I’ll put up a link for the BonBon download tomorrow morning. I want to be around while people try it, in case there are problems or questions. #
Podcast: How Claude Code allowed me to think big. We can have the web we want and don’t have to wait for a big company to do something to help the web, when all evidence of history is the exact opposite. They consume the goodness and we’re left with yet another totally predictable tech mess. Hope this came out okay, I recorded it on a drive to Kingston, NY this morning. #
If BonBon were made by a big company instead of by an aging hippie software developer working with his super-powerful friend Claude, we would have gone on a press tour, probably via Zoom, of all the major pubs to prepare them for a fundamental change in what we think of as the web. We would have produced a huge example site, and worked with leading content publishers like the NYT, Kottke and Heather Cox Richardson to get their work BonBon-ready. We would have a developer conference where web devs and political leaders would speak about the importance of writing on the web. I would have contacted Tim Berners-Lee and asked him to write a preface for the docs, telling about his vision now, over 30 years since the inception of the web, and what he thinks of the future. Then I would speak, and say enough of that, here’s a demo, and then I would give you the URL of the download page, and ask you to keep an open mind to what’s possible. If you do, imho, you will be rewarded. But I won’t be doing any of that. Instead I asked ChatGPT to explain BonBon from a historic perspective, based on a conversation we had yesterday. And later today or tomorrow I will link to the download site from my blog home page, and encourage you to find out what having a new web Writer’s Web might mean to you. Till then, your friend, web writer and reader, Dave Winer. 😉 #
I would have asked John Gruber, the inventor of Markdown, to give a talk. I’d give him a big hug and thanks for having such a great vision, so useful, so many years later. #
Also — I would have briefed Cory Doctorow, because imho BonBon strikes a blow for _de-_shitification which I hope becomes as much discussed as en-shitification. #
AI has enabled hackers to find security holes in widely deployed software much faster, and it is able to find security holes that humans can’t. So there’s been an until now behind the scenes race by the good guys trying find and fix holes be for the bad guys find and exploit them.
At least some of the recent outages were caused by software that was trying to close security holes that were discovered by AI and the actual code that was doing the updating was AI. Probably the first time it’s ever been called on to do this so it probably didn’t understand all the limits of the networks it was working with. But when you’re dealing with a security issue, there really isn’t a lot of time to consider all the options and sometimes you have to take risks. The system going down isn’t the worst thing that can happen in that context.
The danger is real and our economy, if our computer networks were to fail, would collapse in an instant.
I don’t want to be alarmist, but this is a time to be alarmed because even if we had the best people in the top positions in the government, we would still be in danger.
This was a reply to a friend’s post on Twitter about how decisions are made in open source communities.
I guess it depends what the goals of the project are.
If you don’t have a clear statement of the goals, then you have no means to decide on any specific feature.
I welcome community input on goals, and bug reports and feature requests.
But… the decision belongs to the product showrunner. Try to imagine a great TV series like The Wire or Succession without a showrunner, but rather it was a vote of who showed up one day. It wouldn’t fit together with itself as it zigged all across the map.
It’s kind of like the US government, never has a working longterm strategy, and is constantly undoing itself.
In the RSS world, as the community leader, I would have listened closely to developers who had skin in the game, a product, either a CMS or feed reader that had a lot of users or who had taken risks in the past that turned into key features for the community. And I would listen for ideas that came from people with no power, because I have missed ideas that came to me that way, the famous near-miss was Adam Curry’s “last mile” idea which turned into podcasting. You have to allow an open door for people who see something the rest of us don’t.
And every product, open source or not, must be open to competition. So if you don’t like what they do, and they won’t fix it, then go out and make some news sister or brother. That’s how we grow!
But remember to leave the door open for those who follow. Not locked-in, locked-open. 🙂