActiViewer for customers this means that there will be no further development on it and is essentially a dead product To adapt such a produc
64. er problems indirectly caused by using JavaHelp are the lack of support available There are several forums where questions can be posted but there does not seem to be much activity on those anymore This makes it difficult when problems occurred because there is often nowhere to turn So the big question is why use JavaHelp if there are so many problems using it The simple answer is because there no other suitable platform available the only other options is to create a new Online Help system from scratch and that would take too long Not all problems can be blamed on JavaHelp some parts of the interface can be a bit confusing and need to be redesigned Then again the application used to implement the Online Help was not done yet since the interface is still under development 57 14 2 The Development Process The thesis did not completely follow the original development plan mentioned in chapter 5 First of all the mock up was made before any real research was done on the different Online Help platforms so there were no knowledge if it was possible to do the Online Help illustrated in the mock up with any of the available Online Help platforms No second mock up was made because the first one gave a good enough idea of what kind of functionality the Online Help needed to have But if none of the Online Help platforms had supported the functionality in the mock up a second one would have been made with the functionality actually available i
65. est It is easy to compile but it can be hard for the participants to translate their feelings to numbers How do you really know the difference between a 7 or an 8 Also the 1 to 10 scale that was used was chosen so that there were no middle point 5 for example This forces the person doing the questionnaire to take sides otherwise it very easy to just choose 5 which in reality really means It is ok Now at least it is either a little bit good or a little bit bad But it is not always that the person answering notices this and chooses 5 anyway because he thinks that it is the middle number This can explain that there were a lot of 5 and 6 answers for the group using the traditional manual Much of the result is a bit unclear and could perhaps have been made clearer if a larger test had been held The main issue was the lack of participants six participants that were divided into two groups seemed like a good idea and give a good enough result to study But for it to be more conclusive two or three more participants per groups may have helped at lot But since it was at the end of the vacation season at SMW it was hard to get that many people during those two days The reason that the test was held around that time was because it was already overdue and the following week from Wednesday to Wednesday was already booked for an opposition of another group s thesis So in the end the test was as
66. f how it would be to use the help system After completing the prototype a usability test would be held in order to find the main issues with the Online Help as well as determining what could or could not be done with the tool platform The feedback would then be used to create the second prototype 5 5 Second prototype The second prototype would focus more on the actual online manual a part of the Online Help system The online documentation was supposed to be created from the same material as the traditional manual so suitable tools for doing that must be found and researched and then used to create a simple version of the online manual This manual would then be implemented in the second prototype as well as more functionality and fixing the greatest issues with the first prototype 25 second usability test would be held maybe with same people as last time maybe with new ones see chapter 4 1 This test would likely be more structured and take longer than the last one would probably result in more data to compile and analyze 5 6 Third and final prototype The feedback would yet again be used to create the next and final prototype that would have a complete online manual and all Online Help functionality that was envisioned in the beginning of the project This prototype would also go through a final usability test that would be even more structured than last time more thorough and have more tes
67. good as possible considering the circumstances 63 14 6 Future Work There are a lot of things left to do in the field of Online Help As describes earlier in the report JavaHelp is far from the ideal platform of use to implement Online Help But as also mentioned earlier there are few others alternatives to choose from Oracle Help can be seen as a better but more limited choice that is not as versatile as JavaHelp So depending on what the developers want to do this can be a better choice Rather than depending on already existing platforms that do not fully support what the developers want to do it may be a good idea to try to develop an Online Help system of their own This could either be from scratch or built on an already existing system like JavaHelp Doing this would probably take a lot of time and resources so SMW has to decide if this is worth it There are of course already existing solutions within Saab like the Online Help based on Firefox Portable mentioned in chapter 2 6 6 and 8 6 The main thing that needs to be improved in JavaHelp is the window management Pop ups should not be limited to the size of the window that it shows in the Pop ups should be able to be shown outside the boundaries of their parent window It should also be made possible for the Pop ups to automatically change size depending on the amount of text it has to show The current solution in JavaHelp the size has to be defined in the helpset file This seems
68. grad 1 2 3 4 5 6 7 amp 9 10 6 I vilken grad upplever du att hj lpen navigerbar var jag Mkt liten grad Mkt h g grad 1 2 3 4 5 6 7 8 9 10 75 Totalupplevelse Efter att du genomf rde testet fick du prova samma f rfarande med den andra hj lpvarianten Efter att ha f tt k nnedom om de b da s f r du svara p f ljande tre fr gor 1 ver lag vilken typ av hj lp f redrar du ___ Anv ndarmanual operationskapitel Onlinehyalp 2 Vad upplever du den st rsta f rdelen arna av att anv nda den typ hj lp du f redrar 3 vriga till gg 76 Appendix F Results from Usability Test 3 Questionnaire Answers Traditional manuals Question 1 1 2 3 mean val 3 4 5 6 7 8 10 X Numerical 8 3 4 stand dev Question 2 1 2 3 mean val 10 5 2 6457513 stand dev Question 3 1 2 3 mean val 10 gt stand dev Question 4 1 2 3 mean val 10 gt OB stand dev Question 5 1 2 3 mean val 10 O AJA AA gt stand dev Question 6 1 2 3 mean val 10 4 2 5 3 6666667 1 5275252 stand dev gt ano Online Help Question
69. hat do not explain in enough details the functions and control of the interface Other disadvantages are that because the software and thus the interface is not complete when the writing of the documentation starts means that the control manual in particular must be updated several times before the software is done This is a natural thing in development cycle but does make DP K s work problematic since the continuous update of the software also forces a continuous and often retroactive update of the documentation Worth mentioning is that the employees at DP K are in constant contact with the customers of SMW s products and therefore have lots of experience of what the customers think do and like knowledge that would be priceless if they would take an active part of the software development 2 3 New practices SMW 15 currently working on changing the way software and documentation is done Although this is still in the early stages and will probably take some years before it is fully established One of the main goals is to try to focus the development on Usability see chapter 3 To help achieve this goal SMW recently hired a Human Factors Engineer HFE that will help the programmers to obtain enough resources to make the interface usable using their full potential as interaction designers and usability testers This is also something that Saab as a whole is interested in so there are usually a few meetings a year with people working
70. hat if a Pop up is open in a window when someone closes that window using the X button in the right upper corner the Pop up does not close instead it will just become invisible If the same window is opened again the Pop up will become visible again and there is no way of closing it without having to restart the whole application This bug does not happen when the window is closed using the OK or Cancel button in the lower right corner A third window management problem is that the Pop up s size has to be defined beforehand in the helpset file resulting in a number of Pop up templates with different sizes so that different text masses can fit This makes it difficult writing the code because the text each topic ID must be checked first to see which Pop up size is needed It would be better if the Pop up windows could automatically adjust the size of the window to fit the text There were also problems with making the help content window open when the F1 button was pushed Apparently JavaHelp have trouble with what part of the interface that is currently having focus It all works fine if F1 is pressed as soon as a window with Online Help implemented is opened but after using that window for awhile the focus is shifted from the actual window to the components and the F1 key does no longer work Considering that the users are more likely to press F1 after trying to figure out the window rather than pushing it right away can cause problems Oth
71. he interface that was not well enough explained so both of the help types needs to be complemented and rewritten see the next chapter for more on this As mentioned in the result analysis chapter 13 one of the test users said that she preferred Online Help over the traditional manuals even though she seemed much more at ease using the traditional manuals i e she studied it more carefully This could be because she felt that she was supposed to answer Online Help and answering otherwise could be seen as retrogressive It could also be that the question itself was formulated in a bad way the question was Which type of help do you generally like to use with the options traditional manual or Online Help This does not ask for which of the help types she preferred in this test just which type she generally likes to use So it is not completely out of line to say that she may have been uncomfortable using the Online Help in test but she have no problem using Online Help for other applications Another thing that could have affect this result is that some of the participants knew the supervisor that assisted in the test and knew that he really likes the concept of Online Help This could result in that they are more generous with the critic than they would otherwise be The tester mentioned above was one of these people One may wonder if using numbered scales in the questionnaire is the right thing to get the test subjects thoughts about the t
72. he prototype The developers may also become too attached to their prototypes refusing to dispose it or make necessary changes There is also a risk that the use of prototyping can cause the system to be left in an unfinished state and or be implemented before it is ready This can also lead to incomplete documentation Worth noting is that creating sophisticated prototypes with 4 generation languages CASE Tools result in that time saving benefit is lost Mcclendon et al 1999 4 5 Design Guidelines Design guidelines help developers by providing useful high to low level guidance on the design of user interfaces Sometimes the use of specific guidelines are specified as a part the usability requirements It is important to get familiarized with the different existing guidelines as they contain years of experience in the field of interface design It would be foolish to make mistakes that could have easily been avoided if appropriate guidelines had been used Following design guidelines improves the quality of the interface There are different design guidelines that can be roughly divided into four groups First there are general user interface guidelines that can be read in ISO 9241 10 for example There are also guidelines for graphical user interfaces that can be found in a number of different books such as Cooper 2007 or Dix et al 2004 Thirdly there are guidelines for web pages as well and lastly there are application specif
