mgsloan2 · HackerTrans

10 か月前議論

My homedir is just a git repo, similar to https://www.atlassian.com/git/tutorials/dotfiles. The .git folder has a different name so that git commands aren't accidentally applied to it. What I love most about this is keeping my homedir clean - I can see what files are new (untracked) and either delete them or add to .gitignore

mgsloan2

10 か月前議論

Very fun to peruse! Curious why the olivetti typewriter has such strangely shaped key caps.

While it seems like OLPC didn't really achieve the full vision of the project, I have some anecdata of success - my coworker got an OLPC when young (otherwise without access to computers), learned to code on it, and is now an awesome programmer.

mgsloan2

11 か月前議論

Collection of data from code completions is off by default and opt-in. It also only collects data when one of several allowlisted opensource licenses are present in the worktree root.

Options to disable crash reports and anonymous usage info are presented prominently when Zed is first opened, and can of course be configured in settings too.

mgsloan2

11 か月前議論

You could, but it is extremely expensive to train an LLM that is competitive on coding evals. So, I was assuming use of a model someone else trained.

Also, if it is only trained on code, it's likely to miss out on all the world knowledge that comes from the rest of the data.

mgsloan2

11 か月前議論

Sounds like cool stuff, along the lines of structure editing!

The question is not whether it can work, but whether it works better than an edit tool using textual search/replace blocks. I'm curious what you see as the advantage of this approach? One thing that comes to mind is that having a cursor provides some natural integration with LSP signature help

Yes agentic loop with diagnostic feedback is quite powerful. I'd love to have more controllable structured decode from the big llm providers to skip some sources of needing to loop - something like https://github.com/microsoft/aici

mgsloan2

11 か月前議論

I agree the current way tools are used seems inefficient. However there are some very good reasons they tend to operate on code instead of syntax trees:

* Way way way more code in the training set.

* Code is almost always a more concise representation.

There has been work in the past training graph neural networks or transformers that get AST edge information. It seems like some sort of breakthrough (and tons of $) would be needed for those approaches to have any chance of surpassing leading LLMs.

Experimentally having agents use ast-grep seems to work pretty well. So, still representing a everything as code, but using a syntax aware search replace tool.

コメント