> The seniors today who have got to senior status by writing code manually will be different than seniors of tomorrow, who got to senior status using AI tools.

That’s putting it mildly. I think it’s going to be interesting to see what happens when an entire generation of software developers who’ve only ever known “just ask the LLM to do it” are unleashed on the world. I think these people will have close to no understanding of how computing works on a fundamental level. Sort of like the difference between Gen-X/millenial (and earlier) developers who grew up having to interact with computers primarily through CLIs (e.g., DOS), having to at least have some understanding of memory management, low-level programming, etc. versus the Gen-Z developers who’ve only ever known computers through extremely high level interfaces like iPads.

> I think it’s going to be interesting to see what happens when an entire generation of software developers who’ve only ever known “just ask the LLM to do it” are unleashed on the world.

we only have to look today at how different software quality is compared to the "old days" - when compilers were not as good, and people wrote in assembly by hand.

Old software were fast and optimized. Hand written assembly used minimal resources. Today, people write bloated electron webapps packaged into a bundle.

And yet, look who is surviving in the competitive land of software darwinian natural selection?

Casey Muratori says that every programmer should understand how computers work and if you don't understand how computers work you can't be a good programmer.

I might not be a good programmer but I've been a very productive one.

Someone who is good at writing code isn't always good at making money.

Problem with much of this talk is receipts are always nowhere to be found.

But I don't see any receipts from the opposite side either.

You don't see any good software made by people who know how computers work?

You don't see any good software made by people who don't know how CPUs, GPUs, networking work at a deep level?

AI slop books made more money than JK Rowling, too.

Maybe in the future, yea. Most likely not because creating books is much easier now but total reading time can't increase nearly as fast. More books chasing the same amount of reading time.

To be fair, I wouldn't be entirely surprised if they were better than barfs onto a page. She's not exactly Tolkien

How is caching implemented in this scenario? I find it unlikely that two developers are going to ask the same exact question, so at a minimum some work has to be done to figure out “someone’s asked this before, fetch the response out of the cache.” But then the problem is that most questions are peppered with specific context that has to be represented in the response, so there’s really no way to cache that.

From my understanding (which is poor at best), the cache is about the separate parts of the input context. Once the LLM read a file the content of that file is cached (i.e. some representation that the LLM creates for that specific file, but I really have no idea how that works). So the next time you bring either directly or indirectly that file into the context the LLM doesn't have to do a full pass, but pull its understanding/representation from the cache and uses that to answer your question/perform the task.

I built a new server this time last year. My board does 6 channel RAM so I bought 6x32GB ECC DDR5. $160 a stick at the time. Just for grins I looked up the same product number at the same supplier I originally bought from. $1300 apiece. One of the VMs running on that server is TrueNAS, with 4 20TB WD Red Pros. God help me if I have to replace a drive.

Best Buy is actively selling 2x8gb sticks of DDR4 3200 for $80 a stick. I was floored. Ten bucks a gig, $160 for the pack.

We're fucking doomed.

Ten bucks a gig is lower than what some DDR5 memory is selling at.

Perhaps there is an incentive to go back to OS that can operate with 640KB RAM ... /s

> Personally, I want people on the high end of earnings (such as myself) to be taxed more so that a basic income scheme like this can be available for anybody who wants it. Charge me an extra $300/month and give it to some random 24 year old so that he can smoke weed and play his guitar. He'll get more use out of it than I will.

You know you CAN donate money to the government any time you want, right? Do you do that? Practice what you preach, don't hide behind "oh if only the government made me do it."

> One day, that kid will decide that living in a crap shared apartment is getting a bit old and he'll grow some ambition, get a real job, and eventually start earning enough to help sponsor the next round of deadbeats.

This is the critical problem you and others like you make: assuming that everyone is a reasonable, honest, ambitious person just like you are. Many people -- not all, but a big enough proportion to be a problem -- aren't. And when we make it possible to actually make "do drugs and play videogames all day" a viable lifestyle, there's loads of people who will take the government up on the offer. And remember, they can vote themselves UBI raises.

> You know you CAN donate money to the government any time you want, right? Do you do that? Practice what you preach, don't hide behind "oh if only the government made me do it."

You know we can also advocate for higher taxes, given that it's astronomically more meaningful for everyone to give ten cents than for me to give a few dollars, right? Or did you think this was an insightful, valuable addition to the discussion that no one has ever suggested before? Is this the comment section of a local newspaper? Good god.

> there's loads of people who will take the government up on the offer.

Prove it. How many are loads? What, specifically, do you think UBI actually amounts to?

> You know we can also advocate for higher taxes, given that it's astronomically more meaningful for everyone to give ten cents than for me to give a few dollars, right?

And you can ALSO voluntarily pay more in taxes while doing so. It's called leading by example. Warren Buffet and Bill Gates get called out on this and they do the same thing; "oh I'm just one person, my extra tax money is but a drop in the ocean, so why bother." If you and everyone else saying "tax me harder" actually put up, it might amount to something! And at least it would make people respect your position a bit more.

> Prove it. How many are loads?

