100k

If song lyrics count, Eleni Mandell's "Iowa City" is the poem that will always stick in my mind about that town.

  Girls, the boys don't cheat in Iowa City
  Iowa City nothing to do
  Now they're crisp and they're clean
  Iowa City
  Iowa Iowa
  Skies are blue
  
  Not so, Chicago
  Never, New York
  When you're off and you're looking for something
  What will you do?
  Where will you go?

https://www.youtube.com/watch?v=C_mpBLD6_V0

If you like this movie, I recommend Jamelle Bouie and John Ganz's podcast "Unclear and Present Danger" (the name is a riff on Clancy as well) which covers the political and military thrillers of the 1990s and how they dealt with the United States' changing place in the world after the end of the Cold War.

They have done two episodes on this movie:

https://jamellebouie.net/unclear-and-present-danger/2021/10/...

https://jamellebouie.net/unclear-and-present-danger/2022/10/...

Seconded! This was my favorite talk at Strange Loop (including my own).

As someone who was against the Iraq war, it's hard to express how important Paul Krugman's columns were to me at the time. It felt like the world had gone mad, but he provided an island of sanity. And he was right about the war.

I also appreciate Krugman's rediscovery of current music as an older person. It encouraged me to keep exploring.

My favorite Krugman moment is him visiting the White House after he won the Nobel Prize. This photo of him with Bush (who he'd been railing on for 8 solid years) is priceless: https://freakonomics.com/2008/11/bush-congratulates-krugman/

Zeke Faux's Number Go Up (https://www.amazon.com/Number-Go-Up-Cryptos-Staggering/dp/05...) started as an investigation into Tether. He found some extremely suspect clues, but he wasn't able to crack it. Hopefully the feds are able to shed some light on it.

It's quite fun, I played a lot of it during the pandemic. It's sort of like Advance Wars, but when you lose you can choose to send one of your leveled up characters back in time, which lets you use them again in a new game.

Yes, you can use CGO to call Rust functions using extern "C" FFI. I gave a talk about how we use it for GitHub code search at RustConf 2023 (https://www.youtube.com/watch?v=KYdlqhb267c) and afterwards I talked to some other folks (like 1Password) who are doing similar things.

It's not a lot of fun because moving types across the C interop boundary is tedious, but it is possible and allows code reuse.

There's some interesting companies in this space.

https://gofourth.com/ - focusing on grid storage, uses graphite (carbon) and liquid tin for heat transfer

https://antoraenergy.com/ - industrial heat and electricity

https://rondo.com/ - industrial heat

You can find interviews with their founders on the Volts podcast.

I’m happy to see Standard Ebooks here! I’ve read their editions of Nostromo by Joseph Conrad and Vanity Fair by William Thackeray and the quality great. I recommend it if you’re interested in classic literature.

Yes. The concepts apply to all programming languages. The exercises use multiple programming languages. For example, one exercise has you explore the Git source code (which is in C) to learn how to use a program's data structures to understand an unfamiliar codebase. Other examples have you find bugs in Python code or write Java.

Yeah, we need to do better on the visibility of this for owners. And we're trying to scale the system so it's not a problem at all. In the meanwhile, you can ask support to index all the repos in your orgs and we can take care of it.

I hear you, but we didn't break this for no reason. We rewrote our entire code search stack to support features it didn't before, like exact match, regular expressions, indexing forks, navigating to where methods are defined, not timing out constantly, etc. None of that was possible in the old system. Unfortunately, the architecture we chose to make the new features possible made some of the features of the old system difficult. We made the choice to focus on things we believe (more) users get (more) value out of.

It refers to the geometric progression of bucket sizes used in the data structure.

It means that we have a result limit. If a term matches in a lot of files, not all the hits will be shown.

Tree-sitter is very cool! We use it to power semantic analysis for GitHub code search.

For the repositories not in the index, we do show a message to try again later. For documents that have been excluded, we agree we need better visibility.

(I gave this talk and work on code search.)

