Some personal notes on this as a community project user:

• Andre has changed github stuff indicating that the next releases will be more focused on 5.0.0alpha1, this means this release looks like the last one of the 4.x series and for now we can expect just minor bug fixes for the 4.x branch, which is good signal 'cause this release looks very stable (it includes the latest critical bug fixes) and uses responsive layout (new ezdemo design), so it's a nice point to stay until 5.0.0 release (maybe until 5.1.0 if you really want to migrate when everything is really stable).
• Instead of expending time creating extensions and tutorials for the current eZ publish series, 4.x, better to focus on testing 5.0.0alpha1 'cause in a few months most of the current stuff will be obsolete and community will tend to support the 5.x series.
• This may be a nice moment to become an eZ publish developer as long as most of the current established eZ publish developers will not have much more advantages than a good developer at any other PHP CMS. So all a developer needs to do is start studying the 5.x releases from now.
• If eZ systems do it right there will be a lack of eZ publish 5.x specialists, I'm sure most of current developers will sticky with eZ publish 4.x for a long time before getting ready to do real 5.x projects, so the Darwinian principle, adapt or die, will be applied here and maybe we will see some new winners in this scenario in a small period, like 1 or 2 years.
• The summer camp will not be as useful as it could be if it focus on 4.x, it's a big investment on a thing that will be obsolete in a few months, so it can be too risk and not worth it, it would be nice to start creating a base of eZ publish 5.0 experts in September instead of focusing on 4.x, even using 5.0 alpha and beta versions.
• This is a good time to do this 'cause the later all this changes happens the worse it can get.

Modified on Friday 01 June 2012 3:46:17 pm by Thiago Campos Viana

While I have a lot of expectations of the upcoming version 5, I sincerely hope that version 4.x does not die such a quick death as you are envisioning.

First of all, the whole of the 4.7+ kernel will we shipped within version 5.0.

Second, there's so many opportunities to improve and optimize the 4.x kernel, that it'd really be a shame not to implement them. I'm talking e.g. about all of the open pull requests, such as 344 from crevillo, 292 from ballinette or 152 from omh.

The Open Source model can really bear some fruits in this situation I think, as eZ Engineers are busy working on the new stuff, community developers can step in and continue sending pull requests which scratch their own itches.

Last: I think that while it makes a lot of sense to start learning how to use version 5 - and participate in the development process to shake as many bug as possible and make sure all usecases are catered for - existing sites will not migrate all in a rush, and the need for developers competent on eZ 4 will remain for a while.

Just like cobol: it will not go away in a day.

Quote from Thiago Campos Viana :

This may be a nice moment to become an eZ publish developer as long as most of the current established eZ publish developers will not have much more advantages than a good developer at any other PHP CMS. So all a developer needs to do is start studying the 5.x releases from now.

By the same token, it may also persuade eZPublish developers to look at other frameworks since the learning curve will be there anyway.

I'm pretty sure things will be managed well and that backward compatibility will be maintained and that changes will be incremental.

And while I'll be watching code changes closely I also think I'm going to be taking it slow for clients.

Speaking of which I still know of sites that are on the 3.x version of eZ... and up until recently some 2.x sites.  So some users will be slow.

If I remember correctly, the 3 series didn't really hit its stride until 3.6 and the 4 series not until 4.3 IMHO.

If you have issues when indexing content in Solr, maybe it's because you are affected by the regression described in this topic.

Cheers

Hi!

Thank you for 4.7! Works really great BUT, you should UPDATE documentation page regarding "Speedup ajax calls" that if you use it, file index_ajax.php is missing from JSCore extension and should be copied from previous version (2012.1). Just lost an hour hunting ghosts.

Best regards,
Sinisa 

Quote from Siniša Šehović :

Hi!

Thank you for 4.7! Works really great BUT, you should UPDATE documentation page regarding "Speedup ajax calls" that if you use it, file index_ajax.php is missing from JSCore extension and should be copied from previous version (2012.1). Just lost an hour hunting ghosts.

Best regards,
Sinisa 

Good catch, Sinisa.  I think that index_ajax.php was actually purposely removed because it wasn't providing a material performance boost and needed to be constantly updated.  So the documentation should probably mention that it is only applicable for older installs.