FeedCity

A pro-social approach to building better platforms

I've been looking up all kinds of things now that my new Daytona search engine has access to my blogging going back to 1994. I just did a search for Mullenweg, and came up with this blog post by Matt in 2006 about the Feed Validator. The clarity is remarkable, and he's of course right. The people running the validator were actually promoting Atom over RSS, and were trying to tilt the table towards Atom. The goal was interop, not to give Google or IBM control of the syndication format of the open web. When you keep changing your mind about what to flag and what to pass you end up with a standard that's as murky as the tariff policy of the US in 2025. I think you have to assume that was the goal. They didn't like that RSS didn't fall apart when Atom came out. If there's no benefit in changing, people don't change.

An unofficial guide to the HTML caption element

How to understand Trump talking about tariffs. Remember how he talked about Covid. Same thing.

When I announced Chris Lydon's podcast in 2003, I called it a weblog for the ears. We didn't have the name podcasting yet.

Andrew Hickey whose podcast I support, also appears to be strongly against people using ChatGPT. I am a programmer, so you might think I'd see it the same way because what I do is similar to what musicians and prose writers do in that I create "intellectual property" which ...

[Kate Knibbs at WIRED] This is an insightful interview with Bluesky CEO Jay Graber. The headline here overreaches quite a bit, and needlessly describes what Bluesky is doing in monopolistic terms; I left excited about what's next for the open social web, whi...

Some of the news out of Microsoft Build is what developers hope Apple will do: access to AI models in the cloud, on devices, or even from web apps inside Edge. There’s also MCP everywhere. I’m reading through NLWeb, created by RV Guha, whose semantic web work for decades seems a nice fit AI search.

Italian pro tip: always carry a knife because you might have to cut a pizza out in the wild.

---

Thank you for keeping RSS alive. You're awesome.

Email me :: Sign my guestbook :: Support for 1$/month :: See my generous supporters :: Subscribe to People and Blogs

I was going to use this code in the UI of my search engine, when it wanted to display an image over http, reroute it to use an https proxy server. But Chrome wouldn't stay out of the way, so had to give up. Hey I was trying to route around their outage.

Mark your calendars: Friday, 20th June — that’s when Salter Cane will be launching Deep Black Water at the The Hope And Ruin in Brighton

I can’t wait to get back on stage with the band! These songs sound great on the new album but I can guarantee that they’re going to absolutely rock when we play them live.

Support will be provided by our good friends Dreamytime Escorts, featuring former members of Caramel Jack. They’ve also got a new EP on the way.

Doors are at 8pm.

I’m really, really excited about this. It’s been far too long since Salter Cane were last bringing the noise live on stage. I hope to see you there!

Apparently Chrome has changed the behavior of images served over http in web pages. It's changing the request to https, so it gives an erroneous error message about the certificate which the page never claimed to have.

If you use Chrome or Arc with Micro.blog, check out the latest update to our Chrome web extension for bookmarking web pages. It improves on the last version, now better saving HTML to archive the page. Still working on making it compatible with Safari and Firefox.

And with this update, we don't depend on OPML any longer to create the database, all the data is exported in a format that can be easily imported directly into an SQL database. I provide the schema and example JavaScript code that, via SQL, updates a database. Hopefully this creates a better more useful archive.

As the import script runs I just tried looking up Engadget, got zero hits. I expect that link will work in a while after all the importing is done.

Having the 90s and 00s in the index means I can look up old friends like Chuck Shotton.

I've been working on the search function for this blog, and now we have all the DaveNet pieces between 1994 and 2004 in the database. So here's a search for menu sharing which was a technology we made on the Mac that allowed people to write menus of scripts in Frontier that appear inside other apps. Netscape supported it as did MSIE, Quark, Eudora and many others. It was kind of magical, esp when the apps themselves were scriptable (many of them were in the 90s). Next up, I'm going to import the home page of Scripting News from 1997 to 2010. It was mostly a linkblog then.

What it means to be distributed. Mastodon can't go down, neither can RSS. But Bluesky can.

Not gonna lie, I’m close to dropping $200 to try OpenAI’s Codex. But I don’t think most of my code is well suited to it. Not enough automated tests! It figures that would come back to bite me.

I mostly use AI as a machine that can generate unlimited example code. I learn best from editing examples.