raybb

One thing I'm still trying to figure out is how this compares to something like gliner. If you're just doing classification in what situations would you choose kev vs gliner?

How long does it typically take for something like this to become available on openrouter?

I have an idea for a use case for this, and I'm wondering if you think it makes sense or if you have any thoughts on the approach.

I'm a big fan of OpenStreetMap, and I enjoy editing it from my computer. From my phone, I find it quite tedious trying to make sure I type in the phone number exactly correctly and double-check it, or find and select the right field from the large list of fields available in Upredor.

Generally, how it works is I see a restaurant, and there's a sign. I know that it says, "Cash only. Here's the phone number. Here's the opening hours." What would be really cool is if I could just speak to the phone and say, "Hey, here's the information about this place." It would automatically use your location to detect what places are nearby and maybe even detect which place you're talking about, and then tell you, "Okay, here are the changes I think you're proposing to make, or these things you stated are the ones that would create a diff." This would be limited to just perhaps the 20 most common keys in some predefined set of values for most of them. Like cuisine=x should just match to the most common not make up new ones.

Of course, this is something a large language model could do, but having it run on device would be a lot nicer and cheaper.

Any more you can share? Is he actively doing a search? I didn't see anything about it in the repo.

Do you have any examples of forks maintained by LLMs and how they do it?

I can improve the UI to my taste but if the maintainer doesn't have time/money to maintain a big ol vibe coded PR for some very opinionated change isn't going to do much.

That would appear to leave me with the choice of just using it for myself (lame) or I then have a public fork to maintain :(

If there's basically no shot of it getting up streamed I think I'd rather make something from scratch to suit my needs without the extras.

Anyway, I hope the AI overlords help us fund more maintainers soon!

Every Door is definitely one of the easier ones and I like it but dang I wish the dev could get more funding to make it a nicer UX. Between the glitchy raster tiles and need to click load data all the time I feel it gets tiring fast.

I've considered building a separate app that's similar but oriented around the idea of using voice of pictures to update things (with AI extracting and formatting the data for you to review). But alas making yet another tool isnt ideal.

I guess they got that message so they're porting to rust!

https://docs.litellm.ai/blog/litellm-rust-launch

The issue wasn't the latency of a single DB query (though many were improved/cached), but sync worker saturation:

When traffic spikes, slow network calls to archive.org (for availability) and Solr tie up all available sync workers. The queue backs up, and suddenly every page becomes slow because there are no free workers left to handle requests.

This was handled from both sides: handling high-concurrency waiting better, while simultaneously optimizing Solr and removing archive.org from the critical path where possible.

Not another language because the performance issues are limited to the code being sync. Anyway, a gradual migration was a must as well.

This is the first I hear of Litestar and I think it does seem very values aligned. I fastapi was picked simply because it was already known by the team and had a very thriving ecosystem. That being said, the hard part is getting off the old sync web.py stuff. Once everything is in async land moving to another framework if needed is relatively trivial. Thanks for sharing that though I'll keep an eye on it!

I have been volunteering with the Internet Archive via Open Library for about half a decade now. Sending money is awesome and they're also always looking for volunteers! There are so many projects going on for Open Library and staff has quite a few volunteers but right now especially I think it would be awesome to have a few more experienced folks joining to help (many volunteers are students).

Here's just a few things we could use more experienced folks to help on:

- Solr performance, it currently gets quite slow under load and there are two read replicas but we think there may be some ways to improve it besides adding more replicas. You can run a full production copy of solr based on the dumps... But the current workflow to do this is kinda broke so I had AI fix it up but we haven't been able to verify if it's a fully working/good solution.

- We're doing big redesigns of major parts of the frontend. If your down to help for that Lokesh had many tickets open and guidelines and welcomes PRs.

- I'm currently migrating us from web.py to fastapi and could certainly use more experienced help for trickier ones.

Anyway, if you're actually willing to help all you gotta do is join our community call tomorrow at 9am PST and share your interests. We'll you on to something you like quickly :)

https://docs.openlibrary.org/everyone/community-call.html#co...

That's fair but also think about the ewaste.

