2,942 karma · joined August 25, 2010
http://github.com/laughinghan
The author is likely unaware of Common Lisp's condition system because that's a completely random and arbitrary conceptual ancestor. It was predated by throw/callcc in Standard ML in particular, I believe (which, needless to say, draws a direct line to OCaml and thus this post). In fact, continuations in some form or another date back to 1964 according to: https://en.wikipedia.org/wiki/Continuation#History
> due to its elliptical orbit with varying eccentricity, the instantaneous distance varies with monthly periodicity. Furthermore, the distance is perturbed by the gravitational effects of various astronomical bodies – most significantly the Sun and less so Jupiter
https://en.wikipedia.org/wiki/Lunar_distance_(astronomy)#Per...
However, we are not getting impressively close to AGI. That's why we need to stop the AGI alarmism and get our act together on the enormous societal ramifications that machine learning is already having.
He didn't have internal combustion engines, but that's a practicality, other mechanical power sources already existed (Alexander the Great had torsion siege engines). They would never be sufficient for flight, of course, but the principle was understood.
But he could never have even begun to build airfoils, because he didn't have even an inkling of proto-aerodynamics. He saw that birds exist, so he drew a machine with wings that flapped. Look at the wings he drew: https://www.leonardodavinci.net/flyingmachine.jsp
That's an imitation of birds with no understanding behind it. That's the state of strong AI today: we see that humans exist, so we create imitations of human brains, with no understanding behind them.
That lead to machine learning, and after 40 years of research we figured out that if you feed it terabytes of training data, it can actually be "unreasonably effective", which is impressive! How many pictures of giraffes did you have to see before you could instantly recognize them, though? One, probably? Human cognition is clearly qualitatively different.
The danger of machine learning is not that it could lead to strong AI. It's that it is already leading to pervasive surveillance and misinformation. (idlewords is pretty critical of OpenAI, but I actually credit OpenAI with taking this quite seriously, unlike MIRI.)
Once we had Newton's law of gravity though, we knew the distance, radius, mass, and even surface gravity of the moon. Would you say it's fair to say that by then we knew in principle we could go there and walk there?
(P.S. I assume you know this but the way you wrote your comment makes it seem like our measurements of lunar distance are nearly as inaccurate as Hipparchus's, when we actually know it down to the millimeter (thanks to retroreflectors placed by Apollo, actually). The wide variation from 55x to 64x Earth's radius is because it changes over the course of the moon's orbit, due to [edit: primarily its elliptical orbit, and only secondarily] the Sun and Jupiter's gravity.)
Conversely, if it turns out that everyone loved Racket 1 not because of the design elements that the maintainer put so much blood, sweat, and tears into, but because of the high barrier of entry due to the parentheses (the one thing that the maintainer didn't put effort into), how is that at all motivating to the maintainer?
I'm less of a community member and more of an admirer of Racket from a distance (and I don't mind parentheses, really, I've used Clojure professionally and it was great), but I look forward to seeing how this initiative turns out.
I am of course not saying you're wrong that "we don't know". We obviously don't know. It's possible, just like it's possible that we could discover cheap free energy (fusion?) tomorrow and then be in a post-scarcity utopia. But that's worth taking about as seriously as the possibility that we'll discover AGI tomorrow and be in a Terminator dystopia, or also a post-scarcity utopia.
More importantly, it's a distraction from the very real, well-past-imminent problems that existing dumb AI has, such as the surveillance economy and misinformation. OpenAI, to their credit, does a good job of taking these existing problems quite seriously. They draw a strong contrast to MIRI's AI alarmism.
Have you ever read idlewords? Best writing I know of on this subject: https://idlewords.com/talks/superintelligence.htm
We haven't even had the AGI equivalent of the Rutherford model of the atom yet: what's the definition of consciousness? What is even the definition of intelligence?
But the fact that Fermi was doing such a calculation in the first place proves that we knew in principle how a fission weapon could work, even if we didn't know "how far off [they] were". As soon as we figured out the moon was just a rock 240,000 miles away, we knew in principle we could go there, even if we didn't know how far off that would be.
