[development] 'old' core 6 & new 7, working together..? +crossite

[Gerhard Killesreiter]

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

Ad schrieb:
> (I tried to respond on a particular message, but don't know how)
> 
> = = = = =
> 
> Thanks for all the feedback. Ofcourse; backwards programming doesn't
> make sense. But I encountered several sites running on core 5,
> because there isn't even an upgrade for the mods they need to core
> 6, or only a preliminary one.  I'm not going to make a report of
> what I encountered (last 4 days uploading many-many mods); probably
> all here are familiar with the problem.
> 
> There are mods that have something about what I suggested, like this 'outdated' module I just encountered:
> http://drupal.org/project/crossite
> 
> Many usefull mods are written for 5 and never updated for
> 6, see for example this hyper-modern site that untill now only can run on 5:
> http://drupal.org/node/342472

That's unfortunate. But since it is Open Source you are talking about,
you can also do something about it: Get somebody to take over that
module and make a release for Drupal 6. If you can't do it yourself,
pay somebody to do it.

> So the question: what is outdated in these? Ofcourse the new core 7,
> not at all. But the whole package with the mods, will be literally
> outdated for one year, up to some years.
> 
> Why not getting crossite running again, in a new and broader &
> standardized form, for both 6 and 7? Maybe even changing the 7 core
> before the definitive first release, to anticipate for this
> development & make some transitions smoother?  , IF SO:
> 
> 1) one could run some parts on core 6 and others on core 7, for the
> first 2 years untill almost all necessary mods are on 7. Everyone
> could migrate inmediately to 7, so it would be a step forward with a
> smooth transition, not backward;


This would be an IT nightmare.

> 2) The same system (updated) can in future also be used for 7 to 8,
> so its not a one time case and lost creativity;

Just countless hours spent by IT people.

> 3)
> Several other web-programs have some bridge with Drupal, but quite
> amateuristic and no broad development support to get the
> basic/general API ready for all of these specific bridges (,
> communicators, exchangers, integrators a.s.o.).

That's probably because most of these bridges have very limited appeal
to real Drupal users (the various forum bridges for example).

> It might be web-worlds future hype, or already is with integrating
> gadgets, cellphones, twitter, facebook, groups, mediawiki and so
> on. Mediawiki was partly running, but with new releases it stopped
> running again, lacking a protocol for both Drupal and
> Mediawiki.

You can make a fine wiki using Drupal only and I bet there are ways to
set up shared log in across both.

> Yes, it is difficult to get the core optimized for this and make a
> sort of inter... protocol how programs best can bridge with as many
> aspects as possible. But because of the importance, to my opinion it
> should become a main developers community task to get done
> 
> Do we still need to wait untill core 8 for such a development?
> Actually I had a test site on 5 and had to wait 2 years before not
> even all required modules were up to date. So I might start soon in
> 6, but I'm still missing some features from 5.  If possible, I
> recommend to start working on bridging even before the release of 7,
> examining if there better needs to be changes in the basics of core
> 7 first, to optimise for an additional 'general bridge API'.

To whome do you want to make your recommendations? Your proposal has
two flaws:

1) There is probably nobody interested in executing it.

2) It probably won't work since all these projects work all in rather
different ways.

Cheers,
	Gerhard

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEARECAAYFAktDRaEACgkQfg6TFvELooQPjACfZMczuXYKnxxRU9TdCdQWnJAS
qTEAmwct45Ue8ieBOpolBbUakYKc3+OA
=AqpP
-----END PGP SIGNATURE-----