The Day The Standard Library Died (2020)
cor3ntin.github.io
cor3ntin.github.io
The submission title is currently "It is currently faster to launch PHP to execute a regex than using std:regex", and, while the article makes this claim, it doesn't provide any evidence for this. Maybe it's true, but that's not the point of the article (and I really doubt it's true in most cases--just the overhead of starting a new process and communicating back and forth is significant).
Extraordinary claims require extraordinary evidence.
The above sentence was mentioned in the article (without supporting evidence), and is not the focus of the article.
Tldr of article; abi was chosen not to be broken, which is costing performance. Author argues if std library isn’t optimized, alternatives will keep being developed, and std will be considered dead.
URLs make this harder than it should be.
Trying to submit a subsection of a bigger work, with the closest it has to a title, is perfectly acceptable... if it has its own URL. If it doesn't have a URL, then you're screwed.
One of the few benefits of tweet threads: you can link a particular tweet.
If the article had been a series of tweets, they could have linked to that tweet and it would have been fine and not been editorializing.
That's why I'm iffy on calling it editorializing. Even if it is, I blame URLs for not allowing the equivalent of linking a specific tweet from a series.
It's especially bad when there's a news roundup article talking about 3 or 4 different things, and you can't just submit the third one with the title of that section.