Mine is that the useState hook is an actual plague on the entire React ecosystem. Regardless of your opinions of React, if you wanna use something then use it right.
Using LLMs are not the same as using intellisense or compilers and there are a myriad number of reasons not to use it that could fill an entire book and I will look down on any company or freelance developer that uses it.
Yes I’ll fight Linus over this and yes I’ll be right.
-
C++ is an awful language overall.
-
Linus is just some guy with good programming skills and not an authority over anything tech.
-
There aren’t any good text editors in the wild. VSCode is awful apart from extension support. Kate is close to being perfect but lack of extensions are pulling it back. Vim, Helix, Emacs, NeoVim are all flex toys.nvm just tried zed with AI disabled -
AI bad
Linus is just some guy with good programming skills
I somewhat agree. Most celebrated personalities in tech aren’t programming geniuses - they were in the right area at the right time and had some luck. Zuckerberg is a good example.
I think kernel maintainers and rendering engines (browsers and others) have a higher base skill level. Like some web developers are clever and some are stupid, but I don’t think there are stupid kernel maintainers.
The thing with Linus is that he’s been involved, and at the pointy end, for decades. You or I may not agree with him on every issue, but you can acknowledge that he has experience and he has connections.
I have massive respect for the guy. He did so much for the open source community. But for some reason everything he says is interpreted as inarguable truth by most tech guys.
For example his opinions on AI are pure trash and bootlickery. Which is fine anyone can think anything. We just have to think for ourselves a little.
I agree that everyone needs to think for themselves, absolutely.
However, I struggle with this and it’s kind of what I’m getting at:
his opinions on AI are pure trash and bootlickery
Firstly, you’ll be hard pressed to find someone who hate’s AI more than me. It literally keeps me awake at night. I hate the environmental impact. I hate the people it’s enriching. I hate the work it’s stealing. I don’t often use the word “hate” but it really does apply here.
That said, I always try to defer to the opinions of people with more experience than me, and as regards kernel maintenance and Linus, that’s certainly true.
You’ve probably heard of the Dunning-Kruger effect. I know a bit about code, a bit about FOSS, a bit about linux, AI, and communities. So it’s easy to underestimate the yawning chasm of experience and knowledge between myself and Linus.
That doesn’t mean we have to blindly accept the opinions of others (I honestly haven’t accepted Linus’ opinion myself), but I’m aware of the profound discrepancy between my opinion and his and it concerns me.
All that aside, is it possible to acknowledge that something is the correct course of action even if it’s not the ideal? I haven’t been following the Linus / AI thing, but as I understand it a project like that kinda has to allow AI bug reports because threat actors will certainly be using AI to find exploitable bugs.
With the greatest respect, responding to this comment would be to engage in a discussion of whether Linus is right in this particular instance. I’m not willing to do that because I’m sure that’s a conversation that’s been had a billion times.
I was / am happy to talk about how we injest the opinions of experts in a more general way.
it gets down to whether experts are experts at everything. Linus is not an expert at anything outside of kernel maintenance. His opinions on whether we should use LLMs are null and void considering the massive effect on the planet. It’d be like if a nuclear engineer said we should be building more nuclear bombs for everyone to use, just so they could use nuclear engineer more freely. Like, their opinions about nuclear energy might be fine, but their opinions that have an effect on the entire planet are completely wack.
That’s a really good point. There’s loads of examples of this, like doctors don’t give good nutrition advice, et cetera.
He doesn’t really need any latitude to offer an opinion on whether these tools should be used on the Linux kernel.
Whether you should use an LLM to plan your shopping list, or understand your toddlers behavior, is not something you should listen to Linus about.
injest
Could you be serious, though?
My argument against Linus would be that the non-LLM tooling is auditable and known. I know I can trust all the tooling below it and only read your code, but with LLMs they are explicitly trying to replicate good looking code and can generate a ton of it, so it becomes incredibly difficult to actually audit.
Again this is part of the discussion about Linus decisions which has been had ad-nauseum.
I like Nano :)
Give micro a shot
What makes it better/different?
Just try it
Found the differences and am intrigued.
Now I’d just need to unlearn some muscle memory ;)
VSCode is like the worst fauxpen source ever.
Unrelated, but I absolutely adore your pfp.
Thanks!
Agree with half of that. C++ is last century stuff and should be replaced. Any new code in C and C++ is immediately a vulnerability.
Atleast C is easy to understand which makes it easier to find and fix vulnerabilities. C++ feels like it’s duct tape holding it together.
