For the last seventeen years, I’ve been building WordPress-based websites. Since 2008, this has been the primary focus of my business, and I’ve been fortunate to have worked with many corporate and government clients.
Last night, I watched the State of the Word 2021, Matt Mullenweg’s annual address about WordPress. And for the first time, I got the feeling that WordPress is moving in a direction that does not align with my goals and use case.
Matt showed a lot of demos of upcoming features in WordPress 5.9, which is to be released in January. Most importantly, “Block Themes” and the “Site Editor”. Watching them, I got the feeling that none of my clients need these features. Or rather, most of my clients need to not have these features.
Brand identities
Block themes allow you to design large parts of your website within the WordPress administration backend (wp-admin). Using the Site Editor, you determine what the header, footer and other parts of the site look like. You build the visual parts of the site by combining blocks and choosing from a large number of design variations and settings that are available for those blocks.
This introduces two problems:
Design decisions are stored in the database, not in code.
At my biggest client, I’m part of a team that builds and maintains a “toolkit” that allows them to roll out new websites easily and quickly. Essentially, it’s a single theme and a large set of custom-built plugins. Having a single theme makes sense, because there’s also just one visual identity for this organisation, which we need to follow to the letter. Changes to the design need to be rolled out across all the team’s sites, so that they remain consistent.
Currently, we can simply deploy a new version of the theme. It dictates the way the sites look, and that’s a good thing. Out theme has very little settings, and none of them are design decisions. If we’d need to log into all (60+) sites to manually apply the same set of changes in the Site Editor, that would be extremely cumbersome, and prone to errors.
Semantic design
This has actually also been a problem with “Gutenberg phase 1”, the Block Editor. Our team’s clients need to be able to focus on writing content. They do not want to design every page they’re working on. Most of them consist of paragraphs, headings and the occasional list or table. 90% of the core block are never used, or actively disabled on our sites.
The new editor is an obvious improvement over TinyMCE. But we’ve had to do a lot of work to limit the choices editors have. We really don’t want them to pick colors, fonts, etc based on their personal preference. Elements on the page should look a certain way because of their function. Editors should indicate what an element’s purpose is, and the theme should automatically format it accordingly.
Competition
This situation reminds me of “Post Formats”, a controversial feature that was introduced in WordPress 3.1. If I remember correctly, there was considerable opposition to this feature, but it got pushed through anyway. Why? Because Tumblr had something similar, and Automattic considered Tumblr to be WP’s main competitor.
As usual in the State of the Word, Matt shared CMS market statistics. He emphasised the growth of closed-source systems like Wix and Squarespace. Both of those tools offer a Site Editor of sorts, and I can see why they are a threat to wordpress.com (Automattic’s commercial WP offering). But for wordpress.org (the open source platform), Wix and Squarespace are probably not the best targets to focus on.
The reason is simple: Those tools are absolutely horrible to build large sites with. The constant need to design while you’re writing makes it tedious, and is exactly what WP should not do.
Also, Wix and Squarespace have relatively tiny bits of market share, and WP should probably just focus on improving its own product.
A crossroads?
Of course, much of the new functionality can probably be disabled. PHP-based themes are probably not going away soon. But with WordPress pushing Block Themes, at some point the current way of doing things will start to feel old, be abandoned, and eventually die off. It’s already starting to happen to TinyMCE. And at that point, WordPress may no longer be a good fit for my clients.
For a very long time, WordPress has been a general purpose CMS that worked great for small personal websites as well as very large implementations. To me, it seems like with version 5.9, WordPress is finally picking sides. Perhaps it has to, I honestly don’t know.
