-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thursday 11 September 2003 19:42, Dominique Devriese wrote:
> I don't think this is a good idea.  We can't move everything into
> kdebase..

This has also been my way of thinking always. But OTOH, why is that 
really?

Try imagining what it would be like, if we collapsed kdelibs, kdebase, 
kdenetwork and kdepim into one big pile.

First of all it would be a big chunk to download over modem. But if the 
modem users all download all the modules anyway, that argument is moot.

The distros could continue to package stuff pretty much the same way they 
did before.

Stuff that depends on runtime parts from other modules won't run without 
the other anyway. I dare you to come up with actual users who have 
kdelibs but not kdebase installed. And it's impossible to install kdepim 
without kdebase for example, since necessary io slaves are in base.

The modules are all seen as one big chunk anyway - KDE.

In fact, as I think about it, I see only two really serious reasons not to 
do this:

Security releases could be slowed down, due to more testing being needed. 
This could be solved by only releasing the last version plus the fix, 
though.

Developers job would be made harder. It's already annoying to wait several 
compile minutes on testing a fix in kdepim/libkcal because the linker 
will run all over the place. I know I would start having compile scripts 
to compile only the parts I am working in.

So, why exactly is it that we don't just pile everything up? There are 
pros and cons of doing it, but "that's just the way it should be" is not 
enough to justify anything.

Bo.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2-rc1-SuSE (GNU/Linux)

iD8DBQE/YMaImT99lwfUS5IRApajAKCpsm6o1NGKKXBKHhH8H6cdGyg60gCePI+h
IfidHbSqafwDnyeuGayG1BI=
=L3L5
-----END PGP SIGNATURE-----

 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<