HCI Design Principles for eReaders
Contents
1. M lt Previous M gt Next f Home Q Size Options Resolution 800 x 600 pixels Size 175 x 122 x 10 mm Weight 286g Wireless Connectivity None Memory 512MB Internal Expansion Accepts Memory Stick Pro Duo MS and Se cure Digital SD cards up to 16GB Power Sealed Internal Lithium Ion Battery around 7500 page turns on one charge eBook Formats Supported Adobe PDF TXT RTF Mi crosoft Word BBeB EPUB RRP 279 99 A 3 Amazon Kindle 2 Display 6 inch eInk 16 Level Greyscale Physical Buttons Devices Power Switch Vol Vol PREV PAGE NEXT PAGE X 2 HOME MENU BACK Full QWERTY Keyboard amp Symbol Text Size 5 way Directional Controller Resolution 800 x 600 pixels Size 203 x 135 x 9 mm Weight 289g Wireless Connectivity 3G Memory 2GB Internal Expansion None Power Sealed Internal Lithium Polymer Battery up to one week with wireless on on one charge eBook Formats Supported Kindle AZW TXT ADOBE PDF HTML MS WORD JPEG GIF PNG RRP 189 99 B SONY READER POCKET PRS 300 B 1 Ergonomics PRS 300 The Sony Pocket edition has several sets of buttons on its main facia As well as its home f and return D buttons the device also includes dedicated buttons for bookmarks A and font size X as well as a circular directional pad and enter button and a set of 10 numerical inputs These buttons have been arranged in a somewhat symmetrical pattern forcing the direc2. 6 REFERENCES 1 Annette Adler Anuj Gujar Beverly L Harrison Kenton O Hara and Abigail Sellen A diary study of work related reading design implications for digital reading devices In CHI 98 Proceedings of the SIGCHI conference on Human factors in computing systems pages 241 248 ACM Press 1998 2 M J Adler and C Doren How to Read a Book The Art of Getting a Liberal Education Simon and Schuster 1972 3 George Buchanan and Jennifer Pearson Improving placeholders in digital documents In Research and Advanced Technology for Digital Libraries volume 5173 of LNCS pages 1 12 Springer Berlin 2008 4 Catherine C Marshall Reading and Writing the Electronic Book Synthesis Lectures on Information Concepts Retrieval and Services Morgan amp Claypool Publishers 2009 5 Catherine C Marshall and Sara Bly Turning the page on navigation In JCDL 05 Proceedings of the 5th ACM IEEE CS joint conference on Digital libraries pages 225 234 New York NY USA 2005 ACM 6 Rolf Molich and Jakob Nielsen Improving a human computer dialogue Commun ACM 33 3 338 348 1990 7 Jakob Nielsen and Rolf Molich Heuristic evaluation of user interfaces In CHI 90 Proceedings of the SIGCHI conference on Human factors in computing systems pages 249 256 ACM New York NY USA 1990 8 Kenton O Hara Towards a typology of reading goals 1996 9 Kenton O Hara and Abigail Sellen A comparison of
3. in a reading group It can be confusing however to have multiple pages with the same page number A more suitable solution to this problem when on a larger zoom level would be to keep the original page numbers but also include how many of them they are as in Page X Part Y of Z The other more common page numbering system used by Sonys eReaders is to calculate the number of pages within a document on the fly depending upon the zoom level Thus the more zoomed in you are the more pages there are in the book This method however now leads into further prob lems such as the shared reference problem encountered by those collaborating using the same document 4 Previously when two users were reading separate copies of the same doc ument they could say hey look at page X paragraph Y Now to find the particular passage a user would also have to say go to zoom level S then go to page X paragraph Y Another problem faced here is bookmarking If a user places a bookmark on a page then zooms in the old book mark is lost as only one of the multiple pages the original page now spans over is left bookmarked by the system For example if I were to bookmark page 10 on the smallest zoom level then zoom in to 4 times magnification my origi nal page 10 now spans over 4 pages except the system only bookmarks one of these 4 pages Similarly if a bookmark is made when the document is on a high zoom level then
4. the user must move the caret to the beginning of the place they wish to highlight using the 5 way directional controller and start by clicking the 5 way down The caret then turns italic and will highlight text that is selected using the 5 way directional buttons Clicking the 5 way down again will finish the highlight which is subsequently indicated by a faint grey underline D 3 2 Notes Kindle Users can add notes to specific points in the text by first going into annotation mode described above then moving the cursor to the place in the text they want to anchor the note To create a note in a specific location the user only needs to start typing using the full QWERTY keyboard in cluded on the facia of the device They then select the save note option using the 5 way and a new note is created in the specified position D 4 Page Turning Kindle There are two next buttons but they are positioned in the right place for fingers unlike the sony reader where you have to strain your hands to use Has been designed with both right and left handed users in mind i e there are two next buttons positioned on the sides of the devices in such a way that they can be utilized by either thumb and more importantly be used with only one hand Since books are designed to be read sequentially from start to finish the previous button is expected to be used less fre quently This is reflected in its position a single small but to
5. Notes menu is off Once a comment has been made and committed it will disappear from view and can be accessed by tapping once on the icon adjacent to the bookmark or highlight Not only does this functionality deny the creation of comments randomly on a page as each one must be anchored to an existing note but each note i e bookmarks or highlights can only sustain one comment each C 4 Page Turning PRS 600 Changing pages on the device can be accomplished either by using the dedicated buttons on the device itself or by making sweeping gestures on the touch screen To move to the next or previous pages within the document the screen must be dragged in the appropriate direction with a stylus or finger and to repeatedly change pages i e flick quickly through document the screen must be held after the ges ture has been made The default page turning gesture direc tions have been designed to coincide with the buttons on the device specifically gesturing from right to left lt takes you to the previous page and from left to right gt takes you to the next page These directions however logical con tradict the learned behavior of reading physical books For example when reading ahead in a paper book i e going to the next page one would take the right page and flip it to the left Conversely to go back to a previous page one would take the left page and flip it to the right completely opposite to the gestures provide
