Some of us at Netgen attended eZ Conference a few weeks ago in Berlin (and had a great time there :-), so we would like to present eZ developers with a small summary on few topics that we think are important.
Two code bases
It was announced by eZ crew that in the future there will be 2 code bases for eZ Publish CMS: one for enterprise version and the other for community version. Preparation for this have already began, with current code being moved to github (http://github.com/ezsystems/ezpublish) few days before the conference.
In our opinion this decision could be very good for the future of eZ Publish but only if it’s being conducted properly:
- Core of the community version should be kept under control of eZ crew and some prominent eZ community people (community board). This could enable faster and easier creation of new features and bug fixes, together with other useful features that come from stronger community engagement (faster testing of new features, etc). After some time features added in community version would be introduced to enterprise version. Due to this, enterprise version would lag a bit with its feature list, but for solutions where stability is the most important factor, it would be the correct choice.
- Extensions in enterprise version should be kept under eZ crew control. These extensions would be built by them or certified by them.
- Extensions in community version should be under community board control. Useful and mature extensions from project.ez.no should be included into community version (in close cooperation with extension authors, of course). In that way public will benefit from having more features introduced regularly.
In general we should have a choice between less featured stable version and more featured community version (but still good enough for lot of projects).
Solr as a content storage
In future eZ Find versions we should have an option to store complete eZ objects into Solr engine. Currently, the main application for this would be archiving old content that does not need to be editable any more. On very large sites with fast growing content (like news portals, forums, etc.) big database tables are becoming performance bottleneck, so the possibility of periodically shifting old content to Solr could be a life saver. Archived content would be still searchable. Few notes though:
- there should be an “unarchive” function
- links to old content should work
There are also other use-cases for using Solr engine because of the performance gains it brings. One example is shown in our tutorial about replacing standard list/tree fetches with ezfind search (available here).
The last but not the least is the new API announcement. eZ crew will implement in forthcoming versions a new, more abstract API . It will not replace the existing one, just add higher level features for often used tasks (e.g. content publishing, editing, ... ). This API will also be accessible through REST. This announcement was loudly approved by the audience when announced. It will surely provide easier way to implement various adaptable interfaces (e.g. for mobile and tablet devices) and integration utilities.
@eZ please, develop this soon :-)
We are glad we were there and met interesting eZ people (from both eZ Systems and eZ community). And it was very nice to experience city of Berlin: from grand Alex and Brandenburger Tor over charming Ku’damm to artistic and youthful Kreuzberg. Till next eZ conference (rumor is it will be hosted in London)...