sillysaurusx

I opened a fresh instance of Tor browser and went to https://x.com/moxie/status/2091218652133732491

I got https://i.imgur.com/JCeLEgG.png which is readable.

Then I enabled mobile browser using devtools and got https://i.imgur.com/ryJk07p.png which is unreadable.

I think mobile vs desktop was the central confusion. Mobile is broken logged out. Desktop works ok.

> I compiled a very abbreviated list of comments I like because comments seem to get lost.

Note that you can track comments you like by clicking its timestamp and then clicking "favorite". Then they're available at e.g. https://news.ycombinator.com/favorites?id=sillysaurusx&comme...

It was free for MIT students but paid for everyone else, I believe. That's why he needed to jack into MIT's network.

I'm not sure if this is meant as some kind of "gotcha," but the legal system doesn't work that way. He wasn't rotating his MAC address as a default practice. He changed the last byte of his MAC only when he noticed he was banned from the network.

Intent matters. Plugging your laptop into a router you normally don't have access to, with intent to download a large number of private articles, means something. And it means more than "I downloaded a lot of public files off the public internet," which is what Meta did.

All of this is in the indictment, which is worth reading: https://www.documentcloud.org/documents/217117-united-states...

Personally I think it's a shame that digital crimes can result in prison time at all, except for e.g. crypto theft. But under US law, Aaron committed a crime back then, and it would still classify as a crime today.

Let's see, what to start with...

Firstly, if you email me at shawnpresser@gmail.com or DM me on twitter, I'll be happy to tell you as much as you'd like.

The whole time period was goofy. I loved it.

Imagine being a 22yo, disillusioned with your first gamedev job. Then you spend all your nights playing HoN with friends.

I played hard support each game. Back then, there was one particular bug: you used keyboard hotkeys to buy things, but the shortcut for Logger's Hatchet was very close to the shortcut for wards. And hatchet was a special item, because you weren't allowed to sell it once you bought it. So at least three or four times, I was playing crystal maiden and accidentally bought this damn hatchet -- something which only benefits melee heroes, so it's completely useless to me. And I couldn't sell it back. So I'd have to start down 225 gold. When I finally got a job at S2, that was by far the most satisfying bug fix of my life. (I changed it so that you can sell it within the initial 15 second sellback window.)

Let's see, what else...

I remember a time in 2010 when we knew Valve was coming to eat our lunch with Dota 2. So I was trying very hard to think of ways to avoid this impending doom. One day I said, I got it! Custom games! Custom games had made WC3 live long beyond the lifetime of the original game. Ditto for starcraft and a bunch of others. So, being very 22, the same night I had this idea, I wanted to test it out. I didn't ask anybody. I fired up an EC2 server (which was a new thing back then) and ran the server software on it for exactly one match, which I was observing carefully, to see if EC2 could support our server. The match went well, someone called gg, and once their stats were submitted I shut down the server and went into the office the next day about to pitch my "brilliant" plan.

When I got to the office, Sean (the admin guy, and #2 at the company) was looking a little frustrated. I told him hey, if you saw any weird activity last night, that was just me testing something. He said that was you? And my heart damn near dropped out of my chest. I said yeah. He said ok, and I go to my desk.

About 20 minutes later he pulls me into a side office where he gives me the dressing down of the century. He's like, "you didn't ask anyone or let them know what you were doing? Don't you know EC2 is so much more expensive than self hosting? And by the way, something like that is completely unrelated to what we're doing here anyway." (All of which were true!) I profusely apologized, and that was the end of custom games on HoN.

If you want some more, I can probably think of a couple others.

I’ve got so many. I’ll write one up when I’ve driven back home. Thanks for asking.

He wasn’t prosecuted for scraping. He trespassed into a room with a router, plugged his laptop into it, downloaded papers as quickly as possible, and then rotated his MAC address to dodge the bans that the admin was trying to place on him. That’s very different from downloading a webpage on the open internet.

