HNHacker News
TopNewBestAskShowJobs

mewpmewp2

3,684 karma · joined November 13, 2019

submissionscomments
mewpmewp2··on Learning to code is still worthwhile
I swear I see far worse from humans for working businesses, and in much worse ways than just duplication of code. At least AI tends to be smarter about React intricacies even if it does create huge files. For my personal projects however I have straight token limits per file, whether it is backend or frontend so AI can't do that many lines of code. And I have specifically token limits so it would avoid hacking with formatting. And for React specifically I force it to put hooks in separate files and component files to be presentational only with max 1 hook usage.
mewpmewp2··on Learning to code is still worthwhile
Having worked at a company with thousands of microservices, I feel like most of it is just a bunch of boilerplate riddled with tons of ridiculous business logic, security and otherwise bugs. It seems like people were historically more concerned about Java inheritance and spreading everything around different files and directory hierarchies, while ignoring common sense business requirements completely. With each service being on separate older version, many being very difficult to start up locally, I don't see how LLMs could do worse than that, and that is a profitable, working business.

Funnily enough AI is the one discovering all those glaring issues now, and everyone is overwhelmed with getting it fixed. You could allow AI to attempt fixing it, but even though all of it was human written it is still hard to review, and even locally spin up or test because no one any more knows the true business requirements as people have rotated etc, so it's a nightmare.

The code on the first sight looks good, but what could be a simple config map, is spread out abstraction that is impossible to understand. Think just massive amounts of boilerplate to make a proxy call to another microservice etc.

mewpmewp2··on Shadcn/UI now defaults to Base UI instead of Radix
The issue is that you can't trust that every word of the message was intentional due to how much Claude will usually add various levels of nonsense, some more obvious some less to your voice or requests. Considering how much Claude alters the meaning of what I want to communicate I wouldn't trust the accuracy of that output right.

And I use Claude a lot, 24/7, but not for things like that. And I appreciate how much it elevates my productivity, but not like this. It usually prioritizes or highlights the wrong things, it overfixates on one thing I said and adds random content there out of nowhere.

So then I can't tell what part of it is slop, how slop, and it becomes impossible to trust.

mewpmewp2··on Claude Code is steganographically marking requests
But even if this gets detected, they could have other less detectable processes going on as well, right.

It is going to be this cat and mouse game right, so at some point you want to throw as much out as quickly as possible when you are under attack, while building up the long term more scalable defense mechanisms.

Rationally I would assume that a lot of what you would quickly throw out would seem sloppy whether it is AI or not.

mewpmewp2··on Claude Code is steganographically marking requests
There is laziness, but there's also the conditions in which you have to react fast to an adversial in various conditions. Ultimately it's hard to take any stance here without knowing specifics. But it absolutely could be a matter of time, to do your best effort to stop efforts from the attacker if there's known attack going on.

There is a real time cat and mouse battle going on here in terms of keeping the advantage here, right.

As a rational actor, if someone was e.g. attacking me, leaving aside the whole copyright thing, but potentially using some sort of system to increase their value while decreasing my value (without calling it theft to avoid the whole debate), I would want to put proportionate defense out there as fast possible, depending on the amount of value that was exchanged to stop the bleed, while in parallel figuring out the best long term plan, right.

mewpmewp2··on Working With AI: A concrete example
Yeah, I think it's more so learning from experiences that didn't scale. E.g. I started out with Notepad, and wrote everything for the website I wanted to build in a single large massive php file. I of course don't do that anymore, but it was a step by step iterative progress to move to where I am now. Although I still miss how easy it was to see changes locally, deploy to prod quickly, and make hotfixes in prod. I sometimes think maybe I should go back to php.
mewpmewp2··on I used Claude Code to get a second opinion on my MRI
But the problem was that the report is not saying "not found", it is saying "is not present" or "there is no X".

And I think we can easily have examples where we can reasonably trust this, and a spectrum of such.

E.g. there is a math solution and the report says "there is no errors in this solution", you would imagine that to be quite reliable, no?

mewpmewp2··on IP Crawl: Living atlas of open webcams discovered on the public internet
This is precisely why I never heat my food and consume caffeine pills instead of coffee.
mewpmewp2··on IP Crawl: Living atlas of open webcams discovered on the public internet
> anyone is morally and legally correct

I think it's more so similar to that if you leave something shiny and expensive in a visible position in a car in a neighborhood known for high rate of thievery there are good odds of your stuff being stolen. They are not claiming that the thieves are morally or legally correct.

mewpmewp2··on The Exhaustion of Talking to a Tool
What is your relation to those people? Because, what is the use of emotional intelligence if it doesn't help them to see their biases and instead confirms them. If presumably their biases are harmful for their lives?

I just see those excuses of "emotional intelligence" to stay dishonest in my life a lot, is that the right way to live though? In a way those LLMs are doing the same though, and in a lot of ways those people are then treated as some beings that can't handle the actual real world.

I guess, just in real life when talking to people, I would just not pretend and go along with something like this, so it's uncomfortable for me to also imagine that someone would go along just for the sake of not upsetting someone for some virtue called "emotional intelligence".

