The shape of the next world
http://scripting.com/2026/06/23/150827.html?title=theShapeOfTheNextWorld
There was a long discussion last night on Bluesky about whether twitter-like apps should show blog posts in addition to tweet-size things. Should it have a character limit, allow titles, links, bold, italic, editing, enclosures, markdown, etc? This is a permathread, it's been going since 2006. I didn't contribute, because there are no new ideas at this point, except this -- there are readers and writers and they have different needs.
As a reader sometimes I want a concise intro to the idea and I'll decide if I want to read more.
As a writer, I want to write in one place, and broadcast it out the world, and let their reading app decide for them if this is something they want to read based on whether it has a title, is over 300 chars, has links or uses styling, or if the writer doesn't disclaim editing, and the reader doesn't like editing.
We can do a lot better than the hard restrictions our reading environments force on us. It's now 20 years since the inception of Twitter, I think we know enough now to try out some new approaches. There should be a million readers, and they all read the same content flows. They can look at a post and see if it meets the reader's limits, and only show it if it does. If a post has a title and we don't want posts with titles, don't show it. Then writers could all use exactly the writing tools we like, and it wouldn't matter where you read it.
This route has always been there, but now I think people will be open to trying out some new ideas.
I'm old enough to remember the Tall Ships in NY Harbor on this day in 1976, the bicentennial. As a NY kid, I wasn't very impressed. I liked rockets and rock bands, sound systems, had started programming then, was working in BASIC at Rapidata, a time sharing company with its office in the Empire State Building where I had my office on the 39th floor. The windows opened. This was betw Tulane and UW-Madison. I had no clue what was going on, but I had already come close to getting drafted. I had been raised to think the US absolutely was totally special, the best place, the rest of the world was far behind us. We were right to feel that way. It was the US vs the World and we won. I was born only 10 years after the end of WW II, so the feeling of power and righteousness was our foundation growing up, but also the certainty we'd all die in a nuclear holocaust. By 1976 we had had Watergate, the president was a crook, and were about to go through spiraling stagflation. Ronald Reagan. John Lennon killed. We had shit to deal with, worse in some ways than what we have today. Are we still the USA? We are if we decide we are. Anyway, my friend Jerry at the right wants to sing for you: "I'm Uncle Sam that's who I am been hiding out in a rock and roll band." We sing this song here every July 4, and it's always as true as it was in previous years. Freedom is something you practice.
I need new podcasts. The only one I listen to regularly now is the Bill Simmons podcast, but that's because the Knicks won and the NBA is re-forming itself around the Knicks. It's so freaking unusual to have your team, which was once right up there with Charlotte, New Orleans, Portland, Washington, Memphis, in the very the bottom rung of the NBA, to have them be the model everyone is chasing with the qualification that no one expects it to last (I don't care if it does, I love this team, the're as memorable as the 1973 champs), but all of a sudden Bill Simmons is respectful. I can't listen to a podcast of Democratic consultants, or Republican consultants that vote Democratic now. I did listen to them on the lead-up to the election in 2024. But whatever happens in the sport of elections the Democrats as they were before 2024, the one that re-nominated Biden and then switched to Harris and lost a race that should have been an easy win, are over. Those Democrats still think people will vote for well-executed government. Some people will (me, for example) but enough people see the election as Reality TV, so you want someone who looks like a winner in that context. The world has changed in so many ways and the Dems haven't even caught up with the change brought about by blogging and podcasting. Now we have Claude. I probably would vote for Claude too. I don't know. Anyway I'm warmed up now. Onto my day's work with the aforementioned Claude.
I saw a bit of a commencement speech by Eric Schmidt, ex-CEO of Google, where he was talking about AI and getting boo'd by the audience. But he was saying things that were right and should be paid attention to. Most important, and I'm paraphrasing, the AI world is just getting started, and we can change it now most easily, it's malleable. That won't last forever. As Obama says, "Don't boo, vote." Same thing here. AI has already completely changed how we develop software. It's not replacing humans, it's giving us amazing new power. Maybe it will at some point replace us, but don't be so sure that what we do with it might be every bit as new as the things it can do. We have different abilities. And I am old enough to remember a time before personal computers, the internet, the web, mobile devices, all the things that have since become everyday fixtures, and they all had negative aspects, but I would never go back. We're on a train and it's going somewhere. Where it goes is something we all have a say in.
A thought for people who think the US can't be fixed. I've seen very strange things happen, like all of a sudden people figure it out and boom next thing you know they're the NBA Champions. It wasn't exactly sudden, but the last leg of was. A gestalt. Now two leaders figure out how to. The thing about each of those people is determination, and a belief they were right, and they went right up to the edge and fought. I think the country would unite behind such a leader.