C is easy to understand and find vulns? AI is uncovering decades of vulns in C code that humans hadn’t seen since the code’s inception. And they aren’t logic errors but memory management errors. Not a single day goes by without a CVE in a C/C++ project about memory management.
Both languages are walking vulnerabilities.
All authorities are just some guy with good programming skills. The giants on whose shoulders we stand were mostly rich guys who had the privilege of screwing around. I prefer the original definition of genius - not a quality that a person possesses, but an external spirit that comes and visits you.
based
-
if you’re making a user-facing program, you should always design with the possibility that it will be run on some truly dogshit hardware and you should make sure that they still have a good experience. good computer hardware is a luxury on the world scale, even before the rampocalypse
related, but the answer to the question “who still uses {x} these days?” is almost always way, wayyyy more people than you think
Me when I got my commercial NAS with 7 years of support, I was thinking: “who even keeps hardware for this long?”
Me, 7 years later: “What do you mean EOL?!”
I certainly learned to care for and appreciate the hardware these years.
I agree. Designing based on “Who uses x anymore?” Just reinforces the ever-increasing cost of computing.
There’s about 0.2% of people still using internet explorer, per one market share site I found.
Which works out to potentially millions of people.
I do my web development on a 2005 era windows XP machine without the unofficial SP4 because if it will run there (without silverlight) I know for damn sure it will run well on anything made today.
Firefox is a good browser
If a browser does not support unlock origin it is worthless. So Firefox wins by all opponents willfully exiting the race
This was my central argument when they tried to take away my Firefox at work.
Firefox is good, Mozilla is bad
You know what, I’ve been a Firefox user since the days of beta Phoenix. I’ve expressed your view for years now - Firefox was great, but held back by a commercial org that wanted fancy SF offices, fancy SF compensation etc.
But I tried out Helium a little while ago and in truth, I’m not sure Firefox is all that good these days. Helium is basically ungoogled-chromium + extension support + super lean UI.
The upside is that you no longer have to contend with sites that have given up on Firefox testing. The downside is manifest v3. But all in all, I find it a better browser. Firefox is so busy telling its story, and finding revenue, that it just seems to have left behind product market fit.
Mozilla has totally lost their way. But using any chromium-based browser is still bowing down to Google’s near monopolistic control over web ‘standards’.
Using the blink engine, even in a ungoogled browser, still gives google power and is immoral.
It is though
It’s the only non chrome browser
There’s niche WebKit browsers too, like Epiphany. Also Ladybird in development, but that went full AI slop AFAIK.
Safari? I’m not suggesting you use it, but it does exist.
We don’t acknowledge the existence of proprietary blobs here
I’d care if they had a Linux version
How about the KDE ones?
Well, there are forks of firefox like floorp, zen, waterfox, etc. but i guess they might still count as firefox, because they are forks and not built from the ground up, only using the gecko engine from firefox.
Calling them forks is, in most cases, wildly too generous. They are derivatives of Firefox. They develop and maintain a few features, but they ar completely dependent on upstream for general progress.
This would not cause the reaction in the picture lol.
Yeah maybe not here but sometimes I feel like it
Where? On the Chrome/Brave subreddit? :p
Talk to any normie that knows what a browser is, and you’ll discover that many people think firefox is bad (and then some of them will also proceed to glaze brave because they never saw what uBlock origin can do…)
No need to go to reddit, there’s enough trolls here saying Firefox is trash and to use brave instead.
Y’know like, the guys saying accessibility features are trash no one wants to use because they use local AI models.
Firefox has single-digit market share.
Single digit can be 0, 1 or 9,9%.
10% market share wouldnt be half bad. ¯\_(ツ)_/¯(PS: I didnt read the link because I don’t care enough. I use it, and am fine with it)
“Your website has a bug”
“well DId yoU TRY IT in a nORMAl BrOwseR?”Every. Damn. Time.
I do have to put up with an unsupported browser banner on government websites (which work perfectly)
systemd does a lot of things very right and a mayor reason that linux is in a way better place than 15 years ago. 27 unmaintained tools are not better than one that is professionally developed.
systemd has me in two places. Like, on one side, it’s just SO much better and easier than the rest - at least to me, someone who doesn’t work in IT and just wants his stuff to work. On the other, init freedom is essential, and we need the alternatives to stay alive.
APIs generally don’t need many alternatives, only implementations do. There’s no reason devs should be trying to provide init files in multiple formats. If the systemd team actually did something bad, then people could just make a different implementation of systemd without requiring a different format for init files. Arguing for more init systems is like saying we need more APIs to compete with both Wayland and X11, instead of just talking about implementations of those protocols.
