rjtobin

Here's another one, seed text in italics. Clearly some Star Trek used in the training data.

"Engage", said Captain Picard from the bridge of the Enterprise. "But sir", Commander Data began, "this must have been part of something called the Borg."

"You know I didn't tell them that," said Picard, "I have a very hard time talking to them."

- Picard, Data, and Worf while the ship is under attack by Borg drones at Federation border

"The Borg don't know that they're being infected with something dangerous. And they're not too fond of it. There's a saying: 'Borgs eat Borges. Borges eat Borges.' They'll get tired of eating their own."

This is really a derivation of the Binet formula for Fibonacci numbers: that u_n = (phi^n - (-phi)^n) / sqrt(5), where phi is the golden ratio (1 + sqrt(5))/2.

Didn’t know that Allen had a big part in funding KEXP, in addition to all the well-known stuff (Seahawks etc). Crazy the impact one person can have on a city, even if that person is insanely wealthy.

I’m always hesitant to read a translated novel (especially literary fiction). Seems like too much would be lost in translation. Even putting language aside, without knowing the cultural context within China, would Lianke’s books be totally lost on me?

I’m guessing my view here is naive (and, in terms of the cultural context, a sort of self-fulfilling project). Any HNers care to educate me on this?

I get robocalls that leave messages about once a day. These days it's some message in Mandarin. A few months ago it was something about health insurance plans (same message every day).

> You should try it some time.

Worth noting that Patrick Collison was/is a Lisp user. While in high school, he won a fairly prestigious Irish national science-fair-type contest (the “Young Scientist”) with an AI bot written in some Lisp dialect.

IIRC he also contributed some patches to one of the major Lisp projects.

I don’t see why the “synchronized play” strategy described in your last paragraph also doesn’t violate the no signaling of any kind rule. If I have a Two but no One, then by consciously not playing on the first second of play I am signaling something. Am I missing something?

I visited some dorm rooms in Croatia once, and they had a nice solution to this problem. Both doors opened inward, with the non-hinge side of the doors sitting very close to one of the walls with no doors. Instead of a conventional lock, a narrow piece of wood ran the length of this no-door wall that could be rotated to block both doors simultaneously.

This is like the solution in the article, except deals with the "opens-inwards" case instead of the "open-outwards" case, and is less fumble-y than having to attach a rope to both doors.

Romansh, a mystery to everyone except Sean Connery.

One area is scientific computing: based on, for example, the Top500 list, or their recent successes with the Gordon Bell prize (basically best paper prize for the biggest supercomputing conference).

Of course there are a lot of political factors in both of those achievements. But money definitely translated into results.

(I'm not arguing that China is the best nation in this area either, but they are clearly a leader.)

One source of bias here: there are lots of small schools, chances are high that one of them will exhibit a large number of founders due to noise, and this will be the one picked up by the report cited.

Also by Stephen (and previously discussed on HN): https://www.cl.cam.ac.uk/~sd601/papers/mov.pdf

As usual, he injects just the right amount of humour. Eg. this footnote in his thesis:

"*This paper is out of print, in Russian, and quite difficult to get a hold of, so the interested reader is directed to a shorter proof by Conway [Con71, p. 105], which is merely out of print."

Once (also in JFK) I was quizzed a bit by the CBP officer. I told him I was a math grad student, he said "well, then tell me about the Euler problem". I explained that Euler was a fairly prolific guy, and asked if he could be more specific. He didn't relent, apparently he had seen some documentary that was all about the "Euler problem".

Eventually, we moved on to my background, and I mentioned I had done a masters in computational neuroscience. He said something like "oh neuroscience, my great aunt had that", I think he thought it was an illness? Was sort of expecting the reality TV cameras to be busted out at that point...

He let me through though! Usually I clear immigration in Ireland (one of the few places you can do the immigration before you leave), and those folks are always much more pleasant.

I thought so too, but they are after the references. Still I was disappointed that it wasn't a "proof without words" article.

Well one movie can be in many categories. And Netflix has some pretty diverse categories (e.g. "Movies featuring an epic Nicholas Cage breakdown", which appeared once for me).

Though I suspect some of the 76000 categories contain only movies that are no longer available.

Well, not Iran (thankfully)

Just in reply to your first paragraph, there is more than one type of well-paying programming job...