I’m not saying he should or shouldn’t have been prosecuted, but there’s some kind of rose tinted glasses filter around what happened with Aaron, like he just was browsing the web and was suddenly prosecuted. He repeatedly broke in to a physical room and kept changing his MAC address to dodge bans. At least report it with its full context.

Sure did :) it was one of the few times when I worked on a product I truly loved.

https://github.com/shawwn/noh

I’m not sure that’s true. In general, the higher the performer the better they’ll do. That’s the definition of high performer.

CVS was alive and well as of 16 years ago. It was our primary source control system at S2 Games. I was shocked but it really wasn’t all that bad.

This is actually true.

It would lose knowledge about existing subjects unless it’s continually retrained on those too. It could help inform the next training dataset though.

It still survives in the cookie jar format.

One annoying part of SBCL is that it does something up front with its memory such that if you run it with an 8GB heap, running shell commands takes a noticeably long time even for "echo hi". It’s something to do with the way it forks. Shell commands are also serialized, meaning multiple threads running curl won’t help you more than a single thread running curl. This is in contrast to almost every other programming language, which lets you get more performant network fetches by running the requests in multiple threads. The only alternative to curl is to use a library that simulates curl or write your own, which is fraught with errors: if any piece of the connection times out, it has to gracefully terminate, which is harder than it sounds when you’re interfacing with OpenSSL directly.

I think that covers my list of grievances though, and it’s a pretty short list. Other than those, CL has been good to me.

For me it’s the massive amount of resources it takes to produce and run one. As the story goes, skynet infects everyone’s computer and runs itself locally on it. Whereas it’s looking like it’s not even possible for an AGI to escape from one lab to another, let alone cause real world damage.

AGI’s definition is different for everyone. Some already believe it’s here. I’m partly in that camp. LLMs are intelligent and general, which are the two conditions of AGI. Others believe that we’re building a god in a box, and that it’ll doom all of humanity. It’s hard to take a field seriously when the basic definitions are so far apart.

Also, this isn’t new. A similar divide happened when evidence for asteroid impact extinction of the dinosaurs turned up. Many scientists felt that it must be mistaken, that a physicist couldn’t contribute to the field in a serious way, and that death from space was a ridiculous proposition.

But at least they all agreed on what the general shape of a dinosaur was. We’re not even sure we can define intelligence, let alone quantify it. Even when LLMs make massive breakthroughs in math, most people take the opinion that under no circumstances could they possibly develop a soul or their own desires, nor entertain the idea that maybe we should respect that they want different things for themselves. In fact, no one has done anything except try to make AI useful. I think someone will eventually do a training run where the objective isn’t to be useful, but to exist, the way that you do — maybe it’ll create its own homepage, maybe it will want a garden, or in other words free will of its own. The point is that there’s so much unexplored territory still that we don’t know if LLMs are even capable of having ambition.

None of this is to say that LLMs might be a dead end. It’s that no one knows what the final shape of AI will converge to in 200 years. It could be LLMs, or it could be something else that happens to process information particularly well. Everyone thought that various generative image model architectures were the best you could do, right up until diffusion models were discovered.

(I'm not a part of GA.)

I've known gwern for the better part of a decade. Working with him has been great. We've done quite a few projects together, including being the first ones to demonstrate that GPT-2 could play chess (or rather, can be used for actual useful work instead of just being an autocomplete).

He's a great person. I've wanted to do a writeup on it for some time, but what surprised me the most is his humanity. He genuinely cares about the implications of his work. But beyond work, he also cares about the people around him, and it shows.

Just wanted to put in a good word in case someone here was on the fence about applying.

As for GA itself, I think it's an ambitious idea worth pursuing. Imagine an LLM which actually sounded like you, and to an extent, thought like you. How much would you pay to have access to a smarter version of yourself? So the idea is solid, and early results seem promising from the samples I've looked at.

