Human factors and MS "Chicago"
Necochea Cristian Andres
I'm doing a project on Human-Factors in computer design as it
relates to cognitive psychology. I'm concentrating mainly on STM, and
recall of icons vs. text, so far. I'm more interested in the industries
use of H-F in software design (e.g. Macintosh, Microsoft Windows). I've
been told that Windows 4 "Chicago" is the result of extensive H-F work,
but I haven't seen much (or anything) in print on the subject. Microsoft
Corp. itself is pretty hard to get in touch with, but I'm still trying.
My question is if anybody can inform me of any sources that
depict the H-F work involved in developing Chicago, or any kind of work
of that sort involving Macintosh systems. I would have thought that
Apple would have researched the pants-off this area when designing the
Mac, but apparently most of their design was intuitively based...
"intuitively" doesn't really help all that much. Oh well, hope someone
can help. Thanks.
Cris Necochea
<neco...@coral.bucknell.edu>
RADick
(Necochea Cristian Andres) writes:
I'm doing a project on Human-Factors in computer design as it
relates to cognitive psychology. I'm concentrating mainly on STM, and
recall of icons vs. text, so far. I'm more interested in the industries
use of H-F in software design (e.g. Macintosh, Microsoft Windows). I've
been told that Windows 4 "Chicago" is the result of extensive H-F work,
but I haven't seen much (or anything) in print on the subject. Microsoft
Corp. itself is pretty hard to get in touch with, but I'm still trying.
My question is if anybody can inform me of any sources that
depict the H-F work involved in developing Chicago . . . . .
The current beta version of Chicago comes with a 300-page "Chicago Review
Guide" that contains some (slightly) useful information of the type you
seek. Find a local developer for the MS platform (you can't spit without
hitting one) and ask to take a peek.
Bob Dick
rad...@aol.com
Andy Dent
: depict the H-F work involved in developing Chicago, or any kind of work
: Apple would have researched the pants-off this area when designing the
: Mac, but apparently most of their design was intuitively based...
Huh
Where have you been looking?
I've seen heaps of articles detailing Apple research. From memory, a lot
of them were in the IEEE Transactions (on Human Factors?).
My understanding of the Apple vs MS look-and-feel cases was that they
were partially started because of the huge investment in CHI research that
Apple wanted to protect.
Andy Dent (A.D. Software - Mac, DOS & Windows GUI dev.)
94 Bermuda Dve, BALLAJURA Western Australia 6066
Phone/Fax: 09 249 2719 (local) +619 249 2719 (International)
Internet: de...@iinet.com.au Compuserve: 100033,3241
Wolfgang von Thuelen
> ....... I've
>been told that Windows 4 "Chicago" is the result of extensive H-F work,
>but I haven't seen much (or anything) in print on the subject. Microsoft
>Corp. itself is pretty hard to get in touch with, but I'm still trying.
> My question is if anybody can inform me of any sources that
>depict the H-F work involved in developing Chicago, .....
Chicago is based on CUA 91, while Windows 3.x was based on CUA 89.
CUA 91 was developed by IBM and has been the cornerstone of OS/2 2.x
for over two years. Two years ago Microsoft was denigrating OS/2 and
its CUA 91 interface. Now that Microsoft will be utilizing it in Chicago,
they are of course claiming that they invented it. And it looks like
you are falling for their line.
You may be interested in the following two manuals which were published by
IBM.
IBM Systems Application Architecture CUA Advanced - Guide to User
Interface Design, SC34-4289
IBM Systems Application Architecture CUA Advanced - Interface Design
Reference, SC34-4290
--
Wolfgang von Thuelen <wo...@muug.mb.ca> or <wo...@mbnet.mb.ca>
ISM Information Systems Management Corp. CIS: 74147,523
400 Ellice Avenue Voice: +1 204 946 6776
Winnipeg, MB Canada R3B 3M3 Fax: +1 204 947 3837
timskene on BIX
TOG on Interface is an *elent [D [D [D [D xcellent* source on overall
design in this area. As you said, it is largely intiutive
probably because not a lot of research existed when the
Mac interface was being developed.
TOG on Interface
Bruce Tognazzini Apple computer
Addison-Wesley Publishing Co.
ISBN 0-201-60842-1
**********************************************************
**** Tim Skene E-mail: tims...@bix.com
**** Skene Design
**** 4843 Jeanne Mance Phone: 514-277-3366
**** Montreal QC H2V 4J6 Fax: 514-279-1311
**********************************************************
Tim Shea
>The current beta version of Chicago comes with a 300-page "Chicago Review
>Guide" that contains some (slightly) useful information of the type you
>seek. Find a local developer for the MS platform (you can't spit without
>hitting one) and ask to take a peek.
You can ftp a variety of publicly available information on Chicago from
ftp.microsoft.com,including the document referred to above. The directory
is:
/PerOpSys/Win_News/Chicago/
This will describe a lot about Chicago, including documentation, speeches
by Bill Gates, etc., but I don't know that it will tell you much about the
actual human factors_ process_ that went into the design.
You can glean some info from back issues of publications like
PC WEEK. E.g. a couple of months ago there was an article on MS bringing
a lot of outside consultants to Redmond for review of the Chicago UI.
Recently there was an article on the projected costs of upgrading to
it ($1,600,00. US, they said, for a 2,000 employee company -- most of
the cost due to employee training/downtime). Again, I don't know if
this is exactly what you're looking for.
Microsoft stands to earn a lot on Chicago upgrades (20-40 millon
Windows PCs times whatever the upgrade costs works out to a big
pile of money). The key is making it different enough to give people
a reason to upgrade, but no so different that the process is too
costly for large existing installations (of which there are many).
So this is a case where UI decisions and financial outcome are
very closely linked. I imagine this had a significant effect on the
Human Factors process.
Best Wishes.
Peter Castine
>
>Necochea Cristian Andres (neco...@coral.bucknell.edu) wrote:
>: depict the H-F work involved in developing Chicago, or any kind of work
>: of that sort involving Macintosh systems. I would have thought that
>: Apple would have researched the pants-off this area when designing the
>: Mac, but apparently most of their design was intuitively based...
>
>Huh
>
>Where have you been looking?
>
>I've seen heaps of articles detailing Apple research. From memory, a lot
>of them were in the IEEE Transactions (on Human Factors?).
>
Another place to look is Brenda Laurel, _The Art of Human Interface
Design_, Addison-Wesley. (This is from memory).
Hell, just go to a decent library and look up Laurel, B. in the authors
catalogue. Donald Norman is another author to check up.
Alternately, grab the HCI bibliography maintained at U. Ohio and do a
little grep'ing.
--
Peter Castine | Oh, wenn jene alten, musikkundigen Gelehrten die
pcas...@prz.tu-berlin.de | Modernen hoerten, was wuerden sie tun, was
Process Control Center | wuerden sie sagen!
Technical University Berlin | -- Jacobus von Luettich (ca. 1330)
Adam C. Gross
pcas...@jake.prz.tu-berlin.de writes:
>>
>>Necochea Cristian Andres (neco...@coral.bucknell.edu) wrote:
>>: depict the H-F work involved in developing Chicago, or any kind of
work
>>: of that sort involving Macintosh systems. I would have thought that
>>: Apple would have researched the pants-off this area when designing
the
>>: Mac, but apparently most of their design was intuitively based...
>>
I have a personal theory on this matter, although I may just be confused.
I seem to recall Microsoft entering some sort of small deal with NeXT,
where Microsoft would develop NeXT applications and also get to use some
NeXT interface elements in Windows. While I haven't heard anything about
this in a long time, Windows 4.0, from what I've seen, does look a lot
like NeXTstep. Not only does it have that gray color scheme, but it
seems to have the same types of pop up menus, check boxes and shaded
square buttons that made the NeXT interface a pleasure.. and thoughts?
Adam C. Gross
SRI International
Gary Stephens
:
: Another place to look is Brenda Laurel, _The Art of Human Interface
: Design_, Addison-Wesley. (This is from memory).
That reference in full :
The Art of human-computer interface design / Brenda Laurel, editor.
ISBN 0-201-51797-3
Addison-Wesley
Gary
gary.s...@sse.ie
Andrew Lih
people think about MS's new "Windows 4.0 keyboard" that they are now
promoting.
From PC Week (Aug 1, 1994):
Will new peripheral unlock the key to Windows?
By Lisa Dicarlo
Microsoft is developing a modified version of the standard
101-key AT keyboard that adds three keys designed to activate
features within Windows 4.0 and NT.
The keyboard is being designed by Microsoft's mouse hardware
division, but MS plans to subcontract assembly of the
keyboards to KeyTronic Corp. of Spokane, Wash., sources said.
KeyTronic officials declined to comment.
The integration of operating system and hardware features is
part of Microsoft's Nexus project, an initiative to simplify
and optimize the usability of Windows, sources said.
Sources familiar with the plan expressed concern that MS was
trying to establish yet another de facto standard that PC
users and OEMs would have to accept if they wanted convenient
access to the Windows GUI. Microsoft officials, however,
downplayed the impact.
PC OEMs will not be required to buy the keyboard, and users
will have access to all Windows 4.0 features even if they
don't have keyboards with the three extra keys, said Paul
Maritz, senior vice president of systems software at MS in
Redmond, Wash. The keyboard also would not affect the
performance of any other PC operating system, Microsoft
officials said.
The three added keys are an Application key, a Left Windows
key, and a Right Windows key. The latter two keys will be
embossed with the Windows logo, according to sources.
The Application key replaces the right mouse button in
applications that use it, such as Word or Excel. A short
editing menu pops on the screen when activated. The Left and
Right windows keys, located on either side of the space bar,
activate the "set focus" feature in the Chicago user
interface. This eature changes the cursor status, so when it
passes over the menu-bar commands, the commands are
highlighted without the mouse being clicked or added keys
being depressed.
The keyboard will be optimized to work with the Chicago and
Cairo user interfaces, although Microsoft has written software
that will allow Windows 3.1 and NT 3.1 and 3.5 to work with it
as well, said Keith Kegley, a product manager in MS's hardware
unit.
"It sounds interesting, but what is the value add?" asked
Randy Dugger, associate IS manager at Liposome Technology Inc.
in Menlo Park, Ca. "If it makes me more productive and
reduces time to do things, I'd consider it."
PC OEMs expressed no concern that the new keys might force
them to redesign or upgrade their keyboard designs. "We don't
use Microsoft mice, and we haven't lost a sale for lack of
it," said Kevin Roberts, product manager at Advanced Logic
Research Inc. or Irvine, CA. "Other keyboard vendors may
offer different versions of it in the future, and we may use
that if customers want it."
The technical specifications for the keybaord is avialble to
ISV's on the plug and play section of CompuServe's Chicago
beta forum. NMB Technologies Inc. for example, is working on
a second generation Windows 4.0 keyboard that will further
enhance the operating system's functionality. NMB expects to
demonstrate it at Comdex in November, said Dave Snyder,
keyboard product manager for teh Chatsworth, CA firm.
--
`''' Andrew "Fuz" Lih Columbia University
c @@ l...@cs.columbia.edu Mobile Computing Laboratory
\
- <A HREF="http://www.cs.columbia.edu/~lih/">Fuz Page</A>
Daniel Galeus
>You may be interested in the following two manuals which were published by
>IBM.
>IBM Systems Application Architecture CUA Advanced - Guide to User
>Interface Design, SC34-4289
>IBM Systems Application Architecture CUA Advanced - Interface Design
>Reference, SC34-4290
I think they have been superseeded by a paperback from Que, which is
now the official CUA91.
I dont have the title and numbers right now, but I have the book at home.
Best regards, Daniel
--
Daniel Galeus | Internet: dga...@vinga.trillium.se | Ph. +46 31 819850
| CIS: 75170,240 | Fax +46 31 812139
Vinga System AB | SNAIL: Gotabergsgatan 28 |
Sweden | S-411 34 Gothenburg, Sweden |
Magnus Ramage
> From PC Week (Aug 1, 1994):
> 101-key AT keyboard that adds three keys designed to activate
> features within Windows 4.0 and NT.
This is a joke right? The 101-key keyboard at last is a standard across
almost all IBM-compatible (remember that old chestnut we used to say before
"Windows"?) PCs. Indeed lots of non-IBM PCs have them too: Macs, Amigas, etc.
So we have a standard which astonishingly enough fits the whole computer
industry (little bit of hyperbole there - OK the whole industry except Unix
boxes...) And Microsoft are seeking to CHANGE it? They're off their
trolleys!
Now read on...
> The three added keys are an Application key, a Left Windows
> key, and a Right Windows key. The latter two keys will be
> embossed with the Windows logo, according to sources.
>
> The Application key replaces the right mouse button in
> applications that use it, such as Word or Excel. A short
> editing menu pops on the screen when activated.
Oh this is good. Coming next - the one-button Microsoft Mouse? Tog's stout
defence of the one-button mouse on the Mac notwithstanding, I think few would
now disagree that it's one the Mac's dodgier features in terms of the
usability/functionality tradeoff. I would _hate_ having to do right-button
functions on the keyboard. Stone age stuff.
> The Left and Right windows keys, located on either side of the space bar,
> activate the "set focus" feature in the Chicago user
> interface. This feature changes the cursor status, so when it
> passes over the menu-bar commands, the commands are
> highlighted without the mouse being clicked or added keys
> being depressed.
Erm. Am I missing something deep and subtle offered by Chicago, or is this
just a way to avoid clicking on the window title bar with the mouse? Never
heard of this being regarded as particularly heinous by anyone.
What an astonishingly lunatic idea. Hope no-one buys them.
Magnus Ramage
School of Cognitive & Computing Sciences,
University of Sussex,
Brighton, England
(mag...@cogs.sussex.ac.uk)
Yean Wei Ong
l...@news.cs.columbia.edu (Andrew Lih) writes:
> Microsoft has finally gone off the deep end. I'd like to hear what
> people think about MS's new "Windows 4.0 keyboard" that they are now
> promoting.
>
> From PC Week (Aug 1, 1994):
>
> Will new peripheral unlock the key to Windows?
>
> By Lisa Dicarlo
>
> Microsoft is developing a modified version of the standard
> 101-key AT keyboard that adds three keys designed to activate
> features within Windows 4.0 and NT.
(bits deleted ...)
Hmmm ... it sounds a bit fishy to me. Considering the sizeable
influence that Microsoft wields, however, there may not be much we
can do about it (_if_ we want to do something about it).
From the article, it basically seems as if Microsoft is trying to
incorporate more mouse functions into the keyboard. There shouldn't
be too much harm in this, considering the number of keyboard
shortcuts that are already used in most GUIs.
Thanks for the post, Andrew. :-)
Regards,
Yean Wei.
