I’ve seen a few example of software written by two independent groups—one in the US and one in Europe—which ended up giving drastically different user experience. For instance, Microsoft Guidance (written by two people) is a framework for working with large language models and upon my first try, it worked really well. I was able to generate controlled sequences of tokens using Guidance, no problem other than my linter yelling at me for some minor issue. But there’s an equivalent software written by folks at ETH Zurich. It’s called LMQL (https://lmql.ai/). I actually like its syntax more than Guidance’s, and I appreciate their good documentation and tooling support. But… LMQL just doesn’t work on my machine even though I have set up the env variable for OpenAI API key. It’s a nice research project with interesting ideas, but it seems to be lacking when it comes to good programming. I wonder if it’s a pattern. Do Americans (or more generally, programmers in the states) write better software because they’re more closely in touch with the tech industry?
Story Published at: June 4, 2023 at 06:34PM
Story Published at: June 4, 2023 at 06:34PM