osener

Come to think of it, this reminds me of how LLMs fill every empty space in a design with technical jargon that is technically correct but doesn’t serve any purpose. Basically typographic cornice.

Reminds me of https://github.com/emacs-eaf/emacs-application-framework which I've never been able to get to work. I guess this is peak personal software.

Fantastic rundown, thank you!

Wow I missed canvas support completely. This is going to make some custom UI's way easier to make compared to the hacks I had to resort to in the past [0]

I do hope one day we'd get a new cross platform performant display engine that draws like Zed does for more UI capabilities in Emacs. Neomacs[1] promises this and has some of this but looks like its suffering from never ending scope creep and continuously moving back and forth between 80% there and 90% there due to unattended agents running wild.

[0] https://github.com/ozanvos/emacs-webkit-color-picker

[1] https://github.com/eval-exec/neomacs

After a couple of years of detour I’m back to using Emacs since all the fancy lsp, debugger etc stuff in VSCode and Zed doesn’t matter as much for me anymore and I am just more productive with Emacs. But I am still using it like it is 2020. I hear people using coding agents to make crazy customizations to their emacs config and I can imagine Emacs is particularly suitable for this.

Curious what are the biggest things that have come out in the last year or two, and what emacs based personal software have others reading this made?

How did MirageOS end up powering parts of Docker stack?

This is just silly. Nobody is adding icons to Mon, Tue, Wed. Yeah maybe with LLMs it is even easier to fill up empty icon: optional<string> values without thinking. But extreme of not using icons is even worse than meaningless icons that you can at least get used to differentiating.

Here's an example from Chrome on macos:

https://i.imgur.com/I3d8YdA.png

This would benefit from icons at least on new tab, mute, pin, search, split etc

Feels like sunk cost fallacy. Clearly US is so far behind in human facing concerns so might as well not waste any more of the precious resources. Let’s just double down on investing in the prosperity few rich armed with super intelligent robots and maybe they’ll solve the problem by accelerating the pace of making the rest of the world so much worse that this comparison is over once and for all.

Exactly, so even heritage brands use their brand power to peddle the same low quality stuff you can find on AliExpress. And their flagship stuff continuously gets eroded and enshittified as well. YT is full of "why Dr Martens/The North Face/KitchenAid/Black & Decker/... doesn't make the same quality stuff anymore" type of content. Even when you pay for the brand name it doesn't get you far.

It's all the same crap. you buy name brand from Tefal, or a local kitchen shop and it's the same crap with a different label and large markup. On Amazon you get all these letter brands with the same crap. You try to go to the source and buy from AliExpress but the lobbying worked and countries like The Netherlands introduced taxes to make it unattractive and help the local drop shippers keep their jobs. Some countries outright ban it. You think they're protecting the consumer but it's all the same crap but with extra steps.

I can't open those with a coin, the plastic is too soft indeed. But if I stick them to another object using the included sticker I can just rotate the device itself to reach the battery. If you want it to be mobile you can stick it to something small or a 3d printed piece of plastic.

Python REPL definitely hates me

I usually try to keep an open mind but wow this is so bad. I was a loyal Firefox user but between the memory and perf issues and now this, I've now switched to Helium. I miss the "Multi-Account Containers" but that's about it.

They just had to look at Arc for what people want out of the future of browser design and instead they committed this own goal.

For the record I did not mean to say that about your work, nice job.

It is impressive, but all the hype and fake demos are selling it as a model that is as smart as frontier reasoning LLMs in the decisions it makes yet much cheaper and much faster, which is not true.

which ones do you recommend instead?

Reminds me of textmode demos https://www.pouet.net/prod.php?which=24206

It is not without its fair share of faults, but I stay on Firefox because of multi account containers

I like the end result of OpenTelemetry tracing when using Axiom and the like, but the SDKs have been a nightmare. Too much emphasis on automatic instrumentation, Java-isms, everything is stateful and abstracted away.

It can do distributed tracing of otherwise traditional long running microservices, but breaks down when your functions are distributed like in durable execution engines, Cloudflare Workflows, “functions” that span hours/days/weeks and steps that retry many times.

I had to reverse engineer how SDKs work and how tracing UIs display data so I could make simpler functions that fit wider variety of runtimes and more freely parent spans, start spans and end them from different function instances.

I think most of the API and terminology complexity is self inflicted. Would love to see a rebooted developer experience that is less Kubernates-brained.

Would your app run equally well with sqlite?

how is enabling ever bigger and more expensive to produce forms of entertainment making modern society better? remove all that and we won't lose anything. we'll always find a way to entertain ourselves. likely, with more culturally beneficial and social forms of entertainment.

If you find this interesting, also check out their UI library called Bonsai that built on top of Incremental: https://github.com/janestreet/bonsai/

Libraries like React are pretty efficient with skipping work by using Virtual Dom, but constructing this vdom still takes time. Bonsai makes the vdom incremental and it is pretty fun to work with.

I built a desktop UI library with it by targeting (now unmaintained) Revery. It is using a much older version of Bonsai however: https://github.com/ozanvos/bonsai_revery

Smart people have been falling into this trap as long as LLMs have hit production. Supposedly early internal versions of GPT-4 had "sparks of AGI" but the public version was "dumbed down for safety"

https://www.youtube.com/watch?v=qbIk7-JPB2c

In my experience restricting programs that can be expressed is a good thing, even more so with agentic engineering. The more guardrails there are, strong typing/TDD/computer use/..., the solution space shrinks and chance of a robust solution increases. Sure maybe this burns more tokens going in circles but it feels less like a slot machine more like a robot searching for a solution for a well-defined problem.

Devs have very strong opinions about dynamically typed programming languages. But reasons such as "exploratory programming", "expressiveness", "taste" that makes them feel good to program in for humans does not matter for agents. Agents don't care that the language "limits them" and prevents them from expressing the code in a succint way because it would not type check.

Surprised to see Clojure/ClojureScript come up on socials more often all of a sudden. I used it professionally for a few years around ~2012 and like many others moved off JVM and moved into typed [functional] languages.

Is the sudden buzz due to agentic coding? Does it rip through code faster with no type checking and fewer invalid syntax errors and reserved keywords to deal with? are we in for a sexp resurgence?