home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.windows.x:16501 comp.human-factors:2182
- Path: sparky!uunet!cs.utexas.edu!torn!utzoo!censor!isgtec!mike
- From: mike@isgtec.com (Mike Sellers)
- Newsgroups: comp.windows.x,comp.human-factors
- Subject: Re: GUIs Considered Harmful
- Message-ID: <3260@isgtec.isgtec.com>
- Date: 10 Sep 92 13:51:38 GMT
- References: <NEERI.92Sep3175345@iis.ethz.ch> <3243@isgtec.isgtec.com> <1992Sep8.211314.23372@thunder.mcrcim.mcgill.edu>
- Sender: root@isgtec.com
- Followup-To: comp.human-factors
- Organization: ISG Technologies Inc., Mississauga Ontario
- Lines: 111
-
- Note the followup-to line.
-
- mouse@thunder.mcrcim.mcgill.edu (der Mouse) writes:
- > I am not arguing with [Apple's] findings. I am saying merely that it *is*
- > slower _for_me_. Or are you trying to say that my time-sense is so out
- > of whack that I merely *think* I'm typing 10 to 20 cps?
-
- There is at least a moderate possibility of this. People almost always
- overestimate their typing speeds and underestimate their reaction times,
- sometimes by an entire order of magnitude.
-
- > Or perhaps
- > that I merely *think* that I can't make a mouse selection in .05 to .1
- > second?
-
- No, of course I'm not saying that. Making selections with a mouse is
- almost always slower than doing the same from the keyboard, but this is
- not an issue to many (most) users.
-
- > In article <3243@isgtec.isgtec.com>, mike@isgtec.com (Mike Sellers) writes:
- > > On the other hand, the folks at Apple apparently also found that the
- > > mouse and keyboard was actually *faster* than the keyboard alone,
- > > people's intuition notwithstanding.
- >
- > With Apple's choice of subjects, choice of software, choice of UI.
-
- Of course. They tested on their software/UI and with those who have
- similar characteristics to those who are likely customers. Doing anything
- else would be senseless. They weren't doing basic research.
-
- > If the subjects had been shown the full spectrum of UIs available, had
- > experimented with various ones, had each chosen a preferred one, had
- > chosen software to fit that UI, had practiced for a week or so, and
- ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
- > they had then timed the result, I (a) am certain the results would have
- > been different and (b) would say the result is much more applicable to
- > the real world. (And (c) that Apple wouldn't be trumpeting the
- > results!)
-
- Two points here:
-
- First, the effect of practice is a strong one, but NOT one on which the
- makers of software can rely. Increasingly, people are not willing to
- give a program the benefit of the doubt and hope that after a week
- they'll be able to use it better. Thus, UIs and programs have to be
- usable both on the first day, and after a week or a month of use.
- One proven solution to this is a flexible UI that includes both mouse
- and keyboard input.
-
- Second, there have been a few studies done such as you propose. Basically,
- it turns out that the type of UI really doesn't matter! What matters is
- flexibility, directness (whether via GUI or CLI or menus), *reliability*,
- interactivity, and all the "small problems" that so often don't get fixed
- in software ("well, it's not like it crashes the system"). Users report
- higher satisfaction and lower frustration, AND their actual task-completion
- times are lower when using "polished" products. Now, lest you think I
- discount any differences in UIs, there is also an emotional component:
- Users will not use a product that they do not think, a priori, will do
- the job or will not be enjoyable to use. Would you drive a Porsche if it
- only came in a dirt-brown package shaped and appointed like a Yugo?
-
- > ...unless you are taking the interpretation that
- > Apple's results mean that *everyone* is faster using both keyboard and
- > mouse, even when allowed to choose their software to fit their
- > preferred style of interaction...in which case I *strongly* disagree.
-
- No, not everyone -- just the large majority of the *paying* public.
- High-speed typists do not form a large enough segment of most markets
- to be considered alongside the legions of computer-phobes and poor
- typists who are willing to plunk down their dollars for software.
-
- > >> If it takes as long as *one* second to decide which key to press, I
- ^^^^^^
- > >> consider myself "idling". I type well over 100wpm when I don't have
- > >> to stop and decide what I want to type next.
-
- > > Unless you've had yourself independently timed, I doubt strongly
- > > whether _any_ of these numbers are correct
-
- > That is not a *reaction* time, that is an *action* time.
-
- If you have to _decide_ what to do next, it is a reaction time; the
- lowest RT figures I've been able to find are on the order of 0.3 seconds.
- Even for spinal reflexes involving complex actions, I don't think the
- action times are as low as 0.05 seconds for humans. Still, this is
- splitting hairs.
-
- > > Given this, your needs and abilities do not conform to the needs and
- > > abilities of the vast majority of computer users; you cannot judge
- > > them by looking at yourself.
- >
- > I am not trying to. I'm complaining about the syndrome of refusing to
- > admit that anyone but the typical member of the majority even exists:
- > the leap from "the majority of those we studied were this way" to
- > "everyone is this way", or perhaps less obnoxiously phrased, but with
- > the same effect, to "we can ignore everyone who's not this way".
-
- Like I said before, Apple is looking at a particular sub-population.
- It is not in their financial interest to cater to all parts of the
- spectrum of, e.g., typing speeds, as this would add to the cost of
- their products without adding proportionately to the money generated by
- them. It's not that you don't exist, it's just that there aren't
- currently enough of you to make a dent in Apple's (and most other
- company's) markets.
-
- > der Mouse
- --
- Mike Sellers User Interface Team Leader
- mike@isgtec.com ISG Technologies, Toronto, Canada
- "Actum ne agas" -- Do not do what has been done.
- Furare tantum ab Optimis
-