--
Yean Wei Ong Department of Psychology
MSc(AppPsych) student The University of Western Australia
yea...@freud.psy.uwa.edu.au Crawley WA 6009
Opinions expressed are strictly my own. Australia
Bradley K. Sherman
>Microsoft has finally gone off the deep end. I'd like to hear what
>people think about MS's new "Windows 4.0 keyboard" that they are now
>promoting.
>
> Microsoft is developing a modified version of the standard
> 101-key AT keyboard that adds three keys designed to activate
> features within Windows 4.0 and NT.
Over the last 10 years or so, I have been asked, "What kind of computer
should I buy?" by many folks who wanted to get started. For about 3
years now, it has been extroadinarily easy to answer: Get a Macintosh.
I don't particularly care for Mac's because they are closed systems
(best for non-networked use by a single user), but at least the novice
gets stuff going without having a guru standing by. I used to get a
good laugh from the "Chaos Manor" columns in _Byte_, in which Jerry
Pournelle would have a problem with a piece of software and "Harry" the
technical manager for the software house would run over to Jerry's
house and fix it for him. With MSDOS 6, Window 3 & 4, OS/2, Chicago,
Daytona, NT, and who knows what else being heaped on the egregious
Intel architecture, it is a tribute to the gullibility of the US
consumer that Bill Gates stock holdings aren't rapidly approaching $0.
--bks
--
Bradley K. Sherman Computer Scientist
Dendrome Project 510-559-6437 FAX: -6440
Institute of Forest Genetics b...@s27w007.pswfs.gov
P.O. Box 245, Berkeley, CA 94701 http://s27w007.pswfs.gov/People/bks.html
Peter Castine
K. Sherman) wrote:
> Over the last 10 years or so, I have been asked, "What kind of computer
> should I buy?" by many folks who wanted to get started. For about 3
> years now, it has been extroadinarily easy to answer: Get a Macintosh.
> I don't particularly care for Mac's because they are closed systems
> (best for non-networked use by a single user), ... [schnipp]
BTW, it's 1994 and the Mac ain't bad for networking either (and still faster
to set up than a Novell configuration).
I will agree that Mac is a single-user machine, but that is a characteristic
of desktop personal computers.
Apologies for inserting this into human-factors, but I couldn't resist
responding to Brad. Followups to comp.sys.mac.advocacy.
--
Peter Castine | In old days books were written by men of
pcas...@prz.tu-berlin.de | letters and read by the public. Nowadays books
Process Control Center | are written by the public and read by nobody.
Technical University Berlin | -- Oscar Wilde
Clara N. Fitzgerald
>Necochea Cristian Andres (neco...@coral.bucknell.edu) wrote:
>: depict the H-F work involved in developing Chicago, or any kind of work
>: of that sort involving Macintosh systems. I would have thought that
>: Apple would have researched the pants-off this area when designing the
>: Mac, but apparently most of their design was intuitively based...
>I've seen heaps of articles detailing Apple research. From memory, a lot
>of them were in the IEEE Transactions (on Human Factors?).
>Andy Dent (A.D. Software - Mac, DOS & Windows GUI dev.)
> Internet: de...@iinet.com.au Compuserve: 100033,3241
Most of the 'Mac' look was developed at XEROX PARC for a computer called
the Star - I believe some of the programmers transferred to Apple. I
can't remember references, but I could ask some professors if there is
interest.
--
-Clara A. N. Fitzgerald cfit...@s.psych.uiuc.edu
- < - < -< <> >- > - > -
Help stamp out, reduce, and eliminate redundancy.
Elizabeth Buie
b...@s27w007.pswfs.gov (Bradley K. Sherman) writes:
>Over the last 10 years or so, I have been asked, "What kind of computer
>should I buy?" by many folks who wanted to get started. For about 3
>years now, it has been extroadinarily easy to answer: Get a Macintosh.
>I don't particularly care for Mac's because they are closed systems
>(best for non-networked use by a single user), but at least the novice
>gets stuff going without having a guru standing by.
The answer of what to recommend is based as much on the task(s) as it
is on the computer experience of the user. To limit oneself to the
novice/expert dimension when recommending a platform is to insult
those of us who are very computer literate, thankyouverymuch, but
prefer Macs anyhow. I don't care to program the dadblasted thing
(I had enough of that activity in 1975-1985), just to use it as a tool.
Your mileage may vary, but that's exactly the point.
Elizabeth
--
Elizabeth Buie, Computer Sciences Corporation, Laurel, Maryland, USA
eb...@csc.com (This space accidentally left blank.)
Rodney Fuller
>> Over the last 10 years or so, I have been asked, "What kind of computer
>> should I buy?" by many folks who wanted to get started. For about 3
>> years now, it has been extroadinarily easy to answer: Get a Macintosh.
pcas...@prz.tu-berlin.de (Peter Castine) writes:
> I will agree that Mac is a single-user machine, but that is a characteristic
> of desktop personal computers.
I just returned from the AAAI 94 (Amer. Association of Artificial Intelligence)
conference in Seattle where Steve Baumer (high up in the research at microsoft)
talked about some of the AI research they have put into cario and the
hardware platform they expect by the year 2000.
The fact that microsoft is developing both chicago (their single user software)
and cario (their multiuser software) for release "sometime" is a very strange
way to develop operating systems (as I imagine the process). So the above
statement that Mac's are still "single user" machines implies that MS
software has solved this problem. Chicago will not offer much in the way of
networking in terms of os (just what third party vendors can offer) unless
microsoft has redefined the term "single-user" (I hope someone does soon).
Cario (sometime after 1995) might offer an open collaborative system but with
the refelction research they talked about I (personally) doubt it. And with
the chief architect being the same guy who designed VMS (&%$#@%) I really
need to see cario before I can tell.
On the other hand microsoft expects the "average computer" by the year 2000 to
be "100 Mhz, 96 MB RAM, 2 GB HD, 1020 dpi graphics, ..." So it doesn't sound
like they are investigating much human factors other than the normal hand-eye-
screen direct manipulation type. They made an appeal to AI to develope shrink
wrappable products for their OS--which you don't expect someone with their
market share to do (ie there are some weaknesses in development somewhere).
I also talked to someone from sun who detailed the normal "olimpic" approach:
bigger, faster, stronger. I keep waiting for someone to try "smarter,
forgiving, powerful" -- maybe after 2002.
Rodney Fuller
ful...@cgs.edu
ful...@cgsvax.claremont.edu
> de...@iinet.com.au (Andy Dent) writes:
>>I've seen heaps of articles detailing Apple research. From memory, a lot
>>of them were in the IEEE Transactions (on Human Factors?).
> Most of the 'Mac' look was developed at XEROX PARC for a computer called
> the Star - I believe some of the programmers transferred to Apple.
One of my most heartfelt desires is to hear *two* stories about the mac
interface development that mention the same people and events before I die.
I know three of the people on the mac interface team personally and even their
stories don't match up completely. And there are several articles out there--
one of the worst was written by someone who "joined" the team a week before the
product shipped--his book was the "standard story" for a while--and the latest
issue of Interactions has a VERY, VERY nice "rumor smashing" article by the
project manager who tells how the interface was already prototyped before the
famous/notorious PARC visit. And that the PARC visit was simply a way to use
another labs product to visualize the theory developed at apple.
like I said--I just want two of those damn stories to match before I die.
Rodney Fuller
ful...@cgs.edu
luis fernandes
The following is excerpted from an article (in "Toronto Computes!", a
freebie published every month) on a talk that Gates gave at the
recent Comdex/Canada show here:
A Microsoft employee demostrating at Gates' talk said studies
have shown [that] new users take nine minutes to figure out how
to launch a word processor under Windows 3.1, but just three
minutes using Chicago.
"Surely you mean nine seconds to launch the program", a seemingly
incredulous Gates asked. Replied the nervous staffer: "No, these
people had never used a computer and were doing it without any
help. Under DOS with the manual, they'd have been lucky to do it
in nine days".
Bradley K. Sherman
>>From the "Be Very Afraid Department".
> "Surely you mean nine seconds to launch the program", a seemingly
> incredulous Gates asked. Replied the nervous staffer: "No, these
> people had never used a computer and were doing it without any
> help. Under DOS with the manual, they'd have been lucky to do it
> in nine days".
(Reminds me of the S. Kelly-Bootle article where the conversation does
something like:
...
Tyro : "How do I start Wordstar?"
Pro : "Type ws"
Tyro : "How do you spell that?"
... )
Where do you find people who have "never used a computer" these days?
Are these the same people who will serve on the O.J. Simpson Jury?
It is ridiculous to base interface development (or do interface
analysis) on people who have never used a computer. By the time
you get the interface right they'll all be dead.
Tim Shea
[...]
>Where do you find people who have "never used a computer" these days?
>Are these the same people who will serve on the O.J. Simpson Jury?
>It is ridiculous to base interface development (or do interface
>analysis) on people who have never used a computer. By the time
>you get the interface right they'll all be dead.
[...]
If you visit factories in the USA you will find many types of workers. Some
hold advanced college degrees. Some are hourly-wage earners with little
formal education. Some are executives. Amongst many of these people you
will encounter a significant proportion who have never touched a keyboard --
of any kind. I suspect the same would be true in many other countries.
For those of us in the computer industry it's probably hard to imagine this.
But it's true. It's also difficult to picture the anxiety created when a new
computer system is introduced. A typical case is an hourly wage earner
who has never typed a sentence in his life who is now going to be judged
(at least in part, maybe in large part) on his ability to operate a new
computer system. This person may be relegated to menial tasks or put
out of a job if he can not adapt. That adaptation is made more difficult
if the system is poorly designed (as it could well be if the designers hold
the attitude expressed by the poster above). Needless to say this has
real-life consequences for the worker, his family, and the factory.
One way to design systems for successful use by people is to create an
analog of what they do today. If they fill out a form to record material
loaded into a truck, then you design something which is as analogous as
possible to that process. This was recognized long ago by the Star and
Macintosh designers, who were designing systems (in the case of the Star
anyway) explicitly for office use. Of course ever since we've been
stuck in this office-based desktop metaphor. Since most software
designers are themselves office workers, it is difficult for them to imagine
people who don't work in offices, like factory workers, farm workers, auto
mechanics, etc. So the desktop metaphor is self-perpetuating.
When designing analogs you can also use a little magic to break the
real-world, literal metaphor if necessary. A good discussion of this can be
found in the paper by Randall Smith on the Alternate Reality Kit and
the "Tension Between Literalism and Magic" in the CHI + GI 1987
Proceedings.
Another approach is to integrate the computer system into the work
environment itself. This is done today with things like bar code scanners;
the boxes are scanned as they are loaded, giving an inventory of
everything put onto the truck. Although there will be new technology
to learn, this approach seems promising since it integrates with the
existing work practice, with which the worker is probably already quite
familiar. Much of the research currently being done on this
approach is by Marc Weiser and his group at Xerox PARC. There
are a number of good articles on this (Communications of the ACM,
July 1993, Scientific American September 1991, Interactions,
January 1994). The basic theme is embedding more computational
support into the environment.
Again the PARC research is directed primarily at a white collar office
workers, which makes sense given the Xerox customer base. I would
like to see these same principles applied to manufacturing and other
settings however, since these are probably the most fertile ground for
initial deploymentof these technologies. I say this based on their
early adoption of embedded process control systems, bar code scanners,
radio frequency networks, and various forms of mobile computing.
Dave Curbow
cfit...@s.psych.uiuc.edu writes:
> Most of the 'Mac' look was developed at XEROX PARC for a computer
called
>the Star - I believe some of the programmers transferred to Apple. I
>can't remember references, but I could ask some professors if there is
>interest.
This is a bit of folklore -- a bit of the truth, but not the whole truth.
A number of Xerox Star developers (including myself) have gone to work at
Apple since '81 and some of the ideas that Xerox developed have made it
into Lisa and later Macintosh. But, those same ideas (like Xerox
developers) went to lots of other places as well.
You could just as easily say the SunView came from Xerox, Or, that
Microsoft Windows came from Xerox. Likewise, Microsoft Word came from
Xerox PARC where it began as Bravo (or Bravo X, I can never remember).
Yet, all of these companies made refinements to those ideas and invented
some of their own. So that if you were to sit down at a Xerox Star and
any one of the other GUIs, you would instantly see how they are
different. The Macintosh icons are similiar to to the Star ones, but less
so than the Microsoft Windows icons (some of the basis of the lawsuit, I
think.) Star and Mac have a desktop, and dialog boxes, and windows, etc.
But, Macintosh did drag and drop much better than Star. I could go on,
but I think you get the point.
BTW: As I understand it, the Macintosh team had no one from Xerox working
on it.
Dave Curbow
Human Interface Designer
Apple Computer, Inc.
<< My opinions are my own, even though my return address says Apple. >>
Adam C. Gross
ful...@cgsvax.claremont.edu writes:
>> Most of the 'Mac' look was developed at XEROX PARC for a computer
called
>> the Star - I believe some of the programmers transferred to Apple.
>
>One of my most heartfelt desires is to hear *two* stories about the mac
>interface development that mention the same people and events before I
die.
>
>I know three of the people on the mac interface team personally and even
their
>stories don't match up completely. And there are several articles out
there--
>one of the worst was written by someone who "joined" the team a week
before the
>product shipped--his book was the "standard story" for a while--and the
latest
>issue of Interactions has a VERY, VERY nice "rumor smashing" article by
the
>project manager who tells how the interface was already prototyped
before the
>famous/notorious PARC visit. And that the PARC visit was simply a way
to use
>another labs product to visualize the theory developed at apple.
Windowing/graphical operating systems and the mouse were of course
developed at the Stanford Research Institute (SRI) in the late 60's, long
before Xerox PARC. A good portion of the SRI team went to PARC in the
70's and developed their Alto GUI-computer (wasn't that the name of the
PARC machine?)
Peter Castine
K. Sherman) wrote:
> Where do you find people who have "never used a computer" these days?
> Are these the same people who will serve on the O.J. Simpson Jury?
There are millions, nay, billions (thousands of millions, for those preferring
British usage ;-) of computer-naive users out there. Really. The more relevant
question is: How likely is it, that any of them are going to buy a computer
in the next ten years?
One answer to the previous question: Apple estimiated the potential market
for their Performa series (aimed at first-time users) at several millions,
IMS, it was 7,000,000 homes.
> It is ridiculous to base interface development (or do interface
> analysis) on people who have never used a computer. By the time
> you get the interface right they'll all be dead.
If you intend to sell to people who have never used a computer before, you
would be well-advised to base your UI on their problems. If your time from
drawing board to market is longer then the average life expectancy of your
intended audience, you're in the wrong business.
ful...@cgsvax.claremont.edu
> There are millions, nay, billions (thousands of millions, for those preferring
> British usage ;-) of computer-naive users out there. Really. The more relevant
> question is: How likely is it, that any of them are going to buy a computer
> in the next ten years?