Current AI tools are excellent productivity and quality enhancers, but only for people who are willing to adjust their workflows and guide it to give good quality output.
The damage is done by Capitalism running the AI industry and the over financialization of the market.
Don’t be anti AI, be anti unregulated Capitalism.
I am anti commercial capitalistic LLMs which, coincidentally is ALL of them.
“B-But Axo the open weigh—!” They are NOT open, it’s the equivalent of giving you the executable of M$ Edge or a Windows ISO and calling it open, fuck this shitIf I was king of the world, I would rule that all AI models were based on stealing from all the knowledge of humanity. Since it cannot be un-stolen, I would decree that all copyright is unenforceable and all current proprietary data must be public domain forever as restitution for the abuse.
Here your crown sir 👑
Its like money laundering, but for copyright
I was thinking more of the French revolution when all the palaces were emptied of valuables and taken by the people.
Many of them are truly open though, with training data and everything open source
Can you share some links please? I tried searching for them but nothing came out but weights
I agree on their usefulness but as long as you don’t use local and open LLMs you are unfortunately part of the ecosystem that does the damage. Luckily open models are catching on and hopefully some reliable companies take up on this business model.
I can’t wait for the crash, so that we can get more sustainable AI and access to the hardware to run it.
Won’t happen unfortunately. Unless you want to run a 1U with 6 GPU, requiring a custom water-cooling system, and which sips like 10kW/h of electricity while idle. It’s all integrated hardware born to be trash. The ram, cpu and GPU won’t be usable on consumer hardware without hard modifications.
We also don’t need frontier models. The type of models that run on $5000 Mac studios can do today what the frontier models did 1 year ago.
We just need the Mac studio class hardware to come back to $2000 where it used to be.
Also, do not underestimate the ability of Chinese BGA rework shops from being able to salvage AI chips and install them on board that will fit in standard hardware. They proved they could do it with the Crypto only chips, they will figure it out for the AI chips if the supply is there.
The main reason these units use water cooling is for deployment density.
Local models are only months behind. The primary utility of comically large networks is a wider search space for the tiny fraction of weights that really matter.
I agree with you but that doesn’t change the fact datacentre hardware is not consumer friendly like it was in the past
Was it?
Not the point - this isn’t about secondhand servers, it’s about RAM being artificially scarce. Possibly as a deliberate measure to make cloud services sound less wasteful. If a 3090 with 32 GB was only a thousand bucks then OpenAI’s revenue would be even worse.
I have an old server with lots of DDR4 which I can for the most part just plug into my PC. The CPU can be put into another motherboard, and the GPU are discrete cards. Most of that hardware can be re used by consumers.
Most new AI datacentre deployment won’t be like that. An Nvidia Blackwell server has everything soldered. Good luck desoldering the CPU or GPU, the ram is either LPDDR5X which I never saw consumer motherboard use, the GPU is HBM3e and not a discrete card anymore. Whole package is unpractical for even a enthusiast to run.
The original point was that once the AI bubble bursts, we’ll have hardware to use for local inference. My claim is that no, it’ll probably be too specialized. Unless some maniac in China figures out a way to make new consumer hardware for it, but they probably won’t since Nvidia hardware is technically illegal (both export and import at this point). Even then, I don’t see Nvidia or others go back to making consumer hardware at a level where we can do serious inference locally.
1 of those GPUs can probably run 90% of what a small team would need, so you don’t need the full thing.
That’s about 600w of draw and 150w idle. Or about the same as a 5090. You don’t need exotic power or cooling for that.
They’re all soldered together, you can’t just take one
Work has forced us all to adopt it, so everyone did. It was either that or suddenly be the slowest person at work, which eventually gets you canned. People will evangelize all day long about how they refuse to use AI, how it’s dog shit, etc. Meanwhile I’ve managed to dial in my AI workflow so that it never deviates from our code standards, it never tries to invent its own components or styles, and I still manage to review it with experienced human eyes. It’s the reason I can start my work day at 10:30am, walk away from my desk at 4:30pm and still be well ahead of schedule. I’ve malicious-compliance’d the fuck out of my job, it has made my life easier, and I don’t give a fuck what some nerd still devoted to the “art of code” says. We live in a capitalist hellscape and I’m a survivalist.