That's a fair question, and I have no hard data to back it up. Seems to me like it would be true (as another poster said, mostly number of edits, due to increased attention), but of course my intuition could be wrong.

Still, since it plausibly could be a factor it is something that should be ruled out.

I'd think that after a celebrity does, there would be an increase in the number of edits / length of an article. Wouldn't this skew the numbers, increasing the "celebrity" statistic for everyone that has died that year (and hence making the year seem more extraordinary than it is)?

It would be interesting to do the same analysis for 2015 and previous years, using wikipedia snapshots at the end of each year, and see if those years also appear extreme.

The PCIE NVIDIA P100 has peak memory bandwidth of 730GB/s, at 250 Watts, which is almost exactly the same bw/watt as the MI8 (well, there are two PCIE P100s, I mean the better one).

Well, looks like it got $103k in crowd funding, so it's not totally unsubstantiated.

Here are some performances with higher optimization parameters and with the Intel C Compiler. All run with a Haswell Xeon (E5-2698 v3). Using icc, for the larger problem sizes you get performance speedup that very similar to the speedup from the hard-coded asm.

With same arguments as blog post:

gcc -O1 -o fft-test fft-test.c fft-portable.c -lm:

...

   262144    min=15721397  mean=15748901  sd=0.12%
   524288    min=34474976  mean=34546878  sd=0.18%
  1048576    min=99814826  mean=100391926  sd=0.26%

...

Adding -march=native and -O2:

...

   262144    min=14668178  mean=14701671  sd=0.14%
   524288    min=31297081  mean=31395468  sd=0.14%
  1048576    min=94027867  mean=94196746  sd=0.13%

...

With Intel C Compiler (-O2 -xHost):

...

   262144    min=14258167  mean=14297606  sd=0.16%
   524288    min=31126755  mean=31212153  sd=0.23%
  1048576    min=92893600  mean=93491419  sd=0.46%

...

Can provide more numbers if people want them. EDIT: looks like I was using the less optimized C code, will re-run with the better performing C code when I get the chance.

I didn't really think about that case. In full generality, then the corresponding (n+m-k)x(n+m-k) matrix can be written as the matrix whose top-left (n-k)x(n-k) block is A\B, whose bottom-right (m-k)x(m-k) block is B\A, and where everything else is 1.

Note that both involve relabelling vertices. Many graph theorists do this freely (it's the same graph), but perhaps you have some application in mind where you don't want to do this.

Not sure it's fair to say that graph theorists rarely use matrices to represent graphs: for many problems in graph theory some of the simplest solution involves representing the graph as a matrix (eg Hoffman-Singleton theorem).

Just a side note: The join of two graphs does correspond to a reasonably nice matrix operation: take the two adjacency matrices A and B (of dim m and n respectively, say), then form an (n+m)x(n+m) block matrix:

A J

J B

where J is the all one matrix (of the appropriate sizes to make the above matrix square).

Further aside: this block form might seem unnatural, but it pops up all the time in adjacency matrices of graphs. For example, bipartite graphs are exactly those that can be written in the form

0 A

B 0

(Where B = transpose(A))

Context is important here I think. This is part of a series of posts containing career advice for mathematicians at various stages ([1]), and not life advice in general. So I don't believe Tao was addressing cases like Torvalds etc.

This post appears in the "Graduate level" section. At this point many open problems become accessible, while at the same time many grad students (emerging from being a big fish in a little pond) are looking to make a name for themselves. So a common trap is for people at this point to obsess over famous problems, to the exclusion of tackling less exciting but more tractable problems, and Tao is warning against this.

[1] https://terrytao.wordpress.com/career-advice/

To be fair, I think his main point was that the article was trashy for suggesting that millenials receiving help was unusual (and not to bash the parents of the 1 in 4).

Problem is that the camera is on top so if you rotate the phone and stand the phone on a surface, the camera is pointing in the wrong place.

I've ranted about this to friends but they didn't get it. Glad to know I'm not alone on this!

Also the audio jack on the bottom: can no longer use headphones and prop the phone up on a desk when using FaceTime (which is surely a common use-case)

Most of the articles I've read about this refer to a "tipping point": that we're crossing some line where the behavior of the system will change dramatically. But what are the actual consequences of this? Can anyone on HN explain the real-world effects that this might have, apart from slightly higher sea levels?

I tried searching, but the fluff to substance ratio is so high out there.