redfloatplane

Yes, exactly. Unless you have a deep intention to maintain this library or somehow elevate it beyond the version anyone would get given a couple of minutes and Claude, what's even the point in publishing the code or submitting to HN? And what is the benefit for anyone to use _this_ library instead of getting Claude to build their own with whatever constraints and requirements they have?

Again I don't want to come across as criticising the OP, I am just struggling with this myself. I've vibe-coded a few great things for myself but the value is so personal (and the lack of intention to maintain is so strong) that I would rather others build their own version than use mine!

This is pretty much what I'm trying to get at. Why would you even begin with _this_ library when you also have an agent that can build precisely the thing you want with the abstractions, constraints, etc that you want? What value does _this_ library have, considering that it takes you _exactly_ as much time and effort to build your own?

Yes, definitely some of it can be simplified or automated. I've been migrating various apps from a docker-compose to a kubernetes environment. Spending a day writing a solid prompt document and running some agent on the repos involved has drastically sped up my workflow.

The integration hell isn't much easier, but I spend _far_ less time actually writing configurations, yaml files, looking up docs, etc, which frees up my time so that I can deal with the workplace politics involved... actually, maybe I want to go back.

Yes, indeed. I think that's why I haven't published much code in the last years since I vibe-code everything I build now and I have essentially no intention to maintain it once it's 'working'

I think that's the thing I'm trying to say. _This_ library might not exist but anyone with an Anthropic account and the knowledge that there is a readable sensor can make it exist just as easily.

Ha, I'm not into programming for the sake of it. I vibe-code all the time now and don't think I've written a line of code by hand in over a year. I'm really not trying to criticise this library itself and I agree that in many ways it's quite good.

What I think I'm trying to say is that this specific kind of thing now takes everyone the same amount of time to create. It's "cheapened" this specific type of code. I'm not trying to say this particular library is bad or anything, just that converting some small library from one language to another no longer really holds much value, since essentially everyone can do exactly the same thing exactly the same way now, for the same time cost.

OP, don't read this as a direct criticism of you.

I think the age of AI has really cheapened work like this. It's clear this library was vibe-coded; it's clear enough that the python version of the library originally posted yesterday was vibe-coded; I didn't look at the original library but it would shock me not at all that it was vibe-coded. Often just one or two commits and a functional library, emoji all over the readme, "Clean and easy-to-use API", etc.

In many ways this is pretty amazing. Only a few years ago it would have been a huge pain in the ass to come across some valuable library only for it to be locked in some language I didn't understand or wasn't working in at that moment. But in other ways, maybe it feels a bit "cheap" now to do `claude -p "port this library to $LANG, make sure it works, do a good job" and I'm not sure there's a ton of... accomplishment? craft? care? in it.

I agree. We all have thoughts and feelings and emotions; denying that, making that invisible, hiding the mere fact of your humanity is surely a difficult and sad way to exist. It is OK to feel things; it is OK to write about your feelings; it is OK to publish those writings wherever you like.

In my opinion this type of blogging - writing about genuine personal experience, ideals, feelings - is a positive.

At the end of the day the grandfather comment is more or less correct, the internet is a cold unfeeling whatever. But the act of writing from the heart is deeply, genuinely human and in this day and age I feel that the more human things I do, the better. And perhaps by being more obviously human will inspiring others to be more obviously human, and eventually making the internet (and perhaps the world) less of a cold and unfeeling whatever.

Yeah, that's a great idea. I think perhaps after the fourth guess you could also show the guess direction arrows, so the user can know if they're in the ballpark and they won't get too frustrated? But it's hard to keep the simplicity of a wordle-style game!

I like the concept but it's basically a bisect / binary search simulator. Guess a reasonable but definitely high number as a high bound, a reasonable but definitely low number as a low bound, guess the average of the two, then the average of that and the high or low bound, etc.

This is especially the case when the question is asking for a bounded number in the first place (eg a percentage). In fact I'm pretty certain you should _always_ succeed within 4 steps given +-10 on a percentage question and nearly always within 3 steps. ChatGPT says it's provably so but I'm not smart enough to verify. Rings true though.

Certainly made easier by knowing whether it's higher or lower, and especially with the yellow arrows if you're not too far off.

One UX change that might be nice is to have a "spoken" version of your guess live-update below the input. I keep having to count zeroes and it would be nicer to see "Eleven billion".

Insanely smooth user story. From signup to unique solution to my problem in no time. Amazing work, I am blown away.

Reviving my long-dead account to say that I built a perfectly functional small site to help schedule my dungeons and dragons group within about 5 minutes, on my phone, from my bed. If this isn't the future I don't want to go there. Fantastic work.

I got one too! https://github.com/CarlQLange/chatgpt-arcana.el

In the meantime there is also my package, but I'd be fairly confident it's not as good as the other one. https://github.com/CarlQLange/chatgpt-arcana.el

TabNine actually announced that functionality a few weeks ago for their code-completion tool: https://www.tabnine.com/blog/ai-unit-test-generation/

I haven't used it, but it might be what you're looking for. I'm actually pretty sure ChatGPT can write effective tests as well, though.