73. hink that the ISO definition puts to much focus on that the target groups needs are fulfilled An equal amount of effort should be put in making the buyers needs and purposes fulfilled as well in order to call a product usable The ISO definition and much of the literature within this area seems to assume that the buyer already have determined the products purpose before any development have been done 3 2 User Centered Design User Centered Design UCD is described differently depending on who is asked but there is an ISO standard describing it ISO 13407 1999 According to this ISO standard UCD is a multi disciplinary activity It incorporates human factors and ergonomic knowledge and techniques with the objective of enhancing effectiveness and productivity improving human working conditions as well as counteracting the adverse effects of use on human health safety and performance There are four iterative steps of UCD that are commonly used gt Understand and specify context of use gt Specify the user and organizational requirements gt Produce design solutions gt Evaluate design against requirement These are illustrated in Figure 3 1 below The process is supposed to be iterated until the objectives are satisfied How these sequences are performed or the level of detail and effort used depends on the design environment and the stage of the design process 13 Identify need for UCD Understand amp specif
74. his third party to let Chalmers University of Technology and University of Gothenburg store the Work electronically and make it accessible on the Internet Online Help The implementation of user manuals into a radar application from Saab Microwave Systems SEBASTIAN DREYER O SEBASTIAN DREYER November 2009 Examiner OLOF TORGERSSON Chalmers University of Technology University of Gothenburg Department of Computer Science and Engineering SE 412 96 G teborg Sweden Telephone 46 0 31 772 1000 The picture on the cover page is a screenshot from the final prototype showing a window with a Pop up explaining the button the mouse pointer is hovering over and to the right is the online manual help content with the topic explaining the window with the Pop up Department of Computer Science and Engineering G teborg Sweden November 2009 Abstract This thesis was done at Saab Microwave Systems a business unit at Saab AB and focused on how Online Help OH could be implemented in one of their products as well as if it was possible to use the already existing documentation to create the help content to the OH without having to rewrite it all Suitable OH platforms were researched and tested Afterwards it was decided to use JavaHelp because it was the most versatile as well as supported Pop ups The development process was done iteratively with three prototypes Between each phase a usability test was carried out and the resul
75. hrough feedback Try to find a tools for creating help content like RoboHelp W28 Start working on second prototype Should have a fully working online manual More context sensitive help i e F1 key help and more Pop ups Start working on report if not yet done so W29 30 Second prototype done Show supervisors Do a simple usability test and look through the feedback Start working on the third prototype This should be the final version i e contain all necessary functionality W31 Third prototype done Show supervisors Have a final more structured usability test Put together the test result W32 Make final changes to prototype Write the report W33 36 Write report Maybe take a week of W37 Report done send to opponents Prepare presentation W38 Thesis presentation Fix report Table 5 1 The Time Plan 5 2 Preparation First step was to find and read material about Online Help support systems and similar to get familiar with the area The second step was to talk with the people that are making the product and others that are somehow involved to see what sort of thing they expect or not expect from this thesis 24 From this project description would written explaining purpose and goal of thesis as well as some Human Machine Interaction HMI guidelines embodying a good Online Help system After that it would be time to start looking for suitable tools platforms that support Online Help in a Jav
76. ic guidelines for special technologies UsabilityNet 2006 23 5 Planning The following chapter lays out the initial plan of thesis i e this is how thesis was originally planned to be carried out Note that this plan is not accurately describing how the thesis was actually done 5 1 Time Plan The original time plan can be seen in Table 5 1 below The rest of this chapter will explain the different steps in the time plan more specifically Time Plan W18 19 Gain insight into the thesis Meetings with people involved in the thesis Write thesis description and time plan Get access to computer programs and so on Start looking for suitable Online Help platforms W20 21 Go through the different platforms Have meeting with documentation department DP K Contact other people within the Saab Group about Online Help W22 Go through literature and create Online Help guidelines W23 Take screenshots of software and make mock up prototype W24 First mock up done Show supervisors Create second mock up from feedback Show supervisors again W25 Research information about heuristic evaluations Get access to source code for project and gain insight into how it works W26 Decide which platform to use and how the Online Help should be implemented Start working on the first prototype It should focus on implementation of Pop up in 1 2 windows W27 First prototype done Show to supervisors Do a simple usability test Go t
77. ill be describes in the next chapter 3 1 Usability Today the every day definition of the term Usability can be considered highly diverse There are often several unknown aspects influencing people s interpretation of the term such as what user profile the term is supposed to reflect as well as the system complexity in which the user is supposed to operate in The SO definition says The extent to which a product can be used by specified users to achieve specified goals with effectiveness efficiency and satisfaction in a specified context of use ISO 9241 11 But what do effectiveness efficiency and satisfaction mean in this context ISO DIS 9241 11 states the following Effectiveness The accuracy and completeness which users achieve specific goals Efficiency The resources expended in relation to the accuracy and completeness which users achieve specific goals Satisfaction Freedom from discomfort and positive attitudes towards the use of the product ISO 9241 11 ISO DIS 9241 11 is a definition of how to identify information that is necessary when specifying or evaluating Usability of a visual display terminal by measuring the user performance and satisfaction It is a guide that describes general principles and techniques rather than requirements and can be used in procurement design development evaluation and communication of information about Usability ISO 9241 11 This is one explanation one that is se
78. ing text for this Thought you needed to create the geographic boundaries in a separate window other than TDN filter Goes into Shapes symbols and areas lo look for it Should not have EW strobes because you cannot make filter for them Looks for filter in the chapter for the L11B window in the operation manual Have problems understanding why you cannot make geographic boundaries for EW strobes but manage to figure it out Uses drag and drop in Sit display Unsure about Point Differences between Online Help traditional manuals 1 2 3 Likes the search function Much easier to use likes Pop ups Thinks that the font in the Online Help should be the same as the font in the rest of the application Uses the search to great extension Does not use Pop ups unless told to do so Have the same problem understanding EW strobes Quicker to get where you want Better descriptions clear procedures Likes Pop ups Too much programmer language Hard to get an answer to why you have to doa particular task to get something to work Previous experiences background 1 No earlier experiences of similar systems 2 Not much do not know what TDN amp TDUN is 3 Verified C2 system Unde23 There are no filters in this system 79 Online Help Time Tester min Questions 1 2 3 4 5 6 7 1 2 3 4 9 75 0 5 13 2 x xe Be X 6 4 0 x
79. introduced earlier attempts at Online Help and descriptions of platforms that can be used to implement Online Help Next is a theory chapter explaining User centered Design and usability followed by a chapter describing the methods used in this thesis The original planning of the thesis is explained in the chapter after that This is followed by a small chapter explaining the design guidelines chosen The mock up chapter explains the mock up made in the beginning to illustrate how the thesis should look and work Afterwards there is the platform search chapter explaining how the different Online Help platforms were found and tested The next three chapters focus on the three prototypes made and the test performed on them followed by a test result analysis chapter explaining the test results from the final usability test Next is the result chapter giving a thorough walkthrough of the final prototype The discussion brings up different aspects of this thesis as the prototype development process this report test results future work etc followed by the conclusion chapter Lastly there are the reference list and appendices 2 Background This chapter describes how SMW is currently developing their products and manages the documentation 2 1 The Software The C2 systems as mentioned in the Introduction are divided into two parts the SGUI and TGUI The SGUI takes information from the local sensor GAMB unit see Figure 1 1 to show how th
80. is that when they got to try the other help type some of them used it differently than the first type But it is hard to draw any conclusions from this since they had a so called learning bias from the first attempt so it is possible they just looked through it without actually read anything in particular The background of the participants seemed to matter in the way that they completed the assignment Those that worked with testing and verification of systems had an easier time completing the assignment since their occupation means that they likely have a better knowledge of how the design philosophy for this kind of system works 1 their mental model of how the assignment should be solved is better compared to the others A trained operator thus solves the assignment faster through learning than an untrained 45 one This does not necessary mean that system is intuitive and easy to understand it is just that these people have learned how the system works and have an easier time to understand them while those that have not has to use the manual or help system much more In the questionnaire the test subjects had to fill out see Appendix E there were a question about which manual type they preferred to use All of them answered Online Help this can mean that all of them actually preferred it over the traditional manual in this test or that they simply preferred Online Help in general not necessary the one in the test The answers
81. k IDs and associates them to several topic IDs i e making it possible to associate an HTML link with several targets letting the user choose which of these to follow It is also possible to use Oracle Help for the web which is very similar to Oracle Help for Java but this is not part of this thesis and thus not pursued any further To use Oracle Help in the code the three JAR files help4 jar ohj jewt jjar and oracle_ice jar must be added to the project and then a reference to the helpset must be created much in the same way as JavaHelp After that it is a simple thing to add Online Help to components by using the function addcomponent Component component String topicId After this a context menu will be opened if a user right click on such components and doing this will launch the help content It is also possible use F1 to do this but in order to make this work another version of the addComponent function has to be used in that function the boolean parameter needF1Help must be set to true first Oracle 2009 2 6 4 ActiViewer ActiViewer is a program that the documentation department DP K at SMW is using to run their IETMs Interactive Electronic Technical Manuals IETMs are the electronic manuals that are currently shipped to customers as a complement to the traditional manuals It was a request from DP K that ActiViewer should be used in this thesis if possible to show the help contents in the application The manual is divide
82. k my supervisor at Chalmers IT University Staffan Bj rk for keeping me on track and not lose the sight of the goal of this thesis as well as being an invaluable support at writing this report Contents TINTRODUCTION 6 sss vesssvovs0te dosvesss vsesse desvesv sv ns oddosvesk sssesdussessenssessassosvesnisssnasesssssssnsssvseiesssssessosnsvsseele 1 TI PURPOSE evince hie 1 122 CONSTRAINT S 2 1 3 SAAB MICROWAVE SYSTEMS 2 ATR RI ei 3 2 BACKGROUND sos oes ossssossvsssonsssbodossos sorssossssos ssssornssvsnos ssseors s nsosisssssonsssvossiossnoss nosvoss sssnsssssdsoons nsiss 4 S ORTWARE sek eatin 4 22 THE DOCUMENTATION gt 4 2 3 NEW PRACTICES 5 2 4 EARLIER ATTEMPTS WITH ONLINE 5 2 5 OTHER APPLICATIONS WITH ONLINE HELP 6 2 6 THE AVAILABLE ONLINE HELP 6 2 6 1 Eclipse Platform Help System 6 2 6 2 SAVGH Ip saves 6 2 6 3 Oracle 7 250 4A CH VI CW CI EA 9 2 6 5 Firefox Portable Soluti n 10 12 12 32 USER CENTERED DESIGN eLo susse diss Ng die 13 3 24
