Re: [HIG] Should vs. Must (was Re: [HIG] Naming)

[Seth Nickell <snickell stanford edu>]

> Kathy makes a good point about the ease with which mere recommendations
> can be dismissed by time-pressed developers. I have felt for some time
> that we need to take a stronger stance on some of the points, making
> them design requirements. The truth is, it is unlikely that some hand of
> the Gods is going to strike down anyone who "disobeys" a requirement, so
> we aren't forcing conformance even if we do mark some guidelines as
> required. We would just be improving the chances that the most important
> guidelines would be followed.

Hackers will tend to resent statements that are worded more strongly
than we care to enforce. If at some point we start refusing the ship
applications with GNOME that do not meet important interface guidelines
(dia comes to mind....), maybe we can change such clauses. But at the
current state I think that will cause resentment and will have little or
no effect on the likelihood of implementation.

Free software hackers have a long standing tradition of the maintainer
as god who writes his application exactly how he wants it written and
nobody can do a damn thing about it. In the long term it would be nice
to erode this notion some, not just for usability but for things like
interoperability and solving cross-module disputes. But for now.... I
think our best bet is convincing maintainers that they want to follow
the guidelines (and generally respect the thoughts of the usability
project).

-Seth