I feel like the solution to wordpress-the-org being mismanaged is not to build better replacement infra around wordpress, but to build better publishing tools that don’t burn a mysql connection and require php get executed on every pageview.
It is far easier and more likely that the community around WordPress would rally around replacing WP.org than starting over from scratch with a new toolkit. Some people might flee WP altogether, but the majority of people and companies just want to get things done - not seeing them having a lot of tolerance for or interest in having to start again from square one.
To be fair, the first and best solution most people in these parts have is "Fork it, I'm going to make my own Linux distro." or something along those lines.
No, they don't get very far but I will credit their consistency.
Also, Wordpress doesn't scale well (without getting expensive), the code is bad, it is poorly managed, security issues are all over the place. It's about time we got some alternatives.
There is no reasoning with people like this.
Unlike PostgreSQL, MySQL was specifically designed to handle a very large number of short-lived connections. A properly configured MySQL daemon can handle tens of thousands of connections per second. If you're still opening and closing too many connections for that, MySQL supports using Unix domain sockets instead of TCP, and PHP supports persistent connections -- a sort of ad hoc connection pooling system.
There are many issues with WordPress and its ecosystem, for sure, but database connection handling isn't one.
So this is the year 2024, and something amazing called a CDN has been invented. PHP does not get executed on every pageview, nor does a Mariadb connection get burned. If you are building a website or web application at scale with any framework or platform, a big part of your job is to architect it such that everything which can be cached, is. You may also employ tools like Varnish or transients. The underlying design principles are universal and whether you're building on top of WordPress or something else doesn't matter that much - in fact many experienced WordPress developers are accustomed to serving multiple millions or tens of millions of users per month, so the ways for making this work are pretty mature.
I’ve worked extensively with Wordpress; that’s precisely why I hate it so much. It needs to die.
The proper solution, 20 years from now, will not descend from this branch.
Mind you when I made my original statement I had in mind websites that handle millions of monthly uniques and are somewhat dynamic in terms of the content they publish - for smaller scales WP needs basically nothing else and just works, but this scenario is really quite easy with WP if you employ common caching strategies, and is probably impossible with the typical "I wrote this script that generates static HTML last weekend and therefore it's a revolution in CMSes" solution
I guess you MVP would be something that could do simple brochure websites and be easy to extend. Design it so it is easy to install for individuals or web hosting companies. Maybe plan ahead for things like multi-site, blogging, eCommerce. etc
With all the competition and even the MVP is going to be a lot of work, especially since the average user won't be very technical.
Has anyone tried to use Wordpress as a static site generator, with a few dynamic-content functions moved to Javascript against an API for alternative open-source server (or even another Wordpress instance)?