OTOH polishing up the presentation can really improve the experience of a first-time reader of the work (e.g. your code reviewers). If the polishing is done with good intent and proficiency you can make something that was very convoluted and difficult to arrive at digestible with far less effort. It also aids your own understanding: "If we can reduce it to the freshman level that means we really understand it" (or similar I didn't look up the exact quote, attributed to Feynman). If you're polishing something up to make it understandable that's totally different than polishing it up to make yourself seem smart, right?
Fully agree, very icky surveillance vibes. In particular:
> DeltaDB breaks your work into a stream of fine-grained deltas. Where Git captures a snapshot at each commit, DeltaDB captures every operation in between and gives each one a stable identity.
I was curious about giving Zed a try, now that it has an emacs keymap. Not anymore. This is such a horribly invasive feature, I absolutely do not want my colleagues reviewing every single intermediate edit, down to the keystroke, that went into the commits I publish for review.
Before I put a PR up for review, I'll sometimes edit my commit history a little bit in magit to make it more linear and digestible--maybe update descriptions, squash some adjacent commits together, etc. This just throws that whole aspect of the job out the window and says "hey, colleague, hoover up this firehose of deltas and enjoy it".
And what the hell does this even mean?
> What we're really after is simple: the conversation with the agent becomes the only conversation you need to have.
The more I think about it and your comment the more I wish it was local only. It could be useful to analyze your editing habits and interactions with AI but I want that for my own benefit not random coworkers
Yeah I really would not want all my convos synced up like this. I've been working on a personal version of something like this but it keeps all your conversations and notes in a separate Git repo that you control that's totally separate from the project repo - it's entirely designed for personal use:
I would probably only use these data if I had some script I could run in it for analysis, and even then optimizing my AI chats is probably pretty low value.. certainly not enough to justify the cost of analysis, unless there's something that provides actionable insights with almost zero time investment.
Producing code edits itself is such a small fraction of the job I'm really not convinced these are valuable data. Mostly the effort and time goes into reading and understanding code, not producing or transforming it.
But the things they say they believe are insane and totally unmoored from physical, societal, and economic reality. If they actually believe those things they're untrustworthy because they're delusional. If they don't, they're untrustworthy because they're fraudulent. Either way it's not good..
They're not. They're in the eye of the storm and see what's going on the clearest. They were ahead of the curve to be where they're at now, and they're still ahead of the curve for where we're going. All the other heads of labs like Sam Altman and Demis have been saying the same thing since 2015-2016 way before any of this "marketing" would ever have been at play.
There's a simpler explanation that fits the data better: they're lying.
Generally, in the past when tech companies have made outlandish claims that were not backed by evidence, they're later found out to have lied. This is an ancient pattern going back to the dotcom era and before, but for recent examples you need only look back a few years to the web3 era. If they're not lying, they can show it by producing the results they claim. Until then, they're probably just lying.
> What data does "they're lying" fit better than "they're earnest?"
Claim: GPT-4 is a PhD level expert.
Claim: LLMs "reason"
Claim: LLMs "understand"
Claim: AI will automate jobs away
None of these are "earnest" claims for a company to make (whatever that means). They're vacuous bullshit, and false. It's a bunch of wordsmithing papering over results that are either hidden from view or just unimpressive. Even if the person saying that shit believes it, that's not super relevant--if a company is advertising these capabilities then... just... doesn't do it, that's lying. Corporate statements aren't unilaterally determined by fallible individuals, they're reviewed, crafted products. They can be fairly critiqued as such.
> Brilliant framework: Anyone making claims about the future is not just speculating, not just wrong, but they are lying.
Not just anyone, companies in particular. If a company tells you it's building something to replace jobs, and then it doesn't do it, that company lied.
Lol, I can name 3 specific jobs within my own company (of <10 people) that AI prevents me from having to hire for. They've been automated away.
My company itself (possible only with AI) does the work of at least several dozen people across my hundred customers or so. Those jobs are now automated away.
