Here are some notes from the Drupal 8 status update talk that Dries and I gave today to the Acquia team, since this seemed like useful info for the community to know as well. :) It covers both process changes for addressing previous issues that arose in Drupal 7, as well as a status update on Drupal 8 progress to-date. This could be useful to folks who have been wondering where all of the various Drupal 8 status updates fit into the "bigger" picture.
Please comment here if I left anything out, or messed up anything.
Background of D7 challenges and how we’re addressing these in D8
- Two core maintainers overseeing entirety of Drupal core doesn’t scale; appointed “initiative owners” to take on strategic aspects of development.
- Bi-weekly “scrum of scrum” calls with initiative leads to get them unblocked.
- Established http://groups.drupal.org/drupal-initiatives as central place to keep up to date on D8 initiative progress and get involved.
- Working on “Community Initiatives 2.0” as a way of providing better structure to these and other initiatives http://drupal.org/node/1300972.
- Appointed Nathaniel “Catch” Catchpole as Drupal 8 co-maintainer
- Focus on reducing complexity of core and increasing performance.
- In close contact with Angie/Dries to make sure his own patches get reviewed/committed in a timely manner
- Gave Angie access to commit to D8 for D7-affecting bug fixes/tasks
- Effectively raising the number of core maintainers to 3 for the most severe issues.
- RTBC queue for D7+D8 down from high of ~120 in the summer to ~10-15 pretty consistently now. YAY!
- Established “Issue thresholds” of 15 critical bugs/tasks and 100 major bugs/tasks.
- No new features get committed to D8 while we are over thresholds: focus turns to bug fixing/stabilization so we can get back to a “stable state.”
- Means we can release D8 within a few weeks of freezing the code, as we’ll never have more than 30 or so release-blocking issues.
- Worked with accessibility, usability, performance, testing, and documentation teams to establish “gates” that form checklists of things to check for core patch reviewers: http://drupal.org/core-gates
- Created http://groups.drupal.org/core as an “announcement-only” list of major happenings in the core queue.
- Started “Core office hours” as a way of on-boarding new contributors: http://drupal.org/node/1242856
Drupal 8 progress to date
- Canvassed both large Acquia customers as well as the wider Drupal community to identify key strategic initiatives for Drupal 8.
- Identified “initiative owners” to take point on each of these, and to spread out responsibility/authority among more than just 1-2 people (from Acquia ;)).
- Greg Dunlap: Configuration Management
- Larry Garfield: Web Services
- Jeff Burnz: Design
- Jacine Luisi: HTML 5
- Gábor Hojtsy: Multilingual
- John Wilkins: Mobile
- Most initiatives have bi-weekly IRC meetings announced at http://groups.drupal.org/drupal-initiatives to provide status updates/assign tasks.
- Configuration management
- http://groups.drupal.org/build-systems-change-management/cmi
- This initiative deals with how to move content and configuration from dev to stg to live.
- Content: UUID patch committed which allows unique identifiers to allow for content staging.
- Next step is refactoring entities to use it.
- Config: Developed architecture that allows for saving buttons in the admin interface to automatically generate XML files holding configuration data. These can then be moved around with version control. The system intends to replace variable_get/set as well as import/export functionality in Views, Image Styles, etc.
- Initial prototype code at http://drupal.org/sandbox/heyrocker/1145636
- Recent sprint at BADCamp with yched, sun, davidstrauss, etc. to talk about how CMI can work with fields in core.
- Trying to get a couple of pages in core done as “proof of concept”, then initial core patch will follow.
- Web Services
- http://groups.drupal.org/wscci
- This initiative consists of four parts: a unified context system (to replace global $user vs. arg(1), etc.), a unified plugin system (to replace variable_get(’filename’) vs. hook_something(), etc.), turning Drupal into a web services platform (providing REST APIs for getting JSON, XML, etc. out), and revamping the page rendering structure to allow for chunks of content to be re-ordered in whatever way (aka “Panels in core”).
- Initial prototype code of context system at http://drupal.org/sandbox/Crell/1260830
- Recently committed Symfony 2 HTTP library to core so services API can take advantage of it and it does not need to be more code we maintain ourselves.
- Part of a larger imitative to have Drupal play nicer with the larger PHP community, lots of outside interest. http://www.garfieldtech.com/drupal-symfony2
- Design
- http://groups.drupal.org/design-drupal/d8di
- Two-pronged approach: new responsive, mobile-friendly front-end theme for core, and also improvements to Drupal.org to make it more designer-friendly.
- Design.drupal.org prototype: http://groups.drupal.org/node/161204
- Initial design brief proposal: http://groups.drupal.org/node/179079
- Currently Jeff is helping with HTML 5, since is largely blocked until that and snowman happen (so people know what they’re designing for/with).
- HTML 5
- http://groups.drupal.org/html5/drupal-8
- Goal to convert Drupal 8’s native HTML output to HTML 5.
- Other parts of initiative include: native HTML 5 Form API elements, Modernizr/other prototype library in core, CSS clean-up
- Current sprint focused on core template conversion: http://drupal.org/project/issues/search/drupal?issue_tags_op=and&issue_tags=html5%2C+sprint
- Just committed html.tpl.php patch - Drupal 8 now outputs HTML 5 DOCTYPE! YEAH!
- Multilingual
- Roadmap: http://drupal.org/node/1260534
- Unifiying the language system, bringing in good ideas from contrib to core (e.g. content/entity translation), better UX for supporting complex translation interactions
- Work so far primarily around cleaning up underlying language system, but also some solid progress on UX as well thanks to Berlin/Montréal sprints.
- In close contact with CMI/WSCCI initiative leads to ensure multilingual concerns are addressed.
- Mobile
- Roadmap: http://palantir.net/blog/drupal-8-mobile-initiative
- Just kicked off last week; not too much progress yet. :)
- In many ways a “glue” initiative to tie together elements of WSCCI, Design, and HTML 5. Covers things like responsive/mobile-friendly conversion of core themes, a solution for dynamic image resizing, and front-end performance
- Goal: Make Drupal the “go to” CMS platform for mobile. YEAH!
- Other initiatives identified by survey, yet to be announced
- WYSIWYG
- Usability/ease of use
- Media/asset management
- “Better APIs”
- Content import/export & Content staging
- Both part of CMI
- Non-initiative progress
- PHP 5.3 as a baseline, allows for better support for OO, namespaces, and other cool stuff.
- Reducing complexity/code bloat: Profile, Blog, Trigger, and Garland removed from core.
- “Unofficial Framework” initiative http://drupal.org/node/1224666 working towards simplifying/de-coupling subsystems
- Numerous performance improvements, most backported to D7
- “Snowman” http://groups.drupal.org/snowman attempting to build useful profile using Drupal core components
- Usability testing done at University of Minnesota in May exposing D7UX was a success (yay!) but still more work to do, esp around site building tools: http://drupal.org/node/1175694
Comments
The progress is really great.
The progress is really great.
It's sad, though, that I don't see any semantic web initiative, which is the next evolution of the web.
I wish you to keep up the good work!
There's already rdf output
There's already rdf output support in Drupal 7 core, did you miss that or are you thinking about something else? Note that just because something isn't listed here doesn't mean it won't happen, but it needs people to work on it.
D8 should get leaner and lighter
Hi Angie,
Thanks for insightful updates, it gives lot's of hope that D8 will finally go the right direction. Better API, HTML/CSS are getting polished, the core framework will get less complex and lighter. I wished the Theming layer and API's will get easier to adopt and learn for front-end devs coming from other CMS communities, attracting more and more end-users to favor Drupal.
Tom
Right on brother!
Right on brother!