6. The sample respondents of the Kindle is far greater than that of the Sony devices some 15 000 for the Kindle versus less than 100 for both Sonys We assume this is due to the exclusivity of the Kindle being sold on Amazon only whereas the Sonys can be bought from many stores and online retailers and therefore their users are less likely to provide assessments on Amazon It is clear from Figure 1 that the majority of users think highly of all three devices giving scores of mostly 4s and 5s out of a 5 point Likert scale However the reviewers on Amazon are self selected If they had bad experiences they might tell the manufacturers or if they simply gave up us ing the device they would probably wish to forget about it On the other hand excited users would likely be keen to tell everybody about their experiences For an unbiased experi ment respondents need to be selected randomly or in some other controlled way not self selected Nevertheless what the Amazon data shows is that people who like eReaders really like them One might then ask why the present paper if enough peo ple already like the devices HCI is about making things easier and nicer and if enough people like something isn t Figure 2 The eReaders reviewed in this paper LEFT Sony PRS 300 pocket MIDDLE Sony PRS 600 touch and RIGHT Amazon Kindle 2 the HCI work finished No This simplistic view comes from commercial usability if
7. a manufacturer can stay in business selling profitable products to enough customers then the business cost of further HCI research is questionable In contrast in research where we place this paper the issue is whether in principle a better user interface can be designed In future we would anticipate manufacturers us ing this research for free and thus making better products 4 DISCUSSION We will now discuss the main findings from our evaluation of three popular eReader devices shown in Figure 2 For clar ity technical details of the devices are summarized in the ap pendices Appendix A provides brief technical descriptions of all devices considered here form factor battery etc and appendices B onwards provide critical reviews of spe cific models based on the HCI issues raised above in section 2 Considering the analyses of the various devices taken in dividually from the appendices we now summarize the role of the various HCI principles for eReaders in general 4 1 Ergonomics The ergonomic design of eReaders strongly affects not only the ease of use of the functions but in some cases the com fort of the user while engaged with it e g where commonly used buttons are positioned The devices we evaluated have obviously been designed with aesthetics in mind sometimes paying little attention to the position of such buttons The Sony Touch for example has opted for the minimalist de sign approach put
8. allow the size of text to be changed under user control so the same book can be read by a user with high acuity or with vision defects needing large text Some eReaders can generate a synthesized voice so they are suitable for blind users Similarly dynamic search of the text for a word or phrase is a distinct advantage to digital books However printed books behave broadly consistently whereas digital interaction allows for a greater variety and inconsistency Furthermore a digital book may become unreadable due to changes in standards or licensing Hence the design of the hardware and of the software within the devices is crucial to the overall reading experience of the user how do eReaders support the tasks of readers This is an HCI human computer interaction question This paper presents a discussion on the basic HCI design principles of eReaders We examine the fine detail of good eReader design and reflect on guidelines appropriate for im proved interaction We illustrate our design guidelines for eReaders with a short review of three popular eReader de vices e The Sony Pocket Edition PRS 300 Small and portable Designed concisely with reading in mind e The Sony Touch Edition PRS 600 Touch screen in teraction to aid in document mark ups e The Amazon Kindle 2 Wi fi enabled with full QW ERTY keyboard for document mark up and web brows ing An Appendix provides brief technica
9. in teresting part of the page you bookmarked happened to be at the bottom you would lose its place if you changed to a different zoom setting Both the Sony s and the Kindle fall into the electronic bookmark trap first forcing bookmarks to reference pages and then not reformatting them when the arrangement of the pages change A major improvement to these devices then would include a bookmark system that as far as the user is concerned bookmarks pages but underneath actu ally references specific positions within the text The Kin dle s location system is well equipped for this system already yet fails to utilize it in a useful manner When a page is bookmarked despite the zoom level each and every part of the text within that page should be tagged as being book marked It is these tags that should then be used when the document is resized to calculate which pages should be bookmarked This would solve the problem of only one out of many zoomed pages to be bookmarked after a reformat to a higher zoom level We now turn to the reading functions listed in Section 2 6 and discuss in turn the usability concerns with each of the tools that a user may commonly use in their use of a reader 4 5 Bookmarks As we have just discussed the problem of so called page numbers within digital texts can significantly affect what we know as bookmarks This is not the only problem that eReader bookmarks ca
10. it is zoomed out again the bookmarked page is now far bigger than it originally was how then will the user know exactly which part they intended on bookmarking What the system seems to do here is take the first line on the page being bookmarked and take that as the an chor point for the bookmark ignoring any other text on the page Thus whatever page the first word of the original bookmarked page is now on is the only one which gets a bookmark on the new zoom level In contrast to the Sony Models the Kindle 2 has opted to do away with conventional page number format and replaces them with locations that correspond to specific places within the text Instead of page numbers then which are in fact rather useless in the digital world as they are not pages as such the Kindle measures the file in locations that are linked to specific positions of text within the document Each loca tion always corresponds to the same position in the text no matter now the text is being split for the display i e how big or small the font is This means that one screen can contain more than one location depending upon the zoom level and where the text breaks Despite this location system however the device still seems to have trouble with bookmarks As with the Sony s the Kindle 2 also takes the first word of the originally book marked page to be the anchor for the bookmark and ignores any text thereafter This means of course that if the
