allreduce

I almost get your point. Use the resources you have to make solving your problem easier instead of spending your time on beancounting.

But there is no way the convergence on web technologies is not a market failure of some kind. The platform is not even that great to develop for! It's full of weird edge cases, unnecessary build steps and slow iteration cycles. Not even speaking of the often superflous frameworks and huge dependency trees which don't do much of anything.

You could build systems with much better devx and ux and an order of maginitude less memory usage and wasted CPU cycles. A bunch of such systems have been built over the years: e.g. Common Lisp + Comemrcial Tooling, Smalltalk Systems.

And real users perceive software as slow and buggy and generally crap. Just talk with your nontechnical relatives what their opinion is of the software we have inflicted upon their daily lives.

Yeah, you would have to focus on buying things you actually need. Be able to save for and look forward towards a stable future. Don't get a new pocket computer every year because the software is not updated anymore. Use fewer subscription services? Maybe do away with the disposable vapes with lithium ion batteries in them.

And they shouldn't be able to. Police accessing DMs is more like "listening to every conversation you ever had in your house (and outside)" than "entering your house".

That's a deficit of most programming languages. One solution is to pass every error value up and let the caller decide. Rust does this to some extent. This leads to verbose code however.

For modules inside your application, designing a good interface involves exposing the right errors and crashing for the rest. This creates some coupling of course (shared assumptions of which errors need to be handled across modules). Trying to avoid that probably just leads into the circle of hell where you have more abstract beancounting than useful code.

In the end, this is another reason why overreliance on external libraries leads to mediocre, buggy software.

I encourage you to look at what the current models with a bit of harnessing are capable of, e.g. Opus 4.6 and Claude Code. Try to make it solve some mathematics-heavy problem you come up with. If only to get a more accurate picture of whats going on.

Unfortunately, these tools generalize way beyond regurgitating the training set. I would not assume they stay below human capabilities in the next few years.

Why any moral person would continue building these at this point I don't know. I guess in the best case the future will have a small privileged class of humans having total power, without need for human workers or soldiers. Picture a mechanical boot stomping on a human face forever.

> JS independence

> Markdown support

I'm glad we care about improving the web now that it might be consumed by Mr. Flattery The Slop Bot.

I sometimes feel like I have a "high functioning" version of this. I have no trouble expressing myself, but it often feels like a puzzle of translating internal concepts to a context-appropriate word approximation.

For websites I use regularly, I can identify those which render server side because I trust that when they load on a slow connection my state is still there. When I see a loading icon on heavy client side JS "apps" I anticipate something breaking and me having to reload and re-enter whatever I was working on.

You talk like a LLM and the entire concept seems from one. What value do you think are others going to get from work you didn't come up with and didn't bother to fully understand yourself?

Goverment should intervene when the few are harming the many.

We do not seem to live in an advanced society, so in the end it's going to be riots and dictatorship by AI owners or perhaps no humans in control at all. Let's hope the worst does not happen this AI cycle and we get riots and slop only...

> After consulting some sources, I started translating the assembly instructions into plain Polish

Ah yes, obfuscation

Reverting to a world without deployed AI is in fact where "normies", that is most people without capital, want to end up.

The current AI promise for them goes something like: "Oops this chittering machine will soon be able to do all you're good at and derive meaning from. But hey, at least you will end up homeless and part of a permanent underclass."

And the people building it are (rightfully) worried about it killing humanity. So why do we have to continue on this course again? An advanced society would at this point decide to pause what they are doing and reconsider.

The thing is, the "Dropbox is just FTP" guy should be right most of the time when you are selling to experts.

There is no reason to not just ask Claude for a review and then distill this into PR comments. Especially because "every LLM output has to be filtered through a human" is a good policy with the current state of these tools.

However, this industry loves distilling frivolities into web tools and it sells for some unfathomable reason. It is the same with the existing static analyzers etc that some orgs pay for. I do not understand why.

Honestly, no. Not for everyone.

As someone from Germany, I don't want Google to nanny family members computing devices. They don't want it either. It is completely absurd for an ad company in a surveillance state an ocean away to play IT services for everyone. This has already gone to far.

