I’m writing this comment from a background of 2 decades of software development experience and tens of WP projects delivered. This article is so massively over-simplified, that as written, it’s not worth paying too much attention to. There’s still value in the article though, good discussion can never be overrated. I’m taking this 10 minutes to clarify my own thinking about the whole subject matter and perhaps link someone here later.
All that matters from the paramount “value to the customer” perspective is that professionals are on the job that know deep details about how the tool works and are able to assemble things correctly in relation to customer requirements. It’s the same for every framework out there.
This article could really use updates with some actual details about how the author tried and failed at doing unit testing or whatever else irked them for their WP deliverable. At this generic level, similar arguments can be made for and against anything out there, lowering the value of reading the article.
Here, it took a reasonably lengthy amount of time (counted in years) to get to a professional level where you can comfortably say you’re able to handle almost anything, because you know the core (+ carefully chosen satellite plugin set) that well. And by knowing, I mean participate in core tickets, hang out in the development chats, write your own plugins The Right Way (tm), etc etc. Raise your game at every opportunity, you know, the way professionals go about their business. As written, the article raises this question: how well did they really know the tool?
For a business, it’s not a goal in itself to build on WordPress. But in majority of cases there’s massive built-in value for the client to go with WordPress. Less vendor lock-in, strong ecosystem (when properly curated), massive free eternal workforce behind improving the core == minimal bus factor etc etc.
This comment is not meant to dispute the correct assessment presented in the article, that choosing the right tool for the job is critical. WP is definitely not the best fit for everything under the sun but the fit circle also gets smaller the less the executing team knows about it. I stand for people claiming to be pros knowing their shiznit throughout.