The last time I saw someone run the numbers they estimated that you had
to divide the total number of personal computers sold by 5 to account for the
same people buying more and more powerful machines (as upgrades). This would
mean that (in today's numbers) there are about 750 million people who have
access to a personal computer (for at least part of their day) and about 5
billion (not the British number) "computer-naive users" left in today's
world.
That quite a market share to exploit.
Rodney Fuller
ful...@cgs.edu
GW Doreen
> Where do you find people who have "never used a computer" these days?
> Are these the same people who will serve on the O.J. Simpson Jury?
When I need users who are computer novices, I go to a grocery store during
the day (when most computer users are working or in school) and ask people
in the parking lot. It takes a while, but I usually find enough to come in
for testing.
Bradley K. Sherman
>In article <pcastine-100...@maggie.prz.tu-berlin.de>,
>pcas...@prz.tu-berlin.de (Peter Castine) writes:
>> There are millions, nay, billions (thousands of millions, for
>> those preferring
>
>The last time I saw someone run the numbers they estimated that you had
>to divide the total number of personal computers sold by 5 to account for the
>same people buying more and more powerful machines (as upgrades).
Sorry guys, but you don't have to *buy* a computer to use one.
Don't know about Britain or Germany, but most people in the US
have used a computer before they buy one. There are about 200
people in this building who use computers, but only a fraction
own one. My 9 year old daughter and 4 year old son both use
computers, but neither of them will be buying one for some
time to come.
I am not advocating that users need to understand the
difference between a program and a process, or be able to
explain deadlock, or write an awk script before being granted
the privilege of using a computer. I am merely reacting to
the idea that interfaces need to be tested on folks who have
*never* used a computer to be good. This idea is wrong,
though prevalent. Interfaces should enable users to do their
work efficiently with a minimum of frustration. Note, I say
work. If all you want to do is play games, or move computers
out the door of the retailer, that's a different topic.
Furthermore, the days when you can find people who have
*never* used a computer by canvassing the folks in the
supermarket shopping lot are numbered. And I question the
value of research on such a group.
I wonder if Toyota evaluates new designs by finding a
group of subjects who have never driven a car.
Mr Toxic
K. Sherman) writes:
>I am not advocating that users need to understand the
>difference between a program and a process, or be able to
>explain deadlock, or write an awk script before being granted
>the privilege of using a computer. I am merely reacting to
>the idea that interfaces need to be tested on folks who have
>*never* used a computer to be good. This idea is wrong,
>though prevalent. Interfaces should enable users to do their
>work efficiently with a minimum of frustration. Note, I say
>work. If all you want to do is play games, or move computers
>out the door of the retailer, that's a different topic.
>Furthermore, the days when you can find people who have
>*never* used a computer by canvassing the folks in the
>supermarket shopping lot are numbered. And I question the
>value of research on such a group.
>
>I wonder if Toyota evaluates new designs by finding a
>group of subjects who have never driven a car.
I think that the difference here is that oftentimes testing software is
also testing the metaphors used in the design. In a car, a steering wheel
is not a metaphor. People know that a car is a car and that you steer it
with a wheel through years of being passengers, watching TV, playing with
toy dumptrucks, etc.
The usefulness of 'Joe Public' testing is that you can discover whether or
not your cultural assumptions are on the mark. For example, if we were a
society of cats, a trashcan might symbolize 'lunch' rather than 'discard
files'.
--
Wayne Greenwood
mrt...@aol.com
Catherine L. Kelley
Peter Castine <pcas...@prz.tu-berlin.de> wrote:
>
>For crying out loud, instead of saying "everybody I know knows how to use
>a computer" and extrpolating to the entire population of the U.S. (let
>alone the planet), read some market surveys. (Yes, I realize their value
>is limited, but it's better than the former method of estimation.)
>
Good point, Peter.
As of 1989, the Canadian General Social Survey data show that 52.6% of
the adult population over the age of 15 could not use a computer at all.
Broken down by age groups, these data show that 26.3% of people age
15-24 could NOT use a computer, 41.8% of those age 25-44 could not use
a computer, and a whopping 78.1% of those age 45+ could not use a
computer.
Although these data are Canadian, I expect that the numbers are
comparable in the U.S. Also, for the skeptical, the data were collected
by Statistics Canada, which is a government statistical agency rather
than a marketing firm. Stat Can has a pretty decent global reputation,
and I trust their data.
These data are only 5 years old, and I do not expect that there have
been huge shifts in computer use in those 5 years. The numbers are
undoubtedly a little lower now, but there are certainly still many
people in North America who cannot presently use a computer.
I think it's pretty funny to see this discussion about the utility of
testing with novice users, on the same newsgroup that regularly contains
stories about the "idiotic" errors that users make.
-Cathy Kelley
kel...@healthy.uwaterloo.ca
Bradley K. Sherman
> ...
>Or as someone else put it, "programmers are the reason why
>people hate computers." Most people are not programmers and
>do not want to be programmers -- and they consider typing
>anything more than their name to be "programming."
>
> ...
In your wildest dreams you're not going to be able to conceive
of a computer that can do something useful for these users
--except for the embedded computers in the phone systems,
ATM's, and cars which they're already using! Why are we
trying to give general purpose computers to people who will
never understand the difference between 'or' and 'xor'?
I'm not talking as a power-user computer snob. I've
been out there on the front lines for nearly 20 years
now dealing with exactly the technophobes all you
folks are describing. There isn't going to be a
MacOS 777.0 or Windows 3333.1 that allows completely
naive users to do *novel* productive things with a
computer.
If all you want is a typed memo, it is harder to show a
completely naive user how to use Word on a Mac than it
was to show them how to use Wordstar on a CP/M machine.
For someone who has moved past this stage however, and
wants to construct a more complicated document, the
Mac is superior. I think this is evidence in favor
of my argument.
What I am saying is that interface design should be
optimized for the users who use the computers frequently
so as to maximize their productivity. For the rest,
hide the computer entirely.
ful...@cgsvax.claremont.edu
A few posts ago Brad set up a straw man arguement of:
> is Toyota required to test its new cars on people who have never driven?
Two points: 1) Toyota (at least I'm pretty sure it's Toyota) claims
to design and engineer its products "for the human race"--and that
would seem to include people who have not driven. 2) I expect that even the
most familiar of car controls can be improved--if you have ever rented a
strange car in a strange airport and a strange town when it is either raining
or snowing and tried to achieve some goal directed behavior like get
to some prearranged place by a prearranged time--and been completely
frustrated by non-standard controls and signs and vague directions--
then you will admit that car interfaces have a long way to go before they
are as transparent as Brad assumes them to be.
And if I were to look for people who have never driven I would look in
New York City. That place seems to be a haven for them.
> So what? What percentage of people cannot use a soldering iron?
> A torque wrench? What percentage cannot type? Cannot set the clock
> on the VCR? Is this some kind of messianic campaign to get everyone
> using a computer? Do we have to have computer interfaces that
> are aimed at the lowest common denominator?
I would say: yes! We do have to design for the person who for whatever
reason might be temporarily behaving in a totally idiotic manner (I've
never met a "lowest common denominator" and wouldn't know how to design for
one). The alternatives are even less seductive--and could result in a
class society based on user inteface skill in terms of retrieving and
manipulating information.
What if you had to enroll your children in "special" schools that offered
interfaces they were familiar with while those children who knew "big-deal-
interface-of-the-future" got the better computers and better resources because
the teachers didn't have time to help your children with remedial interface
skills. (I know I'm exagerating, but look at what the scientific community has
done with UNIX and listing shell specific skills in job listings).
> "Build a system that even a fool can use, and only a fool will
> want to use it." --/usr/games/fortune
Those fools make good friends.
Rodney Fuller
Known to be very foolish at times.
ful...@cgs.edu
George Boggs
of experienced computer users in any country whatsoever has but the most
marginal relationship to the selection of test subjects for usability
testing of any given product?
Humbly, I state Boggs' First Law: Given a software product that could
predict stock prices 24 hours in advance, prospective users would learn to
communicate in Sanskrit if necessary to access the application.
Whether a software product, or any other artifact for that matter, is worth
effort and diligence to learn is largely, if not wholly, dependent on it's
subjective utility to the user. If one is designing software that will help
the home user archive recipes, or keep track of exciting things like
automobile maintenence records, then one had best design such that a mature
lab rat could wend its way through it the first time. On the other hand, if
one were designing an application with some significant utility to the home
user (i.e., the "killer application" Grail), usability, although nice, is
much less critical.
A product with great utility will sell despite poor human factors (Any Unix
users out there? Hmmmm?). The sales of a product with low utility, even if
it has good human factors, will be largely governed by Barnum's Law.
Furthermore, the selection of test subjects should be governed by their
experience only insofar as it corresponds to the experience of the
projected (read "mythical") target market. The designer of a product
destined to sit on the desks of dedicated Unix hackers would be, oh, I'll
say crazy, for lack of a better word, to select a sample of test users from
the population of persons who had never used a computer.
--- So, in summary:
Not relevant question: How many people in the world have used a computer?
Relevant question 1: Does my product have any utility aside from the fact
that it provided the means of my employment?
Relevant question 2: Who are the most likely persons to find that utility
and what are their characteristics?
--- Final question:
Did Carl Sagan have anything to do with the title of this thread?
G. Boggs
gbo...@advtech.uswest.com my opinions alone, thank you
Bradley K. Sherman
>In article <pcastine-110...@maggie.prz.tu-berlin.de>,
>Peter Castine <pcas...@prz.tu-berlin.de> wrote:
>>
>>For crying out loud, instead of saying "everybody I know knows how to use
>>a computer" and extrpolating to the entire population of the U.S. (let
>>alone the planet), read some market surveys. (Yes, I realize their value
>>is limited, but it's better than the former method of estimation.)
>>
>Good point, Peter.
>
>As of 1989, the Canadian General Social Survey data show that 52.6% of
>the adult population over the age of 15 could not use a computer at all.
>Broken down by age groups, these data show that 26.3% of people age
>15-24 could NOT use a computer, 41.8% of those age 25-44 could not use
>a computer, and a whopping 78.1% of those age 45+ could not use a
>computer.
>
So what? What percentage of people cannot use a soldering iron?
A torque wrench? What percentage cannot type? Cannot set the clock
on the VCR? Is this some kind of messianic campaign to get everyone
using a computer? Do we have to have computer interfaces that
are aimed at the lowest common denominator?
I would bet that 41.8% [sic] of Americans (U.S. type) couldn't
find Winnipeg on a map if you spotted them Moose Jaw and
Thunder Bay. Do we need simpler maps, better education,
or a more motivated Homo sapiens?
[B.T.W., I concur that Stats Canada is one of the top survey
research organizations in the world, but I doubt that their
techniques are adequate to give 3 significant digits against
the 2.? x 10^7 people in Canada in a pseudo-random survey.]
Maybe there are billions of people who do not need keyboards,
mice and GUI's. Maybe the millions who do need them should
be the objects of research, and the rest can do the normal
thing when confronted with a novel situation --ask for help
from a human being
Come to think of it, we'd better throw out libraries,
as being much too difficult for those who have never
read.
--bks
"Build a system that even a fool can use, and only a fool will
want to use it." --/usr/games/fortune
--
Peter Castine
K. Sherman) wrote:
> Sorry guys, but you don't have to *buy* a computer to use one.
This is true. Doesn't change the fact that with millions of computers out
there, but billions of people out there. The differential is several
orders of magnitude. Or, are you going to try to tell me that the average
desktop computer (that's where the numbers are) is used by ca. 1000
people?
For crying out loud, instead of saying "everybody I know knows how to use
a computer" and extrpolating to the entire population of the U.S. (let
alone the planet), read some market surveys. (Yes, I realize their value
is limited, but it's better than the former method of estimation.)
> I wonder if Toyota evaluates new designs by finding a
> group of subjects who have never driven a car.
No, but you need to pass a driving test before you can take a car on the
road. That's a hell of a lot more than what's required to tell other
people you're an experienced computer user, let alone start using an
application program.
David Fay
gbo...@atqm.advtech.uswest.com (George Boggs) wrote:
> Whether a software product, or any other artifact for that matter, is worth
> effort and diligence to learn is largely, if not wholly, dependent on it's
> subjective utility to the user. If one is designing software that will help
> the home user archive recipes, or keep track of exciting things like
> automobile maintenence records, then one had best design such that a mature
> lab rat could wend its way through it the first time. On the other hand, if
> one were designing an application with some significant utility to the home
> user (i.e., the "killer application" Grail), usability, although nice, is
> much less critical.
>
> A product with great utility will sell despite poor human factors (Any Unix
> users out there? Hmmmm?). The sales of a product with low utility, even if
> it has good human factors, will be largely governed by Barnum's Law.
What you say is true only in the absence of competition. In a competitive
marketplace, products quickly evolve nearly identical utility so other
attributes replace it as differentiators of the products. User interface
design is one of the most important.
--
David Fay email: da...@gte.com
GTE Laboratories voice: +1.617.466.2675
40 Sylvan Rd. fax: +1.617.290.0627
Waltham, MA 02254
Door
>Furthermore, the days when you can find people who have
>*never* used a computer by canvassing the folks in the
>supermarket shopping lot are numbered. And I question the
>value of research on such a group.
For the software I design and develop (ordering/bride-registry kiosks,
computerized touch-pad phones, ITV, ATM machines), testing on such a group
is not only valid research, but highly desirable.
Not all software runs on the desktop PC platform, nor is all software
designed for a computer-literate audience.
gst...@cruzio.com
said in part:
> Whoah, watch those generalizations, Tex. Up to now, the computer
> industry has been extremely insular, selling to a relatively
> small group (compared to the whole population) over and over again.
> The promise of computers "for the rest of us" has for the most part
> not happened. People really *do* have VCRs blinking "12:00" and they
> really do by $5000 worth of hardware and software -- and then struggle
> to write a short letter (and never even dream of touching the database
> software, much less customizing it).
Example #1.
My last project was a 12 - 16 thousand dollar computerized fitness testing
system. It was primarily used by health club employees and other, often less
than well educated, health and wellness oriented people.
At least once a month I would have to describe what the colon (:) looked like.
These people were often computer phobic. A large percentage (at least 10%,
maybe 20%) of the systems were put away when the trained person left. The
replacement would look at the PC and say 'Ack!'.
It was often a struggle to get a club owner who maybe had spent 15k to buy this
thing, to *actually* use it.
Example #2. On Sunday, my wife and I will drive 3 hours because her parents
are embarassed to leave the house they borrowed for a month in it's current
condition. The problem is that they have cable TV, cable stereo, and both
systems have their own remotes. They went to do something, got flustered, and
'pushed all the buttons', in an attempt to gain control. Of course they have
the system screwed up royally now, they only get one TV station and no stereo.
