Hacker News new | past | comments | ask | show | jobs | submit | kshacker's comments login

I have 3 phones, for various reasons. Not SIMs, but 3 devices. The usage is radically different between them. 2 of them are used daily but even there one routinely runs out of battery and other does not dip below 80%. The third one gets used when it gets used :)

I think the fine print in a report is not an issue but the incendiary blue sky headline is.


I have played around wit AI code from time to time. I do not code routinely but I have pet personal projects that allow me to do some code and this is where I experimented.

Rule number 1 and the only rule: You need to be a subject matter expert. Be it program logic or be it programming language. AI is only a helper, it will go wrong, frequently, and if you do not understand the reason for the code and the programming language, you will take so much more time than if you did not even use the AI.

Without naming the IDE, but one of the top 3 I guess, I asked it to simplify some code. I had repeated code 8 times. 6 of them were identical, the last 2 had a variation. The AI just did not catch it, and refactored all 8 blocks to use the logic of first block. How can you even do that? The code is similar but different, it looks the same but there are extra lines of code in the last 2 blocks !!

And it took me a while to realize this. I never ingest AI code directly so at first I was marveling at a job well done, and as I read and compared, the horror !! And that was not the first time it happened, but once again I got tricked by the soft spoken well mannered AI to believe that it did a fantastic job when it did not.

Edit: It is just an assistant. You give it a task, it will make a mistake, you tell it to fix the mistake, it will fix the mistake. It still saves you time. Next day, it will make the same mistake - and hopefully that gets reduced as the versions evolve.


AI is excellent for tasks you know how to do, but can't be arsed to spend the time.

Example: I wanted a tool that notifies me of replies or upvotes to my recent Hacker News comments. Grok3 did it in 133 seconds with Think mode enabled. Total time including me giving it the example HTML as attachment and writing the specs + pasting the response to a file and running it? About 5 minutes.

I know perfectly well how to do it myself, but do I want to spend the hour or so to write all the boilerplate for managing state and grabbing HTML and figuring out the correct html elements to poll? Fuck no.


Yes I find this to be it's best use case. Unfortunately for anything I actually need help with the results are often terrible.


From my experience using ai, if you don't write a really precise description of your initial requirements in the initial prompt, and it it doesn't one-shot the answer, I don't bother asking it to fix the mistake.

Unless you're using a LLM with really long context, it's prone that some context loss will happen soner or later - when that happens, pointing errors that dropped out of the context will just result in repeated or garbage output.


This is why platforms(?) like Claude Code, Cursor and Windsurf are essential.

Claude Code forces you to create a CLAUDE.md to direct how it works, with Cursor you can (and should) write Cursor Rules.

The difference with a good spec + AI vs just vibe coding from scratch is like night and day.


I don‘t use these tools daily because I have a hard time to commit to these workflows. But do I need to do this setup once, once per project or every time for one off things I might code?


You really don't need these for tiny one-off scripts, but they're essential for larger projects where the whole application can't fit into the LLM context at once.

Basically they're just a markdown files where you write what the project is about, where everything is located and things like that. Pretty much something you'd need to have for another human to bring them up to speed.

An example snippet from the project I just happened to have open:

## Code Structure

- All import processors go under the `cmd/` directory - The internal/ directory contains common utility functions, use it when possible. Add to it when necessary. - Each processor has its own subdirectory and Go package

## Implementation Requirements

- Maintain consistent Go style and idiomatic patterns - Follow the existing architectural patterns - Each data source processor should be implemented as a separate command

This way the LLM (Cursor in this project) knows to, for example, check the internal/ directory for common utils. And if it finds duplication, it'll automatically generate any common functions in there.

It's a way to add guidelines/rails for projects, if you don't add anything the LLM will just pick whatever. It may even change style depending on what's being implemented. In this project the Goodreads processor was 100% different from the Steam processor. A human would've seen the similarities and followed the same style, but the LLM can't do that without help.


why would I pay for the advanced features when I haven't been impressed with the free features? in fact Claude 3.5, which is what is available, is a nearly worthless product, with value comparable to a free search engine, and not even a very good one. It is usually incorrect and frequently in subtle ways that will cost me a lot of time.

pro AI people sound like someone with an expensive addiction trying to justify it. the free product is bad, so I just need to pay to see the light?

