When I talked with Matt Mullenweg about the three focus areas for 2017 (and perhaps beyond), I kept referring to “releases.” Finally Matt said, “You keep saying release; it's not a release. It's all about the functionality and the feature. That's what's leading it.”
While “releases aren't the most important thing anymore” still seems to be the rule, we're going to have a 4.8 release that will be a package of several items more significant than minor releases. They also fall outside the designated focus areas of the REST API, the editor, and the customizer.
In response to questions about 4.8 and the release schedule as the community is trying to navigate what it is, Matt said:
I think people are thinking of this as a normal release, a train leaving the station that a bunch of stuff (multisite! meta!) has to get on to make it in. I agree that needs a much longer timeframe.
What is really going on is that we have a few simple, already-working-as-plugin enhancements that add a few files, and we want to get those in the hands of users sooner rather than later. We already update TinyMCE all the time. Potential breakage or compatibility should be limited to things that interact with the text widget or the news dashboard module.
In response, Jeremy Felt spoke about the clarity that I think several contributors are seeking:
At 218 tickets closed so far, 4.8 still looks like a normal release covering many components. It is a lot smaller than most major releases, which is helpful. The beta period/dev note period should be easier.
I think exploring this new release cycle is a good thing, and I'm trying to be cognizant of any personal overreactions to the change in a process that I'm used to. All in all, I don't think a quick release will harm anything, and we'll learn some things from it.
At the same time, I don't think we've talked much about process and how it's changing in general. It would be helpful to do that. In particular, I think it would be helpful to start discussing different branching/project tracking strategies so that dropping in a few simple enhancements and shipping a major release is really that easy of a process.
It's not that we're clamoring for a bunch of multisite, meta, etc.,… stuff to get in. It's that we've been making improvements throughout the cycle, and they've been sitting in trunk waiting for 4.8. If that's the wrong approach, we need to adjust our workflows.
Matt acknowledged that the process is new and has potential for further change:
I'm glad people are sharing concerns; it allows us to plan and hopefully anticipate and avoid potential roadblocks. It's also true that we haven't done anything like this in a long time.
If we are successful, then we'll get some great improvements in the hands of users quickly, and [we] can return focus back to Gutenberg (which is coming along swimmingly). The impact on meetups and WordCamps is potentially a pretty large one.
This wasn't the only blog post and discussion around 4.8 with people trying to figure out exactly how things will work, but I think it's a good depiction of what contributors are trying to navigate.
In sum there are primarily three items slated for the 4.8 release, which are all relatively minor compared to a typical major release:
- A revamped news widget to include events information (which some believe could have significant impact on meetups and WordCamps).
- New media widgets for simpler embedding of images, audio, etc. Also, the text widget is going slightly WYSIWYG.
- TinyMCE link boundary and inline editing enhancements, as described by TinyMCE themselves. This would include removing primary support for IE 8/9/10.
The 4.8 landing page also states that it “generally aims to refine existing features and set the foundation for
It's possible that some tickets surrounding widget improvements, like rich text for the text widget, could be leveraged for Gutenberg, but I'm not positive on that.
This process is definitely a learning experience, and I know there are varying degrees of excitement from folks contributing. One disappointed contributor told me, “The new system for sure has led to significantly fewer people contributing, fewer fixes going out for bugs, and a lot less progress on the main release features.”
It's too soon to declare the new release format a success or a failure — it was declared an experiment from the outset — but it's certain that there's already a decent amount of pivoting happening.