By contrast, we don't know what consciousness or intelligence even is. A child could define what walking on the moon is, and Fermi was able to define a self-sustaining nuclear reaction as soon as he learned what nuclear reactions were. What even is the definition of consciousness?
Did you read the article? The point of this new platform is contextual ads, instead of personally-targeted ads.
The structure is what matters. Kings and despots throughout history have frequently ruled fairly and justly. Out of all the decisions they made in that role, they probably made many more fair ones than unjust ones. But their making the right call in any particular case is not a reason they should have had that much power in the first place.
Western social media is stepping up moderation of harassment and calls for violence, but they're making an effort to do so in an accountable and transparent way while doing so. And I don't think even they are accountable enough; I hope that Instagram and WhatsApp really do get split off from Facebook.
> The firewall and grey legislation around vpns is calibrated that low information audience gravitates towards party line while those seeking information has access.
Am I understanding correctly that your position is that this is a good thing, or at least fine? Because this placates the masses which tend more towards violence or other destabilizing actions, while allowing elites to continue to have a robust debate? Is that an accurate characterization of your position?
> The numbers suggests more Chinese people are aware of western and Chinese perspectives than vice versa. Leading me to ask, in general, who is actually more informed?
I'm not worried about American awareness of Chinese perspectives. I'm worried about Americans' missing or inaccurate understanding of American history, e.g. the Lost Cause.
I'm not worried about Chinese awareness of American perspectives. I'm worried about Chinese suppressed or sanitized understanding of Chinese modern history, e.g. June 4th.
I don't give a fuck about Chinese people being able to read Americans' sanctimonious takes on Tiananmen. I want Chinese people to be able to read Liu Xiaobo's take on June 4th, on Charter 08, on 996.
The most notable repo involved is "requests", an HTTP library more widely used than the one in Python's standard library (I'm under the impression it has become bundled with the Python standard install but am unable to find a citation for that).
How about changing the title too "Python requests library and other repos seeking new home"?
???
The West recognizes what? The West now endorses Chinese Internet censorship? Western governments now also enforce Internet censorship? Did the White House start calling to Build The Firewall? What are you talking about?
> Access is possible and not even inconvenient.
The Chinese government spends billions of dollars on a system that doesn't work?
> Someone posted a western brand awareness and usage article yesterday, where 17% of Chinese internet users regularly use facebook and are aware of western perspectives.
Who cares about Chinese internet users' fucking Western brand awareness? I care about Chinese people knowing our own history, warts and all.
I'm trying to understand your position. Is it that having ethical reservations about UN resolutions is childish, and the mature thing is to treat UN resolutions as ethically authoritative? Is it that resistance to treating Taiwan as a Chinese possession is childish? Something else entirely?
Who fucking cares? The PATRIOT Act was in response to the worst terrorist attack on US soil in US history. Three thousand people. I have friends who were in New York then. The Japanese internment camps were in response to Pearl Harbor. If an action is wrong, the impetus doesn't fucking matter, the action is still wrong.
6-4 (Tiananmen) was not ethnic violence, do we agree on that? And the Chinese government does use internet censorship to suppress discussion of it, are you going to deny that?
You are also incorrect that AA / United / Delta removed references to "China" for Chinese cities. I just tried them, Beijing, Shanghai etc all continue to mention China. Only Taipei's autocomplete does not include a country name. Again, I'm in California, so this contradicts your claim that: "Airlines only remove Taiwanese paraphernalia when flights operates in China".
However, I don't think this question is very fruitful:
> let's pretend Cloudflare just got knocked out globally by a wizard spell, what do we do?
The way you solve a production issue is you identify its cause and then contain, mitigate, or fix it. I don't think you'd learn anything useful from a drill where there's no specific cause.
