astine

Joined 18 years ago

Karma
5,379
Submissions
1,948
Member since
Mar 18
Employed bum.

You can find out more about me here: http://www.theatticlight.net/

topcolor=8888ff

View on Hacker News

Recent submissions

Comment · 18 days ago
I don't see any talking about agents "choosing" to go rogue, only agents actually going rogue. That is, behaving in a way different from what is expected or desired. Non-AI based software goes rogue in this sense all the time as well. It's just that AI-based systems seem to have the tendency to go rogue in more spectacular ways.

That said, software systems "choose" all the time anyway. They don't need to have some sort of anthropomorphic "will" to do this. They just need to be given the capacity to arbitrate between a number of options, even if that process is one hundred percent predetermined by the programmer ahead of time. There's no need to appeal to libertarian notions of free-will.

Comment · 19 days ago
It may be the case that companies are liable for anything their products do, but that doesn't mean that they can't go rogue. It just means that companies are liable when the AI models do so.

Not that I think we should be credulous about stories where they go rogue.

Comment · 22 days ago
Comment · 29 days ago
> Disabling them is a means to deal economic damage, not an end-goal in itself.

Russia's goal in attacking Ukraine's energy sector is to make the country ungovernable. They attack the energy sector in winter because they know that causes suffering and hope that will cause the populace to turn against their government or at least against the war. It's not just a matter of causing economic damage, turning off the heat is a means to an end in ending the war in their favor.

Comment · 1 year ago
If you think it's silly because it's not a restaurant you go to, imagine if Coca-Cola replaced their script logo with some generic sans-serif one. Don't you think the outrage would be real?

I can't imagine being upset at something like that. I'm sure there would be people upset, given the nonsense that happened in the 80s, but being personally invested in corporate branding has got to be the saddest sort of parasocial relationship possible.

Comment · 1 year ago
I definitely saw some. It came from people who were in that culture war space where they interpret every trend as being part of a big left-wing plot to impose their values on everybody else.
Comment · 1 year ago
So why not just think in terms of lists, which can be nested, rather than cons-cells, which are only rarely used for anything other than to construct lists?

Because, as he continues:

Worse, any proper list can have improper list as elements. So, you can have a list of cons, or cons of lists, cons of cons, list of lists, or any mix. The overall effect of the cons is that it prevents lisp to have a uniform high level treatment of lists, with the result that development of functions that work on tree are inconsistent and few.

In other words, the fact that the implementation details for lists are so exposed means that you have to be careful when interacting with 'lists' that turn out not to be actual lists. There is no type information, either at compile or runtime, that ensures that what you're dealing with is actually a list and not something else. So you can't _actually_ think in terms of lists; you have to think in terms of cons cells which are probably lists but might not actually be so.

Comment · 1 year ago
You'll have an equivalent of a cons cell in any linked list implementation, but you won't necessarily interact with it as a user. Look at the standard library implementation of a linked list in C# for example (https://learn.microsoft.com/en-us/dotnet/api/system.collecti...). You see methods for looping over a list, for accessing elements, for insertions and removals, but not much for directly accessing a list's nodes and manually constructing a list out of those nodes. That part is abstracted away.
Comment · 1 year ago
I think that the author's complaint is that cons cells are too low level a construct to be exposed so prominently to the user. I think he would prefer that it was abstracted over with some higher level interface the way a lot of other language do it. He mentions Clojure as a Lisp that gets it right in his opinion and that's exactly what Clojure does. Cons cells are still available as a construct, but the default method for interacting with lists is a set of higher level functions that also work on arrays and other sequences and don't require that you think as much about the implementation of the lists.
Comment · 1 year ago
Xah's complaint appears to be that cons cells are too low level a construct to expose to the user. He seems fine with lists in general but just dislikes how they are implemented in Lisp. If I were to implement a list in a non-Lisp language (like C or whatever,) it would likely internally include a struct called "Node" that filled the same purpose as cons cells do in Lisp. However, users of this hypothetical 'list' usually wouldn't directly interact with these "Nodes"; they would mostly be an implementation detail, as appears to be the case for most list implementations other than Lisp's. I think that Xah is wishing that was the case for Lisp as well.
Comment · 1 year ago
Right. Presumably a car idling for ten minutes produces less pollution than a car being driven for ten minutes, but a car that is driven for ten minutes and idled for an additional ten produces more pollution than either of them. Any pollution produced by cars idling in bad traffic is superadded to the pollution produced in transit so improving the flow of traffic should reduce pollution even if the total number of cars remains steady.
Comment · 1 year ago
Correlation may not imply causation, but where there is causation there usually is correlation. So, if there is a rise in homelessness there is likely some factor contributing to that and if a factor is correlated there is a good chance that it's causal. Homelessness may caused by multiple factors but if it's increasing and those factors are mostly remaining static then looking for the one that is also increasing seems like a good bet.
Comment · 1 year ago
a camp fire gets plenty hot for blacksmithing - just wait for the coals and then blow on them

Maybe if you're working with bronze or copper, but iron forging requires much higher temperatures than a campfire can provide. That's why the iron age took place after the bronze age, forges capable of making iron workable were not yet invented. It wasn't a trivial invention.

Comment · 1 year ago
> I also think the gruellingly slow death of legacy media and rise of bluesky and X (and mastodon) is a net positive for society, if only for the reason that ~tweets can be immediately and transparently rebutted, whereas brainwashing ‘news’ programs can’t.

The problem with this logic is that for the most part, new media isn't replacing legacy media; it's simply placing new layer of filtering in front of it. The vast majority of people sharing information on these platforms aren't journalists doing their own research. Instead, they're getting their information from journalists and just applying their own filtering and spin. "Rebuting" usually just involves linking to different news sources. You were always better just reading the legacy media in the first place.

Comment · 1 year ago
Presumably because he owns the site. You would think this would be the one place that PG didn't get much pushback on his opinions. I'm not too surprised though; he hasn't been very involved here for years so the culture has shifted.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton