@SubstantialFrivolity's banner p

SubstantialFrivolity

I'm not even supposed to be here today

5 followers   follows 0 users  
joined 2022 September 04 22:41:30 UTC
Verified Email

				

User ID: 225

SubstantialFrivolity

I'm not even supposed to be here today

5 followers   follows 0 users   joined 2022 September 04 22:41:30 UTC

					

No bio...


					

User ID: 225

Verified Email

I would also suggest that @self_made_human provide some account of how long it took (not counting CPU time of course, just how long he had to spend on it) as well as how many iterations it took to get right. You presumably have some kind of idea how long this task would take you, and then you can compare. Because in the end it isn't just "can it do the thing" which is important (though that is indeed important), it's also "is it less effort/time for me to have it do the thing".

Thanks for the clarification. I definitely mistook you in that case and I'm sorry for that. I'm not sure how to bridge the gap between our experiences and our respective "it's so obvious" understandings of the tech, but at least we are coming at it from a place of cordiality which I do appreciate.

I've enjoyed everything I've read by Tom Holland, he's great. Currently I have read Rubicon, Millennium (weirdly title changed in the US to The Forge Of Christendom), and Dynasty. Recently I was at the used bookstore and saw Dominion and Pax, and I'm really looking forward to both.

he believed Belichick should "wait a year" before induction as penance for Spygate, the 2007 cheating scandal that cost the team a first-round draft pick.

Shit man, he should be disqualified for that reason. You cheat, you don't get accolades. Very simple, or it should be very simple.

Thank you for the correction, I did mean ChatGPT base but mistakenly believed it launched with 3 and not 3.5. But no, there was not a meaningful difference between the two models in my opinion (based on my usage of the two). I appreciate you at least not insulting me, but unfortunately I don't think we will ever see eye to eye on this.

It's not intrinsically incorrect to say "him and I went to the store together," though it's not standard American English.

What's grammatical is defined by what is accepted as grammatical speech by native speakers of that dialect.

We are going to have to agree to disagree here. You seem to be a descriptivist, and I am very much a prescriptivist. So I think that "him and I went to the store together" is intrinsically incorrect, no matter how many people say it that way. They are using an object in place of a subject, which is incorrect grammar.

If only that were true. Native speakers have incorrect grammar all the time. For example, people who say "him and I went to the store together". Not all the rules are something you pick up naturally, and a decent number of people simply do not care about using the language correctly.

Seriously, compare gpt-4 and gpt-3 output, this is not something that can really be disputed by any thinking person.

I guess I'm not a thinking person then, because GPT-4 was not in my opinion any better than GPT-3. As such I won't continue to waste your time with my brainless ramblings.

it's willful ignorance to take the capability of a free tier as the actual frontier when told otherwise in a debate forum

No, that is believing the evidence which is available to me. AI bros have been claiming that (insert paid model here) is so much better for a long time now (since GPT-4). It's never been true, and every time those models become available for free use I have seen that they still have the same problems as the previous model did. At this point claims that the state of the art is better than the free tier have no credibility at all, thanks to years of false claims to that effect. Maybe the claims will eventually be proven true this time, but I sincerely doubt it based on past performance.

Several top level posts have boiled down to posters thinking that the free model one can demo on the LLM's developers website, represents the best that developer is able offer.

It is eminently reasonable for people to take the "try our product for free and see if you like it" offering as representative of what the paid offering can do. That is, indeed, the whole point to such an offering: give people a taste so they want more and are willing to pay for it.

Good to know. I didn't choose this one (it was a Christmas gift from my sister), so I'm open to the possibility of reading another. Any in particular which you would recommend?

I'm working on finally reading the backlog of books I acquired towards the end of 2025. Just finished Monstrous Regiment by Terry Pratchett, the only book of his that I've read. It wasn't bad, but didn't really inspire me to read more. There is some wit there, but I didn't find the book as funny as the author's reputation would've had me think. And the plot, while serviceable, was nothing super interesting (and also the everyone is secretly a girl thing got kind of old by the end). Overall, meh/10.

I also read but didn't finish a book called The Expectant Detectives by Kat Ailes. Honestly not something I would normally read, but my local book store had "blind date" books and I thought it might be interesting to go with something unknown. The blurb on the wrapping promised a "funny, cozy British mystery" which sounded promising enough, but I didn't enjoy it at all. The premise of the book is that an eight months' pregnant woman moves from London to a small rural town with her boyfriend, and gets caught up in a murder mystery when the shop owner hosting her prenatal class dies in mysterious circumstances. Unfortunately, the book also reads like it was written by Reddit. Partly in tone, but also in frequent asides for the protagonist to preach the good word of feminism, and comments about "adulting" that leave you no doubt whatever about how much of an adult the speaker isn't. The worst part came towards the end of what I read, when the protagonist shares with a friend that she's worried her boyfriend is "getting into right-wing politics", and goes so far as to say that she would more easily be able to accept him having an affair than that. It was a really awful book, don't read it. I made it through a third before giving up, and flipping to the back to see who the killer was.

