Re: Browser - [was 2.4 Proposed Modules - 2 weeks left]

[Luis Villa <louie ximian com>]

On Thu, 2003-05-08 at 07:46, Murray Cumming Comneon com wrote:
> > Havoc said:
> > >Jeff has said he isn't sure we should include a browser
> > >in general, and that discussion hasn't really been had.
> 
> I don't think Jeff actually said that. (I think he said he likes the idea of
> a GNOME browser but isn't sure if epiphany or galeon is ready to be that
> browser.) But I did say it.
> 
> > Then I guess that discussion should be started :)
> 
> I think the only real reason for having a GNOME browser is the User
> Interface. It would be nice to have something that was consistent with the
> GNOME desktop (HIG and GNOME simplicity), and Mozilla explicitly does not
> try to do this. I believe that will continue to be the case in future. It's
> not as powerful a reason as before because Mozilla's UI will soon be much
> better without the mail/irc/yadayada stuff bundled with it.

It'll be better, but it still won't respect our icon themes and it won't
respect our HIG. We're going to cripple our long-term potential if we
consider apps that ignore those types of things to be acceptable
replacements for better GNOME-based apps.[1]

> I think UI-consistency is probably reason enough. But we should accept the
> fact that, even if epiphany becomes our perfect GNOME browser, a lot of
> GNOME users would continue to use Mozilla instead. 

For accessibility reasons alone, we currently need to continue to
provide users the option to choose their browser, and I believe (in
fact) we should work hard to make sure that functionality works well;
the current system is a mish-mosh that only works sometimes, though it
looks like Frank's work may be making that better. 

> And I don't think we will ever try to tie it compulsorily into the
>  desktop like MS did.

Despite what I've said about short-term needs, I think, in the long term
(i.e., once ephy a11y is solved), we probably do need to think very hard
about doing this in various ways. Regardless of the legality or
illegality when a giant monopoly does this, I do (personally) believe it
is best for users in the long term. We can't be limited by the lowest
common denominator that links supports.

[Note that this is (1) way out of the GNOME 2.4 timeframe and (2)
epiphany is the best choice for GNOME, IMHO, whether or not you agree
with the above statement.]

>  With that
> understood, I'm personally fairly happy to have epiphany in GNOME. And there
> seems to be a good consenus for _a_ GNOME browser - people get upset at my
> stupidity for even suggesting otherwise.
> 
> > What would be reason to include a text editor and a pdf 
> > viewer but not a
> > web browser ?
> 
> For me, it's probably a question of application complexity and the existence
> of competing products. And trying to draw a line _somewhere_ between GNOME
> utilities and 3rd-party apps.

I think a web browser is pretty much an expectation of something that
comes with an OS now- it is even more basic to most users than the text
editor and the pdf viewer, arguably. [We are as text-centric as people
get, as hackers, and we still probably all use our web browsers at least
as much if not more than our editors. [Which is emacs anyway. ;)] 

To put it another way- many, many of the interesting things in the world
are happening out there on the web. To claim we are an even minimally
usable desktop without including something that allows our users to see
that web is misleading. It's like saying 'we're a complete desktop'
without including a file browser or a menu system. If the answer is
'well, there are good browsers out there already' I'd say the response
is 'there is already a good file manager out there- so let's give our
users the choice of rox or konqueror instead of shipping nautilus.'

Luis

[1] I personally feel that this applies to other things too- if we wait
around for OOo to adopt recent-files, just as an example, we're going to
be waiting a long time. Once we get it into gtk, gnumeric and abi will
Just Work. Ditto with a11y fixes, i18n fixes, etc.