jjordan

Given the amount of usage I'm now getting from Opus 5.5, I'm surprised it took this long since release for their first outage. In my experience, it's about 10x the usage of Fable 5.1 with similar, perhaps greater levels of intelligence.

We need an open phone platform badly.

Don't worry, it's on the roadmap for Q4 2029.

When you use a drill to put a hole in the wall, do you take credit for it, or do you credit the drill? Without intent, a tool, whether it be a drill or an LLM, is just an inert object.

It would be great if I could integrate this into our existing (and rather extensive) eslint config as as a plugin of some sort rather than integrate yet another new tool to our dev environment. Any plans for that?

FYI I've been building an app, Android first, and discovered the magic of Kotlin Multiplatform. The LLM modularized everything so that it's substantively the same code base, but with Swift only where it's needed. Still working through it, but it seems like a great and underrated platform to build on.

Why should they skyrocket? As more autonomous vehicles appear on the roads, incidents with non-autonomous vehicles will go down. Driving will become inherently safer.

Yes, I definitely trust insurance companies to do the right thing here.

There should be laws put in place to ensure drivers will always be able to drive their cars, if they choose, and also prevent insurance companies from skyrocketing your rates for doing so. This is IMO past due.

Closed source doesn't mean not trustworthy. O&O has been making tools like these for a long time.

Hear me out.

Decentralized unstoppable storage, combined with decentralized unstoppable training, sorta like SETI for AI training. The seed of this tech already exists with IPFS and others like it.

We know (some? all?) of the big labs have skirted copyright laws at one point or another. Truly open models would just build on what is publicly available.

Fully open models really need to be a big part of the AI future. That includes all source code, open training data, how it's organized, fed to the model, processed, etc. Until that becomes a thing you're always going to be left wondering what exactly lies underneath the closed model you are using, leaving open the possibility for societal manipulation.

People who like near frontier AI at a lower cost than most.

What ever happened to IPFS? Wasn't it built for just this kind of thing? I remember trying it years ago and it was absolutely unusable, but was kinda hoping it had made a turnaround by now.

Fantastic content. Went looking for this on HN after seeing it on lobsters, since I don't have an account there yet. It's truly sad that some of these gems never get front page coverage, for whatever reason.

Anyway, this will be useful as I've struggled to get into an instrument myself and often found myself asking "why" to a lot of accepted conventions. Appreciate all the time you put into this.

I think it should be noted that the CTO of GitHub doesn't use his own product. No commits since January 2024: https://github.com/v-fedorov-gh

No side projects? Nothing? Just seems odd.

I'm partial to Typesense, especially for smaller data sets, since it runs primarily in memory, is easy to use and is hella fast. For bigger data sets, I hear good things about Meilisearch.

Probably the most refreshing thing I've read in a while. Glad to support them moving forward if this is indeed their modus operandi.

I'm just about done with https://RSSVault.org, a curated RSS feed directory.

Password is vaultearly644988 if you want to check it out.

The companion app, a widget-first RSS headline viewer called Just The Headlines is just about done too. Site is online if you want to see screenshots: https://justtheheadlines.app

I'm in the (required, sigh) closed test phase of Google Play Store approval. Drop me a line on X (https://x.com/JJordan) if you would like to join the test for that.

Feedback is more than welcome.

It might be a quiet A-B test they're running, but there are several reports now of Claude Code reporting that Fable usage is no longer subscription based and requires using usage credits.

Thanks Dave! Actually I was meaning to reach out, as parsing your site's RSS brought broader support for the format to the entire site. Those title-less blog posts were something else :D I didn't know much about the history of RSS going into this, so I might be your newest fan. Thanks for all of your contributions to the free and open web.

If you think that's esoteric, there's also a JSON feed standard. I have found one thus far while building out an RSS directory of over 1,000 entries.

Thank you

Yeah I'm focused on including more of the Indie Web. I include BBC, NYT, etc. because people expect it, but the directory already has some real depth worth exploring. And each feed details page has a cadence rating to tell you how often they post. So grouping a firehose with a specialty blog that posts 6 times a year probably isn't a good idea, but an OPML full of blogs that post ~monthly would be a delight, IMO.

Yes, with several months of iteration. I've wanted to build this for years but to get it to the point where it's a quality destination is an effort beyond the time I would have been able to contribute.

Hey thanks for taking the time.

"White or almost white text would be better." -- There is some white text, headers and headlines mostly. Did you find the lack of contrast difficult to deal with?

"Make “Add” button visible on homepage." -- you mean like in the search results? That's a fine idea.

"Make a possibility to batch suggest feeds" -- I've been toying with how to handle OPML imports. Definitely a feature to come on the app, but I fear with mass submission on the vault I would be inundated with feeds, and the curation aspect would suffer. I want quality over quantity. But maybe there's a happy middle ground somewhere.

"Move explanatory part of RSS to some other part" -- yeah I guess this might be helpful to someone who doesn't know what RSS is, but then they probably aren't visiting anyway lol I'll give it some thought.

Appreciate it. Thanks again.

Wow, what timing. Not to shill, but I'm also nearly done building a directory of RSS feeds. It stemmed from a frustration of outdated and broken RSS feed collections, and the long slow death of the orange RSS subscribe button across the web. I'm not officially launched yet but I would love some feedback and maybe some of the RSS feeds that land in your directory can be added to mine.

https://rssvault.org

Password is vaultearly644988

Pardon the dust. There's an app tie-in that I'm almost done with too, a widget-first app that shows your RSS feed collections on your phone's home screen. It would really be great if we could start to bring about an RSS renaissance!

Do you think Firefox development is unsustainable without this substantial stipend?

To stay out of jail, mostly.

back in my day we called this a good ole' fashioned slashdotting.