83. lanes that are landing and taking of from a fictional airport on the tactical display Six people with varying backgrounds with no earlier experience with this system were chosen for this test Half of them would use the traditional manual the control and operation manual to help them through the assignment and the other half the Online Help The purpose was to see if the Online Help made it easier to complete the assignment and which of these two help types the participants preferred to use The hope was that the test would show that Online Help was the superior choice and could be used to convince SMW s customers that Online Help was the way to go During the test the participants were observed and notes were taken of their progress seeing how many parts of the assignment they completed how they used the manuals and Online Help questions and comments among other things The observation protocol used for the test can be read in Appendix D When the participants felt like they had completed the assignment felt like because unfortunately it was not possible to see the result of the filter unless the application was connected to an actual C2 Unit The participants were questioned about how they solved the assignment what they were thinking when solving specific parts of the assignment especially if they had failed some part Afterwards the test was repeated with the other manual type i e the Online Help users got to use the traditional
84. leted the assignment faster than the ones using the traditional manuals so the results in this test still holds merit The result would probably been clearer if more people had been used in the test The number of completed subtasks differentiated only a little from each other those that used the Online Help completed a few more subtasks than those that did not Note that 17 21 in Table 11 1 means that the group completed 17 out of 21 subtasks i e 7 subtasks per person This could mean that the interface were designed in such a way that it did not matter what help type was used and thus it does not matter how good an Online Help is it does not excuse a poorly designed interface It was noted during the test that the way that the users used the help differed as can be seen in the last row of the observation protocol in Appendix D Some people mostly men used trial and error refusing to use any help until they got stuck Then there were those that were completely different reading large sections of the text trying to get an understanding of the interface before starting to solve the assignment In between were those that read a short section and solved a small part of the assignment then read another section and solved another small part and so on This shows that each person uses the help available differently There are those that will not use the help available no matter how good or easy to use it is unless there is no other way Worth noting
85. like a cheap solution done by Java There is also the issues with what currently have focus in the interface since the users are more likely to press the F1 key after using the window they are working in for a while rather than pressing directly after opening it Therefore it is extra important that the help content opens then the user presses the F1 key else it will be a great source of frustration if it does not work This can be done by either rework JavaHelp so that it recursively looks for the topic ID for the window when the F1 key is pressed or make sure that the focus is always on the actual window and not on components More work needs to put down on the online manual the topic files needs to be divided into smaller parts and the manual itself needs to be reworked to be more like an online manual and less like a traditional manual Time should also be put into researching how RoboHelp imports FrameMaker files and see if this is a better option than importing Word documents This was originally planned to be a part of this thesis but there were problems with downloading the latest FrameMaker version from Adobe s server The version currently used by SMW was too old to work with RoboHelp 64 15 Conclusion The aim of this thesis was to figure out if it was possible to implement Online Help into an already existing product in a cheap and easy way as well as see if it was possible to create help contents with the material that already exi
86. lot of Pop ups see Figure 13 14 for several examples Although far from all the components have Pop up help because most of them are self explaining enough to not need them Connection 2 Interface unite ait Name Settings L TON LL Jr Coma Geos Ts a MEPIS EATS 7 62 42 52 CJ Show detats Connection Speed YA Enc tx 2 az 52 C2Buneryc Disconnected 6 LJ Show details Interface units Dan Status Disconnected C Show detats Connection CJ Show detats Vet Nowe Re Ente TA 75 jez 52 C2Baneryc Orsconnected Update Request Sends an update request to an namazal object va order to refresh mformahion regarding for mstance geograplocal postion sura Desconnocies Li Show detalis Connection T mopem 52 42 52 TON Connect om Status Disconnected Unt Name 2 Desconnected SFD L Show dotats TO LTS TO 62142 52 naconnocred un Status Disconnected Show detals Connection 1 I Show detats Und Nana Bu TH Ba CXBanen 6 2 Interface units Identity Shows which identities are s
87. lues from those using the Online Help More details about the result can be viewed in Appendix F 44 12 Analysis of the Third Test Result The result from the third usability test see chapter 11 3 can be interpreted in different ways For one there is the number of questions that the participants asked during the test Just looking at the statistics says that the ones using the traditional manuals asked three times as many questions but when looking at what they actual asked it was simple questions like where the airport in the test was located These types of questions would probably been asked regardless of help type so in the end these result have no weight or meaning Those that used the Online Help to complete the assignment did it faster than those that used the traditional manuals on average it took twice as long for the traditional manual users to complete the assignment However the standard deviation was much higher for those using the Online Help One of the reasons for this was that one of the test subjects completed the assignment in four minutes in a trial and error fashion read more about this later in this chapter and see Appendix D almost without using any Online Help whatsoever It is reasonable to believe that if that participant had been a part of the other test group the outcome would have been the same there as well The result is therefore not entirely accurate but still the two other test subjects in the group comp
88. manuals vice versa This to see if they completed the assignment differently and how they used the different manual type Note that by this point they already knew how the assignment was supposed to be solved this was only done to get a spontaneous picture of what they though about the other variant of the manual After the test had been completed the test subjects had to fill out a questionnaire with some questions about the manual type they had used There were also possible to give suggestions and other comments The questionnaire can be found under Appendix E 42 11 3 The Test Result The result from the third usability test showed clear differences between the test subjects using the traditional manuals and those using the Online Help The result is summarized in Table 11 1 below It shows that it took twice as long to complete the assignment for those that used the traditional manuals compared to the Online Help users Note that these are mean values the individual results differed from each other which can be seen by looking at the standard deviation The number of questions asked by the first group during the test was three times as many as the second group but there were no large differences between the numbers of subtasks completed by the two groups Traditional manual Online Help Completion Time Mean value 17 8 5 Stan dev 4 582576 6 363961 Questions 6 2 Comp tasks 19 21 17 21 Use
89. n case of trouble or if his knowledge of the domain is limited Nielsen 2005 Heuristic evaluation is not perfect there are risks and problems using this method The main issue is that the evaluators are not users themselves they only tries to emulate the users There is thus no real user feedback unless there are actual users involved in the evaluation Another concern is that heuristic evaluations do not scale well for complex interfaces A small team of evaluators may miss a majority of problems and not even find more serious ones UPA 2005 An important difference between heuristic evaluations and more traditional evaluations is that the observer is allowed to answer questions during the evaluation and even give hints on using the interface The evaluation sessions usually lasts 1 2 hours for each evaluator Longer sessions might be needed for more complex interfaces but if so it is recommended to split the evaluations into smaller sessions Nielsen 2005 4 3 Iterative Design Iterative design is a design methodology that is often used when developing user interfaces It work on the premises that the developers first create a initial design or prototype of the interface that is tested analyzed refined and done all over again until the developers happy with the result see Figure 4 1 20 A 0 0 Figure 4 1 The Iterative Design Process The reason that this methodology is especially useful for user interfaces is becau
90. n mind After this the plan was more or less followed with a few minor deviations It was just the end phase that differed a bit The outcome from the third usability test was supposed to be used to make some final adjustment to the third prototype to finalize it This was never done instead the test result was used to determine how good the Online Help was vs the traditional manual as well as determine what needed to be done in the future see chapter 14 6 Things also took longer to do than expected one of these things were the last usability test Since the two first tests were of a pretty unstructured nature there was no need for extensive planning neither were there a problem to find people to do them The third usability test were a bit different in this case the test had to be planned more carefully since the test subjects would have to complete an assignment given to them This assignment to create a filter for air traffic needed preparation the assignment needed to be formulated in an understandable way a questionnaire and observation protocol needed to created as well Manuals needed to be printed a room to carry out the test needed to be booked and rigged with all the necessary equipment the earlier tests had been done at the workstation not to mention getting at least six people with varied background and experience to do the test on a short notice Considering that the invitation went out on the last official vacation week to t
91. ndard information it is possible to choose a Help Content template that creates a table of content among other things The table of contents is located in a XML file called toc xml that contains labels and anchor ids to each topic The topics are also saved in XML files and links to an anchor id in toc xml they also contain the structure of the topic like main and sub topics linking to the HTML or XHTML files containing the actual content The created help content is easily tested by opening Eclipse Help Content window the new help files should be at the bottom of the table of content Eclipse 2009 2 6 2 JavaHelp JavaHelp makes it possible to implement Online Help into a Java project JavaHelp is not included in the JDK or JRE it must be downloaded separately The implementation can be divided into two separate parts creating the help files needed for the Online Help and implementing it into the code JavaHelp 2004 The Help files There are three important files not counting the HTML topic files that are needed to make the help content work The main one is the helpset file which is read by the application as soon as the JavaHelp system is activated This file defines the help content window for the application and contains a set of data that comprises the help system It contains information where the so called map file is view information that describes what types of navigators that are being used in the help content window st
