My oh my where to start. This meeting had over 110 people in attendance and at times, was pretty crazy to follow as it seems multiple conversations were taking place at once. The following is a cliff note version of the meeting although I highly recommend you view the log file and read the conversation in it’s entirety.
Also want to say right up front that the ETA on WordPress 3.0 HAS NOT been determined. I jumped the gun by saying March and I was surprised to see this month mentioned as a goal for the release of 3.0 but nothing has been set in stone.
-brief- update on teh merge, which has started
Ryan Boren updated everyone on the status of the merge by stating Donncha is working on WordPress MU 2.9.1 and once it’s released, they will get it up to sync with trunk and start pulling everything else down. A MultiSite component has been added to trac to keep track of tickets related to the merge. So, we’re still early days. wpmuguru is gonna work on the wpdb stuff tomorrow and they will hack away at it piece by piece. Ryan said that it’s quite possible that by the end of next week, people may be able to start playing around with the merged files. Here is a link to the Multi-Site specific tickets. They don’t plan on any extras to the merged codebase such as Network/Domain UI. They will also be working on preserving backwards compatibility. Domain management might be added if they have time but it might also be relegated to a core plugin.
I had asked the question on whether shared webhosting providers would ban the use of WordPress 3.0 because of the multi-site capabilities being added into the core software. It looks like this decision will be the responsibility of individual webhosts. I sure hope this isn’t the case but in most instances, going hog wild with MU on a shared webhosting account is a violation of the terms of services for most hosts.
Strider72 reminded us that WP will work the same way unless a “use MU” constant is set in config.
Many people agreed to the suggestion presented by Peter Westwood that when they think the merge is done they should release a preview release – i.e pre-beta for the MU community to hammer on.
StevenHodson brought up an interesting question regarding the merge and Windows based hosting. As probably the one lone person running WP on WinServers will the merge follow WP being able to play nice on WinServers (albeit the ‘pretty’ urls have index.php in them) or will it follw WPMU which wilol not play nice at all on WinServers? Westi replied that if it is possible it will. Steven will be reporting any issues he encounters with Windows hosting and the merge to try and get them resolved.
Ryan stated that Current MU API stays the same regarding site and blog. Network/Domains won’t be exposed in 3.0 so it’s a non-issue for now. We’ll keep it in mind as we go though so we don’t step on the future.
The next part of the discussion centers around the ability of upgrading to WordPress 3.0 and turning a single site into a mult-site instance. Right now, there is no easy way to do that and would require some sort of conversion script. This has been added to the scope list but no guarantees of actually happening. Most agree that conversion should be part of the merge as most will expect that functionality out of the box but as of right now, it doesn’t exist.
Peter Westwood brought up the topic of Canonical plugins as one of the things he’d like to see with 3.0.
One of the things I would like to see with 3.0 is for the first few canonical/core plugins to be launched along with core changes to make them feel built-in
Westi announced the first plugin which he would like to see as a candidate for this canonical plugin structure. It’s called Giving Your WordPress A Check Up. These plugins need a community of developers like the core to survive. Westi has also started on a second one where he takes post by email out of the core and puts it into a plugin to get more love.
So if you would like to get involved in the health-check plugin – which is super easy BTW then please let Peter Westwood know and he’ll give you commit access.
There was no mention of a proper selection process of choosing what existing plugins will be canonical but Mark Jaquith mentioned picking plugins is the wrong way of looking at it, it’s more like picking areas. Jane Wells mentioned that setting up a core plugin system in itself will require some thought and work. Do we need Tracs, mailing lists, etc? It would be good to have two volunteers other than lead devs to help Shepard this process.
Some folks volunteered to help pioneer the effort. Also, Peter put the call out for theme authors: If any of you are theme ninjas then your input to the post-by-email plugin would be much appreciated as he would like to launch a simple theme as part of the plugin for no nonsense email blogging.
WordPress 3.0 Scope
After reading all the feedback on the WordPress.org forum thread regarding the scope of WordPress 3.0, much of it centered around the merge and fixing bugs with a side of custom post types. One feature that accumulated quite a few votes was better menu management. But after much debate about this, it seems like better menu management will happen gradually. Page menu management has sucked in WordPress for as long as I’ve been using the software. Eventually, it will get the attention it deserves.
As an aside, Matt Mullenweg mentioned that Media will not happen in WordPress 3.0 one way or the other. He also mentioned that he is taking the reigns on developing the new default theme, 2010. That ought to be interesting. Let’s hope it’s not blue with rounded corners.
Those hoping for a revamp of the user roles/caps system are out of luck as it’s been punted to not cause havoc with the merge.
In summary WordPress 3.0 is: Merge, menus, custom post types, theme, core plugin integration and fixing bugs. Here is the ticket to keep an eye on regarding custom post type support.
There is the possibility that once 3.0 development gets underway, Jane will initiate a wordpress.org redesign community project sometime in late February.
Matt addressed my concern regarding his comment that MediaWiki for the Codex was being deprecated. I took that to mean that it was being replaced. Matt said it just means it’s difficult to work with.
There are at least two volunteers to help with creating the conversion script with single site to multi-site.
While someone jokingly brought up the idea to do a post editor refresh in light of the recent Movable Type news, Jane mentioned 3.1. So it seems as if that’s a possibility down the road.
That’s about it but as I said, it’s always good to read through the discussion to see the finer points of detail discussed. Start here and scroll up.
How To Participate:
If you want to suggest a topic to be discussed at the next meeting, you can by visiting the WordPress development updates blog. If you would like to participate in the chat next week, install IRC or an IRC compatible client and connect to the following IRC server.
chat.freenode.net or any random server on the Freenode network and then join this channel at 4PM Eastern time or 9PM UTC Thursdays. The meeting day was changed to accommodate European users.#wordpress-dev.
If you don’t have an IRC client: http://pjirc.viper007bond.com/?chan=wordpress-dev&network=chat.freenode.net