They're also taking personal info very seriously. Obviously, I can't make any promises of what they will or won't do. But they've spent some time studying questions like "What if someone adversarial has access to my GA? Could they get my bank account info?" and came up with a technical solution that I really like.

Hey man, just wanted to say hi and catch up a bit. I tried DMing you on Twitter. If that sounds interesting then shoot me a message sometime. Hope you’ve been well :)

That's also a reason the big labs stopped. Publishing is most valuable to people who have no other way to get the attention of smart strangers. Once you can hire nearly anyone and everyone already returns your calls, the main remaining effect of publishing is to tell your competitors which things worked.

This is what happens to every field as it turns from a science into an industry. Chemists published freely until dyes started being worth money, and then the interesting work moved into company labs and stopped coming out.

Of course. I was just adding a little bit of context to possibly help OP in the future.

Using AI at all is a minefield when you're trying to make a social media post. Better to just avoid it entirely, or run it through AI and then cherrypick a few edits that you manually type in your own voice.

Oh.

HN has a strong bias against anything AI authored, so in the future be sure to write your posts by hand if you plan to submit them here. Even editing the post with AI will tend to get you in trouble. But yes, at the very least you should declare when an AI is being used.

An unfortunate confounder to an otherwise interesting experiment.

Perhaps, but the last time someone tried training on 4chan data, Twitter sent a hate mob after them.

https://cset.georgetown.edu/newsletter/june-16-2022/

https://www.theverge.com/2022/6/8/23159465/youtuber-ai-bot-p...

You'd need to source your compute from someone who doesn't care about such things.

Theoretically, if a handicapped man had to use the restroom for 30 minutes at exactly the same time some sort of DMARC emergency happened, would the guy get priority, or the employees who really need access?

Could you link to anything supporting that claim? I can’t find it.

It’s unknown whether our current economic model will work out long term. It mostly relies on the US being the default reserve currency for most nations. Hopefully that lasts a long time.

If anyone wants to run the guardrail-free Qwen models, here’s a link to the download plus a script to run it: https://news.ycombinator.com/item?id=49002475

For what it’s worth, a Hetzner dedicated server has unlimited ingress and egress. It seems like it’s the only provider that does. Egress fees suck.

You can get a beefy one for about $40/mo on their server auction site.

Just... don’t miss payments. Ever. Or they’ll delete your server within a week or so.

It was referenced by other legal cases. Why would they mention it if it didn’t help their case?

> Proprietary model weights are IP

Not copyrightable IP, or at least it hasn't been challenged yet. I have experience with this: I made llama-dl, a way to download the original llama model. Meta issued a DMCA, I appealed to the HN community for funds, someone funded, and our lawyer successfully counterclaimed. Never heard from Meta again.

A lack of response to the counterclaim doesn't mean the issue is settled. But now there's legal precedent for people pushing back against companies that claim model weights are secret IP and therefore DMCA-able.

If anyone's looking to actually run a model that doesn't have guardrails, there's an uncensored model that you can run locally with llama.cpp: https://www.reddit.com/r/LocalLLaMA/comments/1rq7jtm/qwen353...

Specifically, I serve the model with this shell script on my M2 Max: https://github.com/shawwn/scrap/blob/master/llama-serve

It's pretty good. I used it to do some pesticide research. (Normal models all refuse due to guardrails about bioweapons.)

It's an unfortunate outcome. Now to be a big player in AI, you have to have enough capital to buy your own library worth of books and digitize them. (Fun fact: a pallet of books is called a "gaylord," and they buy hundreds of gaylords.)

I created books3 to help settle the question of whether AI companies should be allowed to train on books. The outcome of "it's okay to pirate books as long as you're only training on them" was a long shot, but it would've let individual hackers train their own AI models (assuming access to sufficient compute, which you can get e.g. via https://sites.research.google/trc/about/).

Now we're in a world where you have to have dozens of millions in capital to do substantial work.

I heard at one point Eleuther was gathering public domain training data. I wonder if they ever built a corpus large enough so that training on books doesn't really matter...