11. leading reading technologies perform Although physical books are traditional the written word is becoming more common in digital form opening up a wide range of new reading possibilities Despite the potential of digital texts people are still reluctant to read from computer screens 9 One possible reason for this preference is the eye strain caused by back lit displays This problem may be overcome by the introduction of eInk which is illuminated like print on paper by reflected light EInk technology underpins current eReader developments providing paper like reading from lightweight digital devices EReaders are both light and portable and can store a vast amount of reading material that would be bulky and heavy in print However whether these new tools are a suitable substitute to traditional paper books or whether they are best suited to slightly different purposes remains to be seen For example somebody going on a long vacation may pre fer being able to carry many books without weight penal ties On some vacations the reliance of eReaders on battery power may be a problem as well as the possibility of break ing the eReader if it is dropped In contrast conventional physical books rarely fail Traditional books are fixed and can only serve one sort of use For example a large format book which is easy to read for a user with vision defects would be found large and unwieldy for a child with good vision Most eReaders
12. on in this paper are 2 6 1 Bookmarking Bookmarking in paper books is a lightweight 5 process adding and removing physical placeholders can be so unself conscious that users are unlikely to remember doing it The equivalent tools on digital books are fraught with usability issues 3 and are consequently little used How do eReaders accomplish such a well used feature within their documents Do they provide dedicated buttons for this function Can they be easily organized and deleted How can the user view all the bookmarks in a single document 2 6 2 Annotations Reading is not always passive When engrossed in a research project many users will accompany reading with making notes for example highlighting commenting and underlin ing 8 This process is known as active reading 2 and is a common activity for those who engage in knowledge based tasks 1 Annotations then can be considered a by product of the active reading process and can be easily achieved on paper documents using any of many lightweight methods Assuming that eReader devices have catered for all reading audiences as opposed to those who simply want to read a novel without making notes or placing bookmarks how well do they incorporate active reading tools into their design 2 6 3 Page Turning Although page turning may seem trivial how the device changes pages is a rather fundamental feature that if im plemented incorrectly would seriously hinde
13. provide insights into design rationale Hence this paper was based on an exploration of principles a variant of expert heuris tic evaluation It is clear from our investigations that these devices leave a lot to be desired at least from an HCI perspective if not from a business usability perspective Many of them have focused too strongly on the aesthetics of the hardware design to ensure they have included a visually pleasing symmetrical pattern on the facia Of the three devices we inspected we feel the Kindle 2 has matched our criteria better than the others Its er gonomic design makes it easier to hold and more impor tantly turn pages It caters to users independent of their left right handedness Despite it lacking a touch screen in terface the Kindle still manages to facilitate active reading 2 by incorporating a full QWERTY keyboard into its de sign which aids with annotations and web browsing We were also impressed with its location method of page posi tioning even though in our opinion it could have been put to better use Obviously the Kindle is not a perfect example of what an eReader could be but by paying close attention to the principles and guidelines laid down here it could be largely improved as could any such device Acknowledgements Jennifer Pearson is sponsored by Microsoft Research and we gratefully acknowledge their support This research is supported by EPSRC Grant EP F041217
14. quickly you are moving from page to page 4 4 Page Numbers The slow refresh speed of eInk technology means that eInk based reading devices use pagination rather than scrolling The use of pagination is therefore a critical issue within the use of eReaders as it is the only viable navigational method Whereas paper documents inextricably connect the physical page and page numbers digital texts are free from that con straint and may be re formatted completely altering the numbering system Besides the obvious problem of refer encing physical books to their digital equivalents this sys tem gives a lot more freedom than paginated documents can offer Why then are we still using paper like ideas like book marks in our digital reading design The page numbering on the PRS 300 for example fol lows suit with most Sony eReaders The device does not seem to keep the page numbers of the original document unless they are encoded into the document itself If they are well made PDF s with specific page numbers then the pages stay constant throughout if you zoom in then the device does not reorder the pages it simply splits each page into smaller pages and names them accordingly For example if an original document was 10 pages and we zoom to 4x mag nification page 1 would now span over 4 pages all of which would be called Page 1 This method of page numbering is useful if users are co ordinating with physical copies of the same book e g
15. you buy a CD you can listen to it in your car or on your stereo and you can also rip a copy onto your computer so you can lis ten to it on your MP3 player The popularity of portable music devices like MP3 players would be no where near as high if consumers had to rebuy all their music in digital for mat as opposed to simply copying it from CDs So why are books so different Why can t physical books come with a memory device containing an electronic copy attached Or if an electronic copy is all that is required should they not be cheaper than their physical counterparts considering how cheap they are to produce These issues if left unresolved could seriously hinder the success of devices such as eReaders Although the current digital reading consumer market is at present content with simply being able to store and read hundreds of books on one device they will not ignore these issues forever 5 CONCLUSIONS This paper has explored the basic principles and issues as sociated with eReader design and backs them up with real world examples from three popular electronic reading de vices Our paper has used HCI principles to think about de sign This provides a much wider range of insights than what has become the conventional mode of HCI which is based on empirical experiments Here empirical experiments can only answer one question at a time unless they are care fully designed and then the data itself does not