These people have had cable for years, have a fax machine, and are not stupid.
The combination of a new set of remotes, a different hardware setup, and the
fear of looking dumb has had them paralyzed for a week!
Luckily they don't watch much TV!
So there you go... 2 good examples of people who's needs *have* to be
addressed in a well written general purpose program.
Gary Starkweather
--
----------------------------------------------------------------------------
| Gary Starkweather | SGO ltd. 408.338.2945 |
| Boulder Creek, CA | |
| gst...@cruzio.com | Systems.... Solutions! |
Nick Parker
>The appropriately named mrt...@aol.com said:
>
>The usefulness of 'Joe Public' testing is that you can discover whether or
>not your cultural assumptions are on the mark. For example, if we were a
>society of cats, a trashcan might symbolize 'lunch' rather than 'discard
>files'.
This opens the obvious possibility of using a flat cat (otherwise
known as "feline road pizza") as a icon. Let's see now, what function
would that represent? Compress? Nah, too obvious...
--
Nick Parker - nspa...@ingr.com - Intergraph, Huntsville, AL
Statements/opinions are my own, not necessarily Intergraph's.
Bradley K. Sherman
> ..., in his response to Bardley K. Sherman, 'Computer Scientist'
Not sure why you're both misspelling my name and putting my job
title in quotes. Perhaps it has something to do with:
>Example #1.
>My last project was a 12 - 16 thousand dollar computerized fitness testing
>system. It was primarily used by health club employees and other, often less
>than well educated, health and wellness oriented people.
>
>At least once a month I would have to describe what the colon (:) looked like.
>These people were often computer phobic. A large percentage (at least 10%,
>maybe 20%) of the systems were put away when the trained person left. The
>replacement would look at the PC and say 'Ack!'.
This does not scan. Is the lesson here that the owner of the
health club was so gullible that he paid you c. $15k for a
system that served no purpose? That we need to have interfaces
that don't include the ':' character? Or that well-educated
Americans have been rendered so inarticulate by the 30,000
commercials that they absorb every month that when asked to
type a colon they say "Ack!"? Were these the same sort of
people I see around town who drive a car to the fitness club
and then pay to sit on a stationary bicycle in a poorly ventilated
storefront?
It's hard to believe Americans once walked on the moon.
--bks
Michael Sellers
> Sorry guys, but you don't have to *buy* a computer to use one.
> have used a computer before they buy one. There are about 200
> people in this building who use computers, but only a fraction
> own one. My 9 year old daughter and 4 year old son both use
> computers, but neither of them will be buying one for some
> time to come.
On the other hand, neither does having bought a computer mean
that a person will use it. For some odd reason, people who
buy computer products and don't find them useful rarely return
them -- if it was a blender, it'd go right back to the store.
Computers just become doorstops.
I know from my own surveys and observations that many people
in hospital, laboratory, and even regular office environments
have plenty of access to computers, but will not use them --
and the *biggest* reason is because they require too much knowledge
to get started.
Or as someone else put it, "programmers are the reason why
people hate computers." Most people are not programmers and
do not want to be programmers -- and they consider typing
anything more than their name to be "programming."
> ... I am merely reacting to
> the idea that interfaces need to be tested on folks who have
> *never* used a computer to be good. This idea is wrong,
> though prevalent.
Whoah, watch those generalizations, Tex. Up to now, the computer
industry has been extremely insular, selling to a relatively
small group (compared to the whole population) over and over again.
The promise of computers "for the rest of us" has for the most part
not happened. People really *do* have VCRs blinking "12:00" and they
really do by $5000 worth of hardware and software -- and then struggle
to write a short letter (and never even dream of touching the database
software, much less customizing it). The industries I know best --
medical, legal, scientific, and some general office -- are almost
*untouched* by signficant end-user computing. For example, for years
people have proclaimed great breakthroughs in the medical industry at
trade shows and in press releases, but, while the conference floors are
crowded, computers are still not used in diagnosis, review, treatment,
pharmacology, surgery, or follow-up by more than a thin sliver of
the "early adopter" population. The computers are there (and some
are even bought) but people are not using them because the products
available are just too intimidating and/or useless.
The point of this is that there is a great deal of money to be
made in creating *useful* products for the millions of people who
are currently not able or willing to use computers in their work.
Many of these people, while they are intelligent and accomplished,
have little to no knowledge of computers. In such industries, you
are taking an enormous risk if you do not a) figure out who your
users are and b) evaluate your products based on what *they* think --
not what your engineers think they think.
> Interfaces should enable users to do their
> work efficiently with a minimum of frustration.
I'll go further than that: Usable products (and their interfaces)
should increase users' capabilities and enjoyment without increasing
their frustration or cognitive load. (Yeah, that's nearly a direct
quote from a slide in one of my classes. So sue me. :-) )
> Furthermore, the days when you can find people who have
> *never* used a computer by canvassing the folks in the
> supermarket shopping lot are numbered. And I question the
> value of research on such a group.
It all depends who you are trying to sell to. I can walk into
any hospital, large law practice, laboratory, or office, and find
someone who has not used a computer in their work (or probably
at home) in the last five years. I do this frequently. In fact,
I recently ran a usability test for an office product in the
office of a *large* hardware and software manufacturer. We had
no problem finding someone who was skittish around computers
in that environment, and who (by her own description) had little
to no knowledge of computers.
Of course, if you're selling CASE tools it's a different matter...
> I wonder if Toyota evaluates new designs by finding a
> group of subjects who have never driven a car.
I think it's this sort of smugness that gets to me the most. Computer
products today are not as usable as Toyotas -- heck in most areas we're
lousy compared to a Yugo (if you doubt me, go talk to your customer
service people). Somehow, this is conveniently forgotten by most
manufacturers. It's a lot easier to develop products for people who
are just like you, and if you can do that, that's great. However, most
computer products made today (and for the past 20 years) are made for
people whose knowledge and skills differ greatly from the developers,
but under the dangerous illusion that they are, in fact, the same.
--
Michael Sellers New World Designs
"User Interface Analysis, Design, Evaluation, and Training"
sel...@acm.org or sel...@lclark.edu or (503) 538-2745
actum ne agas
Micke Hovmöller
gbo...@atqm.advtech.uswest.com (George Boggs) wrote:
> Might I pointedly remark that this whole discussion about the percentages
> of experienced computer users in any country whatsoever has but the most
> marginal relationship to the selection of test subjects for usability
> testing of any given product?
>
> Humbly, I state Boggs' First Law: Given a software product that could
> predict stock prices 24 hours in advance, prospective users would learn to
> communicate in Sanskrit if necessary to access the application.
This, while being very funny, is only partly true. The reason is that a
novice user don't have the faintest idea what work will be like once he
has learnt the word processor, the spreadsheet, the communications
software, the news reader, the (..., you get the picture), but lazily
asumes that it won't be much different.
Now, I'm not saying lazy is bad, not at all. Laziness is usually cost
effective because you don't do the tedious task but something better
instead. However, the novice MAY be wrong. There may be great improvements
ahead. And the only way for him to find out is through a better interface.
/Micke
--
Micke Hovmöller
mi...@nada.kth.se
IPLab, NADA, KTH
aj
Mr Toxic
K. Sherman) writes:
>Certainly programming a VCR to record a 30 minute show
>at 9:30 PM on Wednesday night and a 1 hour show at
>2:05 AM on Thursday morning should be easier than, say,
>doing your taxes on a general purpose computer.
>VCR's are the classic special purpose programming example.
>It should be really easy to sketch out an interface that
>*someone who has never used a VCR before* can use, right?
Not only can it be done, but I desperately wish that someone would hire me
to do it.
--
Wayne Greenwood
mrt...@aol.com
Bradley K. Sherman
[Lots of good stuff in several replies.]
Okay let's get down to brass tacks.
Certainly programming a VCR to record a 30 minute show
at 9:30 PM on Wednesday night and a 1 hour show at
2:05 AM on Thursday morning should be easier than, say,
doing your taxes on a general purpose computer.
VCR's are the classic special purpose programming example.
It should be really easy to sketch out an interface that
*someone who has never used a VCR before* can use, right?
And it should be on the market real soon now, right?
I say the problem is inherently to complex for a completely
naive user.
I would consider a wager (if I could think of a
way to test it without a grant) along these lines:
There will be no VCR on the market by 1/1/97 that 7 out of
10 consumers will be able to use on that day to successfully
record 30 minutes of TV starting at 9:30 PM on Jan 9th
on channel X and 60 minutes of TV starting at 2:05 AM
on Jan 10th on channel Y. All programming of the VCR to be
started and completed on the 1st and the VCR to
be left unattended after midnight on Jan 1st.
The consumers must *never* have used a VCR or a computer
to be eligible for the test and may not receive
any help other than from the manual accompanying
the VCR (or from built in help functions).
Michael Sellers
> >people hate computers." Most people are not programmers and
> >do not want to be programmers -- and they consider typing
> >anything more than their name to be "programming."
> > ...
>
> of a computer that can do something useful for these users
> --except for the embedded computers in the phone systems,
> ATM's, and cars which they're already using!
That is demonstrably not the case (a nice way of saying "hoo boy, are
you wrong!"). I've helped create functionally complex systems for use
by users who are not at all familiar with computer use, nor are they
willing to type (one test user, an intelligent, accomplished professional,
literally could not find the "x" key on the keyboard).
For a few good examples, see my article (and several others) in the
July issue of _interactions_.
> Why are we
> trying to give general purpose computers to people who will
> never understand the difference between 'or' and 'xor'?
Because the potential exists to increase their capabilities and their
enjoyment of what they do, and (more to the point) because they are
ever more willing to pay for it.
> What I am saying is that interface design should be
> optimized for the users who use the computers frequently
> so as to maximize their productivity. For the rest,
> hide the computer entirely.
That's a good goal, but not practical right now in most cases
(beyond kiosk applications, and even then it's not well hidden).
Interfaces should be optimized for the users. *Sometimes* those
are frequent users, sometimes they're not. I personally believe
there is a bigger market for software which does not require that
you mold yourself to it and that does not assume you are a frequent
user than for the traditional "rtfm" market.
Michael Sellers
> In article <CuDI4...@watserv2.uwaterloo.ca> kel...@healthy.uwaterloo.ca (Catherine L. Kelley) writes:
> >As of 1989, the Canadian General Social Survey data show that 52.6% of
> >the adult population over the age of 15 could not use a computer at all.
> >Broken down by age groups, these data show that 26.3% of people age
> >15-24 could NOT use a computer, 41.8% of those age 25-44 could not use
> >a computer, and a whopping 78.1% of those age 45+ could not use a
> >computer.
>
> So what? What percentage of people cannot use a soldering iron?
> A torque wrench? What percentage cannot type? Cannot set the clock
> on the VCR? Is this some kind of messianic campaign to get everyone
> using a computer?
This is naive. In our society, torque wrenches and soldering irons
have much less effect on what you can do than do computers. Further,
there is *much* more money to be made in *successfully* exposing the
technically illiterate to computer usage than in helping them understand
mono-use tools. Just ask the folks at AOL!
So, is this messianic? Hardly. It is, however, a well-considered
and well-grounded (in technical and financial terms) drive to increase
the number of people happily using computers by making them more usable.
Making better technology alone just won't do it anymore.
> I would bet that 41.8% [sic] of Americans (U.S. type) couldn't
> find Winnipeg on a map if you spotted them Moose Jaw and
> Thunder Bay. Do we need simpler maps, better education,
> or a more motivated Homo sapiens?
If knowledge of Canadian cities were a multi-billion dollar industry,
if people's jobs depended on how well and how much they could get
done with this knowledge, and if people were clamoring away for better
ways to learn and remember these Canadian cities (this is sounding
more and more like a Dave Barry column) then you can bet we'd get
simpler maps and better education. As it is, the comparison is
not very meaningful.
Michael Sellers
In yet another post, Bradley K. Sherman writes:
> >My last project was a 12 - 16 thousand dollar computerized fitness testing
> >system. It was primarily used by health club employees and other, often
> >less than well educated, health and wellness oriented people.
> >
> >At least once a month I would have to describe what the colon (:) looked
> >like. These people were often computer phobic. A large percentage (at
> >least 10%, maybe 20%) of the systems were put away when the trained person
> >left. The replacement would look at the PC and say 'Ack!'.
>
> This does not scan. Is the lesson here that the owner of the
> health club was so gullible that he paid you c. $15k for a
> system that served no purpose? That we need to have interfaces
> that don't include the ':' character? Or that well-educated
> Americans have been rendered so inarticulate by the 30,000
> commercials that they absorb every month that when asked to
> type a colon they say "Ack!"?
You're joking, right? As I hope you understand, the answer is none
of these things. The point of this example (and zillions like it)
is that this system was created essentially by engineers for engineers.
Rather than taking the characteristics of the customer and user into
account, a typical PC interface was apparently dumped into their laps.
The result? First, numerous calls for help. Each call both eroded
customer satisfaction and cost the company time and money. The
second result was, no doubt, fewer systems sold. A person generally
does not recommend a product for which they require continual
technical support, particularly if it's now languishing in a closet.
Think instead what might have happened to the fortunes of this
product and its manufacturer if the user had been taken into account
from the beginning: Faster, more focused development; easier, more
effective demos; increased sales; decreased training and service
costs; decreased returns; increased customer satisfaction and referrals;
and likely increased market share and product longevity. And no,
I am no exagerrating.
> Were these the same sort of
> people I see around town who drive a car to the fitness club
> and then pay to sit on a stationary bicycle in a poorly ventilated
> storefront?
Does that matter? Are we creating effective products only for
those who conform to our idea of a valuable lifestyle?
> It's hard to believe Americans once walked on the moon.
Not when you consider how many Americans actually did versus how
many had to assist them versus how many never really understood how
it was done.
Don Lindner
In a previous article, kel...@healthy.uwaterloo.ca (Catherine L. Kelley) says:
>As of 1989, the Canadian General Social Survey data show that 52.6% of
>the adult population over the age of 15 could not use a computer at all.
>Broken down by age groups, these data show that 26.3% of people age
>15-24 could NOT use a computer, 41.8% of those age 25-44 could not use
>a computer, and a whopping 78.1% of those age 45+ could not use a
>computer.
I would argue that anyone who has ever set a microwave to defrost something
has, for all intents and purposes, used a computer.
The issue seems to be creating a method by which a person can take on a more
complex task with minimal thought.
I've always perceived these devices as an extension of our own thought
processes. Where my skills are weak, I dip into the communal "thought pool"
for an assist.