Why would Anthropic let me use a model for free that is going to make me more skeptical of their paid offerings unless it is pretty similar to the paid ones and they think it's good?

Just read the manual and write the code yourself. These toys are a distraction.


Free version of Claude sucks because the prompts get too long very quickly. I have the Pro version and I have completed many tasks successfully.

(I know what I was doing and had a great understanding of what is going on).


Like many tools, there is some user skill required. Certainly there are situations where AI assistants won’t help much, but if every single attempt you’ve made to use an AI coding assistant has been “useless”, you are either working in a very niche area or, perhaps more likely, it is user error on your own part.

There are plenty of people who are way too high on the current abilities of AI, but I find that the “AI is useless crowd” to be equally ridiculous.

It reminds me of early in my career working in statistics where the company I joined out of grad school was justifiably looking to move out of SAS and start working in R and Python. Many were enthusiastic about the change and quickly saw the benefit, but there were some who were too entrenched in their previous way of working, and insisted that there was no benefit to changing, they could do anything required in SAS, and stubbornly refused to entertain the idea that there was a benefit to be gained by learning a new skill.

You needn’t become an AI cultist. But with the number of people who are getting at least some benefit to using AI coding assistants, if you are finding it to be worthless in your personal experience, it may be worth stepping back and considering if there is something wrong with how you are trying to utilize it.


What I do is go back through the conversation history, select the response that has the somewhat working code, then submit a prompt with what I want changed. By selectively including context and adjusting temperature, top_p/k and sometimes swapping model or system prompt for a given query will give better results. Combine this with repeating the query multiple times with that same context, then select which result is the best and move on.


I like to think of AI as a force (or skill) multiplier. If you’re low skill, it doesn’t do much. The higher your skill, the more useful it is.


That hasn't been my experience, nor that of others using the AI.

It's a force constant, rather than a multiplier. If you're low skilled, asking it to do a low skilled task, it works fine. If you're high skilled, and asked it to do the low skilled task, you saved a tiny bit of time (less than that of the low skilled person).

But it cannot do a high skilled task (at least, not right now). It can pretend, which can lead the low skilled person astray (but not the high skilled person).

Therefore, all AI does is raise the floor of what is achievable by the laymen, rather than multiply the productivity of a high skilled programmer.


You could also be a mixed skilled developer. Good at regular code, architecture, and algorithms but not as familiar with a given UI framework. Having the LLM generate the html and css for a given layout description saves a lot of time looking through the docs.


Does it really? The thing is that there’s a domain model beneath each kind of library and if they solve the same problem, you will find that they will generally follows the same pattern.

Let’s take the web. React, Svelte, Angular, Alpine.js, all have the same problems they’re solving. Binding some kind of state to a template, handling form interactions, decompose the page into “components”,… once you got the gist of that, it’s pretty easy to learn. And if you care about your code being correct, you still have to learn the framework to avoid pitfalls.

Same things with 3D engines, audio frameworks, physic engines, math packages,…


Using myself as an example -- I'm a long time C programmer (occasionally in a professional setting, mostly personal or as a side-item on my primary professional duties). I've picked up other languages through the years, had to deliver a web based application a few years ago so I did a deep-dive into html5, css3, and javascript. Now javascript has evolved since then, and I lost a bit of what I learned.

So now I want to do a new web application -- If I fall back on my C roots, my Javascript looks a lot like C. Example: adding an item to an array. The C style in Javascript would be to track the length of the array in another variable "len", and do something like myarray[len++] = new_value;

I can feed this into an LLM, or even say "Give me the syntax to add a value to an array", and it gives me "myArray.push(newValue)", which reminds me that "Oh yeah, I'm dealing with a functional/object oriented language, I can do stuff like this". And it reminds me that camelCase is preferred in Javascript. (of course, this isn't the real situation I've run into, just a simplified example -- but I really don't have all the method names memorized for each data type. So in that manner it is useful to get more concise (and proper) code.


I'm sure this is valuable for you, but here is my point of view.