92. ndix Maintenance Manuals e Date Inspection Book a Adracs foithar Onerator or VAL Figure 2 4 ActiViewer 2 6 5 Firefox Portable solution It was discovered while researching the different help system platforms that another company within the Saab Group Saab Systems had developed their own version of Online Help This version was based on Firefox Portable and a XML Client Their reasoning for using Firefox Portable as their runtime environment was that they would not need to install it to be able to use it it can easily run on a USB stick or even a CD This is important because most customers do not allow more than the main software to be installed on their computers Also the software has to be able to run in a UNIX environment and Firefox are thus more or less the only web browser to use Worth mentioning is that before they started working with Firefox Portable they tried to use JavaHelp but they found out that the browser that JavaHelp uses was not up for the task When the HTML file becomes too large the browser becomes sluggish and the response time increases rapidly This was five years ago but considering that not much development has been done on JavaHelp recently this problem may still be there Saab Systems uses a stripped version of Firefox Portable which has had many of its functions removed as well as most of the interface such as toolbars menus and address field see Figure 2 5 The structure of
93. ning and execution of the last usability test got delayed 58 The most time consuming part of thesis was to implement all in interface not only in the code but actually create the HTML content used for each Pop up Most of these Pop ups will probably never be used but since it is hard to know which Pop ups that were going to be used in the final test and is going to be used in the future it was better to be safe than sorry SMW could learn from this and make sure that Pop ups are only implemented where they are really needed If the implementation of Pop ups is going to be used in many future projects it may be a good idea to develop some sort of tool that could help with the creation of the HTML content used for the Pop ups During the development there have been some discussions with the Chalmers supervisor about the use of the term Online Help Since it contains the word Online one may think that it has to do with internet somehow but this is not entirely right The way this thesis implements Online Help does not have anything to do with internet since all the content is located locally on the workstation But there are other applications with Online Help that actual are online they usually store their help content on some company server that the application has to connect to before showing the help There are pros and cons with doing it this way the pros are that the company can easily update th
94. nt to go through a step by step call routing process each time they route a call they want a system that can efficiently route calls and rapidly complete them Here ease of use is of less importance But there are systems where easy of use is a major goal for the users for example kiosk at museums or similar Cooper 2007 The GDD Process Most companies do not use a good enough process to design a digital product if they even have such a process The problem is that the word design has lost its meaning in the industry nowadays it is more a word describing the looks of the interface not its function For the word to have its meaning back the designers must have a broader role than they previous had Today the roles in the companies are too precise researchers researches the market and designers do design To be able effectively and systematically translate the research into a detailed design specification designers have to be let into the loop and take part of the knowledge firsthand to better understand the users GDD combines several techniques like ethnography stakeholder interest market research detailed user models among others This defines a process that can be divided into six phases Research Modeling Requirement Definition Framework Definition Refinement and Support see Figure 3 3 16 Research Modeling Requirements Framework Refinement Support users users definition of user definition of of behaviors development and
95. nuals and tests made during the thesis would give information SMW needed to convince the customers of this as well Determine what runtime platform to use It was decided to use JavaHelp to implement the Online Help JavaHelp was chosen because it had the functionality needed to implement the Online Help illustrated in the mock up One can thus ask if this was the best platform for SMW to use This is not certain but there has been a great interest in implementing Pop ups in future projects and the only platform found that supported this was JavaHelp Find a program that makes it possible to generate content for the Online Help using existing documentation A suitable program was found to do this called RoboHelp an Adobe product using this it is possible to import Word or FrameMaker documents and transform these into a RoboHelp s file structure that can be used to create a number of different help types among them JavaHelp This program is far from perfect considering that it only create one big HTML topic file from the imported document at least Word documents This topic file must be manually divided into several smaller topic files to work and look correct A lot other small modifications also have to be done but still using RoboHelp is a far better solution than doing it from scratch Worth mentioning is that it was never tested to import a FrameMaker document because the version of FrameMaker installed on the workstation was too old and
96. ogram that uses JavaHelp s API this is called dynamic merge Which of them is used depends on the structure of the application JavaHelp 2004 Implementing JavaHelp into the code To be able to use the help content in the application there are two important things that have to be done The first thing is to include jhall jar which contains all the classes that JavaHelp uses in the project The other one is to create a reference to the helpset file so the application can access the help content To enable context sensitive help for windows and components in the application they must be linked to a topic ID that exists in the map file This can be done in several ways in JavaHelp but the most common one is to use the function enableHelpKey to enable help when pressing the F1 key it recommended to use this on windows If F1 is not going to be used but the component needs a topic ID anyway when using Pop up help for example the function enableHelp can be used instead JavaHelp 2004 Using the enableHelpKey function is all that is needed to get help by pressing Making Popup help work needs a little more work First of all a mouseListener must be added to the component that needs Pop up help When a component with an added mouseListener is activated the responding code will create a Pop up menu that shows the HTML file that is linked to the topic ID of the component There are other ways of using JavaHelp than those mentioned abov
97. oject Eclipse does not support the ability to import NetBeans projects while the opposite is possible Neither is it possible to copy the code into an Eclipse project because the code contains classes only available in NetBeans So an alternative way of doing the GUI was needed In the end the solution was simple The Java homepage had tutorials available for the different Swing components Most of these had the code for the different tutorial assignment provided The code for one of these were chosen and modified to act as a test GUI for the help platforms The resulting test GUI was a simple window with a menu bar text field and a button see Figure 8 1 below gt MainWindow BAX File Help Knapp il l Figure 8 1 The Test GUI 31 8 2 Eclipse Platform Help System EPHS Because Eclipse is the main programming environment at SMW it seemed like a good idea to incorporate a help system that used that environment 8 2 1 Testing Following the steps mentioned in chapter 4 5 1 a simple help system was created to test EPHS along with a few HTML files This did not pose any problems the problems began than it was time to test the help system with the Eclipse Help Content window Eclipse Help Content window uses Internet Explorer to show the HTML files and for some reason SMW firewall blocks the viewing of these files making it impossible to see how the help content looks or if it even works This es
98. ontain a complete index with a lot of synonyms to make sure that users find the topics they are looking for There would also be a glossary with explanations for acronyms and abbreviations and clicking on an abbreviation in the text would transport the user to the glossary tab for an explanation The help content would also have a search function that the user could use to search for certain words possibly instead of using the index It would also be possible to bookmark topic pages for quick access or just to remember important topics The context sensitive help would be divided into two separate parts the first would be the possibility to access topic pages in the help content for the window or sidebar that the user is currently using by pressing the F1 key This would work for any window or sidebar no matter what size or number of components in it The other part would be the possibility to middle click on any component maybe not those that are self explanatory i e text field drop down box checkboxes radio buttons buttons labels etc to show a Pop up window shortly explaining what the component is and what it does The size of the Pop up window adapts automatically after the amount of text in it although since a Pop up is only meant to contain short summaries the size of the Pop up windows will be kept reasonable The style of the Pop ups would closely resemble the rest of the interface in order to not look out of place 13 2 The Help
99. p som i Eclipse sedan _ kunna flytta och ndra storlek mm Andam lsenlighet 1 L nkar i manualen f r att knyta ihop olika funktioner mm Manual Andamilsenlighet 1 71 Klassificering Antal Kommentar endast positiva inkl personer Interaktion 6 Enkel interaktion med mycket h g acceptans ndam lsenlighet 0 Effektivitet 3 Snabb tkomst Tillfredst llelse 0 F nsterplacering amp 4 Bra respons p placering och storlek p Popups F nsterhantering Klassificering ej Antal Kommentar Tids tg ng positiva inkl 1 l tt fixat personer 5 dream on Interaktion 9 Sm grejer inte s viktiga 3 Andamilsenlighet 1 B ttre interface mindre Popups 5 Effektivitet 3 Fixa fokusproblem vid F1 3 Tillfredstallelse 3 Manual 3 4 Mesta fixas med en mer genomarbetat manual Fonsterplacering 0 Fonsterhantering 1 4 72 Appendix Usability Test Assignment Introduktion Testet g rs f r att j mf ra anv ndningen av en onlinehj lp mot en pappersvariant av anv ndarmanualen Du har blivit ombedd att g ra testet med hj lp av pappersmanual onlinemanual Om funktionen TDN Testet behandlar funktionen TDN filter TDN Tactical Data Net n tverket for utbyte av data mellan olika C2 enheter Ett TDN filters uppgift r att reducera utg ende data och detta g rs i regel pga att utg ende information till en annan C2 enhet r verlas
100. p system into SMW s products can be the start of introducing an iterative way of developing help system manual that will result in a better product as well as reduced development costs This thesis is a first step to achieve these goals 1 1 Purpose The purpose of this thesis is the following Implement a prototype Online Help system in an already existing Saab product and determine the possibility of using already existing documentation to generate the content of the Online Help This prototype will show how an Online Help system would look and work and thus give SMW a ground to stand on in future development of it To be able to do this a number of other problems have to be solved first First and foremost some background research has to be done on what Online Help is But more importantly how can it be made useful and usable and what is expected from it Another important thing is that since SMW wants to get rid of the traditional user manual in benefit of the Online Help they to convince their customers that this is the way to go This should be taken into consideration while doing the thesis A runtime platform has to be found and tested in order to implement Online Help as well as finding a suitable program to generate the Online Help content from the existing documentation There are a number of platforms that can be used but it is important that the platform will work in a Linux Java environment because it is this environment that
