Home
        The Computer Science of Everyday Things
         Contents
1.  is optimal  precisely because it has no other purpose than  to be efficient  However all Nokia functions have  a shortcut code  which  as it is so named  is  presumably to provide a brief alternative  Again   the Huffman tree lends itself to an improved code   moreover  since Nokia do not use their shortcut  codes efficiently  we can provide both Nokia   s  original and a non overlapping code  this gains  the best of both worlds  whatever the    best    of the  Nokia world is supposed to be   Nokia   s shortcuts  have an average access cost of 3 64 keystrokes  the  combined approach has an average cost of 2 87  keystrokes    Such improvements can be made by applying  purely routine computer science to user interface  design  The next section illustrates what we can  do when more imagination is brought into play     5 2 More efficiency   based on a novel design    Full details of the example discussed in this section  can be found in  12     Most keys on the Nokia have letters on them   and just as they are used for entering names  e g    SMS messages  they could be used for searching for  functions  Reference to any algorithms book will  find numerous ways of making good use of keys for  searching  Here  we explore hash codes    Given a function name  such as Call Waiting  we find its hash code by using the Nokia   s stan   dard keypad  Thus  the first two letters of Call  Waiting     C    and    A     are on the key  and  the third and fourth letters are on  5   thus the  h
2.  routinely design computer pro   grams to be efficient and correct  We need to see  the user as at least as important as a computer   and the user interface as a program for the user   User interface design then becomes a matter of     programming the user     Many problems in user       4Two examples   uniformly from function to function  the key behaves  differently  unnecessarily  in different functions  In Phone  book  while entering a number  deletes digits  whereas  in Clock  again while entering a number  goes up a level  in the menu hierarchy  In Clock  then  it is not possible to  correct numbers  all digits have to be entered correctly first  time     short cut codes are not displayed    interface problems then come down to failing to  provide the user with a correct and efficient inter   face    That the user interface also needs to be com   putable is obvious     but this is an idea that seems  to be ignored in almost all interfaces  Even such  simple tasks as determining a sequence of moving  a cursor from one screen position to another in  Microsoft Word are undecidable  24     Once user interfaces are seen in this light  then  many ideas from computer science can be applied   This paper showed how familiar algorithms can be  applied to user interfaces  elsewhere we have ex   plored applying ideas like declarative programming  to user interfaces  25     Given that many user interface problems can  be attributed to bad computer science  much of  conventional empi
3.  their usability is  often low  Usability problems are not just irritat   ing  but are costly and stressful  and hazardous in  many contexts  whether the users concerned are  airline pilots  office workers using photocopiers  or  home users struggling with video recorders    It is popular to argue that everyday things are  harder to use than they need be  and that improve   ments must be sought in human factors  that is in  the study of users and their tasks  16   and in the  study of ecologies and context of use  15   etc  The    Proceedings of the Australasian User Interface  Conference  Bond University  Gold Coast  Aus   tralia  January 29   February 1  2001     lack of    user centred design    is the conventional  reason for the failure of almost all programmed  systems  10     Indeed  as industry can evidently sell to con   sumers without improving usability  business prac   tice itself creates a source of usability problems that  is again of non technical origin    Usability  then  as a field has become dominated  by marketing  social  empirical and statistical  methods alien to computer scientists  The broad  emphasis on human factors encourages computer  scientists  further  to dismiss usability as irrelevant  to their focus  Indeed  in simple terms computer  science appears remarkably successful  cf  Moore   s  Law    2  below   superficially confirming that  usability problems must be solved elsewhere    But successful interaction clearly requires un   derstanding 