I've worked professionally in many languages like Perl, Python, Kotlin, C# and dabbled in Common Lisp, Prolog, Clojure, and other exotics one. Whenever I forgot the exact syntax (like the loop DSL in CL), I know that there is a specific page in the docs that details all of these information. So just a quick keyword in a search engine and I will find it.

So when I come back in a language I haven't used in a while, I've got a few browser tabs opened. But that only lasts for a few days until I get back in the grove.

So for your specific example, my primary choice would have been the IDE autocompletion, then the MDN documentation for the array type. Barring that, I would have some books on the language opened (if it were some language that I'm learning).


LLMs will be best at HTML and CSS among programming languages because even if it's subtly wrong the browser will still render something.

Still, I'm not sure I need to spend $250/yr to occasionally generate some HTML that I could use a template to generate


My high skilled example from the other day: I wrote an algorithm out line by line in detail and had Claude turn it into avx2 intrinsics. It worked really well and i didn't have to dig through the manuals to remember how to get the lanes shuffled in the way I needed. Probably saved me 10 minutes but it would have been an annoying 10 minutes. :)

But that's a very low level task, where i had already decided on the "how it should be done" part. I find in general that for things that aren't really obvious, telling the llm _how_ to do it produces much more satisfactory results. I'm not quite sure how to measure that force multiplier - I can put in the same amount of work as a junior person and get better output?


> Rule number 1 and the only rule: You need to be a subject matter expert.

Strong disagree. I've been coding for 25+years but never on the front-end side. I couldn't write JS w/ pen & paper with a gun against my head. But I know what to ask and how to make sure a react component does what I want it to do, with these tools.


Or, in other words, you are a subject matter expert and are agreeing with the person you’re responding to. Quote the full argument (emphasis added):

> Rule number 1 and the only rule: You need to be a subject matter expert. Be it program logic or be it programming language.

You are a subject matter in program logic, just not programming language. You are supporting their point, not disagreeing.


I see where you're coming from. But later in the message they go > and if you do not understand the reason for the code and the programming language, you will take so much more time than if you did not even use the AI.

I guess that's the part I disagree on. The programming language is largely irrelevant now is what I'm seeing. And especially the "time to first result" is orders of magnitude smaller when using AI assistants than having to rtfm/google/so every little problem I encounter in an unfamiliar language.

And I am in no way stating that my code will match an expert in that field, of course.


How many thousands of lines does your AI-generated frontend code have?

Do you have to maintain the code?


> head. But I know what to ask and how to make sure a react component does what I want it to do

That is the subject matter expertise that many lack.


Our industry's dark secrets getting exposed like this ... yeah you are right, we don't even need a whistleblower for this :)


There are 2 kinds of vacations: Backpacking, Cruise ships or something where you can not connect, OR sightseeing like London, Paris, or even something local.

Although I have always been an employee, I have been able to mix and match work with sightseeing travels all the time. And this was not remote work, this was weekdays and weekends just checking emails and action items 1 hour in the morning and 1 hour in the night to 1) keep up, and 2) take care of things. Did I have to do that? No, but I like to.

And on one trip, I remember I was coding too. There was a problem I was trying to solve and I was way behind schedule, so I would fix one issue AM and one issue PM and test more and more. Once again, did I have to do that? Did my company force me to? Nope. Just something I wanted to take care of. And the day was available for family and sightseeing and I guess subconsciously plot my next bug fix.

I think opportunities for asynchronous communication can be created if A) you are motivated and B) your vacation schedule is not too extreme. This should be even easier with a product you own and understand and when your business is at stake.


My favorite one is unfortunately backpacking. But thank you for your perspective!


I posted something similar to what you said just a few days back although not exactly in those words: https://news.ycombinator.com/item?id=43083345#43085603

1) I would always struggle to interject in time due to slow processing, and 2) I could write better than others because I could edit.

However, this changed / improved over time. And I have a sneaking suspicion that it changed after I had stents placed giving me more oxygen supply. I believe many parts of my life changed around the same time. Not immediately but over the next year or 2, it has now been 9 years. This could just be life eventually connecting all the neurons in the right way, and of course I do not recommend getting stents as an experiment, but I wonder if apart from this being how we are built, there may be fixes (herbs or surgeries) that may be able to improve this.