16. 5 See sections B 3 C 3 and D 3 for details of how each of the three example systems deal with annotations and marks 4 7 Magnification With the exception of the Sony Touch Edition that does offer an additional zoom and pan function for within docu ments only magnification on the devices actually involves reformatting the text of the documents by increasing or de creasing the font size This is a useful feature particularly for those with vision impairments Unfortunately however the consistency of the magnification of these devices is in consistent and does not extend to the menus rendering the device relatively useless for those with poor vision unless a third party performs all the device navigation See sections B 5 C 5 and D 5 for details of how each of the three example systems deal with magnification 4 8 Improvements There are several areas of eReader design that could be sig nificantly improved by paying attention to the basic HCI principles outlined in this paper Firstly the button choice and placement on the devices facia should be carefully con sidered Commonly used functions such as next and book marks should have their own dedicated buttons and should be placed conveniently to avoid discomfort during long read ing sessions As we discuss in the Appendix the next but tons on the Kindle are perfectly positioned to coincide with the thumb positions of users holding the device in their left
17. HCI Design Principles for eReaders Jennifer Pearson George Buchanan Harold Thimbleby Future Interaction Technology Laboratory Swansea University Swansea 2 Centre for HCI Design City University London csjen h w thimbleby swan ac uk george buchanan 1 city ac uk ABSTRACT As interactive digital documents are becoming more and more commonplace we find ourselves searching for new ways to make good use of them The fast delivery and large storage capacity that digital devices offer make reading from bulky physical books seem obsolete even nonsensi cal EReaders the latest craze in digital reading follows from the introduction of eInk and promises paper like read ing capabilities with the added digital benefits But is the excitement justified Can you curl up with an eReader in the same way as you can a physical book or is the design of eReading devices hindering this process As of yet no one has taken a scientific view of current eReader technology from the systematic standpoint of ba sic HCI principles This paper discusses guidelines for good eReader design and illustrates them with examples of short comings of some of the more popular eReader devices on the market today Categories and Subject Descriptors H 5 2 User Interfaces input devices and strategies in teraction styles prototyping General Terms Design Human Factors Experimentation Keywords eReaders eInk Digi
18. Readers be it PDF or in some other format 2 1 Metaphor A metaphor is an expression that makes the comparison that something is like something else and metaphors have been promoted as efficient ways of helping users understand interactive systems Clearly an eReader is like a book and a book is therefore a metaphor for understanding eReaders As books do not have buttons the metaphor is only partial 2 2 Lightweight Marshall introduced the term lightweight 5 for almost ef fortless activities such as page turning or putting a slip of paper into a book as a place marker Clearly an important part of the success of eReaders is that sufficient lightweight reading activities in the physical world are successfully car ried over into the eReader world and remain lightweight 2 3 Ergonomics The ergonomics of the device hardware is crucial for the overall success of an eReader Making them thin and portable enough to hold in one hand is only a portion of what should be considered before manufacture For example which func tions have dedicated buttons as opposed to on screen menus And how easy is it to access these buttons from the main facia while holding the device as you would a book Are frequently needed buttons or actions e g for page turns comfortable and flexible so the user is unlikely to get strain injuries Are the buttons and actions re assignable so that the device caters for left and right handed
19. articular page or location identifier and complicates the interaction for functions that are dependent upon them e g bookmarks 4 3 Completeness As we have discussed earlier in the paper digital devices that mimic physical ones have certain expectations that are not always fulfilled consequently the sense of completeness within devices such as these are rare For example in a paper book it is possible that in certain situations you would want to add more than one bookmark to a page perhaps in the form of Post its to mark interesting parts on the top and bottom sections This function however is impossible on all three of the devices we sampled in all three readers a page could either be bookmarked or not bookmarked with no facility to add additional ones Another example of this would be flicking through a doc ument a process that is easy to accomplish on paper and useful for quick navigation Achieving this interaction on eReaders though is far more cumbersome Of the readers we reviewed the two Sony models had no tool for such interac tion The Pocket Edition PRS 300 did include dedicated number buttons for navigation to specific page numbers but neither allowed for free flicking through pages within the document In contrast the Kindle made use of its 5 way directional stick to allow for quick flicking through pages Unfortunately the screen update speed of the Kindle is slow and so it is difficult to see how
20. d in the devices hardware Fortunately the page turn gesture can be modified from its default direction via the device settings in the main menu C 5 Magnification PRS 600 The PRS 600 offers two types of document resizing The first which allows users to change to one of 5 different font sizes S M L XL and XXL where S is the original size is designed to re render and display a newly formatted ver sion of an open document page with the desired sized font This feature however can be slightly temperamental on cer tain documents For example some tables specifically those that span the entire width of a two columned document pro hibit larger font sizes for that particular page i e M L XL and XXL In addition some images particularly embedded PDF images are rendered incorrectly with the device refor matting any text within them along with the main body of the document To get around any possible magnification problems the device also offers a standard zoom function that does not reformat document pages but zooms in on specific points and facilitates panning using 4 directional ar rows on the touch screen One major downfall of the device is its lack of consis tency of the zoom function Specifically one can zoom in on documents but not on menus making the device extremely difficult to use for visually impaired users C 6 Usability PRS 600 Another problem with the PRS 600 software design is that the main o