4.  to be asserted    A computer scientist would wonder why the al   phabetic labels are not more sensibly organised  Of  the 12 available keys  only 9 are used  and the allo   cation of letters to keys does not reflect frequency  of use  There are international standards of letter  allocation  but as phones differ  they cannot be an  important consideration     and the silliness of the  international standards merely begs the question  why the standards bodies do not ask computer sci   ence questions either     5 3 Other benefits    Both examples above rely on a specification of the  Nokia device  Once the device is specified  it is easy  to devise and analyse many improvements    The inconsistencies in the actual user interface  of the Nokia suggest that Nokia do not use a pro   fessional computer science approach to user inter   face implementation  That is  given a more or   less formal approach inconsistencies are hard to  implement  whereas in an informal approach incon   sistencies are easy if not an expected consequence   It is arguable that consistency makes devices eas   ier to use  outside of games   certainly consistency  makes user manuals easier to write and shorter   In fact user manuals can be generated  at least in  part  automatically  30      then  not only can user  manuals be used concurrently early in the design  process  but they can have guarantees  if desired   that they are complete and correct     6 Programming the user interface    Computer scientists
5.  to come on in interesting patterns  this is  presumably a test mode  or perhaps a salesroom    eye catching feature   but the manual does not de   scribe it    The JVC HRD580 EK video recorder has the  opposite problem  21   It has features that cannot  be used with the user manual  The manual has  textual descriptions of complex operations  such as  tuning in the VCR to UHF channels   Reading  these sections is time consuming  yet the corre   sponding features on the VCR have short time   outs  So by reading the manual while trying to per   form the operations described ensures the VCR will  time out  then the user will be reading the wrong  part of the manual to explain what the VCR is  doing  What the user is reading becomes counter   productive   One solution to the system manual  synchronisation problem is for the device itself to  present instructions to the user  since it    knows     what state it is in  it can always present the appro   priate sections of the manual  1      There is not space here to analyse these assorted  criticisms further  Each unusable system is unus   able in its own way  the faults are too varied to  easily bring into clear comparison   See  26  for a  systematic analysis of a long list of usability prob   lems with a single product   One reason why faults  persist is that nobody has conceptual leverage on  the problems  manufacturers employ production  processes that result in incoherent usability issues   nobody understands the results     
6. 5 Computer science applied    We need conceptual leverage not on problems  in   teresting as they may be  but on solutions  User  interfaces might be improved by better computer  science  but can improvements be achieved system   atically  In this section we outline a computer  science appraisal of the Nokia 5110 mobile phone   s  user interface  We provide evidence that applying  elementary computer science will improve user in   terfaces     5 1 More efficiency   based on Nokia   s design    Full details of the example discussed in this section  can be found in  28     The Nokia 5110 handset organises most of its  functions in a tree  which is presented to the user as  a menu  Clearly the design assumption is the user  will search the tree to find functions to activate   Computer scientists are familiar with search  and  indeed with search trees    A Huffman tree is the most efficient way to  organise a search tree  given that there is a fixed set  of keys   28   We could simply determine the aver   age cost per function and compare it with Nokia   s  design  but the so called cost of knowledge graph   itself introduced by psychologists to evaluate user  interfaces  shows increasing efficiency the longer    it is used for the Huffman over Nokia   s original  design  after 500 key presses  a user of the Huffman  design is twice as efficient       It can be pointed out that Nokia   s design  may have virtues other than trying to minimise  keystrokes  certainly  a Huffman tree
7. In   teraction  Volume IV  pages 58 86  Ablex   1993     
8. The Computer Science of Everyday Things    Harold Thimbleby    Computing Science  Middlesex University  Bounds Green Road  London  N11 2NQ    http   www cs mdx ac uk harold    Abstract    Technology is fashionable  wonderful and getting  better  Moore   s Law predicts substantial  sustained  improvement  Yet the usability of    everyday things     is low  video recorders being a notorious example    It seems to follow that improvements must be  sought in areas outside technology  such as human  factors  But a premise is wrong  in fact  the  technology     the embedded computer science     is  appalling    Obsolescence  a symptom of Moore   s Law  hides  flawed design  poor products are replaced rather  than fixed  The poor quality of the computer