Improving the interface (w.r.t. scrollbars)

Peter Crowther Peter.Crowther at melandra.com
Tue Mar 6 18:55:45 UTC 2001


> From: Dan Shafer [mailto:dshafer at yahoo.com]
> We have, I think, become stuck in the notion that
> users need predictability across machines and even across 
> operating systems to
> some degree, yet how many people do you know who run more 
> than one machine or,
> at most, two different machines (one at work and one at 
> home)? Vanishingly few.

Most of the people I know run two, and get utterly confused by the
differences between Windows 98 on their home machine and Windows 98 on the
company machine.  They have no idea what is an application vs. an operating
system (common conversation: "What system is it?" "Windows 97." --- if they
know at all), absolutely no concept that you might have to pay for them
separately ("Why haven't I got Excel on my home machine?") and freak when
you point out that you can change the desktop colour.  If they run one
machine, they're deeply scared of it because it's a Work Machine and the IT
team have uttered all sorts of threats, and they won't change anything.
*Or* they think they're Grade-A hackers because they know how to copy files
using Explorer (or equivalent) rather than opening the file in Word and
saving it somewhere else.

Computers are not yet the tools of their users --- quite the reverse, and
the companies who install them seem to want it to stay that way.  Companies
also see IT systems as one-size-fits-all --- the whole notion of a 'standard
desktop' that is so beloved of all large corporates --- and there's a good
reason for this.  The cost of maintaining that computer is very high and
very visible; often higher (and almost always more visible) than the saving
made by a user being able to tailor the system to their needs.  Much of that
cost, particularly on DOS/Windows 3.x/Windows 9x, can be attributed to users
tinkering with the system and breaking it because they don't understand what
they're doing.  And add a further cost for the time that user spends
tinkering rather than working, and then hiding the evidence rather than
accepting it.

Users don't necessarily need predictability, but companies do.

> So perhaps rather than busying ourselves with a difficult if 
> not impossible
> task of designing what _the_ next UI should look like, we should be
> experimenting with the notion of creating lots of different 
> UI components
> (morphs) users can _easily_ assemble into a user experience 
> _they_ understand and prefer?

I agree, but this must not be used as an excuse for releasing works in
progress on the grounds that "the user can sort it out".  The user almost
certainly will be unable or unwilling to sort it out.  I think you'd be
surprised (and depressed) by the number of highly flexible, customisable
systems out there that are still in their original configuration simply
because the user can't/won't fix the problems.

Existing UIs count as works in progress for this, with the added pain that
they can't be sorted out by the user or by anyone else because they're not
flexible enough.

The stated aim of Squeak as an exquisite personal computing environment is
marvellous, and holds the prospect of a system that is flexible enough that
more adventurous users may be willing to alter and able to alter without
lasting damage.  I wonder how to get a better interface such as this into
companies, though?

		- Peter





More information about the Squeak-dev mailing list