The problem is, sometimes the "communal thought process" gets a litte
discordant - downright schizo at times. Now, I can wade through a little
maze of conflicting TSR's, etc. just fine - but then, I'm a seasoned user
and a member of MENSA. I understand the processes behind the graphical
metaphors.
The average user, however, does not. I beleive that the lack of continuity
between competing platforms is hindering the development of the field.
*ahem*
Rather than let this devolve into a rant about copyright lawsuits and such,
I'll just shut up now and see if anyone picks up the ball...
--
DA Lindner | The world ends at 10:30!
ad...@lafn.org | Film at Eleven...
Chris Shenefiel
: Microsoft has finally gone off the deep end. I'd like to hear what
: people think about MS's new "Windows 4.0 keyboard" that they are now
: promoting.
: From PC Week (Aug 1, 1994):
: Will new peripheral unlock the key to Windows?
: By Lisa Dicarlo
.....
: Microsoft is developing a modified version of the standard
: 101-key AT keyboard that adds three keys designed to activate
: features within Windows 4.0 and NT.
.....
: The three added keys are an Application key, a Left Windows
: key, and a Right Windows key. The latter two keys will be
: embossed with the Windows logo, according to sources.
Is this really that much different from Apple's daisy key (or Command key).
When this was introduced, the nearest neighbor to it was the Meta key on Lisp
workstations. If users perceive it as useful, it will sell and become a
defacto standard.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Chris Shenefiel
Motorola
My opinions are definately not necessary those of my employer - unequivocally.
Daniel Engelberg
A few general comments on this thread:
1) I think Brad Sherman has a point to the extent that if all
your users are expert, you should focus most of your testing
on experts. And I expect Brad would agree that if all your
users are beginners, it probably makes sense to focus your
testing on users. A considerable number of the above
arguments could probably be avoided if we specified the
target population.
2) A lot of the discussions have centered on the issue of
how many users are computer-illiterate globally. This risks
being irrelevant. The people who use a given product do not
necessarily reflect the global trend. The issue is the
expertise of the population using the specific application.
(See (1))
3) I think many people have needlessly assumed that we will
test exclusively either on beginners or on experts, but not on
both. It seems to me that since most applications will have a
combination of beginners and experts, and since beginners
will become experts, that functionality and testing should
cover both groups. In other words, design an application
flexible enough to accommodate both beginners and experts.
I don't claim to be the first person to come up with this idea.
4) Brad seems to implicitly argue that there is a trade-off
between functionality and usability. I would be interested
to see a proof or disproof of this theory. Case histories
would also be interesting.
-Daniel Engelberg
HCI specialist
Computer Research institute of Montreal
Peter Castine
K. Sherman) wrote:
> "Build a system that even a fool can use, and only a fool will
> want to use it." --/usr/games/fortune
This asinine quote is belongs right up there with "If guns are
outlawed..." and "5,000,000 <adjective> people can't be wrong."
The point of human factors engineering is to develop a system that the
*intended users* can use efficiently and effectively. For many, many
applications, the intended users include people who do not have extensive
previous experience with computers. Hence the legitimate interest in the
needs of computer novices discussed in this newsgroup.
The fact that many software engineers are convinced that ``can't program
in assembler'' == ``imbecilic end user'' exacerbates attempts at
user-oriented software design and leads to statements such as those quoted
above.
Flame off.
I ought to make a collection of quotes that make me see red and put them
in a kill file. :-}
Peter Castine
Sellers) wrote:
Nor when you consider that ~10% of the American population believes
landing on the moon was a hoax.
Peter Castine
> A few general comments on this thread:
> [schnipp]
> 4) Brad seems to implicitly argue that there is a trade-off
> between functionality and usability. I would be interested
> to see a proof or disproof of this theory. Case histories
> would also be interesting.
I wish I had my dissertation on my hard disk here, I wrote several
paragraphs relevant to this point.
My favorite case-in-point is a comparison of two word processors with
which I worked ca. 1983: WordStar and LisaWrite. LisaWrite was easier to
use, *and* provided greater functionality.
(Who's marketing slogan is :``It's that simple.''?)
I suspect there really *is* a trade-off between functionality and
usability. *HOWEVER*, I also suspect that we are nowhere near reaching an
optimal value for the constant in the equation ``usability = c /
functionality''.
Alan Asper
writes:
>Now, I can wade through a little
>maze of conflicting TSR's, etc. just fine - but then, I'm a seasoned user
>and a member of MENSA. I understand the processes behind the graphical
>metaphors.
GOD, I admire you.
Alan
My opinions are not those of my employer.
George Boggs
> What you say is true only in the absence of competition. In a competitive
> marketplace, products quickly evolve nearly identical utility so other
> attributes replace it as differentiators of the products. User interface
> design is one of the most important.
>
> --
> David Fay email: da...@gte.com
Dave, I concede your point with one caveat. Once a product gets
"commoditized", like VCRs, usability is likely to get the second wave of
buyers - but I don't believe it will pry the first wave loose until they
have a bigger reason to change.
ful...@cgsvax.claremont.edu
> da...@gte.com (David Fay) wrote:
>
>> What you say is true only in the absence of competition. In a competitive
>> marketplace, products quickly evolve nearly identical utility so other
>> attributes replace it as differentiators of the products. User interface
>> design is one of the most important.
>
> "commoditized", like VCRs, usability is likely to get the second wave of
> buyers - but I don't believe it will pry the first wave loose until they
> have a bigger reason to change.
There is a dangerous assumption made here that may need to be exanded:
Sometimes, for whatever reason, the better product does not sell. One should
NOT assume that utility = usability = commodity because, while they might seem
highly interactive, they are not rigidly linked.
For example: Sony's development of a higher magtape based video standard
(Beta) was not successful in competetion with the standard developed by the
Victrola Company of Japan (JVC) VHS because in the first few months of the
market competition there were slightly more tapes made in the VHS format than
there were in Beta. And while this effect is unlikly to re-occur for Sony
Corp. (they now own the rights to the world's largest entertainment library and
can choose any standard and be insured of a large number of compatible
commodities) in the late 70's/early 80's the less advanced format showed an
"increasing return" in platform growth (this growth was exagerated by the
existence of "mom and pop" video outlets who had to maintain very low overheads
to insure market share--something that is not likly to happen today but is
still very relevant to the HCI market as currently structured). This
interaction has been explained by Brian Arthur of Stanford and the Santa Fe
Institute.
Within HCI one may look at the same effect in the rapid erosion of WordPerfect
market share and the growth of MS Word (which may be due to differences in
either user interface performance (usability) or to the differences in
inter-application translation utilities (the ability to adapt to different
commodities). Either factor (or a possible third) would make the software
more adaptable in systems of use--but since this analysis has not been done
on these specific markets this is only my opinion. But one must realize
(especially in the HCI market of "user" software) that while the market leader
may only be a new user interface away from the rapid erosion of their market
just providing a new GUI is not enough.
Rodney Fuller
ful...@cgs.edu
Anita Hsiung
>
>I am not advocating that users need to understand the
>difference between a program and a process, or be able to
>explain deadlock, or write an awk script before being granted
>the privilege of using a computer. I am merely reacting to
>the idea that interfaces need to be tested on folks who have
>*never* used a computer to be good. This idea is wrong,
>though prevalent.
There is a reason this idea is prevalent. Like the proverbial
sidewalk that blends smoothly into the street crossing, while
originally conceived for wheelchairs, it is used quite happily by
bicycles, baby carts or anything else with wheels. Good design should
be based on the lowest common denominator.
>Interfaces should enable users to do their
>work efficiently with a minimum of frustration.
Agreed. That's what we're all after. Unfortunately there are many
analysts who have computers (and quite powerful ones) on their desks
and *still* are afraid of using them. They prefer GUIs that don't
require training, contain a minimum of buttons to click and above all,
easy to use. That means designing for someone who DOES NOT KNOW how
to use a computer.
>I wonder if Toyota evaluates new designs by finding a
>group of subjects who have never driven a car.
This analogy doesn't hold. People who buy or have to use computers
have not been licensed and gone through some kind of training in
school. A more proper analogy would be to design a car that both tall
(like my 6'4" coworker) and short (like me, 5'4") people can drive
comfortably, or enough girth for both ectomorphs and endomorphs.
-- Anita --
--
---------------------------------------------------------------------------
| Anita Hsiung | "Look, I'm not my father, I don't do Kung Fu. I'm |
| BBN Canoga Pk, USA | a cop. That's what I am, that's what I do." |
| ahs...@bbn.com | -- Peter Caine |
Jonathan Innes
As for the Toyota comment, a note of human factors relevant lore...the
MR2 is impossible to drive if you are over 6'2" in height. The
engineers designing this baby (nice car) obviously forgot that not
everyone falls into the same anthropomorphic 'specs' as themselves
(persumably shorter than world average japanese men). Toyota builds
nice stuff, but boy could they use a human factors guy!
______________________________________________________________________________
Jon Innes Psychology Department/Computing Research Lab
Box 3452 voice: (505) 646-6254
New Mexico State University fax: (505) 646-6218
Las Cruces, NM 88003 email: jin...@crl.nmsu.edu
USA
--
______________________________________________________________________________
Jon Innes Psychology Department/Computing Research Lab
Box 3452 voice: (505) 646-6254
New Mexico State University fax: (505) 646-6218
Las Cruces, NM 88003 email: jin...@crl.nmsu.edu
USA
Clara N. Fitzgerald
>Andrew Lih (l...@news.cs.columbia.edu) wrote:
>: Microsoft has finally gone off the deep end. I'd like to hear what
>: people think about MS's new "Windows 4.0 keyboard" that they are now
>: promoting.
>: From PC Week (Aug 1, 1994):
>: Will new peripheral unlock the key to Windows?
>: By Lisa Dicarlo
>.....
>: Microsoft is developing a modified version of the standard
>: 101-key AT keyboard that adds three keys designed to activate
>: features within Windows 4.0 and NT.
>.....
>: The three added keys are an Application key, a Left Windows
>: key, and a Right Windows key. The latter two keys will be
>: embossed with the Windows logo, according to sources.
Any information on what they do? The only sense I can make of
left/right window would up/down in the current stack. Application??
>Is this really that much different from Apple's daisy key (or Command key).
>When this was introduced, the nearest neighbor to it was the Meta key on Lisp
>workstations. If users perceive it as useful, it will sell and become a
>defacto standard.
When was the Ctrl key added? I seem to remember (around '84?) seeing
an Apple (not a Macintosh) which had (in the standard Alt positions) open
apple and closed apple keys.
Windows seems to have lost something by requiring that all Alt-letter
and Alt-Shift combinations be available for menu access. The shortcut
combinations are pushed toward 3 keys at a time.
>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>Chris Shenefiel
>Motorola
>My opinions are definately not necessary those of my employer - unequivocally.
--
-Clara A. N. Fitzgerald cfit...@s.psych.uiuc.edu
- < - < -< <> >- > - > -
Help stamp out, reduce, and eliminate redundancy.
gst...@cruzio.com
(Clara N. Fitzgerald) writes about the MS keys on their new keyboard:
> >Is this really that much different from Apple's daisy key (or Command key).
> >When this was introduced, the nearest neighbor to it was the Meta key on Lisp
> >workstations. If users perceive it as useful, it will sell and become a
> >defacto standard.
>
> When was the Ctrl key added? I seem to remember (around '84?) seeing
> an Apple (not a Macintosh) which had (in the standard Alt positions) open
> apple and closed apple keys.
> Windows seems to have lost something by requiring that all Alt-letter
> and Alt-Shift combinations be available for menu access. The shortcut
> combinations are pushed toward 3 keys at a time.
If I remember, all the *personal* computers of the early 80's had command
keys, all Commodores, Apples, Trash-80's, Adams? etc. (not sure about the last
bit there...). In any case, I have an old OmniKey keyboard with room for (I
think) up to 4 *special* keys. This was designed to be the universal keyboard,
and if AT compatible, maybe I'll dig it out. Amazing - the first time in *my*
experience that old computer gear has come back into fashion.
Andrew Lih
Clara N. Fitzgerald <cfit...@s.psych.uiuc.edu> wrote:
>
> When was the Ctrl key added? I seem to remember (around '84?) seeing
>an Apple (not a Macintosh) which had (in the standard Alt positions) open
>apple and closed apple keys.
> Windows seems to have lost something by requiring that all Alt-letter
>and Alt-Shift combinations be available for menu access. The shortcut
>combinations are pushed toward 3 keys at a time.
Windows has not just "lost something". I contend that Microsoft still
just doesn't "get it" when it comes to graphical user design and HFE.
Windows still sticks with the abysmal use of Function keys
and strange triple key combinations with apparently no rhyme
or reason. Arbitrary function key to action mappings are extremely
non-intutive, yet they remain central to the "keyboard shortcuts" in
a so-called modern GUI, such as Windows 3.1.
Windows doesn't need the addition of yet another two modifier keys.
They haven't even used the ones that are currently available in an
effective manner.
How typical of MS to skip quality engineering and "throwing more
hardware" at the problem.
>>Chris Shenefiel
>>Motorola
-fuz
--
`''' Andrew "Fuz" Lih Columbia University
c @@ l...@cs.columbia.edu Mobile Computing Laboratory
\
- <A HREF="http://www.cs.columbia.edu/~lih/">Fuz Page</A>
MikeR...@winpc027.cis076.square-d.com
In article <32uqvi$3...@cs.columbia.edu>, <l...@news.cs.columbia.edu> writes:
> Path:
cisdfl01!meaddata!babbage.ece.uc.edu!news.kei.com!sol.ctr.columbia.edu!news.cs.
columbia.edu!news.cs.columbia.edu!news-not-for-mail
> From: l...@news.cs.columbia.edu (Andrew Lih)
> Newsgroups: comp.human-factors
> Subject: Re: Human factors and MS "Chicago"
> Followup-To: comp.human-factors
> Date: 18 Aug 1994 01:18:42 -0400
> Organization: Columbia University Department of Computer Science
> Lines: 36
> Message-ID: <32uqvi$3...@cs.columbia.edu>
> References: <319hju$c...@coral.bucknell.edu> <31ofj8$o...@cs.columbia.edu>
<32dumg$q...@delphinium.cig.mot.com> <32tj3s$1...@vixen.cso.uiuc.edu>
> NNTP-Posting-Host: cs.columbia.edu
> Summary: MS still doesn't "get it"
I hope you never get a job designing anything!
You are making the assumption that the ONLY
goal of a user interface is to be intuitive. While a highly
commendable goal , there is another goal that should be strived
for - EASE OF USE. Ease of use and intuitiveness are not always
the same thing. There really is no user interface metaphor that
is truly intuitive.
For example, it's easy enough on a macintosh or windows system to print a
document - go to file and select print. However - in "real life" people
don't have a file menu hovering over the heads or their typewriters.