21. here are two methods of creating a bookmarks on the PRS 600 but unfortunately both demonstrate poor user in teraction The bookmark feature is classed by the device as a Note and therefore exists within the Notes menu Op tions button amp Create Edit Notes on screen menu M on screen icon This seems like a rather long winded method of performing a simple operation like bookmarking a task that is so easily accomplished on paper To get around this the device has a short cut bookmark op tion that requires double tapping in the top right corner However this feature is not supported by any kind of visual cue and therefore can only be discovered by close reading of the full user manual not included or by random discovery To indicate to the user which pages are bookmarked the device displays a small triangular dog ear in the top right corner the display which can be seen only when a book marked page is open To view a list of all bookmarks within a document the user must navigate through a series of menus Options button amp Create Edit Notes on screen menu amp Notes on screen menu The Notes menu is an ordered list of all annotations associated with a particular document specifically it merges every user created mark up within a specific document into the same list i e bookmarks highlights handwriting and comments Each of these different annotation types are distinguished in
22. icon They can be removed by using the eraser tool also found within the Notes menu As mentioned above one a highlight has been cre ated it is automatically added to the Notes list along with other types of annotations in the document C 3 2 Handwriting PRS 600 The device makes good use of its touch screen technology by incorporating a handwriting function into its design By accessing the same Notes menu as other document annota tion tools Options button amp Create Edit Notes on screen menu 4 on screen icon users can draw directly onto the PDF surface with a stylus or finger This is a useful function that would be better accompanied by variable pen thickness sizes C 3 3 Comments PRS 600 Although you cannot simply add a text box to a page and type directly onto it there is a tool that allows you to add hidden text and handwriting to both bookmarks and high lights these are known as comments Sadly however this facility is so well hidden within the device that it can take weeks to find via random discovery This situation is exac erbated by the absence of a full manual with the Reader To add a comment a previously created bookmark or highlight must be tapped once double tapping accesses a different function to bring up a menu that allows users to choose the type of comment they wish to anchor to the note either handwriting or keyboard entry Note Comment creation can only be accomplished when the
23. ind predictions of a paperless office 11 it might be useful to consider the heart of the operation dig ital versions of traditional reading material also known as eDocuments There are many formats of electronic docu ment eg PDF BBeB EPUB most of which can be used on the eReaders we have critiqued in this paper One of the major concerns users seem to have with the eDocument paradigm is the fact that they do not actually own a physical copy of the book In previous studies we have conducted 3 10 on this topic we discovered that many users feel almost cheated by buying an electronic document coming up with comments like I d rather have the physical book because it looks good on my shelf Another issue with electronic documents is the price Of ten if you go to a book store you can buy a physical book on sale for only a few dollars or benefit from a buy one get one free offer However if you head to Amazon to buy an eBook they are more often than not full price making them more expensive than buying a physical copy The overheads of electronic books are practically nothing no printing ma terial or delivery costs are incurred yet these savings are not passed on to the consumer In fact in many cases they are trying to charge more for an electronic copy Further to this what happens if you already own a physi cal copy of a book but now want to read it on your eReader The answer is you must buy another copy When
24. l buttons on its facia They have opted to place 5 thin buttons on the device and place the icons of their functionality just above them effectively reducing their target boundary The positioning of these buttons however has not been thoroughly thought out and looks strikingly similar to the layout you would expect to find on a digital music player The page turning M buttons lt and gt for example have been placed adjacent to each other on the left hand side of the device This puts the most commonly used function i e the next page button gt in an extremely awkward position one that would not suit the thumb or finger position of a left or right handed individual C 1 2 The Display PRS 600 One of the main selling points of elnk hardware is the ab sence of a back lit display making it more comparable to ink on physical paper documents This technology not only saves battery power but also to facilitates traditional paper reading practices that are more difficult to accomplish on illuminated digital devices e g no backlighting reads better in direct sunlight Unfortunately the Touch Edition Sony Reader has undermines this feature due to a translucent top layer that is responsible for the touch screen functions of the device This additional layer is partially reflective and there fore picks up ambient light in the room making the device more difficult to read and feel less like a real book C 2 Bookmarks PRS 600 T
25. l overviews of these devices The paper ends with a summary of the findings from our critique and a conclusion 2 RELEVANT HCI ISSUES Before we discuss what we learned from the example devices we will first discuss key HCI principles associated with good eReader design From these basic principles we can assess the success of available eReaders and as appropriate sug gest areas of improvement As an initial study we have used heuristic evaluation 6 7 to uncover usability concerns that are relevant to a cross section of the most popular eReaders The evaluation is also informed by the research literature on the HCI of reading e g 1 4 Heuristic evaluation will not uncover all the usability issues with eReaders but this initial study will uncover some systemic issues that can be readily addressed and set a useful context for future work What we focus on in this paper is how current eReaders might have different interaction designs that would improve their usability We have applied the issues in this section to other devices not reported in this paper and they recurred in those other eReaders Note for clarity we refer to users in the normal way and reserve the words reader and eReader to refer to the devices themselves We do not pedantically keep referring to conventional books or physical books when the context is clear but we do also use the term book to refer to the readable electronic content of e