101. pened the help content window on the page that represented the chosen topic ID All in all JavaHelp was easy to use and there were no problems creating the help files It is well documented and it is possible to configure the help content in several different ways The downside is that not much have happened recently with JavaHelp the latest build is from 2007 but it seems like nothing major has happened since 2004 This might be because the interest from the community has declined over the years The question is if Sun is going to abandon any future development of JavaHelp and if so this should JavaHelp then be used in this thesis Another problem is that there are some things that are not clearly explained in the documentation To find solutions to this Google must be used but because JavaHelp does not see much use these days it is not always certain that a solution will be found 32 8 4 Oracle Help Oracle Help is 1 own version of JavaHelp much of help files structure is taken from JavaHelp but several new features have been included The programming part has also been simplified compared to JavaHelp 8 4 1 Testing It was a simple matter to create the help files for Oracle Help considering that most of the help files from JavaHelp could be reused without too much modification The same test GUI was used as with JavaHelp The code for Oracle Help was even simpler than JavaHelp to implement and worked like charm Strangely
102. r through the Pop up window be able to link to a dictionary or other useful parts of the manual 9 3 Focus of the Next Prototype After looking over the result it was decided that the next prototype should focus on correcting the window management issue Although it is not possible in JavaHelp to interact with the Pop up windows the way the test users wanted to instead the focus should be on why the testers wanted to move or resize the Pop up windows in the first place The reason may be because the placing of the Pop up was handled automatically by the system and could therefore be a bit random Fixing this should probably reduce the users desire to move the Pop up window To solve the second problem it would probably be easiest to create several different sizes of Pop up windows that the scrollbars disappear Without any scrollbars the users would unlikely want to resize the window 37 10 Designing the Second Prototype It was not only decided to fix the main issues with the first prototype in this prototype but also to try to make a real version of the help content 10 1 Creating the Help Content There were two options of making the help content one was to create everything manually with the material from the traditional manual and the other one was to try to find some sort tool to do this automatically As mentioned in chapter 2 4 SMW tried to make an earlier attempt with Online Help by using RoboHelp to make the help material This w
103. r type Trial amp error 1 5 1 Read all help 1 1 Hybrid 0 5 1 Preferred help 0 6 Question 1 Mean value 5 8 Stan dev 2 645751 1 Question 2 Mean value 5 8 Stan dev 1 1 Question 3 Mean value 5 7 333333 Stan dev 1 1 154701 Question 4 Mean value 4 9 Stan dev 0 0 Question 5 Mean value 3 666667 7 333333 Stan dev 1 527525 1 527525 Question 6 Mean value 5 8 Stan dev 1 1 One user switch from being in one category to another mid test Table 11 1 Summery of the test result During the test the way the testers were using the help material were observed they were then placed under one of three categories e Those that use trial and error until they run into problems and then read e Those that read short excerpts and then solve a small part of the assignment and then read again and so on e Those that read large portion of the text before starting to do the assignment and then try to solve as much as possible before reading again Observation done during the test showed that there was almost one person in each category in the Online Help group and similar in the traditional manual group except that one participant was part of two categories at the same time i e he started as one type and then changed his way of using the manual in the second half of the test 43 The answers from the questionnaires 1 6 showed low to medium values from the people using the traditional manuals and high to very high va
104. reating a chair or a car Digital products are so much more focused on the interaction than most physical products The traditional design used by the industry is therefore of little use when designing digital products This new kind of design interaction design requires an understanding of the user s relationship with the product from before purchase to end of life The field that has most in common with interaction design is that of architecture Architects need to understand how humans that occupies a structure lives and work to be able to design buildings that support and facilitates those behaviors This is very similar to digital products the interaction designers also need to understand how the humans using their product lives and work to be able to design a product that support and facilitates those behaviors The difference is that the field of architecture is a very old and well established while interaction design is new 15 In order to address this kind of new behavior oriented design one to understand goals each user has But what are user goals How are they identified Are they same for all users These questions among others need to be answered in order to do this The goals are not always what they seem an accounting clerk s goals for example may seem to process invoices efficiently This is rather his employer s goal not his his goal is probably to appear competent at his job and trying to be concentrated at his
105. rk has already been done and most importantly there is no need to rewrite the text to make it work Although there were talks about combining the control and operation manuals which of course demands that the manuals be rewritten but that is unavoidable in that case The planning could have been done better with clearer deadlines More time should have been put on the last usability test It was also unsuitable to do the thesis during the summer because the supervisors went on vacation in the middle of the thesis leaving me without supervising for a couple of weeks The term Online Help can be misleading because it is not necessary online but this is what this type of system is commonly referred as so calling it something else would only be confusing The secrecy around the application and the intended customers complicated the supervising a bit since the supervisor at Chalmers could not test the prototypes without visiting SMW himself Other than that it has not been such a great issue working or writing the report as long as nothing classified is mentioned and no compromising screenshots of the prototype is shown 15 1 Concluding remarks on the third usability test After taking into consideration that the questions asked during the last usability test was not of any relevant nature and thus did not prove or disprove that Online Help was 65 better the traditional manuals This concludes that the questions did not contribut
106. rketers are good at understanding the market and position of products within that market their involvement in the actual development is usually the delivery of a requirement list These requirements often have little to do with the users needs or desires and more to do with guesses based on market surveys what users say they want The problem here is that users seldom know what they want Developers on the other hand focuses more on solving technical problems follow good engineering practices and meet deadlines as well as receiving conflicting and or confusing instructions that they have to follow rather than consider on how the users will think and use the product Any real interaction design choices are made when the product is almost done thus making it near impossible to make any real changes in it The result of this is products that irritate reduced productivity and otherwise fail to meet the users needs 14 Another problem is that products are designed in way that makes seem rude blaming users for mistakes that is not their faults and then agreeing to this by pressing OK see Figure 3 2 for an example They have a tendency to interrogate the user by asking question like Are you sure you want to delete this file or Do you really want to empty the recycle bin Applications also require users to think like computers To rename a document for example the user either has to use the Save As menu command to create a new
107. rototyping is a process that is used early in the development to build a model of a system this is called a prototype Prototypes help system designer to build informative systems and is also a part of an iterative process see previous chapter Prototyping can be of use when determining the initial system requirements because it coverts these basic specifications into tangible but limited working model of the desired information system By doing this the users will have something physical that they can touch and see and thus giving them the chance to give the developers useful feedback that can be used to modify existing requirements as well as developing new ones There are several kinds of prototypes from low tech ones like sketches and traditional screens to high tech operational systems like using CASE Computer Aided Software Engineering Tools and systems created with fourth generation languages like Visual Basic Mcclendon et al 1999 They can be classified into three main categories wireframes traditional prototypes visual prototypes and interactive prototypes Wireframes and traditional prototypes are good in early development to demonstrate the basic concept although they are very limited non interactive and usually very broad But its strength lies in that it is easy and quick to create and don t require much technical expertise to do Visual prototypes often come in the form of screen mock ups done with some sort of visual
108. rought up in Online Help Online Help should be consistent i e it is unhelpful if a command is described one way in the traditional manual another way in the Online Help and a third way in ToolTip for example The way the help is accessed throughout the system should not differ either Online Help should be robust i e the help system should still be accessible even if the product fails It has to be more robust than the product itself 27 gt gt Online Help should flexible i e a user should get the level of help he needs This can be done by using context sensitive help or even fully adaptable systems that can feel the user s expertise Online Help should be unobtrusive i e it should not hamper the user from continuing his work or otherwise be in the way There are also some guidelines on how the online documentation the help content should be structured and written gt gt gt gt gt gt gt gt gt gt The use of hyperlinks is a good way of making the documentation easy to read and navigate Use clear structure with headings that work as signposts Organize information after user tasks Keep the sentences short to the point and free from unnecessary jargon The language should be simple but not condescending Do not make any assumptions of what the user already knows while writing the documentation Bring up procedures in order with numbered steps Highlight important steps Use
109. s was not done in this case The first was that many of the existing problems with the Online Help system was already known and those that could be fixed had already been fixed The second reason was that SMW needed test result that they could use to show the customers to convince them that Online Help is a good compliment and in the future a replacement to the traditional manual The hope was that the test would show that the Online Help was the better choice and luckily it did During the test more effort should have been put on listening to the questions the testers asked Too much focus was put on observing the test users action rather than listening to all they said while taking note it was easy to miss a question or forget to write it down Still all in all most of the questions asked even those that were not written down was of general nature But there is the possibility that an important question related to either the traditional manual or Online Help was missed and therefore could have made a difference in the result and thus could have changed the coming conclusion The time it took to complete the assignment differed not only between the two groups but also between the test subjects The time deviation between the participants was probably because some people needed more time than other to get familiarized with the system which is totally normal the main reason these people were chosen were because of their diversities Unfort