Ease of use though is there because the user knows after a while that all mac
or windows programs have their print option in the same place.
I think both windows and macintosh drop the ball on their use of function
keys. As long as the functions keys are CONSISTENTLY used for the same
function it could make navigation easier and quicker for some users.
At one point earlier in life I answered tech support calls for a large
well known computer service and you would be suprised how some people can't
seem to grasp pressing two keys at once, let alone giving their ibm's a three
finger salute (ctrl-alt-del). So for these users - pressing F1 (or whatever is
decided on as a standard) for print is easier than pressing command-P.
Andrew Lih
<MikeR...@winpc027.cis076.square-d.com> wrote:
>
>In article <32uqvi$3...@cs.columbia.edu>, <l...@news.cs.columbia.edu> writes:
>>
>> Windows still sticks with the abysmal use of Function keys
>> and strange triple key combinations with apparently no rhyme
>> or reason. Arbitrary function key to action mappings are extremely
>> non-intutive, yet they remain central to the "keyboard shortcuts" in
>> a so-called modern GUI, such as Windows 3.1.
>>
>
I've led GUI projects at a Large Telco Company(TM) so your hopes have
been dashed. :) But don't just take my word for it, consult texts
(such as Tognazzini, Laurel, Baecker/Buxton) and institutions that
have done research in this area.
>You are making the assumption that the ONLY goal of a user interface
>is to be intuitive. While a highly commendable goal , there is
>another goal that should be strived for - EASE OF USE. Ease of use
>and intuitiveness are not always the same thing.
In fact, intuitiveness and ease-of-use are two distinctly different things.
>There really is no user interface metaphor that is truly intuitive.
Define "truly intutitve." If you are resigned to the fact that no
metaphor can be intutive, then why bother designing at all? I think
I'm beginning to understand why you don't mind confusing users with
arbitrary function-to-integer mappings, as that's what Function keys do!
> For example, it's easy enough on a macintosh or windows system to
> print a document - go to file and select print. However - in
> "real life" people don't have a file menu hovering over the heads or
> their typewriters.
I haven't used a typewriter in years. Want to know why? Because I
now have this nifty tool called a computer that has a mouse pointer
and menu present ALL THE TIME. Introduce a metaphor to a person, tell
them that it will ALWAYS be there, and they'll understand and utilize it.
Sidebar: Do you think back in the 1800's, when typewriters were
introduced, the key layout was "easy to use"? Quite the contrary as
they were laid out to slow typists down to prevent jamming the
presses. The only reason why you feel the typewriter is an familiar
metaphor is not because of any inherent qualities, but merely because
it has been in our society for the last 100 years and is now
ubiquitous. GUIs will be ubiquitous in 10 years, but hopefully people
will put some good engineering and design into them.
UI designers will be the first ones to declare that GUIs are not meant
TO REPLACE documentation and training. GUIs are meant to reduce
training time; reduce the dependency on rote memorization of commands;
and provide a consistent, intuitive presentation so that skills
obtained in one area are readily transferrable to others. Examples
include: Shift-clicking for multiple non-contiguous selections
(Macintosh) and pushpins for "sticking things" on the desktop
(OpenLook). No one could possibly master the latter two skills
without any training, but after learning about them it's unlikely they
will ever forget them because they are either natural extensions of an
existing skill (clicking), or take on the appearance of a familiar
physical object (pushpin).
Now likewise, try telling your trainee for the zillionth time: "You
idiot, I've already told you countless times that F6 was for
multiple-selection!!! Why can't you get that into your head?!"
> Ease of use though is there because the user knows after a while
> that all mac or windows programs have their print option in the same
> place. I think both windows and macintosh drop the ball on their
> use of function keys.
Drop the ball? Perhaps you haven't thought about it very carefully.
Have you considered the usefulness of mnemonics? On the Macintosh:
Command-P means (P)rint. Command-Q means (Q)uit. Command-S means
(S)earch. Command-A means Select (A)ll.
I'll meet you anyday for a shootout where you map those same commands
to function keys, and see how the users perform. :) And don't even
think of mentioning those cute little templates that fit over the
keyboard. They only reinforce the non-intuitiveness of using function
keys.
>At one point earlier in life I answered tech support calls for a
>large well known computer service and you would be suprised how some
>people can't seem to grasp pressing two keys at once, let alone
>giving their ibm's a three finger salute (ctrl-alt-del). So for these
>users - pressing F1 (or whatever is decided on as a standard) for
>print is easier than pressing command-P.
While I can sympathize with newbies, I think it's a matter of proper
10 minute training. In 30 seconds, you can teach a person to hold
down two keys at once. Hey, let's even give him/her 5 minutes to
practice that skill. He/she will have that skill down cold for the
duration of their life, barring any physical handicap. As a
consultant for 3 years, I've taught people how to use a Macintosh and
have never found anyone incapable of this skill.
FJ!!
>: At one point earlier in life I answered tech support calls for
>: a large well known computer service and you would be suprised
>: how some people can't seem to grasp pressing two keys at once,
>: let alone giving their ibm's a three finger salute (ctrl-alt-del).
>: So for these users - pressing F1 (or whatever is decided on as a
>: standard) for print is easier than pressing command-P.
>of course, these users are easy to identify by the absence of any
>uppercase letters in the text they type.
Two words: Caps Lock.
Sure, it's silly. But you can type upper-case without ever having to press
two keys at the same time.
FJ!!
Magnus Ramage
: In article <32uqvi$3...@cs.columbia.edu> l...@news.cs.columbia.edu (Andrew Lih) writes:
: >Windows still sticks with the abysmal use of Function keys
: >and strange triple key combinations with apparently no rhyme
: >or reason. Arbitrary function key to action mappings are extremely
: >non-intutive, yet they remain central to the "keyboard shortcuts" in
: >a so-called modern GUI, such as Windows 3.1.
: Some of the function key mappings are non-intuitive. But Windows
: and Pres. Mgr. do offer the mnemonic type of menu access that's not
: found in Mac. (They all have roughly equivalent acclerators though).
: Mnemonics are the underlined letters you see on menus.
: Pressing that keyboard key invokes that menu item. Advantages:
: 1. A glance at the menu bar will tell you what key you need to press
: to open the File menu (for example), at which point you can browse
: the items on that menu & choose one with a keystroke. Accelerators
: don't support this type of browsing.
: 2. Your chance of finding menomnic access to a menu item is
: greater with mnemonics than with accelerators, since in practice
: items almost always have mnemonics, but often don't have
: accelerators. If you've ever experienced the frustration of having
: to do a repetitive operation on the Mac using menu items without
: accelerators, you'll appreciate this.
: The tear-off menus ameliorate this somewhat.
: --------------------------------------------------------------------
: Tim Shea Internet: sh...@marcam.com Phone (USA): (617) 965-0220
: --------------------------------------------------------------------
W.P. Fleischmann
MikeR...@winpc027.cis076.square-d.com writes:
> a large well known computer service and you would be suprised
> how some people can't seem to grasp pressing two keys at once,
> let alone giving their ibm's a three finger salute (ctrl-alt-del).
> So for these users - pressing F1 (or whatever is decided on as a
> standard) for print is easier than pressing command-P.
of course, these users are easy to identify by the absence of any
uppercase letters in the text they type.
>>bill
W.P. Fleischmann Internet: fle...@umich.edu
Dept of Anesthesiology CIS: 76256,252
University of Michigan
Ann Arbor, Mi 48109-0704
Tim Shea
>Windows still sticks with the abysmal use of Function keys
>and strange triple key combinations with apparently no rhyme
>or reason. Arbitrary function key to action mappings are extremely
>non-intutive, yet they remain central to the "keyboard shortcuts" in
>a so-called modern GUI, such as Windows 3.1.
Some of the function key mappings are non-intuitive. But Windows
Clara N. Fitzgerald
> Clara N. Fitzgerald <cfit...@s.psych.uiuc.edu> wrote:
> >
> > Windows seems to have lost something by requiring that all Alt-letter
> >and Alt-Shift combinations be available for menu access. The shortcut
> >combinations are pushed toward 3 keys at a time.
<MikeR...@winpc027.cis076.square-d.com> wrote:
>
>In article <32uqvi$3...@cs.columbia.edu>, <l...@news.cs.columbia.edu> writes:
>>
>> Windows still sticks with the abysmal use of Function keys
>> and strange triple key combinations with apparently no rhyme
>> or reason. Arbitrary function key to action mappings are extremely
>> non-intutive, yet they remain central to the "keyboard shortcuts" in
>> a so-called modern GUI, such as Windows 3.1.
In fact, intuitiveness and ease-of-use are two distinctly different things.
>There really is no user interface metaphor that is truly intuitive.
> For example, it's easy enough on a macintosh or windows system to
> print a document - go to file and select print. However - in
> "real life" people don't have a file menu hovering over the heads or
> their typewriters.
You seem to be assuming that life is intuitive. Learning in real life
and learning about a computer (which, it could be argued, is
part of real life) aren't totally different processes. We have more
practice at life.
UI designers will be the first ones to declare that GUIs are not meant
TO REPLACE documentation and training. GUIs are meant to reduce
training time; reduce the dependency on rote memorization of commands;
and provide a consistent, intuitive presentation so that skills
obtained in one area are readily transferrable to others.
[argument on favor of functions keys]
> Ease of use though is there because the user knows after a while
> that all mac or windows programs have their print option in the same
> place.
Have you considered the usefulness of mnemonics? On the Macintosh:
Command-P means (P)rint. Command-Q means (Q)uit. Command-S means
(S)earch. Command-A means Select (A)ll.
Don't forget the Edit menu. Although the shortcuts aren't mnemonics,
the keys are well placed (X [cut] C [copy] V [Paste] ), in the same order
as the menu items, and consistent across all Mac programs I have ever
seen. I regularly use Windows programs with at least 4 unique mappings
for these functions, one set of which is triple key strokes with Insert, etc.
I very seldom use these Window shortcuts, but usually use the Mac
versions (I use both computers (plus vi in UNIX) regularly, the Mac
least frequently).
>At one point earlier in life I answered tech support calls for a
>large well known computer service and you would be suprised how some
>people can't seem to grasp pressing two keys at once, let alone
>giving their ibm's a three finger salute (ctrl-alt-del). So for these
>users - pressing F1 (or whatever is decided on as a standard) for
>print is easier than pressing command-P.
{ [editorial paste]
> >of course, these users are easy to identify by the absence of any
> >uppercase letters in the text they type.
or: "OF COURSE THESE USERS ARE EASY TO IDENTIFY BY THE
ABSENCE OF MIXED-CASE TEXT IN WHAT THEY TYPE."
My dad used to do this. He is improving!
Are you really suggesting that these people do not understand how
the shift key works, even if they don't (have to) use it?
>>Bill
W.P. Fleischmann Internet: fle...@umich.edu }
--
`''' Andrew "Fuz" Lih Columbia University
c @@ l...@cs.columbia.edu Mobile Computing Laboratory
Hope the attributions are clear...
Clara N. Fitzgerald
>In article <32uqvi$3...@cs.columbia.edu> l...@news.cs.columbia.edu (Andrew Lih) writes:
>>Windows still sticks with the abysmal use of Function keys
>>and strange triple key combinations with apparently no rhyme
>>or reason. Arbitrary function key to action mappings are extremely
>>non-intutive, yet they remain central to the "keyboard shortcuts" in
>>a so-called modern GUI, such as Windows 3.1.
>Some of the function key mappings are non-intuitive. But Windows
>and Pres. Mgr. do offer the mnemonic type of menu access that's not
>found in Mac. (They all have roughly equivalent acclerators though).
There must be a way to distinguish remembering Control-P for
print, and looking up the underlined F displayed in the menu bar.
>Mnemonics are the underlined letters you see on menus.
After you've pressed Alt-<letter> to drop the menu,
>Pressing that keyboard key invokes that menu item. Advantages:
>1. A glance at the menu bar will tell you what key you need to press
>to open the File menu (for example), at which point you can browse
>the items on that menu & choose one with a keystroke. Accelerators
>don't support this type of browsing.
Yes, you would actually have to use the mouse.
Disadvantage: at least two keystrokes are required to actually
perform an action (unless there are those button things on the
menu, which short-circuit the status bar help...)
Windows is supposed to support keyboard access to all programs
(graphics-based ones excepted). Some designers do not take
care to actually do this. (Sharkmail is a great example; I can
do everything by keyboard, but must do Shift-Tab to get around
the main edit box, and must Tab to the Cancel button because
Alt-C goes to the Cc field. The other buttons respond to Alt-
<first letter>, but there is no sign that they should. Ick) If it
is not enforced, MS should give in and insist that all users
get mice, and use the keyboard equivalences for something else.
As a designer: You have to insure that all the items have
unique letters. This can involve underlining strange places
(C_o_py) or using less descriptive names for actions.
>2. Your chance of finding menomnic access to a menu item is
>greater with mnemonics than with accelerators, since in practice
>items almost always have mnemonics, but often don't have
>accelerators. If you've ever experienced the frustration of having
>to do a repetitive operation on the Mac using menu items without
>accelerators, you'll appreciate this.
True; mac should allow more user-customizable menus (as in MSWord,
I believe) where items can be added or removed, and shortcuts
assigned.
>The tear-off menus ameliorate this somewhat.
I once heard about something called a "mole", basically a mouse
device operated by foot. (A pressure sensitive joystick might work
better (reduce dirt problems) ... I'd buy one.) Wouldn't that help
immensely?
FJ!!
>In article <CuqpE...@cwi.nl> FJ!!, f...@cwi.nl writes:
>> W.P. Fleischmann <fle...@umich.edu> writes:
>> >In article <32vo7s$h...@cisdfl01.cis076.square-d.com> ,
>> >MikeR...@winpc027.cis076.square-d.com writes:
>> >: how some people can't seem to grasp pressing two keys at once,
>> >of course, these users are easy to identify by the absence of any
>> >uppercase letters in the text they type.
>>
>> Two words: Caps Lock.
>Are you really suggesting that these people do not understand how
>the shift key works, even if they don't (have to) use it?
I was just pointing out that it is possible to be able to type
mixed-type without ever having to press two keys at once. As an
excercize, I'm doing so right now, it involves a lot of pressing of the
caps-lock key for a single letter, and I'm glad I'm not typing German
where every noun starts with a capital. I think I can imagine a novice
user that has been plonked behind a keyboard for the first time and
indeed may have problem grasping the concept of modifier keys, which is
what Mike talked about. And if these user know the caps-lock key, they
probably still could type in mixed-type.
I guess my answer to your question is `yes'.