26. left or right to scroll through on screen menus D 2 Bookmarks Kindle Following suit with the Sony Touch s design the Kindle does not have a dedicated button for bookmarking Instead to bookmark a page the user must click Menu button then scroll down 6 items and select Add a Bookmark on screen menu To bookmark a page on the Kindle then requires 8 button clicks a rather large number for such a commonly used function On closer inspection of the user manual I discovered the Kindle does have an accelerator shortcut for this which makes bookmarking far less cumber some ALT button B button When a page has been bookmarked it renders a dog eared corner in the top right of the screen Surprisingly however pages that have not been bookmarked show a dashed dog ear in the top right corner a confusing design that has no apparent function whatsoever D 3 Annotations Kindle Despite the lack of touch screen the Kindle 2 still offers a small set of mark up tools These include highlights and notes and can be viewed in a list all grouped together with bookmarks too from the main menu Menu button My Notes and Marks on screen menu D 3 1 Highlights Kindle To create a highlight the user must first be in annotation mode This is accomplished either by moving the caret into the document area simply by using the directional stick or by selecting Menu button Add a Note or Highlight on screen menu Next
27. n above the Next Page button on the left of the device D 5 Magnification Kindle The Kindle 2 offers 6 font size variations accessible through the size button on the device It also allows users to change the margin size of the text using the strangely worded Words Per Line feature of fewest fewer default Follow ing suit with the Sony s inconsistency the Kindle 2 also sets its menus at a fixed size Unlike the Sony s however the Kindle gives no visual audio or tactile feedback when at tempting to change the font size within a menu
28. n suffer from however Despite bookmarks being a well used function in physical books many eReader designs do not incorporate a dedicated bookmark button into their facia design Instead forcing in ternal menus to be used Furthermore in all three of the devices we evaluated viewing a list of bookmarks cannot be done whilst reading a book rather the user must navigate out of the document and into a menu to retrieve the book mark list See sections B 2 C 2 and D 2 for details of how each of the three example systems deal with bookmarks 4 6 Annotation Active reading 2 is a very common activity and describes not only reading but also the process of thinking and marking up documents by annotating highlighting extracting infor mation etc On paper these actions are easily performed and can be considered to be lightweight 5 as the user actions can be so unselfconscious they are not apt to remember them later It is clear then that in order to facilitate active reading in the same way as paper adequate mark up tools need to be included into eReader designs Sadly however of the three we sampled only one Sony Touch PRS 600 had the touch screen facility to create hand drawn notes while the Kindle made do with highlights and hidden typed notes The Sony Pocket PRS 300 had no annotation facility whatsoever making the device useless for active reading and rendering it somewhat incomplete see section 2
29. ng if they happen to be on the page you are reading Otherwise you must exit reading to view a list of all bookmarks within a book In a physical book you can insert as many bookmarks as you please In a sense the bookmark process generalizes from a single example to being complete In eReaders there are often restrictions on bookmarks Or an eReader may make the last page read an automatic but special book mark This assumes that a user is only reading a book in one place at a time as now the user has to distinguish be tween the last page they were reading and other pages they have bookmarked For example the page they are read ing changes but bookmarks do not So looking up another bookmark loses the current reading location 2 6 Reading Functions EInk displays are slow to update and combined with the simple microprocessors of readers the number of steps a user takes to perform any task is more critical than on PCs which have faster processors and faster displays Thus the features of a reader should be easily accessible with minimal menu navigation and short sequences of button presses Some eReaders have touch screens but these are slow to respond and tricky to manipulate because of screen update delays Also requiring a reader to hold a pen is out of metaphor so more commonly used features should be or be complimented by dedicated physical buttons on the device s facia The features we will focus
30. or right hand Further to button positioning device manufacturers must also ensure that the buttons and functions within the device are consistent Users should not have to wonder what ac tion the button or function will perform every time they use it it should consistently perform the same action Design ers should also strive for completeness ensuring that tools and actions within the device mimic the actions that can be performed on paper unless this is demonstrably inefficient in a digital interaction In doing so the functions within the device will become more lightweight and improve the interaction between the user and device There are certain principle conventions that should be fol lowed in order to ensure good user interaction with the tools For example due to the technology used the screen update speed is slow on these devices Therefore an obvious rem edy to this problem would be to reduce the number of screen updates Sadly in the examples we chose there were several instances where the devices failed to do this In the Sony Touch for example the main menu screen is not one big list you must scroll down after several entries to view the second half despite there being ample room for the entire list on the screen at one time This causes unnecessary button clicks and page refreshes that could easily be avoided if the menu was contained in a single list 4 9 eDocuments Bearing in m
31. people equally well Are the displays legible particularly for users with vision deficits 2 4 Consistency When designing a digital interface thought should be given to the consistency of buttons and functions within it to en sure smooth user interaction with the tools Poor consis tency leads to bad interaction and possibly low rates of use of the tools included in the system It is vital therefore that dedicated buttons should be well labeled and always perform the same function such as back buttons should always go back to the previous screen zoom buttons should always zoom and so on Any variation in the functionality of these buttons could lead to confusion from the user 2 5 Completeness When reading a physical copy of a book there are certain actions and affordances that are impossible to re create on digital equivalents For example folding ripping and flick ing are just a sample of actions that are difficult to perform on an electronic reader device Although there are certain things that cannot be replicated in this way there are other actions and functions that could easily be incorporated into the software but are left out leaving us with a strong sense of incompleteness For example in physical books when we in sert a bookmark we can usually see it along with any others entered sticking out of the book while we are still reading In digital eReaders however bookmarked pages can only be seen while readi