With all of that said, I’d still wipe this capitalist shithole fever dream of a tool off the planet because I’m tired of data centers and AI companies fucking up the entire industry, the planet, and people’s lives. I’ve already witnessed firsthand what it did to my industry and I’m lucky to even still have a job (especially one that pays as well as it does—it helps to be in the industry just shy of 20 years).
That sounds useful. Id like to find a good place that is for optimizing use of this shit (esp locally) while not going full “hey check out my slop project”, or people going “the ai are our gods / will cause an apocalypse”
I use it to corral data a lot of the time. Stuff that is already verified by humans. It’s great for getting quick metrics, or pulling info about something into one central place. I’ve also given it an instruction to use the STE writing rules. It makes it speak more technically and deliberately, as opposed to always trying to sound like my friend.
I’m absolutely not one of those AI worshipper types, but I’ve found ways to harness it (hopefully thoughtfully) at work.
It was either that or suddenly be the slowest person at work, which eventually gets you canned.
turns out, you can just refuse to do so and you’ll still be just as fast as everyone else (source: me). Turns out that you’ll actually do the same amount of work and you don’t have to babysit a robot all day long losing your mind.
I know for a fact that I am not as fast as this. I had it build out entire page flows and pull together Vue components while connected to Figma’s MCP, and it was done in 1 hour. Tests, types, services, everything. I can’t compete with my own AI flow. I had it finishing 5-6 Jira tickets in one go.
Refusing also has its own caveats. Like painting a target on my back for layoffs.
but only for people who are willing to adjust their workflows and guide it to give good quality output
… and competent enough to actually evaluate the output properly.
Instead they spend a lot of money to push the fairy tale how anyone can now code without having a clue.
Why not both? Both is good
Nah, that would be like refusing to use a calculator, because they make them from unrecyclable plastic.
Bad comparison, it’s like refusing to use a calculator because it uses 200 times the amount of energy of existing calculators, was trained on stolen data, and despite all that gets the answer wrong
Calculators are also known to get some answers wrong.
Stolen data is not a tech problem, it’s a legal problem.
And it’s not 200 times more power. Maybe 5 years ago it was. But it is being wastefully used by burning capital.
That’s why the bubble needs to pop. So the true cost/value equation comes into play. It should probably cost 5 times what it currently costs from the big providers. It will cut down on the waste significantly.
The problems with AI is not the tech. It’s unregulated Capitalism that has hijacked it.
200 was hyperbolic.
I’ve seen AI get incredibly basic shit wrong though.
Calculators won’t miscount the number of O in strawberry, genAI confidently will.
GenAI is not a math machine. It’s an inference engine.
Still can’t do basic stuff that existing solutions can though
Yeah on one hand the field of evolutionary algorithms is fascinating and I can see potential. But on the other hand it’s being run into the ground by people forcing it into everything. It’s also not human-readable so people might think they know what they’re getting but really don’t because of the black box nature of it.
Cloud computing is the biggest breach of privacy out there that we’ve adopted whole hearted.
There is no cloud. Just other people’s computers.
Also, I would say that AI has probably one-upped cloud computing in that area.
Having worked at one for a for a few years, I can’t say I agree, nor do I technically disagree either.
Theres nuance.
Theres nuance.
There’s also an apostrophe. Or not.
PHP sucks and every attempt at defending it is pathetic.
I get warm and fuzzy feelings that people here are not even talking about it. Perhaps I should follow that trend.
I am with you all the way. For a high level programming language it’s incredible how close to C it is at times AND how many footguns the entire language has.
And you always hear the “a poor craftsman blames his tools” analogy when someone says PHP bad. But PHP is a hammer that’ll blow your arm clean off when you hit a nail at a slightly off angle.
There is no defending it, and it should have been replaced with better languages several times over.Ok I need to stop myself now I can rant about this language for hours
A hammer with a claw on both sides - respect the double claw lmao
PHP is the best web language, not because it’s a good language, but because the other are worst
AI bad
OOP dogma isn’t just bad, it’s actively harmful. OOP leads to unnecessary boilerplate. Any boilerplate that is not absolutely necessary is noise and bad. Full stop.
Private (Getters/setters): Strictly not a necessity in any sense of the word.
Short functions everywhere.
Short files everywhere.
I prefer to see the steps of execution that are actually happening in the real world on hardware. Not an ivory tower of code that happens to execute.
A bunch of weeks ago the CTO of my company calls me up to ask about my prejudice against the
useStatehook, I’m guessing I’ve just been giving my team too much shit in code reviews about them using useState hooks and they went to complain to him and he asked me to explain what the value of the useReducer hook was.His background is Java, Microsoft Java, C++, and not much else.