I lived in a Hindi speaking area for my first 23 years. Then I lived in a cosmopolitan area for 5 years. And then I lived in US for 14 years predominantly speaking english.

At that point, I moved back to India. And I had to converse with a telephone company call center, even though I selected English as my language in the phone tree, they repeatedly put me to Hindi speaking ones - I just could not converse with them. I tried. I could absolutely understand them, but I could just not speak hindi. The words or rather the sentences will not form.

Not any more. I think now I am truly bi/trilingual, but there was that day / week / month / year, when I had lost the ability to speak in Hindi. Not sure why, and it was not a one time event.

[ The only additional information that may be relevant : In between these years, I learned a semester of Russian. I learnt french. I got familiar with a multitude of south Indian languages - Tamil, Kannada, Telugu, Malayalam hearing friends speak them on a regular basis. And I was also conversant with Punjabi, common in our region in north india. But I never spoke any of these in day to day life for 14 years. And all these languages blurred into each other when I really had to use something other than English. ]

[ Also, I had a reason to switch to English pretty much exclusively. My son has learning disability and would find it hard in school as he will speak in Hindi and teachers would not understand. This is a long time back. So I decided not to confuse him and switched to english exclusively and lived like that for about a dozen years before this telephone company incident happened. I have posted about him before, this is not new: https://news.ycombinator.com/item?id=28665942 ]


I feel languages and cultures also gradually change over time and over a couple of decades it can be fairly different when you've been disconnected from it.

I was born in Hong Kong and lived there for 13 years, moved to the US and I'm now 40+. I was back in HK last year and although didn't have problems understanding people in Cantonese for the most part, there were situations like calling a restaurant (to make a reservation) where I absolutely could not understand the waitstaff's Cantonese. After 30 years I still consider myself more fluent in Cantonese than English but language and culture had evolved in HK so much that the way younger people talk, especially on the phone where it's harder to hear, is almost completely foreign to me at this point.


That is true of all languages, but I think that Cantonese has a particularly high change velocity in terms of vocabulary and pronunciation compared to, say, English.

I think this is because Cantonese speakers tend to read and write in Modern Standard Chinese (more like written Mandarin) and so are much less anchored by the permanence of text. Additionally, Chinese characters provide even less guidance on pronunciation than English spelling. In this landscape, Hong Kong's small media ecosystem is a fertile breeding ground for new language.


I know a guy whose mother came from italy to the US when she was young. He went back some Family over there. He was told “you speak Italian well, but you sound like my grandmother”


Same thing happens in the US too —- at least in the north east. If you go to New York, northern New Jersey, Boston or even Chicago, people in their 40s and younger don’t often sound like their parents; there’s little to no accent.

I work with a younger guy from the south and he definitely has an accent but that could be an anomaly.


This is from 90s. Tamilians in Sri Lanka were fighting the government. This is well known. Many of them escaped and came to India.

One of my Tamilian friends who lived in Delhi all his life went to visit Chennai. He was asked more than once if "he was from Sri Lanka" due to his accent - which was probably not Sri Lankan Tamil accent, but unfamiliar enough to get them thinking.


I grew up in Romania. Spoke German at home, it was my Muttersprache (German minority in Romania). Went to a German school, but learned Romanian in K-12 there, spoke it fluently. Left Romania when I was 18. I can still understand most when I listen (I can watch Romanian movies not dubbed). I can read it and understand most of what I read. But I cannot form sentences anymore, cannot speak it for the life of me. Pretty strange. I guess it needs effort to revive it in one's brain.


Same here! Just that my family emigrated when I was 5 years old, plus my Romanian native mother passed when I was 14. this was some 25 years ago. Now, I do understand Romanian media, but can’t speak past a simple „Buna ziua, ce mai faci“.

Interestingly, some 18 years ago I went to intern in Mexico. After 6 months there, I was quite fluent in Spanish. But when I wanted to think of a Romanian word, all that popped up was Spanish. Pretty wild how the brain works.

(Romanian, Spanish, French, Italian, … are all Latin influenced languages, btw.)


I learned a little Spanish growing up in the USA.

As I learned French as an adult, when I didn't know a French word, the Spanish word would often come to me.