32. ptions menu unnecessarily spans two pages i e you must click to change to the second half of the menu despite the fact there is sufficient room on the screen to place all items within it on one page This adds a needless extra click to an already slow menu system and goes against the minimal refresh screen clicks principle we laid down earlier D AMAZON KINDLE 2 D 1 Ergonomics Kindle There are several notable properties of the ergonomic design of the Amazon Kindle 2 Firstly it is useful to note that its sleek thin shape and lightweight properties make it easy to hold with one hand This coupled with the clever position ing of the two Next Page buttons makes linear reading easy to accomplish with the thumb of the right or left hand Unlike the Sony s the Kindle 2 does not have buttons with icons above them they use descriptive words on the buttons themselves to state their purpose The size of the Kindle s buttons are also a lot larger than the Sony Touch making them much easier to press Unfortunately the thought that went into the design and position of the next buttons was not given to the function of the 5 way directional stick This mini joystick does not promote direct manipulation 12 as it is hard to control the on screen cursor using continuous movement on a slow re fresh screen speed An alternative but tedious and difficult way of navigating with the 5 way is to continuously flick up down
33. r the reading process When using the device for reading only the most commonly used function will be the Next button as it is used every time a user wants to change the page The po sitioning of this next button then will be critical to the ease in which these devices are used Placing the button in an awkward spot will cause un natural hand movements which could be uncomfortable for users and possibly lead to Repetitive Strain Injury RSI 2 6 4 Magnification One feature that cannot be accomplished on paper with out additional tools is magnification This magnification or zoom function should ideally allow visually impaired users to read from the same device and read the same books as those with perfect vision One side problem with this fea ture is how the devices restructure the document to fit on the screen when the zoom level has altered Fundamentally this is not an issue the device will simply put more or less words on the screen than the original document presented Looking deeper here however we realize that this now alters what we know as page numbers for example if the original document had 100 pages and the we zoom in to double mag nification it will now contain around 200 pages What was page 1 will now be spread over page 1 and page 2 Clearly this will now effect features such as bookmarks that rely on a page referencing method of page numbers for relocation How devices deal with this issue can effect
34. reading paper and on line documents In CHI 797 Proceedings of the SIGCHI conference on Human factors in computing systems pages 335 342 New York NY USA 1997 ACM 10 Jennifer Pearson George Buchanan and Harold W Thimbleby Improving annotations in digital documents In ECDL volume 5714 of LNCS pages 429 432 Springer Berlin 2009 11 Abigail J Sellen and Richard H R Harper The Myth of the Paperless Office MIT Press Cambridge MA USA 2003 12 Ben Shneiderman and Catherine Plaisant Designing the User Interface Strategies for Effective Human Computer Interaction 4th Edition Pearson Addison Wesley 2004 APPENDICES A TECHNICAL SPECIFICATIONS A 1 Sony Reader Pocket PRS 300 Display 5 inch eInk 8 Level Greyscale Physical Buttons Devices Power Switch f Home D Return 4 Previous Page jp Next Page amp Next Page amp Move Cursor W Previous Page amp Move Cursor lt blank gt Enter AA Bookmark Q Size Number But tons X 10 Resolution 800 x 600 pixels Size 159 x 108 x 10mm Weight 220g Wireless Connectivity None Memory 512MB Internal Expansion None Power Sealed Internal Lithium Ion Battery around 6800 page turns on one charge eBook Formats Supported Adobe PDF TXT RTF Mi crosoft Word BBeB EPUB RRP 199 99 A 2 Sony Reader Touch PRS 600 Display 6 inch Touch Screen eInk 8 Level Greyscale Physical Buttons Devices Power Switch Vol Vol
35. t an easy menu to locate B 3 Annotations PRS 300 There are no facilities to create annotations notes or high lights on the PRS 300 The manual confirms that these functions can be used within the eBook library on a paired PC but the device would be unable to open or use them B 4 Page Turning PRS 300 Page turning M on the PRS 300 is accomplished by using the circular directional pad below the main screen Y ana P for Next and and 4 for Previous Assuming the device will be held in one hand for reading the position of this directional pad is just about in range of the thumb of a right handed individual holding it on the bottom half Left handed users however may find it a strain to stretch their thumb to the next button after several page turns B 5 Magnification PRS 300 Although there does not seem to be a zoom mode within the device there is a size CB button that toggles between three different font sizes This button however does not change the font size of the home or options menus which are a constant size throughout Despite the device giving visual feedback to state the zoom is not possible on the menu it is an inconsistency that could hinder visually impaired users B 6 Inconsistencies PRS 300 PRS 300 The device offers a specific menu for the options of a docu ment that provides functions for utilities such as document info table of contents etc As there is no dedicated op tions button the
36. tal Documents Evaluation 1 INTRODUCTION Mass printing in the fifteenth century established what we know as the modern book with its physical format of covers and paper pages with now standard features such as page numbers Reading is a complex human activity that has evolved and co evolved with technology over thousands of years Reading has become a core part of human culture To Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page To copy otherwise to republish to post on servers or to redistribute to lists requires prior specific permission and or a fee Copyright 200X ACM X XXXXX XX X XX XX 10 00 the extent that new technologies conform to our cognitive perceptual and cultural constraints they will succeed They may also free us from unnecessary constraints and open new opportunities introducing new ideas that will transform the culture of reading As new technologies explore and push the boundaries sometimes they will not be quite right and sometimes they will not go far enough and some inno vations will perhaps be very good for special purposes even new purposes not previously envisaged but not for reading in general What are the tradeoffs for ease of use and how do the
