Re: IRC channels in gnome development



Hi Alan,

FWIW I mostly like GNOME 3, so I don't want to pile on the flamefest.
But this bothered me:

On Sun 06 Feb 2011 15:27, Allan Day <allanpday gmail com> writes:

> Even if you had records of every discussion, you wouldn't get the
> information you're looking for. Design decisions don't get made
> committee meeting style, and design involves a lot of specialist
> background knowledge which doesn't get explicitly referenced. Fact is,
> we'll probably never be able to give 100% of the rationale behind design
> decisions.

The thing is, we've done mostly well in the programming department.  If
the subtext is this is the case for design in contrast to programming, I
would like to disagree; that would be unjust both to programming and to
design.

Often programming is just as solitary an affair, yet we manage to
communicate in such a way that enables collaboration; and surely
programmers are not more socially competent than designers ;-)

Likewise designers don't work alone.  I'm sure you have been one of two
or three or six designers sitting at a table hashing things out.  In
neither profession do things happen "committee meeting style" -- when
things go well, of course! -- but there is collaboration.

This characterization of design also neglects the great community design
work that has been done recently by Máirín, for example, and done to an
extent within GNOME.

Finally, it's rare that a programmer never does design work, or for a
designer never to code at all.  We all need pointers and records to
figure out how things are done.  Of course it's not always possible!
But it would be an error not to hold transparency up as a goal, IMO.

> It simply isn't true to say that we haven't made an effort to explain
> what we're doing. I explained many of the design considerations in my
> blog post [1] on this subject, and I did that precisely because I wanted
> to help people to be informed.

For this, and all your awesome work, thank you!


Andy
-- 
http://wingolog.org/


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]