So I was able to explain to him the value of the useReducer hook is a lot like the common class object. In useReducer you’re expected to write discreet state mutations and the state could potentially have more than 1 primitive field that gets mutated, and that’s just what object classes are! In an object class you’re writing methods for mutating the underlying state, the reason I think most people don’t feel like these two things are the same is because of the functional appearance of the reducer function and how in object classes you have the
thisas a variable just like any other, so you’re not returning a new state.So just like in react where they’re installing REDUX (like three generations of their families would be massacred if they didn’t) just to write a switch case statement that puts the action in variables in the state. Bad programmers are just bad.
generally agree, probably most devs nowadays do. otherwise there wouldn’t be stuff like
@datain lombok.i think the word ‘dogma’ is pretty good: it takes a good idea (encapsulation) too far. i am not old enough to have been coding when OOP was the new and hot stuff. but i imagine it comes from the idea that it is a good idea to protect private members where needed.
public class Account { private Integer balance; public Integer getBalance() { return new Integer(balance); } }that is not such a bad idea to protect
balancefrom stupid sideeffects.public String getName() { return name; }is a nothing burger, as it returns the same reference and does not protect from accidental sideeffects.
I remember on Reddit several years ago I said that automatic bitlocker is the curse of installing W11 on a laptop.
I got down voted so hard. One comment even said you should be thankful because encrypting by default should be the norm.
But yet if I add any thunderbolt device to my windows before I boot it, it requires me to input my 25 character bitlocker key… that’s hosted online within my Microsoft account.
Bitlocker enabling by default without user input is horrible. Hosting the keys online is also horrible. It could be so easy to socal engineer into your Microsoft account.
Care to elaborate on this? Not saying you’re wrong necessarily, that’s just a pretty bold claim and I’m curious why.
React works by maintaining a virtual representation of the DOM. When you dispatch state changes into the virtual DOM react can determine which components depend on the change and which side
useEffects run.My gripe is that YouTube tutorials teach that the useState is the equivalent of a variable. This encourages average developers to problem solve solutions that end up using React’s Virtual DOM for business logic. The YouTube tutorials have react devs planting 100s of useStates - some of which don’t directly induce a visible-on-screen rerender - and then ⅒ as many useEffects to bind these states together. This means that for the Virtual DOM to come to the conclusion as to what html elements need to be changed and how it has already mutated 20-50-100s of times and the CTO tells me the react app is slow and they’re mad.
If 2 states have the opportunity to mutate each other back and forth then the truth is that you have 2 perspectives of the same state. And React gives you solutions for that - proper solutions - the
useReducerhook is for defining these complex state mutations and the Context API for communicating that state to any component.And this has me quintuple pissed off because these same tutorial grade react devs install REDUX as compulsively as a pack-a-day smoker (that’s #3). They use Redux to file variables in JavaScript memory (that’s #4) like it’s a global variable and rarely if ever use it for the complex state mutations they need instead I’ve seen react devs write components pull state out of redux into a component just so it can mount useEffect and process the state and dispatch the result back into redux. Let’s not even bring up how the react-redux binding library literally uses the React Context API (that’s #5).
As someone who’s worked with react for years, I’m convinced that react itself is the problem and there’s no way to educate new people on the proper way to use it without essentially teaching them everything the abstraction is intended to do for them. Why should newbs have to care about mutations and state dependencies when the api is apparently designed to help them forget about those things?
As a tutorial grade react developer (please return me to the backend where I am mediocre instead of bad), having more senior developers take the time to explain why my code was bad was really helpful.
Also I think I just kind of hate JavaScript. It always feels like everything is more complicated and booby trapped than it should be.
I’m suddenly reminded as to why I went Vue years ago and never looked back lol.
React is absolute trash and people should stop using it
I dont think that’s an unpopular opinion
I can do fullstack but only do backend at work. And that one time I said
useStatefeels like a really dated solution, which was a really recent comment I made, the gasps I heard from the room was palpable and people had to calm themselves down.
Svelte, baby, svelte
Vue > React
Java and JavaScript are blights and should never be used.
I’d add TS to the dumpster fire. It’s a gilded pile of poop
TS dev here. Yep.
There’s worse, but not by much. Looking at you, PHP
TS is JS with makeup
That’s the popular opinion tbh. Even Java devs admit Java kinda sucks.
Nobody hates Javascript like people who use Javascript.




