Rather, there should be tools for value-aligned IT services and technically minded family members to help.

What industry are you working in?

In general, if you want to do something you more often than not have to compete nation or even worldwide. It makes markets more efficient in the minds of some economists, but makes creating or finding valuable things fucking exhausting.

> if AI truly becomes what was promised

I mean they are promising AGI.

Of course in that case it will not happen this time. However, in that case software dev getting automated would concern me less than the risk of getting turned into some manner of office supply.

Imo as long as we do NOT have AGI, software-focused professional will stay a viable career path. Someone will have to design software systems on some level of abstraction.

The problem is not going away until the deskilling of the software profession is reversed.

I don't see it happening for end user stuff like Reddit. The market seems to want the barely working thing for the most cheap in all consumer product categories.

And hiring someone desperate who has skimmed a React and Node JS tutorial page and is now fullstack something is cheap. Or at least the kind of cheap you can reach by greedily optimizing KPIs.

Honestly, this is the first argument against CoCs I've seen that I agree with. They do kinda act as a tool to deflect responsibility.

Nah but I recognize that HR, unfortunately, has to exist for larger organizations.

Unless you have an infinitely wise and patient dictator who can just say "you're an asshole, you go" and always make the right call or something.

Usually it does not reduce latency but increases throughput.

Multithreading is an invaluable tool when actually using your computer to crunch numbers (scientific computing, rendering, ...).

I find it weird to jump to this from the scarce information we have.

"Someone did a CoC violation" is just a way for an org to say "someone was an asshole to such an extent it was driving other people away or getting us into legal trouble", with the manner of assholery defined in the CoC. 9/10 times it is nothing sinister.

Of course right now we just don't know what happened.

It's trivially easy to break those restrictions with off the shelf SDR hardware you can buy rather cheaply.

Locking people out of their phone does not raise the skill or effort ceiling much, as there still presumably would be software restrictions in place.

Seems like Americans are into making up food theories instead of just eating with common sense and moving your body with legs instead of cars sometimes.

Not easy to learn is a bit of a red herring imo. Its also a disproportionate amount of stuff to hold in your head once you have learned it for what it is.

An OS is first of all is a set of primitives to accomplish other things. What classic worse-is-better Unix does really well is do just enough to make you able to get on with whatever those things are. Write some C program to gather some simulation data, pipe its output to awk or gnuplot to slice it. Maybe automate some of that workflow with a script or two.

Current tools can do a bit more and can do it nicer or more rigorously sometimes, but you loose the brutal simplicity of a bunch of tools all communicating with the same conventions and interfaces. Instead you get a bunch of big systems all with their own conventions and poor interop. You've got Systemd and the other Redhat-isms with their custom formats and bad CLI interfaces. You've got every programming language with it's own n package managers. A bunch of useful stuff sure, but encased in a bunch of reinvented infrastructure and conventions.

Hah I just wrote something similar today to periodically push backups to another server from my NAS.

I agree the systemd interface is rather simple (just translate nix expression to config file). But NixOS is a behemoth; Completely change the way how every package is built, introduce a functional programming language and filesystem standard to somehow merge everything together, and then declare approximately every package to ever exist in this new language + add a boatloat of extra utilities and infra.

NixOS is anything but a light abstraction (I say this as a NixOS user).

Tbh it feels like NixOS is convenient in a large part because of systemd and all the other crap you have to wire together for a usable (read compatible) Linux desktop. Better to have a fat programming language, runtime and collection of packages which exposes one declarative interface.

Much of this issue is caused by the integrate-this-grab-bag-of-tools-someone-made approach to system design, which of course also has upsides. Redhat seems to be really helping with amplifying the downsides by providing the money to make a few mediocre tools absurdly big tho.

Sounds in effect like what SEO / "trash article soup" companies did for Google et al the last decades.

> You're writing HTML for browser to consume, not for validator to accept.

I'm not a web programmer, but shouldn't one program against the specified interface instead of some edge case behavior of an implementation?