A global single point of failure event happened recently. I’d bet there’s more concern around the gambling potential of this stock. Apologies, I’ve lost my naivete.

Dumping Windows for Linux and/or macOS wouldn’t have prevented this one. In fact, under similar circumstances, the stock, I mean software that crashed everything would’ve caused Linux and macOS (userspace) systems to panic too.

If you need zero trust compliance then start here. This is not a product endorsement. I failed marketing. You shouldn’t trust anything you’ve just read.

I’m not online often. Media that repost old news do good work anew. Apparently there was a protest against Reddit over its API. Now it’s a public company.

Reddit’s CEO probably knows more about its users than the users know about themselves. A shrewd company (with good marketing) is an elegant government.

It’s amazing, the last time I lurked Reddit was for neophyte notes on Guix. Since then I’ve went back to long form blogging/note-taking offline.

Hmm, maybe I’m a bit of a simpleton. The other day, while researching Facebook’s/Meta’s open graph protocol and metadata, I had a rare bout of curiosity and clicked on the discussion section.

Is the public group page just full of bots advertising to each other? Interesting.

Discussing strategies for detecting AI agents and crawlers has its apocalyptic comedy. How many humans are bots, and bots humans? If captchas are unsolvable and work unprovable, then you are a bot. But guess what? I’m not smart enough to play captcha!

So I’m a bot (that’s a joke) and technology works in unwitting ways. The web wants smart humanoids not dumb humans — it’s robots.txt not humans.txt, silly.

Related; a developer creates an AI friend. You are impressed and buy the thousand dollar one time licence. Your developer friend chats with you from the comfort of a phone with automated assistance. I can’t believe it’s not AI?

Remember deep linking? That is, if you’ve ever done computing/science history. Deep linking was a nefarious act and for ultra brief periods (in some places) making a link (like this) was link stealing.

Basically, homepage links were a–ok, but deep links were bad, really bad, because customers skipped the front and arrived at the exact (product) page. Businesses died overnight (the hypo theory).

Deep linking is smartphone marketing lingo now and very much desired. My hunch is that something analogous to taking a screen shot will be the modern equivalent of that unworkable controversy.

Did you know that cPanel has a back end user–space API? I’ve yet to see a web front end user dashboard that’s reliable and consistent.

Some of you were curious. Text assistants and bots are booming, as shovelers seek the god oracle and magnum opus. Here’s the numbers for a naked robots.txt. Inferences, up to you.

1. Applebot/0.1 [20.1 MiB]
2. heritrix/3.4.0-20200304 [21.8 MiB]
3. YandexBot/3.0 [28 MiB]
4. GPTBot/1.0 [30.1 MiB]
5. IonCrawl [31.6 MiB]
6. CCBot/2.0 [40.6 MiB]
7. DataForSeoBot/1.0 [51.2 MiB]
8. SeznamBot/4.0 [58.6 MiB]
9. facebookexternalhit/1.1 [64.1 MiB]
10. Google-Read-Aloud [86 MiB]
11. PetalBot [91.3 MiB]
12. Barkrowler/0.9 [98.6 MiB]
13. SemrushBot/7~bl [141.1 MiB]
14. BLEXBot/1.0 [152 MiB]
15. AhrefsBot/7.0 [159.5 MiB]
16. MJ12bot/v1.4.8 [285.2 MiB]
17. bingbot/2.0 [373.4 MiB]
18. Googlebot/2.1 [535.2 MiB]
19. Bytespider [543 MiB]
20. DotBot/1.2 [806 MiB]
21. Amazonbot/0.1 [2.6 GiB]

Rule #1 of finding technical feeds; type ‘OPML’ into GitHub search. GitHub is like a social network for software engineers and I’m a conservationist of time.

Are you a safe kid? I’ve just had a memory of a past time in a library.

Two newer meta-ish search engines for encyclopedias (about 30+ wikis):

• https://encyclosearch.org/
• https://encycloreader.org/