Re: PATCH for 2.4.0: Correct default panel size

[Malcolm Tredinnick <malcolm commsecure com au>]

On Sun, 2003-09-07 at 13:16, Luis Villa wrote:
> On Sat, 2003-09-06 at 10:31, Jeff Waugh wrote:
> > <quote who="Loban A Rahman">
> > 
> > > > Good point - I guess we'll just have to leave it that new users to 2.4.0
> > > > will have insanely large panels :/
> > > 
> > > Yeah, I thought our big thing was It Just Works (TM). Hence, the look of
> > > the default desktop should be a show-stopper bug.
> > 
> > "Just Works" has nothing to do with "look", and 12 pixels does not make for
> > a show-stopper bug. If this was important enough to fix, it would have been
> > fixed long, long before we were between release candidate and final.
> 
> Actually, I have to disagree here.
> 
> (1) showstopper has a lot to do with 'look'. showstopper is all about
> visibility and user impact, and particularly about impact on new users
> trying to assess GNOME. 'look' is a huge factor in that. 

My argument (if I was going to make one, which of course I am not) would
be that you have misassessed the 'impact' portion here. So somebody's
panels are eight pixels bigger than they used to be by default. Where's
the huge impact? This is for a *new* user, so in all likelihood they do
not even have a previous comparison point. Make a brief note in the
release notes and move on.

I could even make an argument that 32 pixels is a bigger target to aim
at (usually I want to smack Mr. Fitt and the Corollary Gang upside the
head, but in this case they can provide some small benefit).

> (2) this is a bug that has a /very/ low probability of being caught in
> our testing process (very few new accounts get created against the beta)
> so the relationship between 'it is important' and 'it was found and
> fixed' is virtually nil. 

I had seen this a while ago (I seem to create more new accounts for
testing than most people, apparently). But I had seriously thought it
was an intentional change, since it did provide a slightly more roomy
panel. I had to decrease the size on my laptop, since screen space is
valuable there, but I left it larger on my big monitor at work and quite
like it. There was no chance I was ever going to think "this is a bug,
it should be filed". Pixel-sized values are pretty much always going to
be subjective and never universally correct since monitors come in
different sizes.

> So, I'd like to see it go in, given that even I'm able to verify that it

Ah well, I guess it's over now. It's not a big deal, but if we have a
process, we should use it.

This has certainly been a slightly bizarre educational moment.

Malcolm