science  of everyday things is eclipsed by the hope for fixing  today   s problems with tomorrow   s consumption    This paper reviews Moore   s Law and the  usability of everyday things  it shows that  professional computer science can improve  usability with ease  Improvement will be essential  when ethical and environmental issues become  as  they will  unavoidable design criteria     Keywords Moore   s Law  User interfaces   Mobile phones  Programming user interfaces   Environment     1 Introduction    Almost all modern everyday things     from mo   bile phones  cameras  central heating controllers   calculators  to wristwatches and airplanes     work  because of embedded computer programs  While  most are technically impressive 
9. akes the memory equal the display if it already contains  zero  which it may not  To make the memory zero  you have  to press twice  but pressing it the first time sets the  display to the memory  which loses the number you wanted  to store  The solution is not easy  and there is no solution  that avoids the risk of calculator overflow   If you need paper  to work it out  why not use the paper for the memory       the differences or the curious features shared by  both calculators  Neither comes with user manuals  or other information that warn of or reveal any  problems    Calculators are impressive  especially if a  prepared salesman shows you them going through  some calculations  It is possible to demonstrate  the memory in action  apparently working  Only  some critical thought would determine that it is a  very peculiar feature     4 1 More examples    A cynical reader might take the Casio calculators  as specially selected  but see  29    This section  gives a brief selection of problems to illustrate the  diversity of problems  All should have been easily  detectable  and all are easily avoidable in principle   See  9  18  20  for examples of usability problems  with safety critical devices  where there is an obli   gation improve usability    The Sony television KV M1421U has a remote  control  the RM 694  Not only are their colour   number and layout of buttons different  but the  user interfaces are unrelated  they work in com   pletely different ways     which is 
10. and P  Duguid  The Social Life of  Information  Harvard Business School Press   2000      3  C  M  Christensen  The Innovator   s Dilemma   When Technologies Cause Great Firms to Fail   Harvard Business School Press  1997      4  Commission of the European Communities   Proposal for a Directive of the European  Parliament and of the Council on Waste  Electrical and Electronic Equipment  Volume  COM 2000  347 provisional  2000      5  de Longhi  Pinguino Electronic    578909 02 97  1997     eco      6     Ss     10      11      12      13      14     15     16    17          18     19     P  Duquenoy and H  Thimbleby  Justice and  design  In M  A  Sasse and C  Johnson   editors   IFIP Conference Interact   99  1999     C  A  R  Hoare  Communicating Sequential  Processes  Prentice Hall International Series  in Computer Science  1985     M  Jones  G  Marsden and H  Thimbleby   Empowering users with usability certificates   In S  Turner and P  Turner  editors   BCS  Conference HCI 2000  Volume II  pages 37   38  2000     P  B  Ladkin  Analysis of a technical descrip   tion of the Airbus A320 braking system  High  Integrity Systems  Volume 4  Number 1  pages  331 349  1995     T  Landauer  The Trouble with Computers   MIT Press  1995     G  Marsden  P  Duquenoy and H  Thim   bleby  Ethics and consumer electronics  In  A  D   Atri  A  Marturano  S  Rogerson and  T  W  Bynum  editors   Fourth International  Conference on the Social and Ethical Impacts  of Information and Communicati
11. ash code of Call Waiting starts 2255    As a  user enters a hash code  the new design displays  in  a scrolling list  all functions that match the hash  code  To disambiguate the code  the user can either  use up and down arrows  as in the conventional  Nokia design   or enter more digits  If the user  presses the key  the last digit is deleted  and  the hash code looked up again and the display is  refreshed    Marsden built a simulation of the original Nokia  and the new design  and compared their use   Both  designs used the same screen image     Of the experimental scenarios completed on  the new design  users achieved an average 9 54       3Due to bugs in Microsoft Word it turned out  for  unfathomable reasons  to be impossible to place a graph  in the original Word version of this paper to illustrate  this point  Again  this is a computer science problem   bad programming in a consumer software package   This  published version was formatted using TFX      key presses  in comparison to the Nokia design  where 16 52 key presses were required  This was a  strongly significant result  p  lt  0 001  with users  requiring approximately 7 fewer key presses  on  average  to access the functions on the hash code  design    The overall mean times  not key press counts   for the hash code phone was 33 42 seconds  com   pared to 42 02 seconds for the original  The im   provement  though  was not consistent enough from  person to person for a statistically significant dif   ference