110. se it is almost impossible to design a user interface that has no usability problems read more about Usability in chapter 3 1 from the start Not even a usability expert could make a perfect user interface with his first attempt so that is why all user interface development should use an iterative approach in order to make the interface as good as possible The way it works is that the developers first create an initial design or prototype of the interface The prototype would then go through one or more tests with a number of test subjects The problems that the users would come up against would be noted down and used to improve the user interface This new design would then be tested again and the new problems noted would be used to improve the interface yet again This will be done over and over again until either the developers think the interface is good enough or the number of problems that would be found are so small that it would be a waste of time and money to continue How many iterations and number of test subjects that is needed differs from project to project More complex user interfaces will probably need many iterations before most of the usability problems are found and corrected the number of test subject need not be large sometimes a small number is enough to get the needed information Usability Iteration Figure 4 2 Iteration vs Usability 21 Figure 4 2 above shows a conceptual graph of the relations between
111. sentially made any further testing of EPHS pointless working in the blind is never a good idea The main disadvantage with using EPHS besides the problem of not being able to see the help content was that in order to make the help system work once the product had been shipped a slimmed down version of Eclipse was needed to be installed as well Because SWM s customers seldom let anything else than the main software be installed on their computers makes EPHS a bad choice in this case It was then decided that EPHS was not suited for this and thus discarded 8 3 JavaHelp JavaHelp is another obvious and suitable choice of a help system mainly because it is Java s own version of Online Help and therefore in theory anyway should work exceptionally well since Java is used in the application 8 3 1 Testing To test things out simple versions of the helpset map and topic files were created as well as some simple HTML topic files It was then a simple thing to add a few lines of code into the test GUI The main window as well as several components had topic IDs linked to them by using enableHelpKey method In order to get this to work on buttons and likewise the actual button had to be pressed down before the F1 key was pressed This seemed cumbersome in the long run so instead a small What is this menu was created that appeared when the user pressed the right menu button on a component with a topic ID Pressing the menu item o
112. sted i e not need to rewrite everything from scratch The most suitable platform for the task JavaHelp was used to create several prototypes with implemented Online Help These prototypes each went a usability test to improve the user interaction with the Online Help system The result showed that JavaHelp while being a good choice for Online Help does not seem to be developed far enough to be used in commercial products It is perfectly suited for presentation purposes though to show other how Online Help could look and work The conclusion is therefore that there is no simple way to implement a fully working Online Help in a product without having to rework and modify some parts of the help platform in this case JavaHelp The upside is that JavaHelp is free and does not need licensing to use This thesis will be used by SMW as ground for future Online Help projects next will be a project that will test to implement Pop ups for the use in tables and see if it can be used in a commercial product for real The prototype itself will be used to show others how Online Help can look and work To use the already existing help material to create the Online Help worked well the controls manual can easily be used to create Pop up content and by using an application like RoboHelp it is possible to quickly convert existing manuals to a working online manual These will of course need to be modified a bit to be suitable as online manuals but most of the wo
113. t budget so for now they take one step at a time Another important thing is how well the result from this thesis can be used in other systems JavaHelp can easily be integrated into any system written in Java but JavaHelp as it is right now is not recommended for commercial use The code form the prototype should instead be used if it should be used at all to show how JavaHelp can be implemented in an existing system and use it to show how Online Help would look and work in that particular system 14 4 Writing the Report The main difficulty with writing this report was that it was done alone which means that there was no partner to discuss it with as well as someone to read the written text fortunately the supervisors here at SMW filled this role instead The supervisors could only partly help with this task though since they also have other things to do other than supervising it is not always they have the time to read through drafts of the report or read it thoroughly Thus it was mostly the structure of the report that was commented on rather than the content and language 60 While talking about structure of report it is worth mentioning that have been some minor problems with that The main one was that the supervisor from Chalmers wanted the report to look one way while the people at Saab wanted another focus Saab wanted to focus more on the results from the usability tests Therefore the results from that test ha
114. t feedback on it Mostly it was that the help content was not detailed enough or that the text did not make sense 10 4 Focus of the Next Prototype Even though much of the negative feedback was on the interaction the main issues on that part was on the What is this menu and the lack of a button The first one can easily be fixed by not including it and the other one is of no large importance the interface does not need any more buttons The focus problem has to do with the window management of the application and is already worked upon by others in SMW although since they are working on another version of the application than the one used in this thesis their success and failure will not have any affect here It would be to difficult and time consuming to fix this for the next prototype so this problem will be ignored Neither will anything be done about information in the online manual unless a new version is made since the making of the manual has nothing to do with this thesis In the end there is not much improvement to do on the next prototype instead the focus will lie on implementing Pop ups in the rest of the application and focus on the next usability test to make sure that the prototype can handle it 40 11 Designing the Final Prototype Even though the result from the second test did not result in any high priority issues that had to be fixed there were still a lot of things that needed to be added in the
115. t for use as an Online Help system that probably will be switched to something else within a foreseeable future would be unwise and a waste of time 8 6 Firefox Portable A demo of one of the manuals that Saab System had made was sent to SMW so it could be tested and determine its use in this thesis 8 6 1 Testing the demo The demo received was a CD based version and needed to transfer some temporary files to the hard drive to be able to run The manual looked nice and the headings were ordered as hierarchical file tree clicking on one of the heading would open up the topic on the same page slightly indented inside a frame see Figure 2 5 Clicking on it again will close the frame it all work just like the and signs in a hierarchical file tree All the standard toolbars and menus had been removed and instead a toolbar containing the functions of a standard help content window had been implemented This included 33 topic index search etc The only thing giving away that it was Firefox was the small Firefox icon in the upper left corner The only problem with the manual was that it was just the manual not the application it was meant for It was thus no way to test the actual interaction between the application and Online Help Neither was it possible to see how the code of the Online Help had been implemented in the application Contacting Saab Systems about this revealed that it was unlikely that they would disclose any informa
116. t than ToolTips Real Mission Properties 89 File elp Content Mission Real Keyboard Shortcuts Symbol Legend Target type values Fixed wing Helicopter 0 UAV 0 Missile 0 Figure 7 1 The Help menu Figure 7 2 ToolTip This mock up were presented to the two supervisors at SMW one programmer and one human factor engineer that were involved with thesis in order to get feedback and useful advice that could be used to determine what exactly the Online Help should be focused on The feedback mainly discussed if both ToolTip and Pop up should be used whether the use of them should be consequent or used differently from window to window depending on context In the end it was decided that in order to be consistent the focus should be put on implementing Pop ups on those components that are relevant since ToolTips are not suitable to show longer text masses and cannot be controlled by the user in the same Pop ups Other decisions were that components that had a label attached to them should both have the same Pop up the middle mouse button should be 29 used instead of the right one because it is already used to open shortcuts menus the map in the product No Pop ups should be used on the group box labels a label that defines a whole set of components the user will have to look in the online manual for these Yi AM etrahoe Figure 7 3 Pop up
117. t was beh oaks 40 10 4 FOCUS OF THE NEXT PROTOTYPE 0s0cessvsessvesvevevvscvsvesvesvevsvvsvvscvecssvevevesevsveveesvevevvsevsevsessvvvsvesens 40 11 DESIGNING THE FINAL PROTOTYPE ccccssssscsssscesssscccsssscccsssscccccsssccssssacccessssccscsssesesses 41 11 1 BUILDING THE FINAL PROTOTYPE 41 11 1 1 Adapting the prototype to the next usability test ccc 41 112 THE THIRD USABILITY TEST 42 THE TEST RESUET 43 12 ANALYSIS OF THE THIRD TEST RESULT cccsssscssssssccssssccccsssseccsssscccssssccscsssecsscsssecseses 45 13 THE FUNCTION AND APPEARANCE OF THE PROTOTYPE sccccsssscssssscccsssseccssssceeeees 47 13 1 THE CONCEPT OF THIS ONLINE HELP SYSTEM cccccccseeeseseseseseseseseseseseseseseseseseseseseseseseseseseseeees 47 13 2 THE HELP CONTENT 47 O AAA B TRE E E E E E OA 48 13 2 2 The Navigation Panel EE E EEE EEE EE assests EER 48 13 23 The Topic Window aranera e eE REEE E E NEEE E ENEE EE E EER 5I 13 3 ONLINE HELP IN THE 51 TBST The Help terea n a a a eea e beses 5I 13 3 2 The Context Sensitive Help in the Prototype
118. t was used in the creation of the following prototype The third prototype included a fully implemented OH system with online manual F1 key support to access it and Pop up help in the whole application The third test let the testers complete an assignment by either using the traditional manuals or the OH in order to see if there was an advantage to use one over another The conclusion is that it is possible to cheaply and relatively easy implement OH in an already finished application Although JavaHelp is far from a perfect solution and need more work if it is going to be used commercially It is also possible to create online manuals from already existing documentation such as Framemaker or Word documents by using RoboHelp Sammanfattning Det h r examensarbetet gjordes pa Saab Microwave Systems en affarsenhet inom Saab AB och handlade om hur Online Hj lp OH kunde bli implementerad i en av deras produkter men ven att ta reda pa om redan existerande dokumentation kunde anv ndas att skapa hj lp materialet till OH utan att beh va skriva om allt L mpliga OH plattformar unders ktes och testades i slut ndan best mdes det att JavaHelp skulle anv ndas pga dess m ngsidighet och att det st dde Pop ups Utvecklingsprocessen gjordes iterativt med tre prototyper Mellan varje fas s gjordes ett anv ndningstest vars resultat sedan anv ndes i utvecklingen av n sta prototyp Den tredje prototypen innefattade en fullst ndig OH
119. tance from the test users they especially liked to able to open Pop up windows with just one mouse click Effectiveness Satisfaction The quick access to help with certain components as well as the possibility to get help not only from the actual components but also from the labels next to them led to high effectiveness and satisfaction from the test users 9 2 2 Feedback on what was bad Window placement and management Several of the test users did not like that they could not move or change the size of the Pop up windows There were also cases where the Pop up was blocking certain labels which caused irritation from the testers Some Pop ups had scrollbars because there was too much text to be shown in its standard size some of the testers were bothered by these others was not but it seemed like they preferred bigger Pop up window over scrollbars Interaction One test user complained that he could not use the right mouse button to open the Pop ups he also did not like that the Pop ups appeared right away Instead there should be a small What is this like menu that is shown first pressing this menu should then open the Pop up 9 2 3 Miscellaneous feedback There were also a few suggestions of additional functionality that could be implemented to enhance usability things like an option to have a separate help dialog 36 window on side that shows help text components that user is currently working on o