I mean, come to think of it, where can we identify instances where
people routinely press a modifier button at the same time as a certain
button, outside of life behind the keyboard? You never have to press
two keys at the same time on ordinary telephones. The `shift' key on my
Casio calculator has to be pressed _and_released_ before I should press
the key I want the special function from. Some VCRs need simultaneous
presses, but we've already extablished they are bad examples of
usability anyway. I don't think I have ever seen them on ATMs or
automats for food or tickets for public transport.
So, let's see where would novice users have learned about modifier keys
before they come to a keyboard? Can anyone here come up with something?
Some routine appliance I'm blanking on? Maybe modifier keys that have
to be pressed at the same time as the real key are common occurances in
the US on microwaves or vacuums or something? I really can't think of
any instance of it in my .nl domain.
Maybe "simultaneous-press modifier keys" are a bad idea from a
novice-user's familiarity point of view. Which is not to say that they
are unlearnable of course, just that we may have to be sensistive that
they are a certain interface construct that has to be learned by a user
that comes up to a keyboard for the first time. That includes users of
MS Windows or Chicago too.
FJ!!
Glen Kohler
: W.P. Fleischmann <fle...@umich.edu> writes:
: >In article <CuqpE...@cwi.nl> FJ!!, f...@cwi.nl writes:
: >> W.P. Fleischmann <fle...@umich.edu> writes:
: >> >In article <32vo7s$h...@cisdfl01.cis076.square-d.com> ,
: >> >MikeR...@winpc027.cis076.square-d.com writes:
: >> >: how some people can't seem to grasp pressing two keys at once.
: >Are you really suggesting that these people do not understand how
: >the shift key works, even if they don't (have to) use it?
: So, let's see where would novice users have learned about modifier keys
: before they come to a keyboard? Can anyone here come up with something?
: Some routine appliance I'm blanking on? Maybe modifier keys that have
: to be pressed at the same time as the real key are common occurances in
: the US on microwaves or vacuums or something? I really can't think of
: any instance of it in my .nl domain
Since 1906 or so people have been hitting the <shift> key to create
upper-case letters....
Defining "routine appliance" is tricky in this context, too. If it is
meant to refer to washing machines or vacuum cleaners, it is meaningless
in the context of information-handling tools. In the realm of computing,
multiple-key commands are routine.
--
Glen Kohler, CMT
Health Arts Wellness Services
P.O. Box 9719 Berkeley, CA 94709
510-845-8485
FJ!!
>Since 1906 or so people have been hitting the <shift> key to create
>upper-case letters....
>In the realm of computing, multiple-key commands are routine.
I'm fully aware of that. I was just wondering if people who are new to
that realm would have encountered something analogous to hitting
modifier keys ever before during their course of their daily lives.
FJ!!
Jeff Scott
>W.P. Fleischmann <fle...@umich.edu> writes, arguing that modifier keys
are a computer console specific phenomena:
>I mean, come to think of it, where can we identify instances where
>people routinely press a modifier button at the same time as a certain
>button, outside of life behind the keyboard? You never have to press
>two keys at the same time on ordinary telephones.
...
other examples of things with buttons that don't
require simultaneous presses deleted
...
>So, let's see where would novice users have learned about modifier keys
>before they come to a keyboard? Can anyone here come up with something?
>Some routine appliance I'm blanking on? Maybe modifier keys that have
>to be pressed at the same time as the real key are common occurances in
>the US on microwaves or vacuums or something? I really can't think of
>any instance of it in my .nl domain.
Here it is!!!
PIANOS!!!!! They've all got sustain pedals which you have to press at the
same time to modify the sound! Some even have sostenuto pedals. Some
old harpsichord had dozens of foot-pedals and/or knee-pedals that worked
similarly.
VIOLINS!!! You've got to move that bow across the string *at the same time*
as moving your fingers along that fret board which doesn't even have
easy to find default settings.
GUITARS! Pluck strings with one hand, select notes to play in chord with other,
modifying the frequency of the one hands pluck.
Viola de gamba, flute (here it's a combination of hands and mouth movements),
clarinet, hammer and nail, rifle (one hand aims, the other shoots. Have
to do both simultaneously.).
See we have two hands, so it's an _ancient_ practice to design tools that
allow the user to be more efficient by using both hands at once. Makes tools
easier to use and is very intuitive. Everyone uses tools that require
"simultaneous-press modifier keys". Tools that circumvent this are not well
used. Think of your new design that does not allow the shift key to capitalize
unless pressed separately. Sorry, it fails in the market. A new device for
holding nails in place so you don't have to hold them and risk hurting
your hands. Yes, such devices exist but few use them as it's easier and
faster to risk banging your thumb.
>Maybe "simultaneous-press modifier keys" are a bad idea from a
>novice-user's familiarity point of view.
I'm sure you can tell I disagree with this.
> Which is not to say that they are unlearnable of course,
Well of course not.
>just that we may have to be sensistive that
>they are a certain interface construct that has to be learned by a user
>that comes up to a keyboard for the first time.
I think it's a very very familiar idea and that it really really should be
very very obvious to anyone once they see it once. Of course, I could be
wrong as I often am. But have you really noticed a lot of people complete
novices having serious difficulty when this concept?
I do think that there are some tricky parts to it. The tricky part is that
the stupid ESCAPE key is a modifier that has to be pressed separately but
often has a very similar function to that of the SHIFT, CONTROL, ALTERNATE
and COMMAND keys. I do think it would help to print these keys in a different
colour. On my keyboard, these keys are twice as wide as normal keys so that
is some hint, but the BACKSPACE key is double width too. Perhaps the problem
is that they look too similar to keys that aren't simultaneous-press
modifier keys. OK, I see your point of view. On pianos and such, the modifier
action is fundamentally different in character from the non-modifier action,
so it is more obvious that there is a difference. With a keyboard, even if you
watch someone typing, you may not realize that there is a difference. I
believe that printing modifier keys in a different colour would help the
situation a little bit. OK, I agree with both of you.
- Jeff
--
,_. /)/) /) _|_|_ | jsc...@ucsd.edu "Their feet were straight feet;
| _ |-|- _\ _ _ | | | and the sole of their feet was like the sole of a
(_|(/,| | \_)(_(_)| | | a calf's foot; and they sparkled like the colour
| of burnished brass." - Ezekiel 1:7
FJ!!
>fjv...@cs.vu.nl (FJ!!) writes:
>>So, let's see where would novice users have learned about modifier keys
>>before they come to a keyboard? Can anyone here come up with something?
>Here it is!!! PIANOS!!!!! [...] VIOLINS!!! [...] GUITARS!
Good ones, I hadn't thought of those. Now tell me, how fast do novices
become proficient at using these tools? How quickly are users
comfortable managing the modifiers on these devices?
>I think it's a very very familiar idea and that it really really should be
>very very obvious to anyone once they see it once. Of course, I could be
>wrong as I often am. But have you really noticed a lot of people complete
>novices having serious difficulty when this concept?
I never gave it a second thought either, until I saw:
Article <32vo7s$h...@cisdfl01.cis076.square-d.com>,
(Subject: Human factors in computing),
<MikeR...@winpc027.cis076.square-d.com> wrote:
>At one point earlier in life I answered tech support calls for a
>large well known computer service and you would be suprised how some
>people can't seem to grasp pressing two keys at once, let alone
>giving their ibm's a three finger salute (ctrl-alt-del). So for these
>users - pressing F1 (or whatever is decided on as a standard) for
>print is easier than pressing command-P.
It is when I read this article that I started thinking about just how
intuitive or familiar modifier keys are, and I just couldn't come up
with examples of where a user who was new to the typing realm would
have encountered them before. You came up with musical instruments. Ok,
that is valid, many people learn how to play instruments early in life,
and these people would probably understand modifier keys. Any other
places I am overlooking?
FJ!!
Magnus Ramage
: It is when I read this article that I started thinking about just how
: intuitive or familiar modifier keys are, and I just couldn't come up
: with examples of where a user who was new to the typing realm would
: have encountered them before. You came up with musical instruments. Ok,
: that is valid, many people learn how to play instruments early in life,
: and these people would probably understand modifier keys. Any other
: places I am overlooking?
I think you're missing the point, in that most people this century have got to
know how to type fairly young, and are thus aware of the shift key from that
domain. And most kids seem to at least get word-processor training in schools
nowadays.
Now, I agree that this is a two-key operation _for data entry processes_
rather than control processes, but it goes part of the way surely. The problem
seems the general concept of controlling the system.
Another thought on modifier keys: at least they offer a limited modality with
a clear visibility of that modality (your finger's on the button!) Caps-lock
keys are much more modal, and that modality is not necessarily so visible.
(PCs have caps and num-lock lights, I know, but other computers don't.)
Magnus Ramage
School of Cognitive & Computing Sciences,
University of Sussex,
Brighton, BN1 9QH, England
(mag...@cogs.sussex.ac.uk)
Mr Toxic
writes:
<much about modifier keys deleted>
>>Some routine appliance I'm blanking on? Maybe modifier keys that have
>>to be pressed at the same time as the real key are common occurances in
>>the US on microwaves or vacuums or something? I really can't think of
>>any instance of it in my .nl domain.
>Here it is!!!
>PIANOS!!!!! They've all got sustain pedals which you have to press at the
>same time to modify the sound! Some even have sostenuto pedals. Some
>old harpsichord had dozens of foot-pedals and/or knee-pedals that worked
>similarly.
>VIOLINS!!! You've got to move that bow across the string *at the same
time*
>as moving your fingers along that fret board which doesn't even have
>easy to find default settings.
>GUITARS! Pluck strings with one hand, select notes to play in chord with
other,
>modifying the frequency of the one hands pluck.
The obvious oversight here is that it takes days for a new user to get
usable sounds out of these instruments, months to sound decent, and years
to achieve any sort of mastery. I certainly wouldn't spend the hundreds of
thousands of hours I spent learning the guitar and other instruments on a
spreadsheet or a database.
p.s. Is the saxophone a Rube Goldberg device, or what? Those things remind
me of a game I had as a kid called "Mouse Trap".
--
Wayne Greenwood
mrt...@aol.com
Tony Bass
[... interesting discussion]
> It is when I read this article that I started thinking about just how
> intuitive or familiar modifier keys are, and I just couldn't come up
> with examples of where a user who was new to the typing realm would
> have encountered them before. You came up with musical instruments. Ok,
> that is valid, many people learn how to play instruments early in life,
> and these people would probably understand modifier keys. Any other
> places I am overlooking?
Sometimes there is a double action to discourage something dangerous,
interlocked control to record on an audio tape recorder
child-proof caps on medicine bottles
A keyboard with all the keys looking much alike is probably uniform
enough not to trigger this association normally.
Tony Bass
--
# Tony Bass Tel: +44 473 645305
# MLB 3/19, BT Laboratories e-mail: a...@saltfarm.bt.co.uk
# Martlesham Heath, Ipswich, Suffolk, IP5 7RE DO NOT e-mail to From: line
# Opinions are my own
W.P. Fleischmann
> I'm fully aware of that. I was just wondering if people who are
> new to that realm would have encountered something analogous to
> hitting modifier keys ever before during their course of their
> daily lives.
In my car, the clutch for my transmission, and the ignition key
release button on my steering column (I could live without that
one).
On older car doors, holding in the button to ensure that the door
stays locked when you close it from the outside.
On my alarm clock, the different buttons that I must keep pressed
to determine whether I'm changing the current time or alarm setting.
On my cassette recorder, the red button that I have to press along
with "record" so that I don't accidentally write over a tape.
>>Bill
W.P. Fleischmann Internet: fle...@umich.edu
Peter Castine
> > I'm fully aware of that. I was just wondering if people who are
> > new to that realm would have encountered something analogous to
> > hitting modifier keys ever before during their course of their
> > daily lives.
In article <33apus$p...@lastactionhero.rs.itd.umich.edu>, W.P. Fleischmann
<fle...@umich.edu> replied with several examples, most of which are
less-than-optimal designs (IMHO):
(BTW, the following topics are going off the thread, for which I
apologize. However, I'm hoping they may provoke some reasoned discussion
about design from a more general point of view.)
(Note that W.P.'s examples do address the previous question adequately.)
> In my car, the clutch for my transmission, and the ignition key
> release button on my steering column (I could live without that
> one).
The actions required to get a car with manual transmission to move
backwards (or forwards, for that matter) can only be described as baroque.
This was a design made before technology was capable of something more
straight-forward (such as automatic).
Question: is there a better way of specifying forwards/backwards than the
standard automatic transmission? I think the current solution is pretty
good. (Irony of it all: I still drive a car with manual transmission.)
> On older car doors, holding in the button to ensure that the door
> stays locked when you close it from the outside.
This one is actually not such a bad idea, since you want to avoid locking
your keys in the car. Good reason to make closing a locked door difficult.
(NB: I have nonetheless managed to lock my keys in the car--my solution is
to *never* close the driver's door with the lock button down and always
lock up with the key.)
> On my alarm clock, the different buttons that I must keep pressed
> to determine whether I'm changing the current time or alarm setting.
I've not yet found a digital clock with a means for setting time and alarm
that I really liked. That old wind-up alarm clock was not bad--twist one
knob for time, one knob for alarm. Why did they have to go in opposite
directions? Maybe that decreased ``mode errors''? Still, I like not having
to wind the clock up every morning :-)
> On my cassette recorder, the red button that I have to press along
> with "record" so that I don't accidentally write over a tape.
Some recorders have a single record button that you have to press *and
shift* (to the left or right) to start recording. The gesture is easier,
but you're still not likely to record by accident.
Cheers,
--
Peter Castine | C (n.): A programming language that is sort of
pcas...@prz.tu-berlin.de | like Pascal except more like assembly except
Process Control Center | that it isn't very much like either one, or
Technical University Berlin | anything else. -- Ray Simard
Scott Francis
MikeR...@winpc027.cis076.square-d.com wrote:
>
> In article <32uqvi$3...@cs.columbia.edu>, <l...@news.cs.columbia.edu> writes:
> > Path:
> > the same thing. There really is no user interface metaphor that
> is truly intuitive.
What about the nipple?
Ralph Betza
To open a common door, you need to turn the knob and push/pull
simultaneously. (This confusing interface has been improved,
but old releases of the product are still in use.)
Presidents of the United States are expected to be able to
walk and chew gum at the same time. :-)
Nigel Tzeng
Peter Castine <pcas...@prz.tu-berlin.de> wrote:
[snip]
>
>In article <33apus$p...@lastactionhero.rs.itd.umich.edu>, W.P. Fleischmann
><fle...@umich.edu> replied with several examples, most of which are
>less-than-optimal designs (IMHO):
>
>(BTW, the following topics are going off the thread, for which I
>apologize. However, I'm hoping they may provoke some reasoned discussion
>about design from a more general point of view.)
>(Note that W.P.'s examples do address the previous question adequately.)
