I’ve started to get comments on this blog that I figure are probably AI-written spam, but are sufficiently well-written and sufficiently on-topic that I can’t tell for sure.

I hate the idea of giving spam a place on my site. But I used to really enjoy the discussions in my comment space, back when people did that sort of thing. This leaves me conflicted about what to do. I’m seriously considering turning off comments, and just letting the discussion move into social media. (You can see my social media accounts, if you want to tag me any any response you make. That page also has other ways to contact me.)

I’d be interested to hear from anyone who thinks blog comments are still a good way to do things, and wants to advocate for me keeping the site comments open.

I’d like an app that could take my various social media feeds, the RSS feeds I used to follow (until my feed reader broke), and the various news sources that I subscribe to, and find all the articles that I really want to read, and present them to me in a useful order.

Of course it needs to do a lot more than that. It should watch me read them, pay attention to which posts I linger on, which ones I follow internal links in, etc., and try to learn what I actually want to read. (And then give me more of that stuff in the future.)

It would also be nice if it noticed when I read something by a new person I don’t follow (because it was boosted by someone I do follow), and consider following that person as well.

And by “in a useful order,” I mean the AI should understand which articles are full of background information versus covering the latest breaking news, and present the background information first—unless it’s background information that I already know. In that latter case, it should just offer a link to the background information, in case I feel like I need a refresher. It should also present the information grouped by topic (so, news first, then economics/business news, then science news, then the very narrow sorts of cultural news that I want to read).

I don’t know, but I assume this is happening because I block scripts from sites other than the one I’ve gone to visit from running in my browser.

I wasn’t going to turn on scripts, because it didn’t seem important to visit this particular page, but in the time it took me to capture the image below and write this text, the page seems to have decided that I am human after all!

A screen capture of text suggesting that it's "taking a longer than expected" to verify that I'm a human.

Because I am not as clever as Cory Doctorow, I just frittered away 15 minutes setting up my domain to be verified as my Blue Sky handle: https://bsky.app/profile/philipbrewer.net

I post almost nothing there—basically, just links back to my blog here—but you can go find me there with the other cool kids who are not as clever as Cory Doctorow.

I use micro.blog to send out my newsletter. I’m generally pretty happy with its newsletter system, but it does have a serious mis-feature: There’s a very narrow window for editing the newsletter between when it generates it, and when when it sends it out.

The main thing I want to edit is the front text that goes at the top of the email, ahead of the blog posts that I’ve identified as ones that should go into the newsletter. As near as I can tell, there’s no way to create that text until micro.blog gives me the draft newsletter. By default (the way I had it set up until a few minutes ago), there is then only 30 minutes before the newsletter goes out.

That might be fine, except in practice it turns out that the alert arrives after I’ve left on my main morning dog walk, and then the newsletter goes out before I get back.

A dog standing on a picnic table

As a stop-gap I’ve increased that gap to 3 hours (the largest gap the system allows, it would appear). That’s not perfect—I’d like to be able to write the front-matter anytime in the month before the newsletter goes out, and then edit it repeatedly over the month. But it’s good enough that at least I won’t keep missing it just because my dog gets to luxuriate in a long morning walk every day.

This guy has an app for bulk unsubscribing (and text in Section 230 that perhaps protects it).

Personally, I’d like the opposite of what this app is described as doing: I want a plugin to purge my feed of everything except posts by people I follow. (All the rest of that stuff is “objectionable material” as far as I’m concerned.) That would make Facebook usable again, maybe.

… focused on a part of Section 230 that spells out protection for blocking objectionable material online.

Source: NYT

I’m finally sending out my newsletter! An “issue zero” just went out, but starting next month you’ll be getting actual newsletters. My first cut at a title is “Sword fighting, writing, and a dog,” because that’s what I seem to be spending my time on these days. (I’ll try to get more writing stuff in than I have been just lately.)

If you’re interested, subscribe here: https://philipbrewer.micro.blog/subscribe/