120. tas inom en och samma dialog s F nsterplacering 1 skall den minnas positionen Vid ppnande av en ny skall den mao ppnas p samma plats som f reg ende Att som val kunna l gga en separat hj lpdialog i sidopanelen ndam lsenlighet 2 som hela tiden visar hj lptext f r den komponent man f r tillf llet arbetar med Fr n Popup f nstret kunna l nka sig vidare till t ex ordlista ndam lsenlighet 1 f r komplicerade begrepp eller f rkortningar eller annan l mplig text i manualen Denna skulle i s fall l nkas till akutellt kapitel i F1 dialogen Dynamisk text som f rklarar de nuvarande inst llningarna ndam lsenlighet 1 Klassificering Antal Kommentar endast positiva inkl personer Interaktion 4 Enkel interaktion med mycket h g acceptans ndam lsenlighet 0 Effektivitet Snabb tkomst och bidrog till h g effektivitet och tillfredst llelse Tillfredst llelse 3 Se ovan F nsterplacering amp 1 F nsterhantering 69 Klassificering ej Antal Kommentar Tids tg ng positiva inkl personer Interaktion 1 Andamilsenlighet 4 Funktionsaddering r av l gre prioritet i detta 3 skede av utvecklingen Noteringar tas med vid n sta fas Effektivitet 1 Tillfredst llelse 1 F nsterplacering 6 Flera av kommentarerna hanterar detta mne 1 och b r d rmed prioriteras i fortsatta fr gest llningar F nsterhantering
121. tat Mha filtret kan man d rmed f ner den utg ende informationen Scenario Du har uppt ckt att en av era flygplatser genererar en stor m ngd luftm l vars information dina andra C2 enheter inte r intresserad av och som dessutom r p v g att verlasta n tet Du vill d rmed uppr tta ett filter kring denna flygplats Uppgift Din uppgift r att uppr tta ett TDN filter som Ar aktivt runt flygplatsen Far bort information dina egna flygenheter d de startar och landar Tillser att du fortfarande kan se m l som flyger p h g h jd ovanf r flygplatsen Du n r TDN funktionen fr n huvudmenyvalet Communication Hj lp f rsta hand skall du anv nda dig av den de manual er som finns tillg nliga Som sista utv g f r du be observat ren om hj lp T nk g rna h gt Totalupplevelse N r du genomf rt uppgiften kommer du att f prova att g ra samma sak med den andra manualtypen ven om du nu vet hur man g r s f r du f rs ka att se det ur perspektivet att du inte vet Det g r bra att och k nna i h gre utstr ckning f r att sedan ge ett omd me om ditt totala intryck 73 Appendix Observation protocol Pappersmanual Onlinemanual H ndelse Antal Kommentar Antal tillfr gningar till observat r F ljande aspekter r uppfyllda i 1 Aktivera filtret 2 F rst hur Point fungerar 3 V lja r tt Plattform genomf randet av uppgiften 4 V lja r
122. ters The feedback would then be used to make some adjustments to the prototype before submitting it and later be used in the report this report to explain what needs to be done when creating a real Online Help The rest of the time would be used to write this report which of course will be written alongside the rest of the thesis 26 6 Making Guidelines for Online Help To get a better understanding on how an Online Help system should look and work it was decided the best course of action would be to research guidelines for Human Computer Interaction HCI This was done by studying the books by Cooper 2007 and Dix 2004 These two authors seem to focus on different parts of how Online Help should be structured In the end a summary of the most important guidelines from each book were written down which will be described in the next section 6 1 Design Guidelines for Online Help Different books gives different guidelines but most of the time they do not conflict with each other Cooper 2007 gives the following advice on Online Help gt gt gt gt Online Help should never act as crutch product Online Help should be designed for experienced users not beginners A usable index is generated by exploring the application not by reading help text Try to think goal directed while creating the index Thinking like a user helps a lot Especially since a large number of synonyms are need
123. the Online Help consisted of XML files and all functionality is handled by a Java script which makes sure that the users cannot access the file system of the workstation in any way Everything is run from a start file which uses an index file containing about 70000 words or 300 400 documents in both XHTML and It is easy to search in The Online Help can be started in several ways one is to use the start file directly or by 10 right clicking on a component a What is this button will then be shown that user can click on to open the Online Help The button is greyed out if there is no information about the component Each such component has an URL linked to it so when the What is this button is clicked Firefox will show the topic of that URL The URLs are ordered hierarchy depending on how the components are placed in the GUI Each night a XML file is generated containing the changes that have been made during the day so that the developers easily can implement Online Help to new buttons and such This makes the system easy to use and reduces the workload on the developers They work in a UNIX environment while SMW develops products in a Windows environment that are meant for UNIX environments Because of this it is unlikely that the developing tools they use will work in Windows 0088 As Hep Department Reference Origin date Issue date Issue index Document id LO STSSC 2008 10 21 2009 05 19 12 PM305231
124. the and use business and design structure form and needs domain context technical needs and flow content Figure 3 3 The Goal Directed Design Process Cooper 2007 The research phase uses different ethnographic techniques to provide qualitative data about users of the product This results in behavior patterns that suggest goals and motivations that the users have in order to use the product These can be used later on in the Model phase to create personas Personas or user models are detailed user archetypes that represent groupings of behaviors attitudes aptitudes goals and motivations discovered during the research phase These are used in several of the other phases like main characters in scenario based approaches to generate design concepts in the Framework Definition provide feedback that enforces design coherence in the Refinement phase among other things The Requirement Definition uses scenario based design methods focusing on meeting the goals and needs of specific user personas Each persona is analyzed through an iteratively refined context scenario that starts with a in the life of the persona using the product describing high level product touch points and then successively defining detail at ever deepening points The result of this is a requirement definition that balances user business and technical requirements that the designers can follow During the Framework Definition phase the overall concept is cr
125. tion It has six menu items Help Content Acronyms and Abbreviations FAQ Keyboard Shortcuts Symbol Overview and About see Figure 13 8 Help Content Acronyms and Abbreviations FAQ Keyboard Shortcuts Symbol Overview Figure 13 8 The Help Menu The Help Content menu item opens the online manual mentioned in chapter 13 1 Acronyms and Abbreviations will open the help content with the topic about abbreviations has the same information as the glossary tab mentioned in chapter 13 2 2 Keyboard Shortcuts and FAQ have not been implemented but their name clearly states what they are supposed to do Symbol Overview also opens the help content at the appropriate topic Last there is the About item that are supposed to open a 51 window showing system company information but alas this has not been implemented either 13 3 2 The Context Sensitive Help in the Prototype The context sensitive help works the same way in the whole prototype but looks a bit different depending on where it is used The side panel The side panel is placed at the right side of the interface and contains a number of different panels that can be reached by using a dropdown menu see Figure 13 9 Going through them all would take too much time and would be completely unnecessary since most of them work and look the same Instead the focus will be put on the first panel in the list the Air Track List panel This panel contains a lot of information about
126. tion about this to another Saab company in a foreseeable future company bureaucracy at its finest Because of this it unlikely that Firefox Portable will not be used as the Online Help system in this thesis 8 7 Choosing Online Help platform After looking through the available choices the answer was quite obvious Eclipse Platform Help System did not work well within SMW s network and needed a slimmed down version of Eclipse in order to work Oracle Help was easier to use than JavaHelp but did not support Pop up windows in applications ActiViewer was a no longer supported by their developer and thus a dead end The Firefox Portable version seemed like a good way to go but without any API code or otherwise way to use it there was no point of pursuing it any further The only platform left was JavaHelp this is by no means the most optimal platform but the only other options is either create an own version of Online Help from scratch or use JavaHelp as a base and build something from that Both of these options would take too much time and this thesis was never about creating an Online Help from scratch but trying to do something with what was already available 34 9 Designing the Initial Prototype After it had been decided to use JavaHelp to implement Online Help the work on the first prototype could be started The first step was to create a new branch for the code in the source control program ClearCase both to protect the original code
127. tional and Online had not been elaborated enough Observing the way the participants used the help types assigned to them was like watching a demonstration of human nature Men in particular had a tendency to use a trial and error approach to try to solve the assignment this failed in all but one case This seems to have something to do with the joke about that do not need to read manuals since they think they can figure it out anyway Unfortunately there is a lot of truth in this joke men do have a tendency to not read manuals before trying out something new This test was no different but in the end after failing to solve the assignment they felt obliged to start using the help in order to solve it Then there were the opposite those that started with familiarize themselves with the manual or Online Help and read long sections before doing the assignment Lastly there are those in middle that read what they are supposed to do before each subtask The last one seems to be the one that is used most although the test result showed that the participants were pretty much evenly divided between the three groups This only shows that there are many different people out there and probably far from all can be sorted into these three groups An interesting observation was that when the test groups got to test the other help type i e the traditional manual group got to try the Online Help and vice versa some of the participants used th
128. to the questions in the questionnaire can mostly be tied together to the amount of time it took to do the assignment Especially question 1 How long did it take to find the right information plays a big role in how long time it took to complete the assignment The other question plays a more or lesser role in this as well question 4 for example asks how easy it was to learn how to use the help or question 3 To what level did you experience that you got the help needed to reach your goal The rest of the questions also affect the completion time of the assignment and thus it is easy to see why the Online Help group got better completion time than the other group 46 13 The Function of the Prototype This chapter will describe how the final prototype looks and work but also talk about the concept of this Online Help system i e how it would work ideally The first part will describe this concept the second will handle the help content and thirdly the context sensitive help in the application 13 1 The Concept of this Online Help system Ideally this Online Help system would have consisted of a complete online manual help content and fully integrated context sensitive help The help content would be a combined version of the operations manual and the controls manual Each chapter and following subchapter would have its own topic page HTML file with its own link in the Table of Content It would c
