stanmancan

> That makes no sense unless you're claiming that the models are getting worse at writing code.

A hypothetical scenario is it needs to add two numbers so it writes add(x,y).

Later on it needs to add three numbers so it writes add(x, y, z)

Then it needs to add four…

Layers upon layers. A human would likely notice the pattern and refactor so add accepts a list.

I have had a very different experience so far. The more code AI writes for me the worse it gets. It gets more complicated and it’s harder to follow and understand. It doesn’t refactor, it layers on top.

Even worse is that it’s hard to do anything about it at this point because reviewing code is very different from writing it, so even if I’ve seen it all I don’t have that same deep level of understanding that I do when I write it myself.

All these AI code bases are ticking time bombs. Either AI gets smart enough that it won’t matter in the future or we’re going to have a huge mess to clean up.

Can you elaborate for a Hyundai owner?

Isn’t that what the agents.md in your project is for?

It’s always a hassle trying to find a time for our friend group to hang out, it can seemingly go on for hours. Everyone shares what days they’re available, someone suggests a time, it works for all but one, and the process starts again. That works! Makes the process way easier: when you set your availability it highlights time blocks that works for everyone.

It was a fun use of bitmasks; all availability is stored in 30 minute blocks (0001110000 would represent 1.5 hours available in 5 hours total) and then I use the AND operator across them all to find matching slots.

My problem with hard mode is it leaves a lot up to luck. If it was 3 letters and ended in “at” it could be pat, mat, bat, hat, cat, rat, sat, etc. so all you can do is to pick one and hope for the best.

I only shared because it might very well be cool stuff but if people are tired of AI writing and close it like me then it’s a lost opportunity. Not publishing at all is a bigger loss though, so it’s a tricky balance.

The only way to get better at writing is by writing.

I’d rather read a less polished but clearly human written article than something clearly written by AI. If I can’t tell I can’t tell.

Use it for editing or as a review but the weird hyperbolic AI stuff makes me instantly close an article. I barely got past the intro:

  ONE ATTRIBUTE. EXACT BEHAVIOR.
  
  The markup says what happens.
  
  Start with local behavior. Add the server only when the interaction needs to cross the wire. The attribute is moo= — in case you were wondering what the ox says.

HN needs an AI flag or something. Not sure how well it would work, but worth an experiment.

> and it's likely producing better output to keep a human in the loop rather than trying to solve the world independently and making multiple incorrect assumptions.

I was just discussing this with a co-worker yesterday. I would really like a model (or harness?) that worked with me instead of for me. Walk me through its choices and decisions, let me correct it and guide it along. I would be way more confident in it's output, I would be more familiar with the changes that are being made, and it would make reviewing the final code way easier since I was making the decisions along side it. I'm sure it would also reduce the "brainrot" we're all going to experience the more we hand work to these models.

To be fair, Apple and Google are on opposite sides of the "respects user privacy" spectrum.

Cut down that tree.

Can I have a saw?

No

Okay, it will take much longer then as I’ll have to do x, y, and z.

Pick one: [That’s fine, proceed] [Okay you can use a saw]

Sure, but vibe coding a whole telecom platform that’s considered flawless by experts in the language is a pretty mythical feat if you ask me.

Possibly, but I’m a senior software engineer that’s been writing Elixir for the last 5 years and has been experimenting with and using AI for the last 12-18 months.

Congrats on being one of the mythical developers that manages to get AI to write perfect code consistently!

If you vibe coded an entire telecom infrastructure and an external audit found no issues then it sounds like you might need to find better auditors.

Such as?

Yes exactly! Good for you for keeping at it.

If you haven’t ran before and are carrying some extra weight then yeah it’s probably a good place to start for your first couple of short runs. Even running slowly puts additional stress on your achilles so it’s a bit of a wake up call to your body. Running slowly for 10 minutes a few times when you’re just getting started isn’t going to hurt anyone and if everything checks out then you can start to up the pace.

10km/h is very fast for a beginner and likely unsustainable for more than a couple of minutes, I would expect most people to be starting around 5-6km/h if they’re a healthy weight.

I think the best piece of advice is START SLOW. way slower than you think. And run for way shorter than you think. Even if your lungs and muscles are fine, if you haven’t ran for a while then your tendons certainly aren’t. You won’t know until it’s too late and then you’ll be out of the game for weeks or months. If you can run on grass or a softer surface your body will thank you, running on concrete is brutal until you’re used to it.

There’s probably programs than you can follow that introduce you to running, I’d follow one of those.

The issue with not having a single source of truth is not the fact that you have to update code in 2-3 places, it’s that you have to know to update code in 2-3 places.

Accidental divergence is the problem, not intentional.

The ones most likely to be “exposed to market forces” right now are the junior developers who never got to “have it good for so long”.

It's likely nearly impossible to evaluate that in the short term; I think we're looking at generational damage, much of which won't be apparent for years to come.

Yes the same applies to junior and inexperienced developers.

You can scaffold out a simple app pretty easily. Anything large or complex things break down. If you don’t know what you’re doing you end up leaking secrets like the dozens of examples we’ve seen so far.

A large portion of the content on the internet is now generated by AI.

You can and do have full conversations with bots and not know. I want to interact with humans not LLMs.

There’s no way to combat it. An army of bots can post a specific rhetoric and it can and does sway people’s opinions.

The new version of Digg was shut down because they couldn’t find a way to combat AI. They were at least trying to, other platforms are just eating it up because “user activity” is a win for them.

I’m pretty sure they’re smart enough to remember to put “make no mistakes” in their prompt.

This lands for me. I’m pushing 40 and over the last few years I’ve definitely been eliminating distractions. Anything with scrolling or algorithms meant to suck you in is gone. Deleting apps and blocking websites on my phone to prevent distractions. Phones getting much less use. Just yesterday replaced my Apple Watch with a regular watch.

Sorry if I worded poorly but you’re definitely less likely to run into a collision with v7, but it’s not impossible, which is what I was trying to point out.

Thanks for a more articulate answer!

Plenty of systems end up generating multiple UUID's in a single millisecond.

The issue with UUIDv7 is that you also have significantly less entropy since you only have a 62 bits (sometimes less, depending on implementation) of "random" data. So while the time aspect of format lowers the chances of collisions, generating two UUIDv7's in the same millisecond (depending on implementation) have a significantly higher chance of collision than two UUIDv4's.

It's still incredibly unlikely, but it's also incredibly unlikely you generate two matching UUIDv4's, but it does happen.

TLDR; It's possible to generate matching UUIDv7's, don't assume otherwise.

The scenario being the collision itself, the time period isn’t particularly relevant aside from it occurring much quicker than expected.