9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
From: Laura Creighton <lac@cd.chalmers.se>
To: 9fans@cse.psu.edu
Cc: lac@cd.chalmers.se
Subject: Re: [9fans] architectures
Date: Thu, 12 Jul 2001 15:01:58 +0200	[thread overview]
Message-ID: <200107121302.PAA24298@boris.cd.chalmers.se> (raw)
In-Reply-To: Your message of "Thu, 12 Jul 2001 01:22:39 EDT." <20010712052246.F4087199E1@mail.cse.psu.edu>

my experience with non-techie folks is that once they get over the
conceptual hurdle of actually believing that can use a different
finger than their index finger to point with, they think that
being able to use their machine efficiently to do what they want is
as cool as the rest of us do.

The problem is that naive users are what people design for, because
customers who have never bought your product are by definition naive
users of it.  But naive users have a limited lifetime.  For a few
years I was involved with a project that tried to measure exactly how
limited the lifetime was. We came up with ~3 weeks for most users for
every program we tried to measure.  At this point, they become
frustrated users who want a better interface, or resigned users who
don't believe that there is a better interface because they have only
used computer interfaces designed for the naive.  The big lesson I
learned from this was to teach how to user regular expressions to every
vi user who has been using vi for at least 3 weeks ... but not
before then ... and make their lives a lot happier.  And to cheerfully
and patiently listen to all the complaints of the new users who found the
interface hard to learn, because, after all they deserve the respect
of having their complaints listened to and acknowledged; and then
carefully filing whatever changes they want under `things that they
most likely do will not want to change in about 3 weeks'.

Of course if your interface is truly lousy, the naive uers may quit
before 3 weeks is up.  And if you have botched some detail, the complaints
will continue after the 3 weeks.  But while you must never, ever
treat anybody with such contempt as to reply `you aren't significant
enough to have an opinion' - naive users truly are not significant
enough -- because like butterflies, their life expectancy is measured in
weeks.  Experienced users may also have a list of design defects in
your interface, sometimes because you have botched the interface,
sometimes because they have better vision than you did, and sometimes
because they are using your program to do things that you never do --
what is a problem for them never came up for you.  But it is extremely
rare for them to have the same list of changes as the naive (and an
indication that user interface design may not be what you have any
talent for).  In test after test where we established a user community
of experienced users and then announced that we were going to change
the interface in response to complaints that the interface was hard to
learn we got enormous protests (which we of course saved, that being
the point of this) of the form `when I was learning this, I thought
that XXX sucked too, but now I can't live without it.'

There are some things you cannot teach the technically unsavvy.  For
instance, in 20 years I have never, ever, ever, been able to convince
people that floating point, despite looking like decimal fractions,
ISNT, and that you MUST NEVER USE IT FOR MONEY.  The damn fools listen
politely and then go back to using it, because, after all, they think
they know better than you do.  Moth to the flame.  But I have worked
with 8-year-olds and 80-year-olds, secretaries, hairdressers, and
supermarket-check-out clerks, the most non-technical people we could
hope to find since we advertized for them.  And after they get some
familiarity with vi, I can teach nearly all of them how to use regular
expressions.  And just like the technically savvy, they think that
regular expressions are wonderful.

Laura






  parent reply	other threads:[~2001-07-12 13:01 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-07-12  5:22 anothy
2001-07-12  8:04 ` Matt
2001-07-12 10:12 ` Boyd Roberts
2001-07-12 13:01 ` Laura Creighton [this message]
2001-07-12 20:35   ` [9fans] architectures Jim Choate
2001-07-12 21:41     ` Dan Cross
2001-07-12 22:09       ` Jim Choate
2001-07-13 14:52     ` Douglas A. Gwyn
2001-07-13 15:15       ` Boyd Roberts
  -- strict thread matches above, loose matches on Subject: below --
2001-07-13  2:25 [9fans] architectures Rick Hohensee
2001-07-12 12:55 forsyth
2001-07-12 12:43 rob pike
2001-07-12 19:45 ` Boyd Roberts
2001-07-12 11:15 nemo
2001-07-12 20:28 ` Boyd Roberts
2001-07-13 14:53   ` Douglas A. Gwyn
2001-07-13 15:11     ` Boyd Roberts
2001-07-12 10:30 nemo
2001-07-12 10:18 ` Christopher Nielsen
2001-07-12  8:42 forsyth
2001-07-12 13:56 ` Laura Creighton
2001-07-12 16:13 ` Ozan Yigit
2001-07-12 16:33   ` Matt
2001-07-12 18:12     ` Scott Schwartz
2001-07-12 18:16       ` Martin Harriss
2001-07-12 18:43       ` Dan Cross
2001-07-13 14:52         ` Douglas A. Gwyn
2001-07-13 15:13           ` Boyd Roberts
2001-07-12  7:15 Sape Mullender
2001-07-12  6:16 okamoto
2001-07-12  7:46 ` pac
2001-07-12  9:59 ` Christopher Nielsen
2001-07-11 23:17 Jonathan Sergent
2001-07-11 17:59 David Gordon Hogan
2001-07-11 17:38 geoff
2001-07-11 18:29 ` Dan Cross
2001-07-11 16:27 jmk
2001-07-11 16:03 jmk
2001-07-11 15:17 nemo
2001-07-11 15:07 bwc
2001-07-11 16:53 ` Mike Haertel
2001-07-11 14:36 anothy
2001-07-11 14:59 ` Theo Honohan
2001-07-11 15:02   ` Matt
2001-07-11 16:52   ` Mike Haertel
2001-07-11 22:58 ` Boyd Roberts

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=200107121302.PAA24298@boris.cd.chalmers.se \
    --to=lac@cd.chalmers.se \
    --cc=9fans@cse.psu.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).