tyleo

I learned the fast shoelace knot from this site years ago. Highly recommend it: https://www.fieggen.com/shoelace/ianknot.htm

It’s both functional and a great party trick.

I wrote about this a little bit today too. You’re up against a dopamine machine that writes code for you.

https://www.tyleo.com/blog/the-terminal-star

A lot of good comes out but it can be hard to separate from the parts that just take advantage of your brain.

I mentioned this in a separate thread. It seems like a huge risk to release an online game if you don't do this now. A single bad game can take down a studio. Now there's even more risk because a previously successful game may come with a big bill in the future if you have to do refunds or some late architecture change when you instead want to take a product down to save money.

I'm actually in the industry. I don't think it's as much about supply and demand as it is about expected value of the product.

Most games are expensive to make and most of them fail. Way more than normal software which doesn't have ultra-high marketing costs or diverse staffing needs (Art, QA, game design, etc).

Seems like a local maximum or organizing around an individual’s quirks.

Like all team building I feel like the fundamental question is, “what works for this group of people?”

Rather than “teams with/without calls is superior,” and slamming every team you work with into it.

Yeah, I'm with you and I'm not seeing much sense in the argument above.

We should only need to specify observable behavior... but observable behavior is so broad that it includes common defects... but also we shouldn't need to specify the lack of defects even though they are observable.

It feels like, "the AI should read my mind and do the right thing without me fully specifying it."

Caps Lock <—> Esc should honestly be standardized on keyboards. Esc is used pretty often in my experience while Caps Lock, being modal, only gets the occasional press even when used.

If leaking credentials is observable behavior. I don't get your point.

Memory leaks, deleting the hard drive, spending money would all be observable behavior.

By your reasoning that the "observable behavior needs to be specified rigorously" it seems like you'd have to list these all out. We do, after all, already have cases of AI deleting data.

That sounds harder and more error prone than what we're doing now by rigorously defining these defects out of existence in code.

I use the same Mac to write that I use for everything else. But I find it’s more useful to disconnect from multiple screens and just use the laptop on my desk if I need to focus.

Those idle screens taunt me with a desire to use them for Slack or Hacker News when I’m trying to work.

That doesn't sound right. For example, there's plenty of software with the correct observable behavior which leaks credentials. So what needs to be captured goes beyond observable behavior.

If you want to know the truth I find it worse to ask questions that folks can answer themselves.

When you don't know the answer to something you should try to figure it out before asking. That builds important skills. Like it or not LLMs are now a useful tool for that. The first instinct shouldn't be to ask someone else if you haven't expended any effort.

For example, if you don't know 6*27 I'd consider it rude to ask if you have a calculator on your desk but you're too lazy to use it.

While they failed to achieve the goal outright, I'd argue that each is a concrete step towards it. The languages we have today are more productive than the languages we had decades ago.

I think it's an open question of whether we achieve the holy grail language as the submission describes. My guess is that we inch towards the submission's direction, even if we never achieve it. It won't surprise me if new languages take LLMs into account just like some languages now take the IDE experience into account.

The underlying mechanism is still the same: humans type and products come out.

So something which must be true if this author is right is that whatever the new language is—the thing people are typing into markdown—must be able to express the same rigor in less words than existing source code.

Otherwise the result is just legacy coding in a new programming language.

As a meta note, I'm seeing more downvoted responses in this comment section for reasonable points of view on both sides than I've ever seen for any HN topic.

It's interesting that this is so polarizing.

Yeah, asking questions you can find an answer to yourself also disregards the reader. It cuts both ways.

I don't know what's worse:

1. Asking a question which could be answered by an AI

2. Pasting an AI response to something

If 1 is fair game, I'd say 2 is too.

Lots of these places measure employee token use with managers having dashboards. It seems like performative code production rather than making anything useful.

Speed without judgement always compounds badly.

I feel the same way. I hear a lot of complains about wealth tax but it always seems like the problems mainly pertain to billionaires. I don't see why we should optimize for that small minority.

If we moved to a wealth tax I'd be the first in line to pay it. So long as everyone else had to pay it too.

What makes them unenforceable?

I used to be against wealth taxes but as inequality gets out of hand I've more and more felt like they are the right move.

Hell, I'll be the first in line to pay the damn tax so long as billionaires are right in line with me too.

I’ll be honest, I haven’t heard of this until now. Scrolling through the list of blogs, I also haven’t heard of many of them.

A lot of blogging is reputation and consistency. The other part is audience. It looks like he's been at this for 3 years. There might just not be an audience for it. If that's the case, and the hustle doesn't bring you inherent joy, it probably isn't worth doing.

It reminds me of a Stalin* quote: "Quantity has a quality all its own."

* Note that it may be misattributed to him

FWIW I write a blog and do talks so it’s not confidence: https://www.tyleo.com

In my case I just use AI to fix typos and save people time if I repeated myself in a message.

I said in my comment: shortening, cleaning things up.

You’re foaming at the mouth a bit about AI here rather than reading what I said. I don’t see how shortened or fixed grammar is slop. The article argues the opposite.

That’s interesting. When I use AI to help me write chat messages it’s almost always, “make this shorter,” or “clean this up”

Every one of these posts about boos at commencement speeches has one of these comments near the bottom. I feel like I’m failing some pop culture quiz. What does this mean?

It’s not that simple. Many of the newer products are designed cloud-first.

I’d love to see you rise to the challenge. If you think there’s an opportunity to make money here, “lots of ways to make this work,” perhaps take it up yourself.

So an indie dev just trying to make a multiplayer game with no budget and no graceful shutdown plan may be acting irresponsibly?

Is the world really better if those people face additional burden? Does it create a fairer system that helps indies succeed? Does it encourage creativity?

The subject matter is interesting but the blog post is obviously written by AI or at least heavily edited :(

And the Hacker News title should be changed to the slop title in the article, "It's Not The Terminator. It's worse"

I generally don’t fill my context with enough stuff that this becomes a problem. I don’t think more data = better on the token side. Instead I’d be researching with focused prompts or subagents and surfacing only relevant context to a primary agent.