Re: A call to action

[Richard Tibbetts <tibbetts MIT EDU>]

Just a comment on this thread, from a developer who became
disillusioned and estranged from his internal ui design process.

A bit of background: I spent the summer working for MIT Athena to
integrate GNOME with Athena. This involved making GNOME build on all
the athena platforms (IRIX and Solaris being the problem ones), making
GNOME play well with some concepts that exist only on Athena, and
porting legacy apps.

There was a member of our team who took it upon himself to work on UI
design, and had qualifications in that area. Outside our team we also
worked with two usability specialist who were going to set up formal
testing of all sorts of things.

However the other "real" developer and I (who were doing, in our
opinions, the actual work) saw our opinions of these people degrade
quite quickly. And while there were some questions of competence, the
real issue was in the style of communication.

In general UI people, especially in GNOME, talk to developers about
small parts of their project that are already written. Frequently UI
design groups like the hitsquad are asking developers to rewrite a
great deal of code. And, most importantly, the UI designers seem to
hold opinions for a fleeting moment compared to the ammount of time it
takes to implement them.

The solution to this problem, it seems to me, lies in producing a
document similar to the Apple Human Interface Guidelines. This will be
a significant document which covers 99% of UI design decisions, and in
each and every case gives an unambiguous GNOME solution to the
problem. Whether this is written by committee or by star chamber or
any other method doesn't really matter. It has to not suck in any
clear way, and once it is done it has to be followed and enforced.
Once you have that, it becomes possible to say that (1) all developers
of official GNOME apps should follow the guidelines and (2) when the
guidelines are ambiguous people should ask a specific body rather than
just making somethin up.

just my opinion,
tibbetts

-*- http://www.mit.edu/~tibbetts -*- finger tibbetts monk mit edu -*-