Sorry you've had a frustrating experience with the product. When we look into complaints about missing results, it's almost always one of two things. First, the repository is not indexed yet, but searching triggers indexing. Future searches work. Second, the files or the repo hit our documented limitations (https://docs.github.com/en/search-github/github-code-search/...). I think these limits are pretty reasonable, but we need to do a better job making it clear why content isn't in the index.

Thanks for sharing, that's good feedback.

We'll definitely blog about it (https://github.blog/). His handle is https://github.com/aneubeck (it's on the thank you slide of my talk btw)

Please see my other comment about why this is difficult: https://news.ycombinator.com/item?id=38638214

I work on this, and I agree we should do more to improve ranking. Exact duplicates are supressed, but often forks have different versions of some files, so they come up in the results.

If you don't want to see forks, you can exclude them. Here's your same search, but converted to a regex and not including forks. I only get 3 results.

https://github.com/search?q=%2Fload%5C%28%22%40rules_foreign...

Yes! My colleague who created it has started working on an open source version so we can publish it. I am not sure when it will be ready, but I'm excited because it is extremely interesting and has a lot of potential use cases.

(I'm the speaker.) This is frequently requested, and we've tried to answer it in GitHub's feedback forums. The reason it's not implemented is it's quite complicated and there are other things we'd rather work on (sorting by recently was mostly used by scrapers so it's not high yield for us, unfortunately). It's complicated due to some consequences of things I discuss in the talk.

First of all, the old system didn't really implement sorting for "recently changed" files. It implemented sorting for "recently indexed" documents. This was a decent proxy when we only rebuilt the index every two years. But as I discuss in the talk, we rebuild the index all the time now, so it would be pretty worthless as a sorting option. Another reason is due to the content deduplication I also discussed in the talk. When a blob is shared between repositories, what time do you store? Finally, it's complicated because of how Git works. If we did implement this, we'd like it to reflect when the path changed in the branch, which is what people mean by "when did this file last change". But Git doesn't have an efficient way to retrieve that information, so we'd have to build a system to keep track of it.

In short, it's hard to do right and easy to abuse. :-(

Nice! Seems like a useful tool for digesting videos.

Oh hey, that's my talk! Thanks for submitting it. It was a huge honor to present my team's work at Strange Loop.

GitHub - code search | Staff/Principal Software Engineers | REMOTE (US/UK/Germany) | Full-time

Helping developers navigate and understand that data is a core part of GitHub’s mission, and an area with significant potential we’ve only begun to realize. We have high aspirations for how we can help developers understand and navigate code, and for making collaboration more seamless. The foundation for these efforts is our highly scalable code search engine, which we are now expanding to help power semantic/vector search to better support many of GitHub’s AI experiences.

We are hiring for two open roles (Staff and Principal level) in three countries, so there are six listings. All roles are remote, but currently we can only hire for these roles in these countries. These are hands-on engineering positions, writing Rust for a large scale distributed information retrieval system.

https://githubinc.jibeapply.com/jobs/2388?lang=en-us - Staff/USA (USD $118,100.00 - USD $313,300.00 /Yr.)

https://githubinc.jibeapply.com/jobs/2390?lang=en-us - Staff/UK

https://githubinc.jibeapply.com/jobs/2389?lang=en-us - Staff/Germany

https://githubinc.jibeapply.com/jobs/2385?lang=en-us - Principal/USA (USD $134,800.00 - USD $357,500.00 /Yr.)

https://githubinc.jibeapply.com/jobs/2387?lang=en-us - Principal/UK

https://githubinc.jibeapply.com/jobs/2386?lang=en-us - Principal/Germany

The name is terrific.

Another thing you can do with LLMs that I think is pretty interesting is use them to train a cheaper and faster model. Then use the faster model in your application.

I got interested in Al-Andalus when I visited Tarragona, where the Moors pushed out the Visigoths and ruled for 400 years. A short book I recommend is Moorish Spain by Richard Fletcher (https://www.amazon.com/Moorish-Spain-Richard-Fletcher/dp/052...). It's a quick and engaging read covering the conquest and slow rollback of Muslim rule.

I helped with a study of the compression strength of concrete blocks when I was in high school. It was a long time ago, but IIRC, the blocks augmented with monofilament (it was fishing line) performed well.