Slightly related to the article. I have a personal cargo bike. The most fun that I have with it is giving friends a ride home from a party. People instantly start giggling and laughing. It's goofy, you get stares and people curious
I want a system that enforces planning, tests, and adversarial review (preferably by a different company's model). This is more for features, less for overall planning, but a similar workflow could be built for planning.
1. Prompt
2. Research
3. Plan (including the tests that will be written to verify the feature)
4. adversarial review of plan
5. implementation of tests, CI must fail on the tests
6. adversarial review verifying that the tests match with the plan
7. implementation to make the tests pass.
8. adversarial PR review of implementation
I want to be able to check on the status of PRs based on how far along they are, read the plans, suggest changes, read the tests, suggest changes. I want a web UI for that, I don't want to be doing all of this in multiple terminal windows.
A key feature that I want is that if a step fails, especially because of adversarial review, the whole PR branch is force pushed back to the previous state. so say #6 fails, #5 is re-invoked with the review information. Or if I come to the system and a PR is at #8, and I don't like the plan, then I make some edits to the plan (#3), the PR is reset to the git commit after the original plan, and the LLM is reinvoked with either my new plan or more likely my edits to the plan, then everything flows through again.
I want to be able to sit down, tend to a bunch of issues, then come back in a couple of hours and see progress.
I have a design for this of course. I haven't implemented it yet.
Similar ideas have been kicked around over here. One problem is that this seems like a set of features for GitHub rather than a stand-alone product (so no way to make money from it).
I'm not concerned about making money from it, I just want to use it. I'd like to check to see if I'm re-inventing the wheel. I'm curious if others would like a similar experience.
I’m not running fleets of gpus so I’m not your target market, but can you tell me more about who is doing this type of large scale training outside of companies like anthropic, OpenAI, xai…. The big teams. I’m just curious?
I am getting really tired of github. outages happen that's a given. but on so much stuff they don't even care or try. Github is becoming the bottleneck in my agentic coding workflows. unless I make Claude do it intelligently, I hit rate limits checking on CI jobs (5000 api requests in an hour). Depot makes their CI so much better, but it is still tied to github in a couple of annoying places.
PRs are a defacto communication and coordination bus between different code review tools, its all a mess.
LLMs make it worse because I'm pushing more code to github than ever before, and it just isn't setup to deal with this type of workload when it is working well.
> I am getting really tired of github. outages happen that's a given. but on so much stuff they don't even care or try. Github is becoming the bottleneck in my agentic coding workflows. unless I make Claude do it intelligently, I hit rate limits checking on CI jobs (5000 api requests in an hour). Depot makes their CI so much better, but it is still tied to github in a couple of annoying places.
Have you ever considered that this is the problem? GH never planned for this sort of pointless and unpaid activity before. Now they have a large increase (I've seen figures of 100x) in activity and they can't keep up.
It doesn't help that almost none of the added activity is actually useful; it's just thousands and thousands of clones of some other pointless product.
It helps with latency too or schedule padding. Bus schedules are unreliable because of all the stops which slow them down and encourage bunching of busses on a route with a lot of service.
Bus bunching is often blamed on traffic or scheduling, but in my experience in NYC, a lack of enforcement and/or accountability plays a role too. I live near one end of a bus line and commute to the other end 5 days day a week. On a daily basis, there are large gaps where buses miss their scheduled times. Then, as they approach the end of the line, they arrive and depart in groups of three or four, which only worsens the problem.
I started writing a program that needed to have a table with 1 million rows. This means it needs to be virtualised. Pretty common in GUI libraries. The only Rust GUI library I found that could do this easily was gpui-component (https://github.com/longbridge/gpui-component). It also renders text crisply (rules out egui), looks nice with the default style (rules out GTK, FLTK, etc.), isn't web-based (rules out Dioxus), was pretty easy to use and the developers were very responsive.
