I generally agree but I would have appreciated a more-direct response vis-a-vis Yandex.
In re: your broader point I don’t think they are “that smart” per se; or perhaps better put, I think they are smart enough to be aware their platforms are used by all kinds of agents for all sorts of purposes 24/7/365...but up until now also deemed it prudent to turn a rather universal blind eye towards those activities unless they were, say, so obviously fake that ordinary users could identify them as such.
Whether that universal blind eye was smart is tbd, but nothing in their response since then has seemed very clever.
Can you clarify something? You seem to be arguing two points without differentiating them clearly.
One point is that it should be obvious where the money came from (etc); the other point is that that origin should not be cause for concern.
Is that accurate?
If so I would agree with you on the first point but not the second: the same logic that eg allows one to make the inferences in the first point would also lead one to infer certain things about Yandex (analogous to the situation with Kaspersky).
Do you honestly think that the funding under discussion—and the personal relationships that subsequently developed—had no impact on either Twitter or Facebook deciding to establish their respective partnerships with Yandex?
Fair enough but on a site inhabited by intelligent, skeptical people it’s hard to follow those rules when the topic has become “was pg being disingenuous, willfully blind, or simply naive on that day, 6 years ago, when he announced partnership with his mysterious Russian money man?”
Without making any specific accusation I would simply point out that the user with whom you have been conversing is present in [1] and playing the “it’s no big deal” card even then.
In the Information Age most people with an interest in an issue will have already heard “your” opinion on it, because most people have off-the-shelf opinions that have already been widely circulated.
Until etiquette and discourse norms evolve to incorporate this understanding conversations will be frustrating for everyone!
His earlier post (linked from this) has stronger reasons.
I think it’s advisable to keep an open mind on the issue, and not simply on the actual question itself; it would not be surprising to me if the solution comes from something like adopting a different framework.
For example, time bounds and complexity classes smell a lot like conservation laws: “transforming such and such arrangement into such and such arrangement requires at least this much computation.”
That said, it’s possible (a) we’re currently failing to consider some “term”, (b) generally ignoring this term doesn’t cause problems when proving lower bounds but (c) the complex border and your “sparse touches” correspond to situations where the missing term plays a more significant role.
That’s a case where P probably isn’t equal to NP but keeping an open mind at least leads in more interesting directions, imho.
I also think people don’t take seriously the possibility of P being equal to NP but with intrinsically high degree. I say this not to be cute—“what if p is np but still de-facto intractable?”—but because I don’t think anyone has a great intuition for, say, what kinds of algorithms have polynomial solutions of minimum degree, say, 8...at least not in the same way we have good intuition for which algorithms are linear, nlogn, n^2, n^3, and so on.
It’s hard for me, at least, to feel overly confident in the “we’ve been working on finding a fast algorithm for seventy years and gotten nowhere” when our algorithmic intuition vis-a-vis higher polynomial degree seems so under-developed.
Even if you don’t consider it likely that p equals np, you can still follow this line of thought and consider the possibility that these “sparse touches” may be cases where the slippery problems like graph isomorphism (etc) correspond to problems with polynomial running times of (unexpectedly) high degree...and thus we keep finding these sparse touches along a seemingly-complex border because we don’t yet have a solid intuition for the capabilities of polynomial algorithms of high degree.
And so on and so forth. P probably isn’t NP but being dogmatic about it is neither fun nor interesting.
Or, you know, a utility-maximizing actor found it utility-increasing to place certain words into certain sequences in certain contexts due to the expected effect said arrangement would have upon the expected audience?
Silicon Valley will eventually be seen as yet another textbook case of a new industry getting away with externalization of costs initially unrecognized as such.
Enjoy it while it lasts, I guess, but don’t expect to escape liability for the foreseeable consequences of the systems you build for all perpetuity.
There is nothing defective about your NZ citizenship...and why would there be, given that it was acquired in a scrupulous, perfectly above-board way?
Don’t take a chance on that seductive St Kitts!