thunky

> LLMs generate compartmentalized modules which exponentially explodes the number of failure modes

It sounds like you're arguing against modularizing which is an unusual stance. Especially for a million line codebase.

> Guess we’ll need even more human software engineers to fix it.

This is what we've been doing for years before LLMs. Hiring a ton of devs to spit out code, which became another's burden. It just took longer and cost more.

That's two purposes.

I think the first one is the whole story anyway.

When he's gone we can just undo it as easily as he did it. And tear down all of his other tributes to himself.

If we don't want idiots to do idiotic things we probably shouldn't make it possible/legal in the first place. But the corruption and malfeasance goes way beyond Trump so here we are.

> Besides, in this area, clients getting one whiff of AI code would be an immediate deal-killer

If you implement AI code review suggestions then you're using AI code, just the hard way.

Also I don't see why your clients would be ok with having AI review and influence your code. Seems hypocritical, or like you're trying to cheat them.

> Turns out the gains were eroded by having to check the work and turn verbose prose into an actual note.

LLMs are great at this. I'd be very surprised if it wasn't possible to improve the actionable notes.

Most human-made doctor notes I've seen have not been comprehensive, anyway, and the doctor has had to spend a good chunk of a rushed appointment scribbling notes rather than interacting with the patient.

Same thing you would do for a human decision:

You don't blindly follow it, and you also don't order a DNA test of the person that made it. You look at the output to include supporting evidence.

And you have a process for appeals.

You're accusing GP of saying something they didn't say and simultaneously telling them they don't "get it".

That's distasteful.

Great. Just don't get upset when a politician you don't like wipes his butt with the consititution.

The article is fine. LLMs are just turning everyone into divas.

Did you read TFA? Because all of your questions are answered clearly there.

The good news:

Your worst complaint has nothing to do with the overall content or accuracy of the post.

Just style bashing.

A Million Little Pieces was a popular book about addiction that inspired a lot of people. Turned out the story wasn't true, and so a lot of inspiration turned into anger.

Always thought that was interesting, because if the lies weren't exposed the readers would have remained inspired.

https://en.wikipedia.org/wiki/A_Million_Little_Pieces

All of the gripes OP has with Django are arguably worse in Rails.

This implies that "create" is the hard part that we should not offload to machines.

But IME all of the other things are the hard part.

So basically it's fine to get the LLM to think for you and give you the answers as long as you understand Rust syntax well enough to type it all out yourself?

This also seems like it will produce worse code, because it's preventing the LLM from doing the easy/boring part that it's good at.

With respect, you're likely being taken advantage of.

Your advisor doesn't have any fancy computer with more info than you have. You're paying for an illusion and it's not cheap:

A 1% annual fee reduces your wealth by 20-25% over a 30 year period.

At least do yourself a favor and check to see what your total return is with him vs simple index funds.

If you need a professional sounding board you can pay by the hour.

> This is the misunderstanding. Golfnow is for advertising.

Yes that is the misunderstanding. Lets agree to disagree.

> With respect, I don’t think you understand what you’re talking about.

Probably not. I'm just telling you my observation.

And you're pointing out a problem without offering a solution. Your advice to OP is "don't link to Golfnow". So I suppose they should do what you do and just link to the course homepage... which links to Golfnow? What does that accomplish?

In any case, the golfnow.com is not what most people use to find courses. OP even said they are building their search app because "Google filters for golf course results are terrible". Not because golfnow.com searches are terrible. Golfnow is for booking, for which there are other alternatives that courses can use.

Obviously it is, because Golfnow is certainly not a monopoly.

There are small golf courses near me that don't use them and that's fine. Please don't shame the ones that do.

> Because it's convenient to have all buyers operating through a single network

I just don't see Golfnow as a "single network". I golf with a lot of people that have never heard of them. If a course isn't on Golfnow and I want to golf there, I still book with them however they want me to even though it's not as convenient.

Just like Booking.com has competition from Google, Google also makes it easy to find and link directly to any golf course to book.

I'm not saying you're wrong about some courses not liking them but I have a hard time understanding why it's so hard for courses to just not deal with Golfnow and take bookings another way if they think it's so harmful.

> Golf Now can be incredibly bad for those courses

Golfnow offers a service that courses must value, otherwise why would they list with them? And golfers get good deals.

It's not clear to me what's so bad about this.

I agree, which is why I prefer to say software developer rather than engineer.

Sure but lets not pretend this is engineering.

> I think the tool itself might be at fault.

I don't think so. I don't think it's about being "unworthy to work with" (your quote not mine).

I think the uncomfortable truth is that AI is making it harder for all of us to prove that we're productive.

Your examples indicate that you're working with unproductive people.

The truth may just be that you're wasting time and energy engaging with a non-productive coworker's poor outputs.

Atlassian could have fired her for no reason and had no problem.

I think you're helping GPs point: there is a lot of efficiency gains to be made to match the processing power of the brain, given it's size and power draw.

Its not fair that US has to bench their leading scorer due to a mistake either.

So there's no way to please both sides.

But beating someone with one hand tied behind their back should not be preferable.

I never said anything about code reuse, taking credit, or approval. So I'm not sure what your points are related to.