Now I'm learning German, and often the French word comes to me when I don't know a German word. However, I've also struggled to speak French again because the German often comes to my head first.

It has been amusing to observe that one doesn't learn a third language as a strict subset of the second. For example, I know a lot more French than German, but I know the German "Tierarzt" (veterinarian) but not the French equivalent.


Don't worry, you will get speaking skills with practice quickly. I think you can talk to OpenAI app in Romanian. Not sure. After not speaking for 3/4 of my life I understand only about one half of audio news and bit more of text.


I'm the same. Spoke Hindi in India for the first 17 years of my life, and then the next 15+ I've lived in foreign countries and barely talked to anyone in Hindi, just English. I can still understand Hindi movie clips on Youtube, but reading Hindi articles on Wikipedia is hard when I try occasionally (I have to read out aloud and then listen), and forming sentences is extremely difficult because I just cannot remember most of the words. My brain keeps bringing up English words, and Japanese words because I spent a lot of time listening to Japanese, instead of Hindi words. It's the same feeling as "the word is on the tip of my tongue" where you keep remembering other words instead of the one you're looking for.


Wow. English+Hindi+French+Russian+Tamil+.... So cool. You should probably add some Arabic, some Chinese, some Xhosa, some Gaelic and some native American languages to this list to appear endlessly awesome :-)


I can not speak most of them. But having friends speak them for years makes you start catching the words, phrases, sentences and even the context though not equally well, not all the time, and not the same for each language - this is for the South Indian languages. I did formally study French and Russian (1 semester). And lol, I took Russian when the USSR still existed :) Those were the days.

Edit - I think I am done. I want to learn Spanish since it will be useful locally, but my brain can not take any more confusion. When you do not use these regularly, but have a faint notion of many many words, the brain does get confused on a regular basis. And I was not trying to brag, expressing helplessness.


I have noticed it with a lot of Indian people I know, an aptitude for language. My understanding is that many learn Hindi, their home tongue, and english at school.

Of course I only have exposure to my indian friends who moved to Australia for work, so they're biased towards people who went through university and learned English in school.


Yes, a lot of Indians that emigrate to foreign countries will have gone to schools where classes are taught in English, except for the local language class that is taught in that language. There are so-called "English medium" schools ("medium" as in the method of conveyance of knowledge in this case). There are also schools that are $local_language medium, but they were generally for poor people / lower social classes.


Only around 40% of Indians speak Hindi at home. Many Indians don’t speak Hindi at all. But a lot of Indians will learn Hindi and English in addition to what they speak at home.


I speak a few languages, and find myself losing the grip on the native one.

People I know that know more languages than me seem to be worse at all of them.. and by worse I mean they can hold a fluent conversation, but have extremely poor spelling and grammar.


Perhaps it's language dependent? My Indian colleagues mostly speak English with each other, even if they have a common native language.

On the other hand, Chinese people speak Mandarin with each other when there are no English speakers in the group. It was partly why I learned Mandarin...


English often is the common native language between Indians if you have people from different parts in India.


> The words or rather the sentences will not form.

I had the same problem after having speaking in English for a long time and rarely any Hungarian. I had issues finding the right words.


I wonder what the impact of Apple withdrawing from China will be. I know we are talking about UK, but this made me think.

Not only their sales will reduce, but hey Chinese manufacturing cuts down. By how much? Will it be impactful? I would think so but wonder if it is quantifiable.


Almost all iPhones are made in China. They cannot pull out without shutting down.

They make on average 60,000 ios devices there every hour, 24 hours a day, 365 days a year.


Your math adds up to 525,600,000 iOS devices per year. That can't possibly be right


> In 2023, Apple shipped 234.6 million iPhones, capturing 20.1% market share and growing 3.7% year over year, according to IDC data. [0]

So, probably not 525.6 million iOS devices a year, but safe to assume it's going to be 300+ million for 2025.

35k devices an hour, give or take.

[0]: https://www.forbes.com/sites/johnkoetsier/2024/01/16/apple-1...


As medwezys pointed out, you forgot iPads. That’s another 40-70M units per year.

My numbers are a rough estimate from memory, but they’re not wildly off.

