Re: 2.4 Module List and Rationale (aka GEP10 and 11)

[Luis Villa <louie ximian com>]

On Sat, 2003-03-15 at 11:36, Ricardo Fernández Pascual wrote:
> Hello,
>     I and the rest of the Galeon team, we would like to propose Galeon
> as part of the application list for the GNOME 2.4 Desktop, instead of
> Epiphany.
> 
>     Galeon 1.3 is now quite stable for daily use and nearly feature
> complete. It has come a long way to comply with Gnome 2 standards of
> usability and accessibility.
> 
>     We think that Galeon is the best web browser for Gnome in the long
> term. It offers feature parity with other desktops and improves in some
> areas, like usability and offers better integration with Gnome than
> other viable options like Mozilla alone.
>
>     We would like to know why it has not been considered already in GEP
> 11. Specially, we are interested in the concrete problems that make
> Galeon not an option and the ways that these problems could be solved.
> We are in no way deliberately ignoring the Gnome HIG and specific
> suggestion of the usability team, in fact further efforts are made to
> comply more closely with it.

So, let me say first that I haven't really looked forward to this,
despite knowing it was going to have to come for some time. As I told
Jeff once, I'd rather pound my pen^thumb with a hammer- it would be more
fun. But... it seems like a discussion we have to have. :/

Galeon clearly meets most of the standards of GEP 10- as you say, it
would be a marked improvement over Mozilla in most situations,
especially as we work to integrate the 'net and the desktop.

It is my personal opinion that Epiphany is superior to Galeon on two
counts:

(1) overall usability: Epiphany's removal of many options that felt
unecessary to me and to others, as well as Epiphany's creative attempts
to deal better with bookmarks, make Epiphany feel like something that is
more in keeping with the usability direction that GNOME has chosen.
Galeon ATM feels like it is following the letter of the HIG but not
necessarily the spirit, particularly in the re-addition of many
'advanced' options that were axed from earlier versions of galeon 1.3.x
and the choice to make the preferences dialog /less/ consistent with the
rest of GNOME instead of more consistent. 'Remove options again' doesn't
really seem like something you want to do, though.

That said, this is something that you guys are making great strides on-
there are lots of 'make more HIG compliant'-type entries in the
Changelog, and you've made some nice efforts towards bringing back
features in a more user-friendly way than their galeon 1
implementations. This makes the choice less clear-cut than it was a few
months ago, which is great- competition is what makes us strong. :) 

(2) 2.3.2: Developer attitude. When I proposed that galeon be part of
the gnome 2.2 release, the reaction of some galeon developers was
strongly anti-HIG and strongly critical of GNOME2's direction.[1] Last I
was in #galeon, you still had a link to the 'why havoc is wrong' article
in the topic. That's not, um, 'supportive.' :) Contrast this with Marco,
who has very publicly and strongly supported the direction GNOME is
going in and the usability choices GNOME has made. I'm not sure what
your definition of fun is, but mine doesn't include 'working regularly
with people who think you're ruining GNOME by taking away their
options.' I'm afraid I don't have the concrete solution you're asking
for, other than 'be as outgoing as Marco in displaying your desire to
make things more usable.' :/ [Yes, that's a crappy answer, but I don't
have a better one offhand.]

So... those are my two big issues- letter of HIG law v. spirit of GNOME
usability, and developer attitude. That's why I (personally) would
prefer to see epiphany in the platform than galeon.

[OBVIOUS SHORTCOMING IN GEP 10[2] TO FOLLOW]
These are totally touchy-feely-can't-be-resolved-by-pure-code-alone
questions, unfortunately. Really, really, really unfortunately, GEP 10
doesn't really provide a mechanism for 'how do we decide what to do' for
this case- basically it says 'release team decides'. This is, frankly,
fairly crappy and more than a touch undemocratic. But I'm at a pretty
substantial loss as to how else it should be organized- clearly,
outright voting on the platform would be very messy. Saying 'well, we
trust release team' is nice but sort of scary. One suggestion I heard in
IRC was 'put representatives from each team, plus one from each new
module, as responsible people on the GEP.' This is a reasonable proposal
on the face of it, but it means that a set of at least partially
completely random people get to make decisions. (in the current GEP, for
example and no offense to noah, that would be 'hey, I wrote a new
character picker, so I get to choose which browser gnome uses!') So...
anyone who has better suggestions for how the decision on these types of
issues should go... let the list know :)

Luis

[0]On a personal note, I'm taking a leave from Ximian and North America
for the next six weeks, starting late tomorrow. See you on the other
side- hope this discussion is a vigorous and constructive one in the
meantim.
[1]See the galeon-devel archives from Oct. of last year, particularly
the 11th and 12th.
[2] screw GEP 10- GNOME has never had a mechanism for this :/