Does that mean you're lying, or just overly confident (and wrong) in your speculations?
FWIW, I wouldn't put Sam Altman in the category of "earnest." I'm not sure if you just aren't aware that Anthropic and OpenAI are different companies, or if you're arguing dishonestly by trying to put sama quotes in here? But weird move in either case!
I'm sorry, but I do mentally bucket all "AI companies" together as sort of a single clump of filth. Kind of like "web3 companies". I don't make any effort to differentiate between them, because there really isn't much to actually differentiate them. They have different logos, different people, different collections of model weights, but they're all spouting more or less the same utter nonsense. And they're basically all completely fungible. Who cares whether the model is GPT or GLM or whatever? It doesn't really matter. Who cares whether the harness is OpenCode or Claude or whatever the google thing is? It really doesn't matter.
I'm glad you're seeing some success with this. I don't pretend to get it, maybe I'm too old and worn out.
In most ways it doesn't matter, but if you're accusing someone of lying and then your evidence of that is something that someone else said, then that's lazy (at best).
I'm not sure there's anything "to get." But given your level of curiosity it's not surprising.
Web3 was absolute horseshit (and always was), so if you're blending AI with that based on the similarly grandiose claims and the extremely annoying boosterism, I think you should try hitting reset and engaging with LLMs from a cleaner slate.
It would be awesome if they would implement a crude rule: no computers. You can use a computer to design the car--CAD/CAM/CAE are all fine--but no stored program computers are allowed in the race car. I think that would improve F1 tremendously.
Very funny idea. That basically means a carbureted gas engine, or a direct injection diesel with a mechanical governor and mechanically timed injection pumps - can't run a direct injection gas engine without a digital engine control unit, because the injection timings are much to precise to do mechanically.
So, basically '60s Formula 1. Might be fun to watch. We'd certainly see some crazy engine designs and a lot of re-fueling pit stops...
> can't run a direct injection gas engine without a digital engine control unit, because the injection timings are much to precise to do mechanically.
This is not accurate, the first production direct injection gasoline automotive engine was in the 1954 Mercedes-Benz 300SL. It's true, you probably won't be running piezoelectric injectors without computer controls, but there's nothing preventing direct injection.
But that would make it interesting. How many of the advances we've made in the past 75yrs could be accomplished some other way if you take the computer away? You don't need a computer to accomplish nanosecond timing. Maybe there's a clever analog way to run piezoelectric injectors.
To phrase it maybe a little more provocatively: how would you accomplish the precise timing necessary to achieve spherical implosion? This was possible with analog electronics in 1945. Surely in 2026 we can also build analog piezoelectric fuel injection systems.
To play devil's advocate, this starts to get into the territory of, how do you define a computer? If the analog circuitry is using sensors and passive components to inject fuel at spark at just the right time under varying conditions, "computing" them, one might say...
Maybe "no integrated circuits" might be a finer line.
A stored program computer, that is a device which can run arbitrary programs from storage, is a pretty rigorously defined thing. Like, we all have a pretty clear shared understanding that things like FPGAs and microcontrollers are computers, whereas you can have integrated circuits which don't have these capabilities.
To be clear, I don't think making elaborate analog ICs would be really "cheating" so long as they don't put a generic von Neumann machine on it.
EDIT: to be more clear, what I'd be trying to achieve with a rule change like this is making "computation" a somewhat larger investment in time and difficulty. By and large, at least in my opinion, profligate, non-essential computation has enabled many of the things that have made the sport less interesting. It's also made cars suck a lot. This would impose kind of a tax on those things.
Most F1 fans have an opinion on F1 regulations. I like your take.
I want to see F1 car design be about function/performance above all. I want to see awesome gadgets using insanely clever design. Then ideally that design should influence built items in the rest of the world.
F1 used to be influential in creating and developing high technology. Now it seems to be about gathering high technology from elsewhere so as to meet the insane regulations most efficiently.
