[Sugar-devel] Sugar Development

Jonas Smedegaard dr at jones.dk
Wed Dec 30 09:45:03 EST 2009


On Wed, Dec 30, 2009 at 02:07:07PM +0100, Martin Langhoff wrote:
>On Wed, Dec 30, 2009 at 9:11 AM, Tomeu Vizoso <tomeu at sugarlabs.org> 
>wrote:
>>> And a polite reminder that there are now at least three places you 
>>> might find tickets ... sugarlabs.org, laptop.org, and launchpad. 
>>>  ;-)
>>
>> Not really, every distributor of Sugar has their own tracker. I would 
>> suggest anyone wanting to help with this to focus on only one 
>> tracker.
>
>I grumble a bit about that -- the proliferation of trackers adds to
>our burden. Other than packaging issues, most bugs reported against
>any distro are very likely to affect the core Sugar and all the
>distros (since most of the stack is shared).
>
>We'd need a polished distributed bugtracker. And a pony.

Yeah, and we need a simple build system, not both autotools, cmake, 
sCons, ANT etc.  And a single programming language instead of both 
Python, Perl, C, C++ etc.

I suspect it is a dead run to convince all distributors and all 
upstreams to use same issue tracker. A monoculture is *much* easier to 
grasp, but also radically different from how FLOSS works (and works 
well, IMHO!).

So more realistically, I'd say that each derivative, either a 
distributor of some upstream source or a for(re)distributor of a 
distribution, should take responsibility on "giving back" their 
accumulated knowledge on bugs not only related to local oddities.

As I posted earlier: First work directly on the thing you are passionate 
about (i.e. the hacker approach of "an itch to scratch") and then 
if/when you have surplus energy then expand beyond your own selfish 
needs/wants: reflect on how you might improve "giving back" to your 
upstream.

Or concretely:

  0) File discovered bugs in local-to-you issue tracker
  1) Deal with bugs in local-to-you issue tracker
  1a) Fix bugs caused by local-to-you code changes
  1b) Pass on upstream issues that turn out to have a larger scope
  2) Hunt down issues remotely tracked that might affect you too
  3) Hunt down issues unstructurally discussed but not yet tracked
  4) Hunt down issues not yet discovered at all (e.g. evil code patterns)



If what you mean simply is that Sugarlabs, OLPC and SoaS (and those 
three alone) ought to share a single issue tracker, then sorry for my 
rant - I'll keep out of such discussion.



  - Jonas

-- 
* Jonas Smedegaard - idealist & Internet-arkitekt
* Tlf.: +45 40843136  Website: http://dr.jones.dk/

  [x] quote me freely  [ ] ask before reusing  [ ] keep private
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
Url : http://lists.sugarlabs.org/archive/sugar-devel/attachments/20091230/64a6cbcc/attachment.pgp 


More information about the Sugar-devel mailing list