In a snap-announcement, the WordPress core development team has decided to throw out their published dates for the release of WordPress 5.0. The announcement, which gave just 2 days notice, has upset the wider development community.

First up, WordPress 5.0 is going to be great! After spending time with the beta, and getting over the shock of change, we love the new editing experience. While there are still a few rough edges, this is going to make creating content so much more fun and so much more creative! It unleashes a ton of technologies that WordPress has been slow to adapt, and we’re happy for that too.

WordPress 5.0 brings with it one of the biggest changes to the WordPress user experience for several years. Moreover, it introduces several layers of development complexity and new languages and paradigms.

WordPress 5.0 already has over 100 bugs filed that won’t be ready in time. These are scheduled for 5.0.x “fast follow” releases. The WordPress core team is releasing a major release with a ton of known issues.

By bringing the release date forward more than a month, and giving just two days notice of that, there has not been time to work on these, and third-parties that need to ensure compatibility will not be ready. Not because of lack of work, but due to the fast-changing core code base.

5.0 will be released on a Thursday US time, which means Friday for most of the world. It is politically timed to coincide with a WordPress conference happening this weekend. I’m guessing most attendees are now going to be having a stressful time, or having their products flailing while they’re out of office.

It’s also December: a time when many agencies and development shops have started winding down, and authorised significant time off for key staff. I’m guessing some of that is being cancelled as I write.

In short, the WordPress 5.0 code, the people, plugins, themes and systems that support it are not ready for WordPress 5.0. We believe releasing this in December is a terrible idea. That informs our recommendation to you, our community, and our plans for releasing it.

As Yoast wrote:

[O]ur advice boils down to: if you can wait, wait.

Performance Foundry and WordPress 5.0 – Recommended Timeline

Here’s our plans for the 5.0 release, which is designed to give maximum flexibility to our clients, while maintaining high uptime, performance and security.

It’s pretty much the same as we normally do for major WordPress releases — wait for 1 or 2 “bug fix” releases, then roll it out. It’s just more interesting due to the scope of change, and the upcoming Christmas holidays.

December 6th: WordPress 5.0 is released

According to our best data, WordPress 5.0 is scheduled for release on December 6th. Early adopters will download and install.

December 7th: Chaos ensues

We anticipate that forums and support channels will light up like a Christmas tree, and companies and volunteer theme and plugin authors will be swamped. Due to the huge demand, a rapidly changing code base and end of year leave, fixes will be slower than we’d like.

We do expect some instability for plugins for users on both 4.9 and 5.0 as plugins and themes rapidly update.

During December, WordPress core will release at least one, but likely more, patches to 5.0 as they solve already-known and unexpected issues. Plugin and theme authors will have to stay on the treadmill as these introduce new changes.

Through December: New projects

Performance Foundry has been building all new projects for the last few months with a target of releasing on WordPress 5.0. Where feasible, we’ll release those projects as planned and adjust and learn from it.

All newly-migrated sites will be matched 1:1 with their current WordPress version. Where that version is below 4.9.x, we’ll update it to that point, but not to 5.0.

Through December: Existing projects, retainer and hosted sites

We strongly recommend all our existing clients with sites currently in production not to upgrade to 5.0 right now. If you’d like to update early, please email support and we’ll do that for you.

We’ll be watching developments very carefully through the month, and we expect bigger players like Yoast, WooCommerce, etc to be fast in their updates. However, everything’s going to be shifting very quickly, and we just can’t recommend it in a production environment. This is true for everyone that relies on their site, but especially true for e-commerce sites where December is an important time.

Late December: Finalise upgrade paths

By late December, we’ll have a good idea about the state of the ecosystem – both core and all the third-party tools. We’ll finalise our plans for an early-, mid- or late-January roll out. For most clients, we’ll book upgrade windows on a week-by-week basis; and we will work with retainer clients to ensure we time these changes with their quieter periods in Q1 2019.

January: Upgrade all possible sites

When we roll out a major WordPress update, we take several other housekeeping steps at the same time. That includes creating a staging environment for the upgrade, upgrading themes, and doing some additional security screening.

We take a screenshot before and after the upgrade, and these are automatically compared. If there’s changes greater than 5%, we flag the site as needing extra attention, and don’t upgrade that. Normally we see around 20% of sites needing a bit of extra love.

The other 80%, which pass the test suite, get a green light for upgrading in production. We take a snapshot of the database and files, then upgrade the files and run any post-upgrade scripts. We reach out by email to let you know that the work is completed.

Then it’s time to look back over the sites that failed our test suite, and start working through them. Often we can fix things up with a few tweaks; but sometimes that means losing a favourite plugin or having some custom work booked in. We work with all our clients on budgets and timelines for that, and try to reduce any disruption as much as possible.

February – March: Finish the upgrade path for all sites

We expect to be working on this right through February and March, as there are likely some sites that need a lot of extra love. By the end of Q1, we’d expect to see all sites on our servers upgraded to 5.0.

Your thoughts?

If you have any thoughts or concerns about this upgrade, email support and let us know. We’re 100% behind the move to WordPress 5.0 — it’s going to be great!

It’s just not going to be great this weekend.