Definitely the best option today (I would say it's probably the first option that I haven't hated in some way). The only other reasonable choices I would say are:
* egui - doesn't render very nicely and some of the APIs are amateurish, but it's quick and it works. Good option for simple tools.
* Iced - looks nice and seemed to work fairly well. No virtualised lists though.
* Slint (though in some ways it is weird and it requires quite a lot of boilerplate setup).
All the others will cause you pain in some way. I think the "ones to watch" are:
* Makepad - from the demos I've seen this looks really cool, especially for arty GUI projects like synthesizers and car UIs. However it has basically no documentation so don't bother yet.
* Xilem - this is an attempt to make an 100% perfect Rust GUI library, which is cool and all but I imagine it also will never be finished.
I wouldn't bother watching Makepad. They're in the process of rewriting the entire thing with AI and (it seems to me) destroying any value they has accumulated. And I also suspect Xilem will never be finished.
Beyond egui/Iced/Slint, I'd say the "ones to watch" are:
* Freya
* Floem
* Vizia
I think all three of those offer virtualized lists.
Dioxus Native, the non-webview version of Dioxus is also nearing readiness.
> * Makepad - from the demos I've seen this looks really cool, especially for arty GUI projects like synthesizers and car UIs. However it has basically no documentation so don't bother yet.
Have you seen their intro that also has an image viewer building tutorial?
Unfortunately, it seems GPUI is no longer in development.
Iced looks great, but the developer is very open about it being a personal project (https://book.iced.rs/philosophy.html). That is appreciated, but for companies or even individuals wanting to adopt it, it's a big hurdle to get across.
I’m currently writing an application that uses virtual lists in GTK: GtkListView, GtkGridView, there may be others. You ruled out GTK because of its looks I guess, I’m targeting Linux so the looks are perfect.
Not just because of its looks to be fair. Not being native Rust is a pain, and GTK only really works nicely on Linux. At least without a ton of effort to fix everything (I think some apps like maybe Mypaint have done that, but I don't want to).
Def agree. It feels unnatural to be using gobject in Rust. Refcell everywhere. But the end result (at least on Linux) is fast, well integrated and looks nice.
Yeah, I need cross platform, and GTK looks quite foreign on Windows/macOS IMO. I toyed with custom themes, but couldn't find any I liked for a cross platform look (wanted something closer to Fluent UI).
I've been somewhat involved in a project using Iced this week, seems pretty reasonable. Not sure how tricky it would be to e.g. invent custom widgets though.
I believe latest Iced versions do have a `Lazy` widget wrapper, but I believe that effectively means you need to make your own virtual list on top of it
Custom widgets aren’t particularly hard to do in iced, but I wish some of those common cases would be committed back / made available.
Except the above virtualised lists, another case I hit was layered images (sprites for example). Not very hard to write my own, sure, but it’d be nice to have that out of the box as in eg. egui
That's a great idea, and I was just thinking about how it would pair with self hosted CI of some type.
Basically what I would want is write a commit (because I want to commit early and often) then run the lint (and tests) in a sandboxed environment. if they pass, great. if they fail and HERAD has moved ahead of the failing commit, create a "FIXME" branch off the failure. back on main or whatever branch head was pointed at, if tests start passing, you probably never need to revisit the failure.
I want to know about local test failures before I push to remote with full CI.
automatic branching and workflow stuff is optional. the core idea is great.
> automatic branching and workflow stuff is optional. the core idea is great.
I'm not sure if I fully understood. But SelfCI's Merge-Queue (mq) daemon has a built-in hook system, so it's possible to do custom stuff at certain points. So probably you should be able to implement it already, or it might require couple of minor tweaks (should be easy to do on SelfCI side after some discussion).
I can't take this comment seriously unless you are buying snow tires. If you have snow tires, and you still can't get where you want in the winter, sure get 4wd.
I had a RWD pickup with snow tires and went anywhere I wanted to through two utah winters and many vermont ones too.
https://www.visidata.org/
reply