Re: GTK+ 2.10.7 released



On Sat, Jan 06, 2007 at 04:36:43PM +0100, Richard Boaz wrote:
> 
> in all seriousness, how many of the gtk+ developers actually have to  
> use gtk+ in the real world?

I'm not a Gtk+ developer, I don't endorse Gtk+ developers'
decisions and actions and they don't endorse mine.

> problems, in  
> general, do not get fixed when they are not seen or suffered, in my  
> experience.

Agreed.

> it's a question of attitude, and in my experience, anytime anyone  
> comes along and points out how gtk+ could somehow be made better,

Do you mean
(http://mail.gnome.org/archives/gtk-list/2007-January/msg00003.html):

    And I have no time and energy to file all these bugs.

or
(http://mail.gnome.org/archives/gtk-list/2007-January/msg00032.html)

    'konqueror' WEB browser doesn't appear to leak, while
    gtk+-based Mozilla, Firfox, Seamonkey leak a lot, and I'm
    not sure whether it's 'Gecko' or gtk+ issue.

?

> i think what Sergei has been saying, while becoming a bit convoluted  
> after a few iterations, is obvious, and you said it yourself: be  
> accurate when placing information on a FAQ: could it be made more  
> accurate? if yes, then make it so.  why all the arguing about a yes  
> or no question?

So what is wrong with the FAQ entry explaining why top is
not a good tool to find bugs in Gtk+ apps?

In my opinion one can misinterpret it only intentionally,
and that is exactly what's going on here.

> whenever someone makes a suggestion as to how gtk+ could somehow be  
> made better, the gtk+ developers

Here we are again.

> always tend to come off with an  
> attitude that it can't actually be made better and that the person's  
> commentary is way misguided (and will spend endless emails wearing  
> the person down until they just go away).
> 
> and this is crap,

Quite the contrary, this is the only sensible thing to do.
People are coming with all kinds of crazy ideas all the time
and they see their niche problem, not the conseqences and
the big picture.  You have to filter them.

So you say `It won't work because of foo, it will break bar,
it will not be compatible with baz, it cannot be implemented
on quux...'  If the person coming with the idea it cannot
convince you, cannot make compelling arguments and suggest
solutions, it is crap.  And if the idea is really good in
spite of that, someone else will come with it later, so not
much harm done.

Now, this all is irrelevant to this case.  Or if it's
relevant, where are the suggestions?  I can see only 1037th
method of software distribtuon and complaints.

> as a developer of scientific software, i am required to make my app  
> available on three platforms: LINUX, MAC OS X, and Solaris, soon to  
> be adding windows, running on every continent in the world including  
> antarctica (meaning: few resources/connections).

As a developer of scientific software that runs at least on
GNU/Linux, Mac OS X, MS Windows and FreeBSD I'm essentially
content with the current methods of distribution of Gtk+.
I always encourage people to use packages that come from
their OS vendor.  They will get security updates and
everything.

> some two years ago  
> i asked about compiling static exe's to solve the obvious  
> distribution problem here, and the answer i got was that this was not  
> a good idea.

Well, separate Gtk+ package and dynamic linking beats static
linking on download size easily -- except scenarios such as
that the application is never updated or it's hello-world.

> i have no guarantee that the end-user  
> knows anything about something called gtk+ and all the nightmare  
> requirements involved in installing it.

Is the problem that Gtk+ dependences *exist*?  Or does your
application have -- unlike Gtk+ -- some magic building
method that everyone immediately understands and can
perform?

> the real world beckons, oh ivory tower; shall thee truly not heed the  
> call?

Part of the real world experience is that you invent a great
solution and then find ohter people don't have the problem.
I happens to me all the time too.

Yeti


--
Whatever.



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