12. bility benefits     4 Buggy everyday things    There are many gadgets representing bad computer  science  but most are so bad that they are impossi   ble to describe clearly  Perhaps the best example   because it is in principle straight forward  pervasive  yet inexcusably bad  is the handheld calculator    Hand held calculators are a mature technology   Calculators have well defined requirements  of ac   curacy and performance and so on  There have  been many generations of calculator designs  and  the manufacturers have had many opportunities to  step their production to fix problems  The only  limitation on calculator quality is manufacturer in   clination  Research into calculator user interfaces  has a twenty year background  25  29     Casio is the leading manufacturer of hand held  calculators  Two of their basic models are the SL   300LC and the MC 100  which look very similar   The SL 300LC is shown in Figure 2     e Pressing the 9 keys 100 17 5   e g   to cal   culate 17 5  tax on   100  on the SL 300LC  gets 117 50  The MC 100  with exactly the  same keystrokes  gets 121 21212  Two similar   looking  confusable  calculators  made by the  same manufacturer  do different things     Since the     key is unpredictable  from calculator  to calculator  and even in different contexts in the  same calculator   users soon avoid using it  thereby       Figure 2  The Casio SL 300LC  Note the highly   visible self advertising memory keys in the middle of  the top row of button
13. for   mation Technology Standards  Butterworth   Heinemann  1993     H  Thimbleby  The frustrations of a push   button world  In Encyclopedia Britannica  Yearbook of Science and the Future  pages  202 219  Encyclop  dia Britannica Inc   1993     H  Thimbleby  Treat people like computers   In A  Edwards  editor   Extraordinary People  and Human Computer Interaction  pages 283   295  Cambridge University Press  1995     H  Thimbleby  A new calculator and why it  is necessary  Computer Journal  Volume 38   Number 6  pages 418 433  1996     H  Thimbleby  Design for a fax  Personal  Technologies  Volume 1  Number 2  pages 101   117  1997     H  Thimbleby  Visualising the potential of  interactive systems  In Tenth IEEE Inter   national Conference on Image Analysis and    Processing  ICIAP   99  pages 670 677  1999     H  Thimbleby  Analysis and simulation of user  interfaces  In S  McDonald  Y  Waern and  G  Cockton  editors   BCS Conference HCI  2000  Volume XIV  pages 221 237  2000     H  Thimbleby  Calculators are needlessly bad   International Journal of Human Computer  Studies  Volume 52  Number 6  pages 1031   1069  2000     H  Thimbleby and P  B  Ladkin  A proper  explanation when you need one  In M  A  R   Kirby  A  J  Dix and J  E  Finlay  editors    BCS Conference HCI   95  Volume X  pages  107 118  1995     I  H  Witten and H  Thimbleby  User mod   elling as machine identification  New methods  for HCI  In H  R  Hartson and D  Hix   editors   Advances in Human Computer 
14. ho do not understand com   puter science     6  Because of the market demand  practising  computer scientists  even ones not working in  user interfaces  are trained to a low standard     7  Manufacturers put their best programmers  onto tasks that must work correctly  In a  mobile phone  say  these will include battery  management and radio communications   Customers will notice quickly if the phone  does not work at all  but most customers will  blame themselves when they find the user  interface tricky  Thus manufacturers put their  worst programmers onto user interfaces     8  Computer scientists who develop a product are  familiar with its user interface  Their knowl   edge and experience will make it  and in gen   eral  all devices  much easier for them to use   Therefore computer scientists discount the ev   idence of usability studies     9  Because it stimulates consumption  users are  encouraged to blame themselves rather than  the products they buy  Researchers  too  are  consumers with similar stakes in the usability  of the everyday things they own  It is unusual  and depressing to question the design of prod   ucts you have chosen to buy  especially if it  brings into question the standards of your own  profession     7 Sustainability and ethics    With all the reasons against improving user inter   faces  where will improvement come from    Moore   s Law apparently underwrites manufac   turers    confidence in faster and more powerful pro   cessors for their pr
