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

Re: blends and Google Summer of Code?

Hi Yaroslav,

thanks for the remark.

On Wed, Mar 27, 2013 at 08:36:59AM -0400, Yaroslav Halchenko wrote:
> > is now the right place to go.
> to me the goal is not quite clear.  First you push into a direction of
> using debtags, but because (as we discussed at some point I believe) the
> tag-set is quite limited ATM we cannot really use it ATM and it would
> require more of discussion/coordination with debtags people to make it
> useful for our purposes, than of any actual software development.
> Then you describe "The task of the GSoC project is to rewrite the
> blends-dev code to enable architecture dependent metapackages." which is
> a clear statement but has nothing to do with tags ;)

Right.  Finally this is also the main point and I'd call the project
successful if this is solved.

> So if it is just about architecture dependent packages -- drop debtags
> from description, may be?  or discuss with debtags people (Enrico?)
> first on possibility of formalizing/extending debtags space to cover

I wanted to explicitly mention Debtags because I'm hoping that some
"fresh blood" could get new / better ideas than me and might find an
easy way to include Debtags information.  For instance in an UDD based
method to create metapackages you (could) have tasks and debtags
information side by side.  (The "could" part is just discussed in
another thread here on the list - read more in a couple of minutes.)

So we could in principle merge the metapackage creation procedure (to
make sure those people who say "I prefer debtags over tasks - and so I
do nothing for the Blends framework tools" will lose a (non)argument).
IMHO this can be discussed in the GSoC term.  The problem with
discussing this in advance is that those few people who are involved are
way to busy to discuss anything - at least this is my experience from
the past.

Feel free to edit the wiki to make this idea more precise.

Kind regards



Reply to: