Why doesn't Walt Frazier have a freaking podcast. Come on. (Jon Stewart did a series of podcast-style interviews with him.)#
If everything goes well, the RSS feed for Scripting News will now have a channel-level image element because it's part of a network that requires an avatar-like image. This required me to go through some very old code that my system still depends on. It's remarkable how much time this seemingly small feature took to implement. One of these days I have to move this code to one of my more modern servers. One reason it took so long is that a random package that does something that never changes, had a breaking change in it. The breakage culture of the Node.js world is just plain fucked up, no other word for it. #And btw one of these days I'm going to clear the time to write a useful and up to date RSS feed validator. The one the W3C uses is a total embarassment. I'm not even going to link to it it's so awful.#
I've heard that Andor is great stuff. I'm on episode 6 of season 1, and it's the usual Star Wars bullshit. It was fun in 1978. But now? It's so freaking boring. Tell me it's worth continuing to watch, that it gets an actual plot at some point.#
I had a neighbor and friend a long time ago named Ann Doerr. I used to joke with her how her name was a combination of two of the main types of logic gates in computers. I think that's what attracted me to the Star Wars Andor show.#
I now have the special ChatGPT function I've been waiting for. Codex. Give it access to my entire GitHub collection and let it go. I stopped myself from authorizing, wanting to sleep on it. I know I'm going to do it, but.. Gulp. Do I really want to dump all my thinking for Sam Altman?#
- I've been enjoying the new David Frum podcast. He's very good at thinking and does a lot of it, and expresses himself very well. I read all his columns and have read two of his books, even though I don't share much of his pre-Trump politics, I find it interesting to hear how he parses things. I also subscribe to The Atlantic. I listen to him. But in this week's episode it's clear that the listening is one-way. #
- In today's episode we heard from the elite journalist David Frum, interviewing elite editor in chief Marty Barrett, formerly of the Washington Post, and they both make what I believe are wildly inaccurate generalizations about the quality of the news reporting by professionals and the ugliness and inadequacy of bloggers and podcasters. #
- The problem is, they don't read widely enough to know that there are bloggers and podcasters with good intentions, ethics, deep knowledge and experience who are also well educated -- they don't know we exist and therefore can't hear about the huge mistakes they are making with their own self-perception of their role that we can see because we are not in their world. (BTW, I have been in their world, I was a contributing editor at Wired, a research fellow at Harvard and NYU, each for two years, and I have collided with Frum-like elites, the usual response is a hand-wave of dismissal, written off as not worth listening to, I guess.)#
- Frum, at the end of the episode talks about how the press covered Biden's incapacity, that led to the second Trump term. They should have stopped, once we were all informed of what happened, and let the electorate decide, it wasn't up to them, or the celebrities they gave space to in their op-eds. In the United States, the people decide. Instead they put up a roadblock and wouldn't let the issue drop. We all knew, we were all informed, probably long before we read a single article about his debate performance. They should have stopped. That was their malpractice. #
- Their responsibility was to give us the information, and it was also their responsibility to stop right there. Once informed, they were wrong in continuing to harp on it. #
- We saw them do it with Governor Cuomo of New York. And Senator Al Franken. And Donald McNeil at the NY Times. And on and on. There's so much reporting that's lazy and incompetent, but they'll never report on that, because of their conflicts. And they never listen to their critics. And as a result they play a role they are not entitled to play. #
- They need to wake up to the idea that while reporting from their bubble is still important and is heard, they misuse it regularly, and we are fed up with it. If they want to control the government, do the right thing, and run for office. Don't do it from your byline in the Washington Post or NY Times. #
- FediForum is a virtual conference that starts on June 5. If they had asked me to keynote, this is roughly what I would say. #
- Imho this stuff is pretty freaking simple, esp since there are well over 20 years of prior art to use, and not that many ways to do what we're trying to do. I'm impatient, so here's a quick set of observations with my opinion, take it for what it's worth, ymmv, etc. #
- You shouldn't be reinventing so much. Always look around for prior art. That will make it possible for you to interop more quickly at both a software level and at a human level. #
- Don't invent stuff you think you may need later. Save that for later. You have no intuition for what's needed and more important nothing to test against. You will get it wrong 100 percent of the time.#
- Don't invent stuff you think you may need later that makes what you're doing now more complicated. I see a lot of that in ActivityPub. #
- The only reason you're doing all this work is for interop. If you've been doing it for years and you don't have much meaningful interop you're doing it wrong.#
- Write the software first. Use formats that exist, or if you must invent new stuff, make sure it's simple. Work on its simplicity as you would any other feature. Factor! If there are going to be 100 interoperable products and you make it a little more difficult to implement, that work will be multiplied 100 times. And it be a barrier to entry, so you may not get the most powerful interop possible with the most interesting products. #
- Read and follow the Rules for Standards-makers. There aren't that many. But if you're breaking them, you're not going to end up with a standard. If your goal is to appear to be making standards, you should also read RFSM, and don't do anything in it. I've seen people do that btw. #
- Have you looked at the world outside the tech stuff to see how important this all has become? All the time we're wasting is very costly in everything that depends on the social web actually existing as opposed to just being talked about.#
- Now, what would I request if I could influence you??#
- Mastodon and Bluesky should support inbound and outbound RSS, and do it really well. Right now they do outbound only, and the implementations are incomplete at covering the functionality they have now, and there needs to be more (see the next item). #
- They'd support the basic features of the textcasting spec, including Markdown because it's a great standard, very much of the same school as RSS. If it had existed when we did RSS 2.0 it would have been part of it. #
- That's it. Really not much work. #
- From there we'd have 25 years of interop to explore. #
- And we could really call this the open social web.#
- I'm ready to do it any time you all are. #
- Yours in not wasting time.#
- Dave#
- PS: I asked ChatGPT to visualize me giving this talk.#
- PPS: Comments or questions on Masto or Bluesky.#