15. ld not expect  the law to apply to the USA and to Japan  It is  too coincidental  2   More likely  then  the law is an  observation about capitalism rather than a surpris   ingly uniform rate of technological breakthrough    Stimulating consumer demand by enhancing  performance expections is a very creative activity   and requires insight into market behaviour   Coincidentally most creative professionals are not  technical  the people employed by industry to    enhance products do not understand or realise the  limitations of the embedded computer science    4     User interfaces are complex and problems in  usability rarely show up immediately  By the time  a user becomes dissatisfied with a product  Moore   s  Law    ensures    new products are available  There is  little need for manufacturers to make old products  better if  by the time their limitations are discov   ered  users prefer a new product to a fixed older  product  Thus Moore   s Law encourages loss in de   sign quality  manufacturers can assume customers  prefer to upgrade to a faster or more featured prod   uct  In other words  provided the product lasts  long enough     a few months     it need have no  long term usability features  this makes it easier to  design  and to get away with poor quality design  methods  Even if a product fails completely  con   sumers would probably rather pay to upgrade than  wait for it to be repaired    Moore   s Law encourages fatalistic computer sci   ence  Your current prog
16. obvious  e g   from  their statecharts  27   One concludes either Sony  never specified the user interface  or did not worry  about the gratuitous differences    The Panasonic Genius microwave cooker has a  clock  The user interface allows the clock to be  set to any number  0000 to 9999  Despite the  impression that it may be a 24 hour clock  it only  runs when set to a valid 12 hour time  Nothing  gives any indication why the clock will not run  when it is set to a time such as 2230  23  31     The Nokia 5110 mobile phone has a menu   driven user interface providing access to    all     the phone   s functions  see   5   However some  functions  such as keypad lock  are not in the  menu structure  Thus a user without the phone   s  manual may be unable to find such functions   There is no reason why all functions should not be  in the main menu    The De Longhi Pinguino  5  air conditioner can   not be used without the user manual  There are  delays between pressing buttons and things hap   pening  so a user might press a button twice  to  really do something  but this resets the mode of the  device before it has even started  When the device  is running  its    on    light is off  A front panel light  labelled    memo control    is described in the manual  as meaning that there is a three minute wait  if it  is flashing  or that the air conditioner is in cooling  mode  if it is on continuously   It is possible to  press the buttons on the front panel to get all the  lights
17. oducts  Manufacturers can an   ticipate adding features to systems without seri   ously impacting their performance  Equally  they  can anticipate users preferring to replace their cur   rent products with new ones     users prefer to  replace rather than have inferior devices fixed  By  the time users find problems with user interfaces   the market will have moved on and the users can  be promised their problems will be fixed by new  features    The result of this is continual replacement  In   dustry benefits  and consumers willingly become  fashion followers  Mobile phones  for example  are  not sold on their effectiveness for communication  but on their status value    There is a serious consequence  what happens  to all the obsolete devices  In the UK  900 000  tonnes of mostly toxic waste electronics is buried  annually in land fill sites  This is not sustainable   Some notable efforts have been made to recycle   for instance shipping computers to the third world      but this is merely shifting the toxic waste  from one back yard to another  Manufactuers  may be persuaded by legislation  4  to use more  environmentally friendly materials and processes   but this does not address the core problem     The computer science of everyday things has  been so bad for so long that we all take it for  granted  It is time to design user interfaces so they  are robust  so they can be upgraded  These are  conventional computer science issues    There is not space to cover further e
