[Sugar-devel] On Sugar 0.84 - status of the "Chat/collab leader" issue...

[David Van Assche]

Well, at least on Gnome, Mission Control not only works well, but its
far more stable and does what its supposed to. Its been very heavily
tested by Nokia (Maemo), Collabra, Google, openmoko and other heavy
hitters. I don't really agree that we have something that works with
sugar presence. In the majority of cases, where  we've had testing
sessions, though admittedly, with badly callibrated xmpp servers, I
would go so far as to say that it was attrocious in terms of
performance and stability. Once connected, collaboration worked great,
but the stuff that happens before that, which is what sugar presence
is supposed to be taking care of does not work well at all. If you
take a look at telepathy-inspector and the advancements in telepathy
itself, of which mission control 5 is one of the major overhauls, its
massive improvement over the passed. And one of the main issues was
that sugar presence used its own bindings, blind sighting a lot of
what telepathy is doing, which is why currently it simply doesn't
work. Without a xmpp server, you'll have a field day getting any kind
of collaboration to work, and even with a really carefully setup
ejabbberd server full of optimisations, I at least, have not been able
to get the presence part to reliably do the same thing every time.
Some times people show up, sometimes they dont sometimes 10 minutes
later, sometimes with totally weird settings and names.... Its quite
clear to me that what may have worked ok in 0.82, now does not. And to
me that makes total sense, if u look at the timeline, the code, the
blueprints, and most importantly, the actualised telepathy dbus
bindings (The presence part has changed completely and looks nothing
like it did when 0.82 and earlier were coded.)

But dont take my word for it, take a look here and you'll see what I
mean: http://people.collabora.co.uk/~danni/telepathy-book/chapter.accounts.html

kind regards,
David Van Assche

On Wed, Nov 4, 2009 at 10:38 AM, Martin Langhoff
<martin.langhoff at gmail.com> wrote:
> On Tue, Nov 3, 2009 at 9:54 PM, David Van Assche <dvanassche at gmail.com> wrote:
>> moving to mission control 5 and letting go of the admittedly
>> antiquated sugar presence now
>
> In planning future work in rpesence and collab stuff, I have a small,
> humble suggestion.
>
> Figuring out if a presence service / collab infra works and scales
> properly on both wired and wireless networks is hard. Very hard. We've
> been gotten it wrong several times by looking at the theory (instead
> of hard-nosed testing).
>
> Right now we have something that -- while less than ideal -- at least
> works for a number of scenarios.
>
> If you play with a major component replacement
>
>  - test it for scalability & stability over wifi before doing a lot of
> integration work
>
>  - do the integr work on a branch
>
>  - test that the integrated thing works stable and scalable
>
> Of course that's ideal world stuff. However, the heart of the matter
> is: approach mission control tentatively... and at least _some_
> significant testing needs to happen before it's merged...
>
> We've gotten this wrong a few times -- I am not keen on repeating the
> adventures... :-/
>
>
>
> m
> --
>  martin.langhoff at gmail.com
>  martin at laptop.org -- School Server Architect
>  - ask interesting questions
>  - don't get distracted with shiny stuff  - working code first
>  - http://wiki.laptop.org/go/User:Martinlanghoff
>



-- 

Stephen Leacock  - "I detest life-insurance agents: they always argue
that I shall some day die, which is not so." -
http://www.brainyquote.com/quotes/authors/s/stephen_leacock.html