Next up is Isles of the Emberdark by Brandon Sanderson. I don't expect it to wow me (as the novella it is based on didn't), but I also don't expect to hate it either. We'll see.

Windows is so much worse than OS X.

Strongly disagree. Windows is going downhill, but OS X has always been inferior to Windows.

Deep Rock Galactic. You are a team of dwarves, going on mining missions in procedurally generated caves while fighting off waves of bugs. The missions are reasonable length, the dwarf classes complement each other nicely, and the tone of the game is fun, dark comedy stuff. Definitely worth your time. Rock and stone!

They aren't supposed to be fast days, they're supposed to be abstinence from meat. The only required fast days in Lent are Ash Wednesday and Good Friday (and in my experience it's common to see Friday fish fries not run on Good Friday for that reason).

I want you to know that I'm not ignoring your question. I've been trying to think about it, and ultimately I'm not sure. Perhaps someone whom I respect personally would have to explain the reasoning to me? But short of that it's hard to say, despite my honest attempt to come up with something. I realize it's kind of a lame answer, and I apologize for that, but unfortunately it seems that lame honesty is all I have in response to your (quite reasonable) point.

Agreed that some such people exist. But they are not the majority in my experience. So if LLMs really can provide the 10x boost people claim, then you'll see most of the engineers seek it out for themselves, and you can address it individually with those whose output starts to fall way behind.

this is just not really how corporate salary workers function

I disagree. I'm willing to acknowledge that there exist people who are going to resist change, not care about productivity gains, etc. But that isn't all of the workers or even close to all of the workers in my experience. Most engineers I've worked with love tools that make them more productive, and will use them no matter what management does. Some will even use tools that management forbids in an attempt to be more productive. It's not theoretical, I've seen this behavior.

So as a manager, there really is no rational reason to push people to use LLMs. One needs to focus on the outcome, not the process. Most of your engineers are going to be stoked to get a 10x performance boost if that is real, at which point you can talk to the laggards and say "hey you need to keep up with the standard the rest of the team is setting". But at no point do you need to go out of your way to push a particular methodology.

The whole concept of FOMO exists because there really are situations where you could be missing out.

Sure, but neither does that mean that one is missing out just because one has FOMO. You need to temper that instinct with some thought, and I see no sign at all that managers are doing that.

I'm definitely feeling a lot of pressure from the higher ups to get on top of using LLMs.

That's something which drives me crazy about the whole thing. If LLMs really are that good of a tool, you don't need to mandate them. If someone can get a 10x speedup on his work, he's going to use the tool without management breathing down his neck to do it (indeed he'll probably use it even if management forbids it). All management needs to do is let nature take its course, and if some people are suddenly doing 10x the performance you have them coach the others on how to get that same speed-up. It's completely irrational to require people to use LLMs, rather than focusing on the results. It's nothing but FOMO really, and it's so aggravating to have to deal with.

So much cologne. What can men do against such based posting?

Really? When a woman passes me with an enticing perfume I think something very much along those lines. Perhaps with a bit more eroticism.

Different strokes, I suppose. I find perfume to be actively off-putting. Thankfully, my wife is not a perfume kind of gal.

If the free tools suck (and they do), it's pretty unreasonable to expect anyone to pay to see if the paid tool is good. That's a waste of money 9/10 times, and there's not a compelling reason to believe that LLMs are the 1 time in 10.

I had previously gathered the impression that AI coders were still at the level of an enthusiastic but sloppy apprentice.

They very much are. They still make syntax errors from time to time such that the code won't compile, let alone getting the logic of the code right. You have to carefully review all AI-generated code for mistakes (which negates the time savings), or you will get buggy code sooner or later.

Normally one doesn't, but it can happen. My front teeth are in pretty rough shape from a great deal of neglect in my 20s, such that my dentist has warned me it's possible the teeth could break off entirely if I'm not careful. I don't wish to play with fire, so I wouldn't bite into an apple at this point. But in general I agree with you that there's not a need to cut apples to eat them.

The thing that always really annoyed me about Enterprise was the decision to come up with a canon explanation for why the Klingon look changed. Just no. It was done for real world reasons, the audience is capable of rolling with it, move on. There's no need for a convoluted explanation, or indeed to acknowledge the change at all (unless it's to wink at the audience like in Trials And Tribble-ations).