18. of both humans and computers  The  computer science of everyday things has ignored  usability  we give examples below   and its stan   dards are woeful  This factor of low usability is  ignored not just by designers and usability workers  but also by computer scientists  There is a serious  educational problem  both at the level of educating  technologists to program user interfaces better  and  at the level of encouraging consumers to demand  higher standards of lifetime usability  which could  easily be provided     2 The    impressive    Moore   s Law    Technology is getting better  and has increasing  performance over time  We can represent this by  the graph  see Figure 1  using a curve of posi   tive slope  it is not necessary to worry about the  precise shape of the line  or exactly what    per   formance    is measuring   For any particular task  the user has  some minimal level of performance p  will be required  From the graph it is clear that  there is a crossover when the lines intersect  at  performance   p and time   t  Before time t   technology is delivering inadequate performance   after t  technology can deliver more than adequate  performance    Norman  17   crediting Christensen  3   makes  the following argument  Before the crossover time    Performance       Figure 1  Perfomance of technology increases with  time  curved graph   After time t perfomance exceeds  the crossover value p     t  manufacturers need only promise technical ca   pability  which i
19. on Technolo   gies  Ethicomp   99  1999     G  Marsden  H  W  Thimbleby  M  Jones  and P  Gillary  Successful user interface de   sign from efficient computer algorithms  In  ACM Conference CHI   2000  Extended Ab   stracts  pages 181 182  2000     G  Moore   Moore on Moore   s Law   Corporation Distinguished Lecture   http   www uvc com     Nanometres and Gigabucks      University Video  1996     R  Nader  Unsafe at Any Speed  Pocket Books   1965     B  A  Nardi and V  L  O   Day  Information  Ecologies  Using Technology with Heart  MIT  Press  1999     D  A  Norman  The Psychology of Everyday  Things  Basic Books  1988     D  A  Norman  The Invisible Computer  MIT  Press  1998     E  Palmer     Oops  it didn   t arm     a case study  of two automation surprises  In R  S  Jensen  and L  A  Rakovan  editors   Eighth Inter   national Symposium on Aviation Psychology   pages 227   232  1995     B  Reeves and C  Nass  The Media Equation   Cambridge University Press  1996      20      21     22     23     24     25     26     27      28     29     30     31     N  B  Sarter and D  D  Woods  How in the  world did we ever get into that mode  mode  error and awareness in supervisory control   Human Factors  Volume 37  Number 1  pages  5 19  1995     H  Thimbleby  Can anyone work the video   New Scientist  Volume 129  Number 1757   pages 48 51  1991     H  Thimbleby  Computer literacy and usabil   ity standards  In C  D  Evans  B  L  Meek and  R  S  Walker  editors   User Needs in In
20. pensively  travel on land at up to  1000km h  though most transport is undertaken by lorries  travelling around 110km h  much less if we average over the  journey rather than take maximum legal speeds      lems  they should learn how to use things properly   Our society is certainly a complex place  and people  do need to be technology literate because that is  how the world is  But this practical response can  be used as an excuse to make systems over complex  because users will take it upon themselves to learn  how to use them  Indeed  manufacturers often com   modify their learning material  thus making further  profit by providing systems that require additional  training    The prioritisation of safety in car design came  about because of consumer action  stimulated by  Nader   s damning expos  s  Currently consumers of  everyday things are unaware usability problems  are avoidable  People working in industry are  themselves consumers of everyday things      they may even be employed because of their  eager following of technological fashion  they  are not immune from the cultural assumptions   Although consumers could be focussed  8  towards  usability  the main driver for change will be from  technologists  Thus  the rest of this paper is aimed  at computer scientists  arguing  i  the computer  science of everyday things is embarassingly bad   ii  the computer science is easy to fix  and to  do so would have valuable benefits     achieving  environmental as well as usa
21. powerful than our competitors  and we can af   ford to finance just so much to get the leverage  De   spite Moore   s Law  users still take approximately  the same sort of time to perform tasks    If the growth of Moore   s Law meant anything  real for users  we would already have had a huge  computer revolution  doubling every 18 months    is a factor of a million after 30 years  Social life  has been transformed by transport  yet the perfor   mance gain is meagre compared to computer tech   nology   s  200  perhaps as little as ten   The trans   formation transport has made within such a small  factor suggests that the tera factors for computers  are measuring something of little significance to hu   mans  Humans do want to get to other places and  they do want to move manufactured goods around  the world  but they do not need computers to have  more memory or higher density processors  What  they do want is to get tasks done faster and better   Computers are not doing that with anything like  the spectacular improvements the proponents of  Moore   s Law would like us to believe  except in  a few specialised areas  10     Human activities have not been speeded up   and certainly not speeded up by astronomical fac   tors  The growth described by Moore   s Law focuses  on technology not impact  encourages industry to  postpone worrying about usability  encourages con   sumers to try to keep up with changing fashions   and all despite continual obsolescence    7   Ironi   call
