Microblog

Drinking a Delirium Tremens by Brouwerij Huyghe.

Bookmarked https://v6.robweychert.com/blog/2021/10/robtober/.

Bookmarked https://samharris.org/the-fireplace-delusion/.

[A]t some point you have to stick with something, because any gains in efficiency are lost in the time spent looking for a solution.

Facebook’s behavior isn’t just anti-competitive; it’s anti-consumer.

I wasn’t a huge JSON Feed fan, originally. It felt unnecessary, although the same can probably be said of anything that goes beyond RSS/Atom. Anyway, I’ve already embraced (correctly formatted, which can be tricky, still) h-feed, so JSON Feed seems like a logical next step. It supports tags, too! Found Aperture/X-Ray doesn’t support (or even discover) all JSON Feed versions, which is a shame, because under the hood, it does. Only had to adapt two or so version/MIME checks to make it work.

It’s probably always been there, but I only just now noticed what looks like a tracking pixel in a FeedBurner RSS entry.

Love this. (Been obsessing a bit with ’80s music lately. Thank you, Turbo Kid.)

How about a Tailwind-like set of utility classes that can be stacked on top of each other, without having to actually code? Either that, or, you know, use … CSS.

“Should Users Be Able To Select More Than One Block Style?”

Was kind of expecting it’d return the u-photo from a page’s h-card or something. Still, looks like a nicer, i.e., higher-res, alternative to DuckDuckGo’s favicons.

This. I’m not a software engineer, either. (Well, I’m not active in IndieWeb spaces, like, at all, but I have implemented and am greatly enjoying some of the typical “IndieWeb” building bricks. They’re open protocols, after all.)

I am fairly active in IndieWeb spaces and I disagree with others in these spaces all the time.

Could be they check if an item’s content starts with its title, and decide to display it as a note if so (this is how, again, popular PHP library X-Ray does it). One way to have both notes and checkins not look like articles is to remove their title tags from your RSS (and not use a p-name in your HTML, but I believe that is already the case) altogether. That, in my experience, is the trick (and perfectly valid).

Which is why I used—wait for it—Dompdf.

My dream for this thing is, eventually, we could start building social media images via regular WordPress PHP templates. The trick is that you need something to screenshot them, like Puppeteer[.] An average WordPress install doesn’t have that available[.]