I think it's likely that the edges of generative AI haven't really been found yet. I think it's hard to say whether any given (textual) requirement definitively can't in some way be fulfilled by ChatGPT (or a similar LLM) without at least another several months of messing around.

I certainly haven't made up my mind about what this might do to the world in general - more a question for science fiction authors than programmers, I think. In any case, I am fairly unworried (and more excited) by these developments. ChatGPT can't stop me from sitting on a mountain and eating a banana. Yet.

Ha, I knew I wouldn't be the only person adding ChatGPT to emacs (https://github.com/CarlQLange/chatgpt-arcana.el).

I am beginning to think that ChatGPT is the singularity. I built that emacs package, but really, ChatGPT built it using me as a conduit - there isn't a single function in there that ChatGPT didn't write the majority of (although the poor code quality is all my own).

It could go further. Using something like langchain, I'm pretty sure you could get ChatGPT to instruct itself in building a python library or something:

  User request: build a python library to print out ascii charts in the terminal

  ChatGPT2ChatGPT: What are the steps I need to take to build a python library to print out charts?

  For step in steps:
    ChatGPT2ChatGPT: Carry out this instruction

And so forth. I'm sure some human interaction is required, particularly given the context limit. But even so.

Another thought I had was that using ChatGPT to help me build things is so convenient that I am unlikely to pick technology that it doesn't know about. Perhaps this kind of thing will cause a chilling effect in the creation of new tech. Who knows. Truly exciting times.

The pricing of the API is truly world-changing. I've sent hundreds of requests today in the process of building this script, and have yet to spend more than a single cent.

Wait 'til I tell you that the original article was also basically just marketing. It's conceptually recursive

Yes, we always heard people tell us that Ireland was an amazing place for hiking, but being Irish ourselves, we didn't quite see it that way - until we walked it. It's really spectacular landscape, although some access issues can mar the experience a little bit. It's certainly a lovely place to go regardless!

Another keyword or two and we've got ourselves ISO50001!

I'm so glad you liked it! Best of luck with the project and with telling people about it!

Ah, that's my partner Ellie that you've seen! I am male.

I was thinking about it, but to be honest, I'm not sure I'd have anything new to say beyond what's in my comment!

The trick is to sneak it in under a bunch of self-effacement! ;)

Wow, I thought I would resurrect my HN account for this.

I wrote this article a little over a decade ago, when I was still in college, and I knew almost nothing about the world. (I still don't, but I didn't then, either) It's fun to see that it still resonates with people despite the naivety and the slight sense of "I am headed for success" that makes it hard for me to re-read.

My initial reflection on the article was that I had fallen off the "do things, tell people" wagon a little bit, having let go of my twitter account and with it much of the easiest path to doing things and telling people. I still work in tech but I haven't created anything I'm particularly excited about in a good while, so I felt that I was failing my own ideal.

Then, of course, I remembered that I hiked all of Ireland's long-distance trails (https://toughsoles.ie) and have a reasonably successful, if niche, youtube channel (https://youtube.com/toughsoles). Funny that my mindset, especially regarding this article, is so tech-focused that I automatically discount something I've been doing for five years.

All that said, I'm starting to look for new opportunities in the larger-than-startup space. Shoot me an email at carl@flax.ie and let's chat.

Hey, thank you!

Appreciate the feedback. We'd never even edited a video before we started walking, so we're definitely still figuring out what we're doing, creatively and even just in terms of using the tools. It's also a function of doing it as we're walking - often we just don't have that much time to put a video together (very short backlog). I think that the style of our videos differ from one to the next - some with music or montages, some with very vlog-style cuts, and so on. For instance the Cavan Way video (https://www.youtube.com/watch?v=tvOQU_ZWoWQ&index=1&list=PLX...) is quite different from the One Second Per Kilometre video (https://www.youtube.com/watch?v=dXJvOvkJbcM&index=6&list=PLX...) :)

We do more than just the videos, too - we keep a pretty active Instagram/Facebook account, and write much more in-depth blogs on https://toughsoles.ie. It's hard not to be overwhelmed by the work and the newness of it all :)

Thanks again for the feedback! It really helps. Our audience is so small still that it can be hard to get honest, uninvested feedback.

Thanks! It is pretty fun! It's pretty hard, too, though, particularly because we don't have a car - it's camping and public transport all the time.

Something I was thinking about earlier today is that the longer I'm away from day-to-day programming, the more I want to think about bigger, open-ended problems. Lately I've been thinking about code synthesis, for instance. I wonder why distance from a desk does that :)

At any rate it's amazing to be outside all the time. I would highly recommend getting outside and walking your nearest trail to every single person reading this! In Ireland, we have a gigantic network of walks, including short 2-3 hour loop walks: http://www.irishtrails.ie/trails.aspx?c=-1&t=-1&l=-1&g=All&f...

It depends on how you classify "successful" :)

I quit my job in order to hike every trail in Ireland with my girlfriend and make videos about it: https://www.youtube.com/toughsoles. We're not making a profit by any means - so, not successful in that regard. But we are seeing the country that we belong to, and I think we're slowly raising awareness about the incredible places that exist tens of kilometres from people's front doors.