22. rams will run 25  faster in  six months with you doing absolutely nothing  In  particular because memory gets cheaper  programs  get bigger  Moore   s Law removes pressure to im   prove program quality  programmers can always  postpone addressing bloat    Suppose  by way of example  a program is so  complex that it will take 3 years to run  If its pro   grammers delay 18 months  and then start  there  will be cheaper  faster computers  The program  would still finish on the same date  The program   mers could have 18 months idling  and still achieve  the same goal  Whatever you want to do  you may  as well do nothing to start with    Enthusiasts of Moore   s Law talk huge numbers   Processor speed has improved by s  memory has  increased by m  and  wow the product has twice as  many zeros  But if we have more memory we need  a faster processor just to take advantage of it in  the same time  If there is memory not accessed  it  is being used inefficiently  thus improvement might  be better quantified by the ratio m s  which as   sumes the computer is used to capacity in both  speed and memory  Talking huge products s x m   while superficially impressive  ignores usage    Since the ratio m s  whose dimension is time  is  more or less constant  it explains why  despite the  naive view of Moore   s Law  computers still take  as long to boot   Hence Machrone   s Law  the com   puter you want always costs  5000   It comes down  to a business issue  we want computers that are  more 
23. rical usability work is misplaced   at least while it tries to understand or make user  interfaces look better without changing the under   lying way they are implemented    There are numerous reasons why change will be  difficult  including     1  Some human factors researchers working in  usability are explicitly trying to find ways of  camoflaguing problems in user interfaces  19      2  Most human factors trained professionals do  not understand computer science  and there is  a professional division between the two sub   jects     3  The accurate computer science description of  typical user interfaces is too hard to under   take  creating the impression that computer  science cannot speak on the topic  In con   trast  user problems are relatively easily ob   served and measured  This makes human fac   tors salient  but it describes symptoms of com   puter science problems rather than helps un   derstand or avoid their cause     4  Because all user interfaces involve people  very  similar empirical methods can be applied very  widely  In contrast almost all computational  approaches are particular     for example  dis   crete systems methods would be inappropriate  for virtual reality  techniques for speech are  inappropriate for graphics     and much harder  to generalise  In comparison to human factors  experts  computer scientists seem to know rel   atively little about user interfaces in general     5  User interface design is done by industrial de   signers  again w
24. s        solving    the usability problem with their own work  around  Thus with hard effort they succeed  and  because of their investment in solving problems it  becomes harder to question the overall appropri   ateness of the device for their tasks    Both calculators have memories  which  appear  to  function identically  The button recalls  the stored number and displays it  but pressed  twice in succession it sets the memory to zero   The button adds the displayed number to  memory  and subtracts from the memory     e We can assume that the memory is there for  a purpose  to store numbers  and perhaps es   pecially to store numbers that have been cal   culated  since other numbers are likely to be  easily available   How can a number calculated  and displayed be stored in memory  It is very  hard to work out what to do   Memory  de   spite high visibility keys giving the impression  of simplicity  is very difficult to use     Arguably  memory should save paper and help  users do sums more reliably  Yet most users   especially those that need calculators  would  need a scrap of paper to work out how to avoid  using paper to write down the number  Evidently  memory is not provided to make the calculator  more usable but is a feature to increase p  see  Figure 1     Casio has been making calculators for a long  time  and the two calculators are not    new    in any  way  It is not obvious how Casio can justify either       2 Pressing adds the display to memory  and only  m