Well let's see, anyone who's lived their entire life on welfare (we have families of multiple generations who have done so at this point) would qualify. So would all the homeless people more content to live on the street and do drugs than go to rehab.

> What, specifically, do you think UBI actually amounts to?

Enough for people with low ambition to live on! And their vote counts just as much as the productive members of society paying for them.

> And you can ALSO voluntarily pay more in taxes while doing so.

You're just repeating the same fundamentally silly thing I argued against like saying it louder will make it somehow less childish and silly. Did you have an actual point here, or do you think "people's respect" has any slight value in the context of the discussion?

> Well let's see, anyone who's lived their entire life on welfare (we have families of multiple generations who have done so at this point) would qualify. So would all the homeless people more content to live on the street and do drugs than go to rehab.

So you have no actual clue, but you think hand-wavy bullshit that feels good will suffice for numbers. Fantastic. I'm glad you have such strong opinions on things you clearly know absolutely nothing about.

> Enough for people with low ambition to live on! And their vote counts just as much as the productive members of society paying for them.

Is "low ambition" angry-posting on social media? Do you think vibe-coding React bullshit, or whatever, is "productive"? Do you think engineers at Meta, busily finding new ways to make teen girls depressed are "productive"? Or is this just more stuff you have convinced yourself is true because it makes you somehow "better" than other people?

The problem I have with the way the EU doles out these punishments is that they like to spring them on tech companies after years and years of radio silence and then suddenly it’s “hey TikTok, we just determined you’ve been breaking the law for years, pay us a couple billion please.”

Like, where were they years ago saying “hey TikTok, we think your design is addictive and probably illegal, you need to change or face penalties.” If TikTok continues to operate in the same manner despite a warning, sure, throw the book at them. Otherwise it just seems like the EU waits for years and years until a company is a big enough player and then retroactively decides they’ve been breaking the law for years. Doesn’t help the impression that they’re running a non-EU tech company shakedown campaign.

Its never really like this.

Tiktok spend a lot of money talking to EU regulators. They know shits coming down the track because these directives have to be put into law by eu members. that takes time.

> Doesn’t help the impression that they’re running a non-EU tech company shakedown campaign.

But thats not the point, companies shouldn't be doing stuff they know is harmful. Thats literally the point of regulation.

> Like, where were they years ago saying “hey TikTok, we think your design is addictive and probably illegal, you need to change or face penalties.”

That is basically what happened today. No penalties have been issued at this point.

Also Commission had sent various requests for information to TikTok in 2023 before they opened these proceedings in early 2024 (https://ec.europa.eu/commission/presscorner/detail/en/ip_24_...) - this didn't come out of the blue.

> Otherwise it just seems like the EU waits for years and years until a company is a big enough player and then retroactively decides they’ve been breaking the law for years.

Lol. It's never like this.

These companies are given plenty of warnings and deadlines. After years and years of ignoring them these companies get slapped with a fine and start playing the victim.

BTW at this point DSA has been in effect for three years

Fines on US tech companies bring in more money to the EU than the EU's entire tech industry combined.

Haha you may be right.

But instead you may see it as a discount on the money these companies are making from European citizens.

You answered it yourself. They can't extract billions if the company is still small.

You’re getting downvoted but seriously, it took them this long to figure this out? I also suspect they won’t outright ban TikTok, but instead levy a multi-billion dollar fine and let it continue operating.

The fines are typically accompanied with a requirment to change the illegal behaviour.

Lots of Californians do not benefit from social programs but support them anyway.

Do they have a choice?

Yeah, they can vote against it. Most don't

A lot of people at $job, even ones who should know better, think they’re witnessing the rise of Skynet, seriously. It kind of makes the AI hype in general make a lot more sense. People just don’t understand how LLMs work and think they’re literal magic.

Skynet doesn't seem to require magic. I'm probably supposed to know better, but I'm a little concerned about it myself.

I'm imagining a strange future reality where "AI" that can't really innovate and shows no convincing signs of creativity still manages to take over large swaths of the world merely by executing basic playbooks really well using military tech previously provided by (now defunct) governments. Like a grey goo scenario except the robots aren't microscopic.

> You always need to ask. Having something always waiting in the background that can proactively take actions and get your attention

In order for this to be “safe” you’re gonna want to confirm what the agent is deciding needs to be done proactively. Do you feel like acknowledging prompts all the time? “Just authorize it to always do certain things without acknowledgement”, I’m sure you’re thinking. Do you feel comfortable allowing that, knowing what we know about it the non-deterministic nature of AI, prompt injection, etc.?

Another way to think about it:

Would you let the intern be in charge of this?

Probably not but it's also easy to see ways the intern could help -- finding and raising opportunities, reviewing codebases or roadmaps, reviewing all the recent prompts made by each department, creating monitoring tools for next time after the humans identify a pattern.

I don't have a dog in this fight and I kind of land in the middle. I very much am not letting these LLMs be the one with final responsibility over anything important but I see lots of ways to create "proactive"-like help beyond me writing and watching a prompt just-in-time.

> It’s in the interest of the model builders to solve your problem as efficiently as possible token-wise.

Unless you’re paying by the token.