andreasvc

It is strange that you characterize Cython as a limited sub-language. Cython supports almost the full Python language, plus adding a superset for producing native C code. In addition, Cython can be used to wrap C code, so what you describe as "exportpy" is possible as well. Cython is a mature project, being in use in various parts of the scientific python ecosystems such as in Pandas and scikit-learn. Nim sure looks interesting but from what I hear, the language and the compiler still have rough edges. In terms of stability, maturity, and ecosystem, I'd say Cython has a clear advantage.

There's always something worse happenning somewhere. That's not a reason not to care about things like this.

The popularity of Searle's Chinese room argument has always bothered me. It's not clear what he actually believes in. So the mind is not a computer, what is it then? Something with "special causal powers"? What does that mean... And presupposing a rule book that passes the Turing test is of course a very dubious premise. I think arm chair philosophizing is ultimately cannot give use clear answers to questions like this.

The idea is that you can only access the pictures if everyone of the group is physically present. Sort of how multiple people need to turn a key at the same time to launch a nuclear missile.

Ah, I see your point.

You're not contradicting anything I said, I wouldn't claim utopias are the only reason for people to become violent. I do believe that true believers such as with religions or utopias are particularly dangerous. Did you reply this way because you approve of communism?

It wasn't equated. It was mentioned as an example of something that is practically impossible to eradicate. Accepting this is better than for example sacrificing all of our civil liberties in exchange for security theater.

There has been plenty of communist-inspired violent terrorism. Germany's RAF, Italy's Red Brigade, Japan's United Red Army, Peru's Shining Path, Columbia's FARC, Sri Lanka's Tamil Tigers, etc. etc. Saying these are not real communists is just a No True Scotsman. As soon as people start believing in utopias there's a danger that they will cross a line.

I would encourage anyone to use TLS. However, I take issue with using it as an indicator of the trustworthiness of the sender or its host.

If anything, a spammer or phiser has more motivation to deliver their mail over TLS because of this development, compared to some legitimate host. Whether the mail was delivered via TLS provides no information on its legitimacy. That is why it is problematic that it could give a false sense of security.

I don't understand how that follows. From all I've heard, people suspect that the difference between cheating and test mode is large.

What amnesty means is clearly defined, as you quote from the article. I don't see why the amnesty offered by a company would be so different from that offered by the government. A government could similarly make your life unpleasant, just as a company, and in neither case is it customary to exhaustively enumerate things. Ultimately it is based on trust, and governments have also been known to renege on their promises. I'm not saying this promise is trustworthy though, just that the fact that it is from a company doesn't disqualify it outright.

Proprietary software is equally if not more based on trust. Trust without being able to verify.

I don't get why you're so dismissive. Your ISP still sees that your IP connects to the IP of some server, at what time, for how long, how much data, etc. That's plenty of information.

Why especially in open source? The leap of faith is larger with a binary blob. (Although I admit it doesn't make much difference in practice, most people don't audit source code).

What good would undo do you if you didn't realize something bad happened?

All of P is in NP. The result claims that it is close to P, when before it was probably expected to be much harder than P.

Again, I could hand-wavingly claim that /dev/random "works". What does "one that works" mean? You may not care for philosophers, but making claims about morality is definitely within their territory, plus it's a philosophical claim in itself to state that philosophy is not relevant.

The argument about finite brain configurations is not compelling. There's no reason why only brain states that have occurred would be relevant. Morality deals with hypotheticals and counterfactuals: what would be the moral judgment in this or that situation. I think it would be uncontroversial to claim that there are an infinity of such hypothetical situations, and each model of morality can deal with them differently, ergo you get an infinity of possible models.

Look I can see that it worked out well for them in the end. What I'm specifically disputing is whether the vaporware-thing was a causal factor contributing positively. If they actually had the software already and didn't need to fool anyone, things might've worked out even better. In that case things worked out despite the vaporware, not because of.

Are we talking about the same thing? You have made a definition of morality that philosophers and other people will all agree on? Where do you get this "finite number of consistent models" from?

It is vaporware, the whole novelty of it is that it is an AI, which it is not. "How it works" is the whole point.

How does your N=1 anecdote show that vaporware makes a lot of sense? I'd say they got away with it, but not that it was a crucial factor in their success.

I am saying it's not a technological problem, it's a social problem.

Sure computers can make such decisions, be it with /dev/random or some sophisticated classifier. But we'd want justifications, and it's difficult to define what that entails. Moreover, what does it mean for a computer to be "better" at this than humans? The whole problem is ill defined.

While I agree, it should be noted that there has been serious controversy about the question whether language and language use can and should be characterized by rules, or whether it is based on some kind of emergent phenomenon or dynamical system (connectionism). The controversy is still not really over. Personally I don't think it's an either/or question, but about levels of description.

I've never heard vaporware characterized as something that "could make a lot of sense"; how could it?

Plus I'm doubtful whether the data would be very meaningful. A bunch of people adversarially trying to figure out whether the AI is real is not representative or generally useful data.

If you want to avoid giving information about whether you are an AI or human, you simply respond "No."

Besides, the CAPTCHA's that are sufficiently hard to solve for computers are already hard for humans as well.

Look you can describe anything with rules as long as they are specific enough. In the pathological case each word is its own rule, but English clearly has way more regularity than that, which explains why there is a seminal academic work describing English phonology with a set of rules.

Why is that? This was actually mathematicians doing "linguistics".

Uh that doesn't follow at all. Just means that the rules are a bit more complicated than "this letter is always pronounced that way". Neither does it mean that "all sequences of letters can be pronounced in any way".

Look up Chomsky & HalLe's the sound pattern of English for an attempt at completely describing English phonology with rules.