25. s easy  since technology is getting  better all the time   After time t  products get  distinguished not by their performance  which is  more than adequate  but by more subtle     and  harder to supply     properties like usability  For  technologies like wrist watches  we are long past the  threshold  and they are now fashion items  that is  chosen on criteria mainly other than technological   But for many everyday things  like word processors   we should also be well beyond the crossover  So  why aren   t word processors much better    Manufacturers can increase consumers    expecta   tion of p  by providing or promising new    essential     features  it is in their interest to increase p  because  this postpones the crossover  Thus many everyday  things have numerous features  a factor influencing  their usability    The slope of the graph obviously reflects  Moore   s Law  Moore   s Law  eponymous of Intel   s  Gordon Moore  13   says that computer technical  infrastructure improves by a factor of two every    18 months  By infrastructure  take your pick   network bandwidth  processor speed  storage  capacity  progress is relentless  We often    do not wonder about Moore   s Law beyond the  excitement  everyone hopes for better products  tomorrow    The law has been followed closely and by many  companies  If Moore   s Law was an observation  about scientific progress  we would expect more  variance  We would not expect IBM  Intel and  Motorola to be neck and neck  we wou
26. thical issues   but see  11   which shows that car radios are a  widespread  unethical  example of dangerous user  interfaces  and see  6   which discusses ethical de   sign more generally     8 Conclusions    We are caught up in a culture that accepts unus   ability  Sadly  competent computer scientists seem  aloof from the user interface  This paper showed  that a professional computer science of everyday  things would help improve usability  Computer sci   entists routinely prove properties of their programs  and worry about their efficiency and correctness   why not apply these sorts of concern to the user  interface  This paper showed that benefits  of us   ability and a better environment  will follow    It is obvious that both computers and human  users both    run    programs  in languages such as  CSP  7  the differences that might seem obvious  in languages like Java disappear   In particular   users deserve as much attention in design of their  programs  i e   the user interfaces  as the comput   ers get in their programs  This paper showed that  standard computer science can be applied to the  user interfaces of everyday things    User interface design is very much a proper part  of computer science  It is time to show users that  higher standards are possible     References     1  M  A  Addison and H  Thimbleby  Intelligent  adaptive assistance and its automatic genera   tion  Interacting with Computers  Volume 8   Number 1  pages 51 68  1996      2  J  S  Brown 
27. y market churn  the corrollary of Moore   s Law   encourages a low standard for the computer science  of everyday things  which in turn makes a causualty  of usability     3 Consumer  in action    The automobile industry was changing rapidly in  the 1960s  and in some the problems of computer  science at the turn of the 21st  century reflect it   In the 1960s  some cars were badly designed and  unsafe to drive  As Ralph Nader exposed  14  22    the prevailing cultural assumption was that drivers  had accidents  and therefore drivers were respon   sible for the behaviour of cars  If an accident is     driver error    it is not the manufacturer   s problem   but is  say  a human factors or contextual prob   lem  If a parked car rolls down a hill  an exam   ple from  14    the driver should have applied the  parking brake properly  they should have turned  the wheels so the car would roll against the kerb   and so on     they should be better trained  the  handbrake should have better affordance     all  perfectly constructive observations  That the car  has been badly engineered and has an unsafe park   ing brake is thereby disguised  Similar remarks  could be made about    pilot error    and the generic  excuse  the    human factor       Thus we are persuaded that usability problems  are human problems  If users have usability prob        1Before the Industrial Revolution people moved between  the speed of walking or of a horse  say 5 to 70km h  today  you can  unsafely and ex
    
Download Pdf Manuals
 
 
    
Related Search
    
Related Contents
Splice the fibers  Tecumseh AEA1336AXA Performance Data Sheet  View User`s Manual  Règlement ST - L`Agence du court métrage  L-50 Series Service Manual  100432-an-01-ml-Scha..  VAKI AQUACULTURE SYSTEMS  CV-RS1 - 日立の家電品  77 Useful Linux Commands and Utilities  Diapositiva 1    Copyright © All rights reserved. 
   Failed to retrieve file