Re: [Evolution] a challenge to the developers

[Patrick O'Callaghan <poc usb ve>]

On Mon, 2009-11-16 at 08:13 -0500, Adam Tauno Williams wrote:
> On Mon, 2009-11-16 at 08:24 -0430, Patrick O'Callaghan wrote:
> > On Mon, 2009-11-16 at 12:22 +0200, Tor Lillqvist wrote:
> > > > So here is my challenge to the evolution development team:
> > > I am sure the team is fully challenged already. You are directing your
> > > challenge at the wrong address. It would make more sense to challenge
> > > the team's employers (various Linux distros). It's they who decide
> > > what the team works on. Companies like a challenge to be spiced up
> > > with a reward. You know, like, "make sure these bugs get fixed and we
> > > will purchase 1000 support contracts/subscriptions/etc".
> > While bounties can be useful in limited contexts, I don't think a free
> > (in both senses) software project can be guided only by interested
> > parties offering money.
>
> A 'free software project' is ultimately, like *everything* else in life,
> guided by the contributions; whether those contributions are motivated
> by fiduciary remuneration or anything else. 

In the case of free software the "contribution" may be money, or code,
or documentation, or simply reporting bugs.

> > My personal gripe is that Evo occasionally freezes so badly that only a
> > forced shutdown will kill it. I've reported this repeatedly with
> > previous versions, both on this list and via BZ, sent in traces etc.
>
> Maybe nobody else experiences this?  I've not ever had Evolution
> hard-freeze that I can recall.  Occasionally I've seen it run into an
> apparent file-lock bug with Beagle, but that is just about the only
> freeze I've seen.

I don't use Beagle so that's not it. Certainly the problem could be
something very specific about my setup, but without feedback it's very
hard to tell. I did have a useful interchange with Srini about one class
of problems of this sort, caused by a slow LDAP server checking
addresses during Junk processing. Working around this made a noticeable
difference, but some freezes still happen, and have done in every
version of Evo I've used, right back to the 1.x series.

> It doesn't handle network issues very gracefully, but
> none of the apps I have do [ as a developer I know that doing so is
> really very extremely hard ]. Watch Outlook after a network issue, it
> typically goes bat-crap crazy; way more developers work on Outlook than
> Evolution.

As I mentioned in a recent thread, handling network outages in an email
client doesn't need to be extremely hard. Email protocols are designed
to be robust in the face of flaky networks. If a remote server doesn't
respond, just timeout, abort and tell the user. After all, that's how we
*already* handle it when we kill and restart Evo and I've never lost a
single message through doing so.

> > It would be nice to hear someone on the devel team say what the worst
> > bugs are thought to be and explain what they propose to do about them.
>
> I'd prefer they write and debug code.

False dichotomy. It's one or the other.

> > Unfortunately the various roadmaps, version by version, say nothing
> > about prioritizing bugs, only about adding features.
>
> Because that is what a roadmap is.  Where is there a roadmap, for any
> project, that talks about bugs?

And this is good?

> > It's also rather depressing to see that the latest Release Notes are for
> > 2.24 (http://projects.gnome.org/evolution/documentation.shtml) and the
> > latest roadmap (http://www.go-evolution.org/Evo2.26) concerns the
> > planning for 2.26. That version was released more than 6 months ago, so
> > the planning must have happened several months earlier. It's hard to
> > avoid the impression that Evo is struggling. I really hope I'm wrong.
>
> Dude, pay attention to the list.  They just did a complete port off of
> Bonobo, which was a big change to the underlying architecture.  Clearly
> someone is investing a lot of energy into it.

I do pay attention to the list. I also post a good deal to try and help
other users (http://markmail.org/search/list:org.gnome.evolution-list
says I'm the 4th most frequent poster here, which surprises me but there
you are). I've no doubt the Bonobo stuff is a significant effort and I'm
sure we all hope it leads to a better Evo down the road (though it's
worth noting that it's happening because Gnome itself is moving away
from Bonobo; if it weren't for that, would the change to Evo be
happening at all?). The question is whether there are enough resources
to do what needs to be done, including the continuing work on Exchange
compatibility, which personally is of no interest to me at all but is
obviously important for many users, who currently have no other free
alternative to Outlook in a corporate setting.

So my impression that Evo is "struggling" doesn't mean people aren't
working hard. You can be working hard and still be struggling. One way
to ameliorate that impression is by being more communicative. Learn some
lessons from Fedora, or Ubuntu, or even Mozilla.

> And Evolution remains, as it has been for many years, THE premiere Open
> Source mail client and PIM.  There is nothing else that comes close to
> the functionality and stability provided by Evolution.  The closest is
> probably Thunderbird [known around the office as 'Thunderous Turd'], and
> talk about issues.... That is a distant second to Evolution.

I don't want to get into a point-by-point comparison here. I regularly
look at other MUAs as alternatives to Evo, including TB, Kmail, Claws
and Mutt. I stay with Evo mainly because it's what I'm used to (inertia
is a powerful motivator) and I try to make constructive criticism to
help it improve. But the elephant in the room is this: given the recent
opening up of the MAPI protocols by M$, what happens to Evo if (or more
likely *when*) TB acquires an Exchange plugin?

poc