On Mar 13, 2006, at 9:01 AM, Jani H. Lahtinen wrote:
Anselm R. Garbe wrote:
Actually this was not intented as insult or something. It simply
gives an idea of achieving good software quality after some
time. Often people request features which could easily replace
existing features (like tags replace not only pages/workspaces,
they even make attach/detach/sendtopage totally deprecated,
apart from the 'detached clients layer' and the 'pager') - the
introduction of tags is a good sample how you can making
software more powerful due removing features.
In case where features overlap, or when a new feature would be a
simple extension of existing ones, that is a good principle.
I think it goes beyond that. There are only so many features you can
cram into something before it becomes difficult maintain and
difficult to use. Sometimes I think you just need to say, "This
feature isn't necessary -- Let's leave it out." I personally think OS
X (and much Mac software in general) is a great example of this.
Despite what a lot of open source advocates say about feature bloat
in Windows and commercial software in general, OSS projects are often
the biggest offenders. I think the main reason for this is that it
takes a tough man (or woman) in charge to say "no" when necessary.
I'm glad Anselm can say no.
(I recently declined(!) several patches
implementing icon support for the bar and in general for
liblitz).
I personally don't want any unnecessary graphics, but some might.
Would it be a good idea to have some place for storing such patches
for those who might want them (and leaving it for them to get them
actually compile)? ... or then not.
This seems fair enough, provided that Anselm isn't expected to spend
one minute making sure they work or maintaining them. To be honest
though, I'd expect that they'd end up in such a state of disrepair
that it isn't even worth cluttering the site with the link.
- John
_______________________________________________
[email protected] mailing list
http://wmii.de/cgi-bin/mailman/listinfo/wmii