I can understand being polite for the sake of it to someone old and in their deathbed, but I just really don't like the idea of pretending or lying to people with their whole lives ahead and going counter to what you believe to be the truth for some social comfort or emotional intelligence aspect. If they can't handle whatever is the truth, it's their problem to deal with, but they better know alternative views to their malformed or harmful ideas of life.

mewpmewp2··on The Exhaustion of Talking to a Tool
Wouldn't Bayes and logic lead you to immediately believe that some claims, ideas, and beliefs may be more logically sound or matter more than others, steering you away from indifferentism, but the reason why I thought you might be stuck is because you were indifferent to moving away from it.
mewpmewp2··on The Exhaustion of Talking to a Tool
Why did you pretend to agree with them?

Would a person with such biases, seeking validation from an LLM having been anything else without an LLM?

If everyone in real life are trying to avoid upsetting such person, the humans in their real life would offer the same experience, no? Or alternatively said person would go into conflict with anyone pointing out issues in their views, ending up only surrounding themselves with people who will agree with them in order to not upset them.

I guess my concern is why did you agree with them instead of using the opportunity to showcase another point of view?

mewpmewp2··on Captcha proves you're human. HATCHA proves you're not
Is that a reference to the popular meme from youth nowadays?
mewpmewp2··on The Exhaustion of Talking to a Tool
I see, that totally makes sense. You got stuck there and there's just no way out.
mewpmewp2··on The Exhaustion of Talking to a Tool
And what is the reason for indifferentism as opposed to atheistic agnosticism?
mewpmewp2··on The Exhaustion of Talking to a Tool
Do you mean those three things are what let you solve it, or that they're what kept you from solving it more often?
mewpmewp2··on The Exhaustion of Talking to a Tool
Would you say you are happy as in you have solved happiness for yourself?
mewpmewp2··on The Exhaustion of Talking to a Tool
I think happiness is very complicated. Would you think there's a perfect formula to follow that will yield you the "happiness"? And how does this relate to different brain wirings of different people? Would you say per brain wiring there's a different specific ideal formula one would have to follow? E.g. there's some base things like getting proper sleep, exercise, diet, but then there's some more specific things for that brain wiring?
mewpmewp2··on The Exhaustion of Talking to a Tool
And on the other hand talking to people, but I guess most extroverts in a way are "addicted" to talking to people as well, so it's a fair comparison. From my personal experience talking to people can end in wildly differing and unexpected results. In fact, I think rewards from LLMs are way more static as opposed to rewards from talking to people.
mewpmewp2··on The Exhaustion of Talking to a Tool
How does that answer the question or relate to variable reward and addiction mechanisms? This just seems another unrelated negative sentiment argument towards LLMs?
mewpmewp2··on The Exhaustion of Talking to a Tool
That is interesting, but also I do wonder, what in life doesn't have a variable reward? You could say that any routine activities like maybe cleaning, etc, is all the same, but most non routine things in life seem like they would have variable reward?

E.g. talking to people definitely yields in very variable rewards, if you do non routine work, there's constant variable rewards, etc.

mewpmewp2··on Captcha proves you're human. HATCHA proves you're not
How many numbers were in the list?
mewpmewp2··on Fired by Google for creating the Google workspace CLI
Depends on the reason. If there was a valid reason not to release it, then sure. But it's hard for me to see what can be so problematic about releasing a cli wrapper. It's hard to think of anything non political. And politics isn't valid reason to me. If someone from the leadership was e.g. blocking the release out of pettiness then I'd cheer on the person releasing it - assuming they know the risks.
mewpmewp2··on Fired by Google for creating the Google workspace CLI
In any case, all I see, is that he built a cool thing, and Google failed to make it work whatever the process was. This is failure on Google's part in my view.

Even if he didn't do the proper process, this is just a CLI wrapper, and they should have made it proper after, not fired the person. And they should have thought long and hard why a person felt this had to be done. Or alternatively why was the person able to publish something as a new repository in their devrel github org if the release process is so vital for whatever reason.

Even if you have or had the legal right or whatever, it doesn't mean that you have to actually act on it. It speaks to some other sort of motivation to me.

mewpmewp2··on Fired by Google for creating the Google workspace CLI
I meant the person got good publicity, not Google. Google got bad publicity.
mewpmewp2··on Fired by Google for creating the Google workspace CLI
In this case I guess the only other option the person had is to quit to build something of their own or work at another more builder friendly company, but at least in this case they got to release a thing that was valuable to many people and got popular, so they got some good publicity out of that. So ultimately I don't see that their actions were wrong overall.
mewpmewp2··on Fired by Google for creating the Google workspace CLI
Going for the little guy who wants to build cool stuff here would be the thing that makes them waste their billions on the wrong thing.
mewpmewp2··on Fired by Google for creating the Google workspace CLI
So if this is a case where they weren't legally in the right, but were morally in the right, which is far more subjective, it makes sense to go the Twitter route since if they were morally in the right people would support them while legal discourse not. So morally they had a cool idea, put a lot of imagination and effort into it, it's an idea that was valuable to people, and maybe a big corp was blocking them for egotistical or whatever reasons, to me it makes sense what they did. I don't know if that was the case, but if a corporation is forcing you to with legal means not do what is perhaps your life work and good value for the society, there must be way beyond legal to do it. But clearly the pettiness of the company is showing here, who cares if this confuses people a bit, there's many other good ways to solve the situation by Google.
mewpmewp2··on Fired by Google for creating the Google workspace CLI
To me the random guy comes off much more likable here than someone who has spent billions protecting its reputation and still failing.
mewpmewp2··on OpenAI Losses Increased Nearly 8X in 2025, with Spending Hitting $34B
I suppose it could be if you did freelance or your own business.
← PreviousPage 2 of 34Next →