37. the list by a different icon and the lists can be ordered by page number type or comment One major flaw of this type of list based interaction however is the lack of overview of where bookmarks exist in the document In paper books relative placement of bookmarks can be easily seen as well as how much has been read left to read etc C 3 Annotations PRS 600 The Touch edition Sony reader contains a within document notes system that can be accessed via a series of menus specifically Options button Create Edit Notes on screen menu This then scales the document view very slightly to provide a small clear space for the notes menu that sits along the top of the viewing window Within this menu users can either add highlights to the book con tent access free hand mode using the stylus delete using the eraser add remove bookmarks and access the notes list In reality notes and bookmarks are different things to the majority of users It seems odd never mind long winded then that Sony has decided to combine them into a sin gle entity and effectively bury one within the other The following subsections describe the functions that Sony has grouped under the common name Notes C 3 1 Highlights PRS 600 Existing text within the open document can be highlighted by first accessing the highlighter function hidden within the Notes menu Options button Create Edit Notes on screen menu amp on screen
38. ting only 5 thin strip buttons on its facia Unfortunately however the most commonly used button on the device i e the next button has been placed second from the left in the row of buttons an extremely awkward position for users to press while holding the device with one hand regardless of which hand they use A similar position has been used in the Sony Pocket design In contrast to this the Kindle has thought harder about their choice of button placement choosing to place their two next buttons on the sides of the device perfect for clicking by the thumb of an individual holding the device in either hand See sections B 1 C 1 and D 1 for details of how each of the three example systems deal with bookmarks 4 2 Consistency The consistency of the software within devices is certainly something that can be improved upon We have touched upon several cases within our evaluation of the three ex amples in which the consistency of certain functions do not follow platform conventions Examples of these range from button inconsistencies as in the case of the Sony Pocket s Return button see section B 6 to function inconsis tencies like all the zoom function s inability to increase the size of menus Even the page numbers within several of these devices seem to be inconsistent see 4 4 recalculating page numbers at every zoom level undermines the traditional con nection between a word or sentence and a p
39. tional inputs used for next and previous functions to be placed in the middle of the device below the display screen This is not the best place to position a feature as commonly used as the next button as it can only be easily accessed by the thumb of a left or right hand if the device is being held by the bottom edge For quick navigation to menu options or page numbers the PRS 300 has 10 numeric buttons down its right hand side Unusually however these 10 buttons have been com bined into 5 sets of 2 that resemble on off switches B 2 Bookmarks PRS 300 The PRS 300 model has a dedicated bookmark button M Clicking on this button within a document will cause a dog ear to appear in the top right hand corner indicating that the page has been bookmarked Clicking the button on an already bookmarked page deletes the current book mark This dog ear visualization however does not give an overview of which pages are bookmarked as only the currently open page can be seen on screen To view a list of all bookmarked pages the user has one of two choices The first option that can be accessed through the main menu ft button MA All Bookmarks on screen menu lists all bookmarks in all books stored on the device in one big list The second and possibly more useful option is to view only the bookmarks within a single book This function however must be accessed from the Options menu for that particular document which is no
40. user has one of two choices to access the menu Firstly selecting a document from a list either by title author or date on the device will automatically show the options for that document before allowing reading to commence The second option which can be accessed from within a currently open book requires the user to click the B 6 1 Document Options return button to access the options screen Obviously this button is not labeled as an options function and does not always behave in the way you would expect i e re turn the user to the screen they were on prior to the current screen Under normal circumstances e g if you are in the Date and Time menu clicking return m takes you back to the previous screen e g the main Settings page How ever in some instances it behaves as a menu button e g if you were on page 5 and then use the number buttons to take you to page 55 the return button should take you back to page 5 Similarly if you were on the home screen then continue reading the return button should take you back to the home screen again However in these two examples the return button actually takes the user to the options screen for a book instead of returning them to the previous screen C SONY READER TOUCH PRS 600 C 1 Ergonomics PRS 600 C l 1 Buttons PRS 600 With the added benefit of a touch screen Sony has been able to make the screen the main focus of the device by limiting the number of physica
41. users experience with reading as well as the features that are bound by it gall Amazon Kindle Sony Touch a Sony Pocket 1 Star 2Stars 3Stars 4 Stars 5 Stars Figure 1 Graph showing the percentage of cus tomer reviews giving each star level source Ama zon com Alternatively a page can be zoomed to look larger but not be reflowed This retains the original page format and page numbering but now requires the user to pan across pages to read them often this is the only option available for pictures It happens that the key design decision is typically forced by the format of the document being read some documents cannot be reflowed so most devices must provide both mechanisms 2 7 Summary In this section we identified some salient HCI principles that can frame a critique of current reader design This was followed by a brief summary of some of the concerns that may arise in commonplace features of the reader software We will now take a short glance at some readily available data on the user satisfaction reported with these devices 3 USER DATA Before we go into detail of our evaluation of our chosen ex ample devices is useful to consider consumer thoughts on the products We looked to the Customer Review sec tion of Amazon com for the 5 star Likert ratings given by buyers of the devices gathered from the US Amazon site www Amazon com Figure 1 shows the data we gathered
Download Pdf Manuals
Related Search