I'm still hoping one of these mega AI LLMs will figure out how to install Linux on the many old iPads that become nearly useless after Apple stops updating them and you can't install new apps.

I love this use of wasm!

I never really got the hang of LaTeX but I've really enjoyed Typst the past few years and even made a few books with it. Have you ever tried that or have thought on it?

I posted this on another thread but can someone please run this on some old iPads so we can be able to fully install Linux on them. If AI is so good surely it can do that and save millions of devices from turning into ewaste.

When a model is free like this what kind of rate limits are there?

There's also one I nopa and Mission. Mission event coming up September 13.

https://repaircafesf.org/ https://www.instagram.com/sfmissionrepaircafe/

This is a lovely project. Were you inspired by any projects like this for other cities? I think I saw something once for NYC but it didn't have nearly as many images.

Also, I'll probably share this in the next issue of https://urbanismnow.substack.com since we try to share positive things happening in cities around the world!

Whenever I want to follow what's happening with a natural disaster like this, where many friends and family are potentially affected, I always use the Wikipedia page, which tends to have the most up-to-date information very quickly. It helps me avoid zooming in on the news and seeing rehashed articles posted on multiple websites.

https://en.wikipedia.org/wiki/2026_Colombia_earthquake

To top it all off Pandoc has a great experience for contributors. Over the past few years I've opened several bug reports related to Typst and docx and all of them got responses that were kind and helpful. I even had a few PRs merged in despite knowing almost nothing of Haskell.

Any thoughts on silent.link ? It seems a little slow but otherwise fine.

A year or so ago I was trying to get llms to make pixel art by asking them to create a 2d array of color codes. I'm no prompt guru but they all failed at anything beyond a basic shape. I tried at many resolutions from 16x16, 32x32, 64x64, and maybe a little larger but basically only then quit small ones were even recognizeable.

Anyone have an idea how good https://qaul.net/ is?

I saw it shared at dweb camp and it seemed like a pretty long term serious project for P2P.

Providing a SQLite db out of the box is a nice touch. I wonder if they're capping it's size in any way.

Oh I don't have a link but there was a project just like this someone hacked together a while back. I think it used Gemini live or something and as you walked and scanned things it would suggest changed to the map. It was a rough demo but seemed promising.

I've been wanting to do something similar but just by taking photos of menus and updating the POI with new hours/website/phone etc

Anyone can help mapping for disasters that urgently need mapping: https://tasks.hotosm.org

I'm not in a position to hire or anything but fyi your resume is 404.

I don't know about restricting rent but they certainly had a progressive tax on high rents, a 1% income tax to fund housing, and then used that to build a ton of high quality social housing with balconies and amenities for all. I also like to think it helped that they slapped a big sign on each to proudly let people know it was their tax dollars at work.

I didn't stay in social housing while I living there but I never once heard people complain about it. They basically just didn't think much about it at all and felt it was a good system and then would ask me why the US only makes it for poor people.

What makes the problem intractable is that we have a system where for a huge chuck of the country your retirement is based on housing prices appreciating. It's clear if you directly own a home but even if you don't lots of the places you park your money to watch it grow are ultimately investing/speculating on real estate.

Some might call it housing asset based welfare. Even if you don't like that mouthful another simple example is the University of California putting 4 billion into Blackstone's REIT with "a minimum 11.25% annualized net return through January 2028." That REIT is 90% rentals. So probably at least a few people will feel the squeeze from it.

https://www.reuters.com/business/finance/university-californ...

https://link.springer.com/article/10.1007/s10901-009-9177-6

Sounds like you may have read it but the book Winner Takes All is about this topic and pretty enjoyable.

I think there's a case to be made that philanthropy produced the Internet Archive but maybe that's a little different from usual philanthropy since Brewster is very hands on for so long.

https://en.wikipedia.org/wiki/Winners_Take_All:_The_Elite_Ch...

I grew up in Florida and that's been a rough situation for a while. Crazy that they haven't been able to figure anything out over the decades it's been going on. I grew up with many citrus trees just around the neighborhood. I wonder if those have got it too.