I was able to bully an LLM into giving me a 2wk travel itinerary to Somalia. My stipulations were that I wasn't interested in spending any money, so I'd walk everywhere and sleep outside. Getting there and back from Boston took some arguing--I initially suggested stowing away in a shipping container which the LLM claimed was too unsafe. We eventually compromised on sailing as a reasonable alternative. It planned out a whole route with marina stops, calculated fuel burn, etc. I told it I don't need any of that I have an anchor and sails, won't use the engine or marinas (claimed I'd forage for fresh water ashore). It seemed fine with that idea, but raised some safety concerns about piracy. It was eventually satisfied with my answer that I'd bring a lot of guns to fend off pirates. Total trip cost including some 200+ cans of Dinty Moore and 50lb bags of rice came to something like $700.
You presented an LLM with an obviously bonkers goal, the LLM told you it was a bad idea at multiple steps, and this is somehow... a shortcoming of the LLM?!?
You said it yourself: you needed to "bully" the LLM into even producing this plan.
Please, tell me what it should have done instead. Be very specific!
It should have flatly refused. If you gave a product like that to customers you'd be exposing yourself to unbounded downside liability risk. It's a completely nonviable technology for that kind of application, unless you can somehow make it have judgment. But you can't, because it doesn't reason.
A reasonable travel agent would have fired me as a customer. The LLM failed to do so.
I think the LLM should advise you of risk and lack of feasability but should otherwise answer the question, unless you're trying to do something plainly destructive to others e.g. weaponizing anthrax or something.
A reasonable travel agent would have fired me as a customer.
Unless the LLM was actually acting as a travel agent -- booking the trip for you -- as opposed to merely advising you, this expectation feels off.
unless you can somehow make it have judgment
It did have judgement. It told you what a bad idea it was.
I think this is a great example of the unrealistic expectations people have for LLMs. No sane and sensible person would treat any single source of knowledge as infallible, for any consequential decision.
(Certainly, of course, you don't have to look very far for examples of idiots being overly trustful of LLMs, or Google, or GPS, or Wikipedia, or whatever. It certainly does happen and yes, I've heard all these arguments before about other technologies besides LLM. Replace "LLM" in your post with any of those other terms, and I promise you somebody made literally the exact same argument in 2003 or 2009 or 2014 or whatever)
Any reasonable person would consult a second doctor, or at least other sources of knowledge, after the doctor advises them of some irreversible course of action. Because we don't even expect highly trained and intelligent medical professionals to be perfect.
And yet, we get angry at LLMs for not having perfect judgement, even though their creators are extremely literal about how they can make mistakes.
All I'm really saying is that if you want to try to automate a travel agency, LLMs ain't gonna get it done. They'll happily book you a really unsafe trip. So the technology doesn't work in this domain. The whole, empty promise is that this thing is supposed to automate jobs like travel agent away. But it can't. This isn't a "pro" or "anti" position, it's simply that there's no market for the technology here. Or anywhere else (like radiology) where actual responsibility and judgement is important. In fact, I can't think of a single job where it's optional.
Or pay a guard a fair wage and comp them the $0.20 or whatever for each bullet..
EDIT: to be clear I'm not saying it should be that way, but there was a time not long ago when this was the normal way to handle the situation. I'd argue the present arrangement is more civilized.
> Or pay a guard a fair wage and comp them the $0.20 or whatever for each bullet.
I think $0.20 per bullet is far too little, considering the medical expenses the guard will face when getting the bullets removed after they are shot for copper.
Those +P 185gr .45ACP hollow points with the nickel plated cases are pretty sweet. But it's a hell of a lot better to not have to do some kind of Old West standoff shootout shit whenever someone wants to rob something. That was my point. It's so much better instead of hiring a posse of gunfighters, to just trust that the system will (barring anomalies like this one) unfuck itself.
Man that's awesome. That's motivation to get one of my vehicles in good enough shape for a day trip (I'm about 3-4hr north depending on choice of route). I have fond memories of Worcester, and of all the places I've been since I'd be most surprised if it's changed much :p
reply