Some things Claude is extremely tedious at. But then it blows you away how it can read thousands of lines of complicated code in a few seconds (in parallel) and find tiny little things that any good obsessive programmer would want to fix (like me). And be amazed at how we, our species, made such a thing. Where is the pride? I was once prideful that my civilization created a great piece of machinery like my Subaru Forester, and now just a few years later, we've come up with a decent simulation of a super-human brain that's not just a demo or a robot vacuum cleaner it actually does amazing science fiction type stuff. Take a deep breath and feel a little awe to go with the cynicism. It's good to be ready to be riled up, but sometimes the truth isn't as bad as you'd like to think, sometimes it's utterly amazing. ;-)
Another truth, the user interface of WordPress could benefit from a total overhaul. Too many expedient choices over too many years that paper over bad design choices with yet more bad choices. But this kind of problem is relatively easy to fix. Make a list of all the features. Don’t organize the list yet. Keep adding. Then play around with logical groups, give the groups names. Voila, there’s your menu structure. And since it’s 2026 and not 2010, do something innovative with AI. Let the user explain what they want to do, confirm it, and then forget about the menu structure and just do what they asked you to do. Over time the UI will become more literate and less organizational. You remember how Nixon could open up China and could because he was such a hawk. WordPress getting a AI/UI overhaul will seem right because it so desperately needs an overhaul and everyone knows it. Another truth, don’t feel bad WordPress, every 20+ year old end user product desperately needs a user interface overhaul because that’s just the way it works. (I have never created a product that lasted as long as WordPress has. I have created concepts that have.)
BTW, I was just contacted by a developer who's implementing all the protocols I mentioned yesterday. And I should mention that Manton Reece, developer of micro.blog and a longtime friend, going back to the Frontier days on the Mac, has inbound and outbound RSS and he covers every freaking API out there, he's a monster. And I said yesterday he doesn't get enough credit for what he's contributed. We're aiming for interop instead of chasing the silos. And it's fine to chase silos if you're into it, I was done with that in 2017. We're going to make it work the way it would work if we weren't trying to lock anyone in, quite the opposite, I want people to use Manton's product. I'm not being commercial here. I'm trying to get the web back on the path it should have been on all along. If I make some money that's cool, if not that's okay too. BTW, this all-together will be the Two-Way Web, specifically Two-Way RSS. And of course textcasting. Don't forget that. It's a rule, textcasting everywhere conceivable.
The other day Matt joked about how old I am, in public, and I am pretty old. But Matt, I was paying attention then as I am now, and connecting the dots. No one else working today, I'd venture, knows what it's like to create and run a modem-based dial-up Twitter-like system on an Apple II with a 10MB Corvus hard drive. Yet it worked, and people loved it. If you weren't alive in 1981, you wouldn't know anything about this. I remember talking with Doug Engelbart when I was running UserLand. If you don't know who he is, look him up. He blazed a trail we were turning into a highway, and we're all using his inventions all the time. Every chance I got to sit down with him I did. I wanted him to work with us, to critique everything we were doing. He had a lot of knowledge that disappeared when he passed on a few years later. That's the sad thing, at my advanced age, that I am trying to avoid. And btw, as surprise, Claude really understands this stuff. I've never seen anything like it with a human, and I've worked with some great humans.
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 
Claude is a brain, very different from ours and when we work together we humans have access to capabilities that work really well with building large software products. And that's a huge understatement. Most remarkable thing. Most of the discussion between people who use the AI tools and those that condemn them are not productive because the opponents of AI don't understand the breadth of what these machines do and the potential to do much more, things that we as a species have never done. Think of it as an alien life form that wants to merge with us. I'm glad to be alive at this moment, and able to explore it as part of my development team. I recommend starting an academic dialog, among people who don't have conflicts of interest, or very well-disclosed and disclaimed conflicts, to accurately record this discussion based on facts, for the record, so when people ask how conscious were we when we did this transition, there will actually be some footprints to follow.
I bet Ward Cunningham is really good at using Claude, he is a big believer in pair programming. I even did a session with him in Frontier, doing stuff with the outliner.
When writing code with Claude you really have to be skeptical when it says it just found the problem, but you have no idea what it's saying, chances are pretty good it's just a word salad excuse for not having read all the code necessary to have an fact-based opinion. Actually debugging software isn't about opinions, it's about proof. When you start clutching at straws until one works you just added another level of bug that will eventually bite you in the butt and you'll still have to solve the original one. Uncorrected, I'm pretty sure you wouldn't want to trust the code it writes, but I guess that's why people have two or more instances playing different roles? For now I'm the one that questions its sanity, politely though. ;-)
