Release cycle too short?

See also http://drupal.org/node/132496 (edit: happy rehashing).

--
The Manual | Troubleshooting FAQ | Tips for posting | How to report a security issue.

Johnyp, you're cruisin' ... for a brusin' now. :P

I agree with you. I think that the code freeze for Drupal 6 should be extended until sometime around November, and the release happen sometime around spring.

But, hey, I realize my two cents are more like two pfennigs. :) (That would be equivalent to an exceptionally small fraction of one US penny.)

Modules aren't the only thing lagging behind, documentation is outdated in a lot of places, too.

I was talking to a friend recently, and he suggested not allowing any new features to be added until full documentation had been written. Imagine that ;p

--
John Forsythe
Need reliable Drupal hosting?

I'm developing a new community site based on ver. 5.1. I hope this version will supported long time...

Regards

This has all been done to death before, but I'll add my 2 cents...

I know this is a controversial topic - but I personally feel that Drupal release cycle became unbalanced. Prior to 4.6 it was somewhat long, but that allowed for greater support for the current version, module migration to new version and more time in between major site updates.

I don't think that is the case - the 4.6 to 4.7 was one of the longest cycles ever due to the far reaching Forms API changes and a bit of an exception. The quicker 4.7 to 5 cycle was a more normal length one, and a much easier one to manage with far less changes needed. Also the Drupal 5 release was the first one I've seen where a huge number of modules were ready to go at release time. With previous releases it seemed that the updating work only really started after the release.

Upgrading to new Drupal versions is getting easier and easier every release.

And now - even though Drupal 5 just came out, some 4.7 modules still haven't been updated to v5, version 6 is already on the way!

What makes you think that all those modules need is some more time? Most of the modules that haven't been updated by now are probably modules that the maintainer doesn't need to update ie they may only be needed on some of their legacy sites (if at all) or the maintainer has better alternatives now. A counter argument [playing the devils advocate here] could be made that that dropping core support for 4.7 might actually help those modules get updated by creating that need.

5 still seems to suffer a bit from performance issues ...

Releasing Drupal 6 is how those performance issues with Drupal 5 get fixed. Fixing them requires redesigning and altering stuff, and Drupal 5 is in stable maintenance mode and only gets bug fixes. A new release is how Drupal fixes design problems.

... yet the same features that people wanted ever since v4 are still not in the plans for the core.

How do devs decide on when to start working on the new version?

As to what ends up in a new release, well that depends on what gets done. Dries is the project leader and basically OKs all the changes to core before they can be committed (a couple of other people can also do that) which is a big job - so nearly all of the actual coding is done by others. Dries can't commit a new feature or the redesign of an existing feature that nobody has written and he can't force anyone to write it. All he can do is state what kinds of things he'd like to see get done so that the other devs get an idea of how likely their changes are to make it in or not.

Developers decide what do do based on what they want to do and what others (eg their clients or sponsors) pay them to do. Generally the best way to get someone to do something they don't necessarily want to do, or to rearrange their priorities is to pay them.

Maybe rushing them out the door should not be such a priority compared to supporting modules and planning the new version from the user input?

There is no planning beyond Dries stating things he'd like to see. Planning would require a level of control over the developers he just doesn't have. There is no rushing out the door though - some of the changes that make it into a new release have spent years being designed and worked on. If an idea isn't polished enough, it waits for another chance to make it into the next release.

And as for supporting 4.7 after the release of 6, that will happen if someone steps up to keep supporting it. The only thing stopping support for an older version is that noone wants to do it. If a group of people care enough about it but don't have the skills to do it themselves, they could always organise some sort of sponsorship to make it happen.

--
Anton
New to Drupal? | Troubleshooting FAQ
Example knowledge base built with Drupal