>
>> In my car, the clutch for my transmission, and the ignition key
>> release button on my steering column (I could live without that
>> one).
>
>The actions required to get a car with manual transmission to move
>backwards (or forwards, for that matter) can only be described as baroque.
>This was a design made before technology was capable of something more
>straight-forward (such as automatic).
>
>Question: is there a better way of specifying forwards/backwards than the
>standard automatic transmission? I think the current solution is pretty
>good. (Irony of it all: I still drive a car with manual transmission.)
Well...a manual transmission has a cruddier user interface but allows
the user complete of control of when and how to shift...leading to
either better gas milage or performance (when you do it right).
The only automatic I've used (and somewhat liked) is the one on the
Mitsu Eclipse. In "performance" mode the shift points are changed and
when you stomp on the gas it seems to optimize acceleration (shifts
near redline to keep the car in the optimal range after the shift).
>
>> On older car doors, holding in the button to ensure that the door
>> stays locked when you close it from the outside.
>
>This one is actually not such a bad idea, since you want to avoid locking
>your keys in the car. Good reason to make closing a locked door difficult.
>(NB: I have nonetheless managed to lock my keys in the car--my solution is
>to *never* close the driver's door with the lock button down and always
>lock up with the key.)
>
My solution was to get the dealer to include an alarm system and I
press a single button as I walk away :) I did this because the acura
doesn't lock except with the key...which I find clumbersome...and I
can't lock myself out (unless the battery dies AND I've locked the
keys inside).
[snip]
>Peter Castine | C (n.): A programming language that is sort of
Nigel tzeng
Alan Asper
ni...@access1.digex.net writes:
>
>Well...a manual transmission has a cruddier user interface but allows
>the user complete of control of when and how to shift...leading to
>either better gas milage or performance (when you do it right).
I like to think that if there is one system for which
we do not want a walk-up-and-use (or sit-down-and-
use) interface, it's an automobile. Hence my preference
for manual-transmission cars, and a loathing of
cruise control; few things are worse than getting stuck
behind Fred and Ethyl MasonicLodge driving/coasting
their pale lemon Sedan de Ville at a steady 64 mph
in the left lane during rush hour, oblivious to
anything save the mellifluous tones of Paul Harvey and
The Rest of the Story: "And that little boy that nobody
liked grew up to be... Bob Dole."
Sorry, but you've never driven in Missouri, I take it.
Alan
My opinions are not those of my employer.
MikeR...@winpc027.cis076.square-d.com
> I think you're missing the point, in that most people this century have got
to
> know how to type fairly young, and are thus aware of the shift key from that
> domain. And most kids seem to at least get word-processor training in schools
> nowadays.
>
Yes - kids have been familiar with computer keyboards since early in life
nowadays - but guess what - there are a large number of ADULTS out there
that haven't especially among the lower economic class. The higher
the income and younger the person , the more likely they are to have used
pc's but to think that EVERYONE is familiar with a computer keyboard
is jusy plain wrong.
I never said to get rid of modifier keys. I DID say some people don't
grasp the concept quickly. This is especially true under windows where
sometimes the modifier key is the alt key and sometimes its the control
key. If you don't believe this is true then go monitor tech support calls
for a few months. You base your opinions on what you have read in your little
textbook - I am basing my opinions on the real world
MikeR...@winpc027.cis076.square-d.com
In article <Scott.Francis...@198.99.205.138>,
<Scott....@SuperMac.com> writes:
> Newsgroups: comp.human-factors
> Path:
cisdfl01!meaddata!swiss.ans.net!solaris.cc.vt.edu!news.duke.edu!MathWorks.Com!y
eshua.marcam.com!charnel.ecst.csuchico.edu!olivea!decwrl!svc.portal.com!portal!
huey!NewsWatcher!user
> From: Scott....@SuperMac.com (Scott Francis)
> Subject: Re: Human factors and MS "Chicago"
> Followup-To: comp.human-factors
> Sender: ne...@supermac.com
> Nntp-Posting-Host: 198.99.205.138
> Organization: SuperMac
> References: <32uqvi$3...@cs.columbia.edu>
<32vo7s$h...@cisdfl01.cis076.square-d.com>
> Date: Tue, 23 Aug 1994 20:24:36 GMT
> Lines: 11
I'm not sure if it's intuitive but I definitely like the design
Craig A. Finseth
Actually, is isn't intuitive. The head-turning and sucking reaction
is _hardwired_ into a newborn's system. This wiring self-destructs
over a couple of months, and is replaced by _learned_ behavior. (If
it didn't self-desctruct, you'd turn your head and try to suck
whenever your cheek was brushed.)
A newborn is learning so much and so fast, and starting from such a
small experience base, that it is not clear that a concept such as
"intuitive" can even be applied in this situation.
Craig
eddy Gorsuch
Alan Asper <aas...@sbctri.sbc.com> wrote:
[...]
>cruise control; few things are worse than getting stuck
>behind Fred and Ethyl MasonicLodge driving/coasting
>their pale lemon Sedan de Ville at a steady 64 mph
>in the left lane during rush hour, oblivious to
You can drive 64 mph (and consider it slow) in rush hour?!?! Some
people are spoiled and don't even know it. Around here, any movement
at all is considered a good speed during rush hour.
eddy
--
ed.dy \'ed-e-\ n [ME (Sc dial.) ydy, prob. fr. ON itha; akin to OHG ith-
again], L et and 1a: a current of water or air running contrary to the main
current; esp)X : a small whirlpool 1b: a substance moving similarly 2: a
contrary or circular current - eddy vb
Mr Toxic
Finseth) writes:
>Actually, is isn't intuitive. The head-turning and sucking reaction
>is _hardwired_ into a newborn's system. This wiring self-destructs
>over a couple of months, and is replaced by _learned_ behavior. (If
>it didn't self-desctruct, you'd turn your head and try to suck
>whenever your cheek was brushed.)
>A newborn is learning so much and so fast, and starting from such a
>small experience base, that it is not clear that a concept such as
>"intuitive" can even be applied in this situation.
Yes. Intuition is often defined as 'instinct', which it is not. Intuition
is based on experience and exposure to culture. Instinct is why dogs walk
in circles before they curl up for a nap. The only software that will ever
be instinctive will be CaveSoft's 3-D virtual reality hunting and
gathering program.
--
Wayne Greenwood
mrt...@aol.com
Daniel Galeus
>Windows still sticks with the abysmal use of Function keys
>and strange triple key combinations with apparently no rhyme
>or reason. Arbitrary function key to action mappings are extremely
>non-intutive, yet they remain central to the "keyboard shortcuts" in
>a so-called modern GUI, such as Windows 3.1.
I must disagree with you here. The Windows interface allows for several
way of implementing shortcuts.
- Function keys ("F1" for Help, etc)
- Control keys (Ctrl+P for Print, etc)
- Accellerator keys (Alt+F+P for File/Print)
For rhyme and reason you may use Control keys or Accellerator keys
since I find it easiest to remember since they use mnemonic-like
features ( *F*ile / *O*pen ).
I prefer using Accellerator keys exept for the standard Windows Control
keys (Ctrl Z/X/C/V/N/O/P/S/B/I/U). The accellerator keys use mnemonics
Best regards, Daniel
--
Daniel Galeus | Internet: dga...@vinga.trillium.se | Ph. +46 31 819850
| CIS: 75170,240 | Fax +46 31 812139
Vinga System AB | SNAIL: Gotabergsgatan 28 |
Sweden | S-411 34 Gothenburg, Sweden |
Tim Shea
>I am convinced that usability testing will reveal that using several
>shortcut methods is significantly worse than sticking to one, whatever
>(reasonable) method that may be.
Why are you convinced of this?
Tom Maynard
>Subject: Re: Human factors and MS "Chicago"
>I am convinced that usability testing will reveal that using several
>shortcut methods is significantly worse than sticking to one, whatever
>(reasonable) method that may be.
I have found this to be generally true for novice users, whose responses tend
to be "give me ONE CONSISTENT way of doing things or I won't remember it!"
Intermediate to advanced users area seem to want to find the fastest way.
However, if the primary method for the novice is visible and obvious, they
will ignore the shortcuts until they are ready for them.
-Tom Maynard internet:may...@marcam.com
Daniel Galeus
>> I must disagree with you here. The Windows interface allows for several
>> way of implementing shortcuts.
>>
>> - Function keys ("F1" for Help, etc)
>> - Control keys (Ctrl+P for Print, etc)
>> - Accellerator keys (Alt+F+P for File/Print)
>>
>> For rhyme and reason you may use Control keys or Accellerator keys
>> since I find it easiest to remember since they use mnemonic-like
>> features ( *F*ile / *O*pen ).
>Unfortunately, more isn't always better. The fact that shortcuts can be
>implemented in several ways also mean that they WILL be impelented in
>several different ways. This will surely confuse novice to intermediate
>users. It won't confuse experts, but hey, experts can navigate in emacs
>wich only goes to show that experts can learn anything, and thus it's
>proof of nothing.
>I am convinced that usability testing will reveal that using several
>shortcut methods is significantly worse than sticking to one, whatever
>(reasonable) method that may be.
I see your point, but I do not entierly agree.
There is a tradeoff between "speed" and "ease of use". Using menus
is easy, but not very fast. Using shortcuts is fast but not very
easy to remember.
Learning curve:
Hard / not fast = Using menus
Harder / fast = Accellerator keys (Alt+F+P for File/Print)
Even harder / faster = Control keys (Ctrl+P for Print, etc)
Hardest / fastest = Function keys ("F1" for Help, etc)
For novice users it is important that the UI is visible. The menus
in Windows are for novice users. By always presenting more
efficient ways of doing the same thing, the novice user will
soon learn to use shortcuts.
Example:
By presenting the shortcuts after the menu, the user will learn
other, sometimes more efficient, ways to select the same menu
alternative. For example you learn a lot about the shortcuts
when you select "_File", "_Open (Ctrl+O)".
1) You can press "ALT+F, O" to do the same thing
2) and you can also press "Ctrl+O" for the same thing
// daniel
David W. Griffin
there is always a standard way the novice can use. For example, in Microsoft Word,
you can print by pulling down the file menu, or by a command key combination (I think)
and by pressing the icon bar print button. The novice need not know all that. He
knows that no matter what Mac application he is in, he can pull down the File menu.
Later, if he wants, he can learn the other ways. Windows should do the same if it is
ever to catch up with the Mac in user ease.
Peter Kay
: The Macintosh has several ways of doing the same thing too. The key though, is
It does, even windows has design standards! Print should always be under
File..
And they are providing little tips for the users in the next version of
Windows (Chicago).
Personally I use OS/2 - partly because of the interface - once you have
worked out the object-orientated data-orientated approach it is easier than
any other i/face I have tried.
The mac is nice admittedly, it does the job. I just find myself wishing that
it would do a little more - it only does the bare minimum and isnt too good
if you want to do more powerful operations.
Its integration beats the PC down flat though (When are people going to
learn that all they want is for a computer to work - sod the bells and
whistles). I don't suppose Mac users know enough either to word wrap their
lines properly??
____oooo______ooooo________oo____ooooo_____________________________________
__oo____oo___oo___oo______oo___oo____oo___ Peter Kay, IT Institute, _
_oo______oo___oo_________oo__________oo___ Salford University, Salford. _
_oo______oo_____oo______oo__________oo____ Beginner OS/2 WPS programmer _
__oo____oo___oo___oo___oo_________oo______ P.J...@iti.salford.ac.uk _
____oooo______ooooo___oo________oooooo_____________________________________
Tim Shea
[snip]
> The novice need not know all that. He knows that no matter
>what Mac application he is in, he can pull down the File menu. Later, if
>he wants, he can learn the other ways. Windows should do the same if
>it is ever to catch up with the Mac in user ease.
"Yeah, and those Germans are fools if they think they'll ever get
past our Maginot Line, eh Pierre?"
(Sound of rumbling tanks grows closer...)
"Pierre? Pierre?...."
Ivan O'Brien
Windows 3.1, for all its faults (and they are many) does have this level of
consistency. You can never say "all", but nearly all apps I have seen have
File New, File Open, File Print and File Exit, for example. They all have
Help About, Help Index, etc. Some of the intermediate menus (View, Text to list
typical ones) are less standardised, but that's largely a factor of the
different applications.
In my experience (I use Unix (including emacs!), Windows, Mac) the Windows user
interface is great for just about everything EXCEPT handling files. This is
because of the DOS filesystem, which various file managers etc try to conceal
from the user, and is always the most difficult thing to teach a new user.
Shortcuts are more consistant in Windows than anything else I've seen...
Ivan
Clara N. Fitzgerald
>>The Macintosh has several ways of doing the same thing too. The key though,
>> is that there is always a standard way the novice can use.
>Windows 3.1, for all its faults (and they are many) does have this level of
>consistency. You can never say "all", but nearly all apps I have seen have
>File New, File Open, File Print and File Exit, for example. They all have
>Help About, Help Index, etc.
>Shortcuts are more consistant in Windows than anything else I've seen...
What about Edit? The actions (Cut, Copy, Paste) are generally consistent,
but the shortcuts (try Borland's Cntrl-Shift-Insert for Paste!) are different.
Windows has tied up all Alt and Shift-Alt keys in menu navigation, but not
enforced non-mouse access of applications, which they profess to support.
>In my experience (I use Unix (including emacs!), Windows, Mac) the Windows user
>interface is great for just about everything EXCEPT handling files. This is
>because of the DOS filesystem, which various file managers etc try to conceal
>from the user, and is always the most difficult thing to teach a new user.
I don't handle files in Windows; DOS is easier. How is one supposed
to move groups of files? (Select visually, then deal with a dialog box
asking for the target path? Either do both visually, or neither!)
[I must admit that my computer does recursive moves and copies in DOS]
Windows is not really an object-oriented desktop, although it tries to
appear to be. The Program Manager's groups do not reflect the directory
structure and do not show all files that are on the computer; Macintosh
does both, allowing hierarchical groups and icons representing text files
(not just executables) to appear on the desktop. The only external 'File
Manager' functionality needed is search facilities.
>Ivan
--
-Clara A. N. Fitzgerald cfit...@s.psych.uiuc.edu
- < - < -< <> >- > - > -
Help stamp out, reduce, and eliminate redundancy.
Helen C. O'Boyle
Clara N. Fitzgerald <cfit...@s.psych.uiuc.edu> wrote:
>Are you really suggesting that these people do not understand how
>the shift key works, even if they don't (have to) use it?
Yes. While working as a computing assistant in college, I saw any
number of people (students, profs, etc.) who were sure that the only
way to get shift to work was to press and release it SIMULTANEOUSLY
with the key one wanted to shift. Needless to say, they very often
ended up with the letter in lower-case, depending on keyboard
responsiveness and their exact timing.
--
* Helen *