129. track amp EW Jam Neutral v Assumed friend Strobe identity is set to none Friend Unknown Pending Pending Suppress volume Suppress volume Limit the geographical area for air track and points Limit the geographical area for air track and points Area shape Area shape Upper left corner Upper left corner Lower right corner Lower right corner C Limit the attitude span Limit the altitude span Min m Max m Min m Max m Filter outside specified volume 7 Filter outside specified volume Create Cancel Create Cancel Figure 13 15 Pop ups 55 14 Discussion The main purpose of this thesis was mentioned in the Introduction to Implement a prototype Online Help system in an already existing Saab product and determine the possibility of using already existing documentation to generate the content of the Online Help This purpose has been fulfilled and will be brought up in more detail in the rest of the chapter There were also few other tasks mentioned in the introduction which needed to be addressed before the actual prototypes could be built These will be discussed here How can it be made useful and usable Chapter 3 explains how it should be done and if the Online Help prototype achieved this or not is explained in chapter 14 3 below What is expected from the Online Help It was expected that the Online Help would be a good replacement for the user ma
130. ubject for C Show Damne _ pet I Secret None of the targets set as secret wil be outgomg Action will be vald for all fitters in the Bet Figure 13 14 Pop ups in the TDN Control Panel window 54 There are also windows within windows that have context sensitive help clicking New in the TDN Control Panel window will open up the New TDN Filter window That window has F1 key and Pop up help as well It is also possible to open the Pop up windows by using the right mouse button instead of the middle one The difference is that a menu with a single menu item with the text What is this will be shown first by clicking on the menu item the Pop up will be shown see Figure 13 15 This functionality is only implemented in TDN Control Panel and its underlying windows as well as the Weapon Engagement Status and Mission Control panel all for testing purposes E New TDN Filter Fitter Name Activate filter Name Activate filter Suppress source Suppress source OwnC2 v OwnC2 Suppress category Suppress category v Point rvam Strobe identity Point Both S Z Hostile Suspect The Point checkbox needs to be checked if the combobox in the group box Air
131. ults of Usability tests are often measureable and quantitative and are thus suited when comparing different design variants in order to determine the most effective solution These tests are especially good at determining e Naming Do the naming of buttons and labels make sense Are there certain words that may be more suitable to use than others e Organization Are similar items grouped in categories that make sense Are these items placed so that the customer easily can find them e First time use and discoverability Are items that are often used easy for new user to find Are instructions clear Are instructions needed e Effectiveness Is it possible for customers to efficiently complete specific tasks Do they make mistakes doing this Where and how often It is worth noting that Usability tests are most suitable for first time use of a product It is quite difficult to determine how effective a solution is the 50th time a user uses it Considering that products are often designed to be used by intermediates and experts this might question the effectiveness of these kinds of tests But there are techniques to solve this for example diary studies where the subjects keep diaries describing their interactions with the product Cooper 2007 4 2 Heuristic Evaluation Heuristic Evaluation is considered a discount usability engineering method meaning that it is a cheap quick and easy way to evaluate user interface design Because of this it has
132. unately this means that the results get a bit harder to interpret with such deviating numbers fortunately the differences between the two groups clearly showed that those using the traditional manual took much longer time to 61 complete the assignment because it took much longer to find the right section to read plus that the Online Help gave much clearer guidance There was also the odd test user that did not use the Online Help at all to solve the assignment this means that he would probably not used the traditional manual either if he had been apart of that group Although he did spend some time with the traditional manual when he got to try that out but this may have been because of other issues that will be discussed shortly This skewed the result a bit but even if he was removed from the result the other two in the Online Help group had better completion times than the best test result in the traditional manual group The test users completed the most of the subtasks Those that took longer time or had problems with certain subtasks was mostly because they did not examine the interface or read the help good enough not because which help type they used There were one subtask that most people failed on it was a checkbox called the box was checked as default but the subtask here was to understand if should be checked or not Most did not get this even after reading the available help giving a hint that the help both tradi
133. upport phase is not entirely a phase per se it is something that is used during the entire development to help answer developers questions as they arise This helps the development team to prioritize their work and make trade offs to meet deadline Not 17 doing this could result in compromises that threaten the integrity of the product s design Cooper 2007 18 4 Methods This chapter will describe how theory from previous chapter was adapted and used in this thesis 4 1 Usability testing Usability testing also wrongfully called user testing it is the actual use not the user that is being tested is a collection of techniques that lets developers test their products to measure the characteristics of a user s interaction with a product with the goal of determining the usability of that product Usually the users are put through standardized test that they have to solve to measure how well they can complete their tasks as well as what problems they encounter during the test These tests often reveal places where the users have problems understanding and utilizing the product but also places where the users are more likely to succeed To be able to conduct Usability tests there have to be something fairly complete to test it on This can be production software or a traditional prototype it does not matter What does matter is that the prototype simulate how the actual product will look work and feel The res
134. ve been a great focus in the latter part of the report Doing a report that way is more scientific but clashes with the focus on creating and designing an Online Help system those report structure is a bit different with more focus on the prototype and less on the tests But since neither side can be followed fully without important information missing the end result of the structure is compromise of the two As mentioned earlier some thing of this thesis were classified this means that care had to be taken while writing the report so that none of this sensitive information was mentioned anywhere This was not really a problem while writing the text since there was no real reason to mention whom the product was meant for nor the specs for it The problem was the screenshots taken of the prototype these could contain sensitive information was not immediately apparent things like the project name that which can be used to figure out the customers country So these pictures have to be looked through carefully by experienced people at SMW before anyone outside SMW could read the report 14 5 The Result from the Third Usability Test The test focused on comparing the traditional manual with the Online Help system This may not have been the best of choices from a user centered design perspective Normally the usability test would have focused on finding faults with the developed product not comparing it to others but there were two main reasons that thi
135. with HFE and interaction design from all over Saab discussing this The introduction of Online Help into SMW s products is one of these goals Currently there is a person that documents all that happens in the interface which DP K then uses to base their work on like manuals There is talk about trying to continuously update the documentation with the software during development The thought is that one of the employees at DP K will be a part of the development team and help with writing the Online Help and user manual i e the source material to these two At the end of an iteration the Online Help will be tested verified which will hopefully increase the quality of the documentation In addition the rest of the team could use the Online Help to learn more about the system Small changes made in the interface can be noted and implemented into the documentation by the constructors The Online Help will thus be continuously be inspected by the entire team 2 4 Earlier attempts with Online Help This thesis is not the first attempt that SMW has made with Online Help a few years ago an attempt to incorporate Online Help into one of SMW s products using JavaHelp and RoboHelp a publishing tool to create help content This was a part of a customer project there they wanted to test how Online Help could be used in an application This was never meant to be more than a quick test because of tight deadlines and limited budget The test showed that it
136. work while performing routine and repetitive tasks even though he is probably not even consciously aware of this This means that products built with business goals in mind alone will eventually fail users goals needs to be addressed for product to be successive Many applications today fail to do this their interfaces makes the users feel stupid causes them to make mistakes require to much effort to operate efficiently and importantly they don t provide an engaging or enjoyable experience The companies priorities are wrong They focus too much on implementation issues rather than the needs of the users Even when they focus more on the users it is not always they can do anything to change the product because the conventional development process says that the coding is done first and designing the interface done later This gives little room for changes If this is not the issue then the developers tend to pay too much attention to the tasks the users are doing rather than the goals they are performing to do those tasks It is worth noting that goals are not the same thing as activities or tasks Goals are an outcome that is reached by doing tasks and activities There is often a misconception that ease of use should be prioritized when designing a product this is an important guideline but following rules that are disconnected from user goals are not good design For example people using an automatic call distribution system do not wa
137. y the context of use Specify user amp organizational requirements System meets functional user amp organizational requirements Evaluate designa against requirements Produce design solutions Figure 3 1 An illustration of the UCD iterative process The purpose of the UCD process is to make the product easier to use improve user satisfaction and reduce comfort and stress improve the productivity of users and operational efficiency of organizations as well as improve product quality appeal to the users and provide a competitive advantage ISO 13407 1999 But as the name UCD states the point of ISO 13407 is to stress the user s role in the development Cooper 2007 has made his own interpretation of how this standard should be followed and puts his focus on the goals of the users This Goal Directed Design is described in the next chapter 3 2 1 Goal Directed Design Goal Directed Design GDD Cooper 2007 is the process of putting the needs or goals of the users as the main focus of the project so that they will be satisfied effective and happy to use that product and therefore willing to open up their wallets and buy it and recommend others to do the same This sounds like an easy thing to do to design a product that users enjoy using Unfortunately it is not that easy and the reasons for that are many In the center of it all is the tug of war between the developers and marketers The ma

Online Help

Contents

Download Pdf Manuals

Related Search

Related Contents