300M or 500M, the point remains: it’s an absolutely staggering scale and cannot be moved elsewhere in any short period of time. Setting up comparable production would take many years, just as it did the first time.

I imagine Apple/Foxconn have already begun this work. The unexpected shutdown or impediment of US/CN trade is a risk that must be accounted for, given the situation with Taiwan.


Apple has more devices than iPhones, so the OPs numbers are not unbelievable


Google pulled out but their phones are made in China. When push comes to shove money always wins still in China.


Classic communism am I right?


And now we are going to require body cams in everything and AI will sort it out for us


This is from a few years back, but etched in my memory. I am an introvert in social situations. I can present to a room, sorta, and I can manage public speech with some prep, but I just listen when there is a party going on. No talk, no intervention on jokes, just listen. My brain just works too slow to butt in to people's conversation, and over time you get the tag of introvert or whatever. I can understand conversation well, but I need an extra second to respond and that does not fit into most conversations.

But then a few years back, we had a sendoff for a colleague, a bunch of us got together, and right across from me, or adjacent, was another colleague who was sci fi fan. We had discussed some series before in office, and this was an opportunity to continue. Bit by bit our conversation went deep, and at one point of time I recall a pointed comment from the host (the person who organized the sendoff), hey what happened to kshacker :) because I think we were talking loud and non-stop. And I realized that too, it was an unusual setup.

Since then, I left the team, that dude left the company, so I havent seen him for a few years, this was definitely pre-covid, but this discussion somehow made me think of a 2% moment, if not a 2%er.


kshacker, I am just a stranger on HN, but if you haven't already... hit him up for me would you?

Why for me? I think you two having another conversation would make the world a more wonderful place as I suspect you 2 will be enjoying each other's conversation. The thought of that makes me happy.

> This is from a few years back, but etched in my memory. I am an introvert in social situations. I can present to a room, sorta, and I can manage public speech with some prep, but I just listen when there is a party going on. No talk, no intervention on jokes, just listen. My brain just works too slow to butt in to people's conversation, and over time you get the tag of introvert or whatever. I can understand conversation well, but I need an extra second to respond and that does not fit into most conversations.

I have this too. I've trained myself to lower my filter. The unfortunate thing is that when I'm in conversations, I'm a lot dumber than I normally in. The fortunate thing is: I talk to a lot more people now.


> The unfortunate thing is that when I'm in conversations, I'm a lot dumber than I normally in.

That's OK. Everyone takes time finding their groove. There used to be days where I did not engage in any conversation until the afternoon, and so my speech always felt rusty. Now I'm at a point in my life where I'm having a chat at 6am, and that sort of thing continues throughout the day. But boy oh boy am I not very interesting to talk to for that first hour.

Anyways, I think most extroverts don't really care that someone appears a little jagged in conversation. They simply benefit from having a conversation.


Did you consider reaching out to that person and saying hi?


I'm an introvert in that I enjoy my own company a lot, and I get exhausted by social events. I'm also on the spectrum somewhere, as I find it hard to read social queues and such, which often makes me feel really awkward.

My sister has a million friends and used to host parties all the time, and invited me often. One day she said "people always love it when you come", and I was a bit surprised by that. I asked how so, and she said "you're so good at making people feel seen".

I didn't do it consciously I think, but I realized I would find people who were quiet and start talking to them. If I got the vibe they wanted to be left alone then I'd move on, but most of the time that was not the case. I'd ask about what they studied or worked with, or hobbies, as I do like to discover new things.

She since got kids so not so many parties anymore, but I got reminded of this a couple of years ago when I was at a wedding for a family member.

We're in Norway and another family member had gotten a US girlfriend who was there. As the evening wore on I went over and started talking. Her work was interesting and in contrast to what I knew from Norway, she lived in Nevada and I'd been to Las Vegas once so we talked about the contrast of that city vs the rest and so on. A couple of hours flew past and we were the last to go.

As we were about to leave she gave me a big hug and almost teary eyed said "thank you, you have no idea how good it was to just talk".

Again I was surprised, but she told me that she'd been with the groom's family for a week and nobody had really just talked with her besides her boyfriend.

So yeah, introvert, socially awkward me at it again it seemed.


Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: