When Claude has all the information available it can figure out stuff a human mind would never be able hold in our minds at the same time, but it often doesn’t remember to get the information first. When you get to the level I’m at with this, it’s hallucinating all the freaking time because it didn’t load the part of the data set that had the answer. It was right there, it was supposed to know, it just forgot to look. My job is to recognize when it has done that and tell it to go read handoff.md again. I mentioned this on Twitter, and got all kinds of help, but the terminology isn’t well known to me. Still diggin, as they say. #

I’m loving Star City. New episode last night, wow. #

With all the Democratic Socialists winning over standard Democratic party incumbents, there’s a fair amount of angst on the cable news. If they’re scared, they should step aside. We tried it their way in the Biden Administration. If we ever get lucky enough to have a president who’s sane and wants to reboot democracy, it’s going to require doing some things that an oldtime president wouldn’t want to do, like Obama or Biden. Both of them gave up without even trying. Forgive them, but let’s not make the mistake of electing their successors. It’s time for clear-thinking people to take office, fully aware of what they signed onto, and then if we elect them, they do it. And when the Repubs throw bullshit at us, say it’s bullshit, and say it that way, not the mealy-mouthed way Jeffries does, or even Elizabeth Warren. What we need now is a strong dose of Bernie Sanders. Did I ever think I’d say that? Hell no. #

Podcast: My (latest) AI Aha Moment. #

I was on Slack chatting with a friend from WordCamp Canada last year, and by accident (I guess) Slack sent me the first message they sent after coming home about all the things we’d do. It reminded me of how possible things seemed then, and for a moment I got lost in planning it out, and I absolutely loved what I saw there. But it was sad, because I am sure it will never happen, not until someone inside the community gets the idea, and there really is only one “someone” here. Heh. I’ve been around big companies and communities before, many times. Anyway, I figured I should post this here now, because I have moved toward a WordPress-less web, or WordPress-on-the-Side, but I want to be clear that WordLand remains in place, free for anyone to use. It’s a great way to write for WordPress. And if this project to make web content APIs a web standard, I’m totally on board for helping the world understand how potent an idea it is.

So here’s the text of the message with light redaction in places. 😉

• it’s funny when i got email notice for this post, it sent me the first message you sent after the wordcamp in ottawa. those were more optimistic days. i still have my wordpress work, but now have pivoted to working on the web without the wordpress connection since matt seems to want to go in a different direction, based on all the stuff they’ve started re AT Proto, which imho is a terrible bet. it would be like the Knicks trying to move from NYC to Mississippi. Why would you do that when so many developers know wordpress and you all have such an excellent api?
• you asked me back then if there’s part of your project i’d like to work on, and i said i’d think, and i had an idea just the other day, thought i should share it.
• the wpcom api is fantastic and very few people know about it, even the people in the wp developer community. but it really is the answer — how will services that work on the social web coordinate? where will users store their data? where will the published results be available to read? wordpress really is the best choice. i have no stake in that, i don’t own wordpress, have never made a dime off it, i’m just me, saying that and i have some credibility in this area.

• so here’s what i suggest

  1. a new simplified version of the api and some example apps, both are already done
  2. a new protocol so that any service can be part of it, we need a way to identify servers that aren’t using jetpack or wordpress.com
  3. a new name and website, and positioning — something to roll out.
  4. work with independent developers to make their products work
  5. co-marketing
  6. investment and distribution

• i know a lot about all these things, having done them before with some amount of success.

• i’m about to embark on it again with my new product, btw called rss.chat. i think you can imagine what it is based on that name?
• but i haven’t forgotten about this opportunity. i don’t know how much of 1-6 the open source project, but #2 is clearly in your purview. not something people are going to want automattic to do (though I’m sure they could). with the new protocol look what we have! a way to distribute apps on the web so that developers don’t have to compete with BigCo’s, they can be a person with a hobby, and who knows they may have a big hit and get rich.

This is always good for a chill.