Perhaps along similar lines to what you're thinking of, something I could see being useful is to look at components that you've already thought to implement a 'global kill' for, like WAF, for instance. Maybe you could run drills where every machine running WAF starts blackholing packets, or maxing out RAM, or (as happened here) maxing out CPU, the kind of thing where you'd want to execute the 'global kill' in the first place. That way, you can ensure that the 'global kill' switches are actually useful in practice. Something like that seems more grounded to me, making the assumption that something specific is going wrong and not just "magic", while still avoiding too-specific assumptions about what can and can't go wrong.
For the record, many of your comments here have been thoughtful, and I've upvoted them. I've also downvoted many where instead of responding to other people's thoughtful comments, you just insult them instead. Those are also the ones that other people seem to be downvoting. I don't think anyone is shooting the messenger here.
The problem is that the individual bears a very small risk of something very bad happening: "consider the hypothetical case of a gay blogger in Moscow who opens a LiveJournal account in 2004, to keep a private diary. In 2007 LiveJournal is sold to a Russian company, and a few years later—to everyone's surpise—homophobia is elevated to state ideology. Now that blogger has to live with a dark pit of fear in his stomach."
https://idlewords.com/talks/haunted_by_data.htm
The individual of course gets no benefit from the small chance the company can monetize on this data trove. So even though chances are they aren't harmed at all by this data collection, arguably the expected value of the benefit/harm to the individual is negative (harmful). But that doesn't change the data collector's calculation, of course. That's why government regulation is necessary.
(This is a genuine question. Many of your comments have added to the discussion and I've upvoted them. But I've also downvoted many that haven't.)
What? Sure there is!
1. The reason to believe native apps' sandboxing won't exceed the browser is that any sandboxing that works on native apps would also work on the browser app itself.
2. There's also 2 reasons to believe native apps' sandboxing may always be inferior to the browser:
(a) The Web has wider reach, and people are already more confident/careless visiting strange websites than downloading and running strange apps, so exploits targeting the Web are more valuable and therefore more resources are spent battle-testing it.
(b) Native apps currently have deeper access to the device which makes it easier for them to do bad things, and (similar to reason 1) will never have less access to the device than the browser app which is also an app.
(I'm aware there's arguably a slight exception here about Mobile Safari and W^X, but I don't think that disproves the overarching reasoning.)
> almost universally, the web version is safer to use. Nobody should be installing Facebook, Twitter, or Reddit on their phone.
Not only is this true, this would be even more true without JavaScript—if those sites were still usable, which they definitely could be, they just choose not to be. (Well, maybe except Facebook Live, but that could be an optional standalone app.)
> you can already do that on the web today. You can already turn off Javascript.
Of course, this isn't really true, precisely because so many websites that could function fine without JS (including things like news sites that should just be static content!) instead choose not to.
Which of course is the real problem with yellowapple's idea. Lots of services cripple their mobile website and push you to install their app instead; if we removed JS from the Web, everyone who could would just start doing the same on desktop too, right? Upstarts trying to maximize growth probably will work great on the Web, but as they get more established they'll start pushing people more and more towards their native apps, and existing established players will do that from Day 1 (of the new, JS-less world), including everyone mentioned so far—Facebook, Twitter, Reddit, GitHub, major news sites, because people will deal with the one-time friction of installing the app in order to access the network or content.
That way, the page displays correctly for you, but the server has no idea your actual fingerprint.
There's some trickiness to get this to work right; the collection of fake fingerprints would have to have a certain amount of persistence, because if it was regenerated every pageload, the server could probably tell that only one fingerprint kept showing up repeatedly. Maybe each fake fingerprint should have a completely realistic-seeming browsing session, happening in parallel with your real one, with half the collection continuing on browsing even after you're done? Except wait, ads could just separately target every fingerprint, and it doesn't matter if 99% of them are fake as long as its accuracy for your real one is still good. To defeat that you need the randomized activity using your real fingerprint.
The ideal would be if this was done through a proxy server, which would then know every fingerprint ever sent to a website. It could then provide you with a random collection of past fingerprints that have actually visited the same website, so every visitor gets a collection of fingerprints randomly drawn from the same "bag", rendering visitors indistinguishable.