Re: [GNOME-India] foss.in 2008...



On Mon, Jul 14, 2008 at 4:03 PM, "Sankarshan (সঙ্কর্ষণ)"
<foss mailinglists gmail com> wrote:
[...]
> Easy dear. Now, have you successfully and emphatically defended your
> procrastination in front of a panel of eminently uninteresting folks ? ;)
> aka, are you ready to join in this fun foss in08 agenda ?

Yep! :D

>> If we're talking about getting people contributing, I think we need a
>> set of minimum requirements. Not in the don't-come-if-you-don't-know
>> way. More like
>> read-these-nice-easy-tutorials-before-you-come-so-we-can-really-get-to-the-next-level
>> way. Yes, yes, hyphenation embargo for me.
>
> Good catch. So, that generally means:
>
> + we need to list out what can be demonstrated / talked about
> + we need to hunt around and keystroke in a list of preliminary reading
> + we need names to put to the potential demo-monkeys

[...]

> More than mere morsels. A thing I wish to see each time there is a GNOME
> event but don't get is performance-related debugging of GNOME - do you know
> anyone who can start taking IRC sessions leading up to a bling filled
> foss.in talk ?

This is a fantastic idea. I think this is good enough to make the
theme of the event, and if we can pull that off, we can make a _real_
impact. I suspect that there is a lot of low-hanging fruit that can be
picked and can make for a really solid contribution, especially w.r.t.
our own context.

For example, I was looking at these little single-piece Celeron-based
boxes that are being targeted towards rural areas, and the current
footprint of a full-fledged GNOME system would probably bring these
things (~256 megs of RAM) their knees, even though there's no good
reason for this to be so.

Things to know as prerequisites would include valgrind, memprof, gprof
and gdb (nemiver, anyone?), presumably. Also need a reasonable
understanding of how memory allocation happens, specifically w.r.t.
gobjects, X, etc. (I haven't the faintest, FYI). I think
rasterman/someone had a talk around somewhere. Will try to dig it out.

The Mono-based apps are a whole different ball game, too, but I think
they could be considered. Mono itself has some decent tools for
memory/runtime profiling.

Of course, we need people (IRC or in person) to help interested
members bootstrap. I'll ask around for ideas/suggestions.

Cheers,
-- 
Arun Raghavan
(http://nemesis.accosted.net)
v2sw5Chw4+5ln4pr6$OFck2ma4+9u8w3+1!m?l7+9GSCKi056
e6+9i4b8/9HTAen4+5g4/8APa2Xs8r1/2p5-8 hackerkey.com


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