Assorted links


Tyler, thanks for the attention to my Better Than Net Neutrality post. I'm a long time reader, so this is very cool.

For anyone looking to join a conversation, I originally posted this idea on my (and others, it's a group effort) blog Sweet Talk Conversation. The link to the original post is in my name.

I'm pretty sure Tyler linked to the below in the past.

That article does mention that getting access to the infrastructure is difficult, so I think leasing at cost makes sense. However, the article also mentions that most companies price assuming 30% penetration, so allowing 4 companies will clearly alter that. I'm not sure why to stop at 4 companies, though.

Europe seems pretty significant competition in wireless because they allow so many MVNOs to resell the already built infrastructure. That's essentially what the Telecommunications Act of 1996 did for ADSL (remember that?) and T1 lines and there was an explosion of businesses reselling those. At the time, consumers had a lot more choice than they do today.

@#6 - if the profits of ISPs are taken up with regulatory issues, as they are, you have to wonder whether "more competition" will increase innovation. Possibly not. Recall that during Ma Bell, which everybody hated due to lack of competition, you did get high prices which actually spurred innovation--as well as the Bell Labs (TC has made a similar point out net neutrality a while ago). Again, people respond to incentives. Throw some money at a problem, reward your inventors, and you get inventions.

Fact: "The economist William Nordhaus has calculated that the inventors and entrepreneurs nowadays earn in profit only 2 percent of the social value of their inventions. " citing Nordhaus, William D. 2004. “Schumpeterian in the American Economy: Theory and Measurement.” National Bureau of Economic Research Working Paper W10433.

Does 2% capture of the value-add to an innovation sound fair to you? Sounds like theft to me. The taking of IP is theft--that's my neo-Proudhon slogan.

But Europe is lagging significantly in 4G wireless deployment. Unbundling network elements and requiring ISPs to sell access to competitors at cost will reduce the incentive to continually invest in uparades to the infrastructure. It doesn't make any sense, except for people interested in regulating the market..

re: natural monopoly. I still don't understand why it isn't more efficient to run a single fiber and have 4 companies share the bandwidth, rather than have them all run their own separate fiber.

Providing a working data pipe is the only thing of substantial value. Whichever entity does that is going to fairly trivially add on the other trimmings that go along with being an ISP, namely connecting the pipe to the Internet, manning a call center, and managing end-user billing and collections.

So four ISPs are never going to share the bandwidth of some fiber owned by someone else, because whoever that someone else is IS the ISP.

As the article states, Disney realized the cash cow they have in the 'Princess Franchise' years ago. What they don't lay out is the systematic study Disney made of which cultures have myths/fairy tales that involve their own princess type characters that could be remodeled and rebranded. The goal was/is the steady creation of Disney versions of multi-cultural princesses to keep the franchise ever fresh, and gain market share in countries where they have little penetration. Princess and the Frog, Tangled, Brave, and Frozen are all outcomes of that project. The original writer/director of Brave, Brenda Chapman, left the film in part because of pressure to make Meridia a more conventional princess.

The particular innovation of Frozen was recapturing a bit of Alan Menken's lost magic, by embracing the idea that they were making a full fledged musical, and pairing the princess with a show-stopping song. What Disney also realized is that the audience for animated features, especially the princess movies, aren't looking for innovation and variety. Expect more variants of Frozen/Tangled.

Frozen and Tangled are a pretty big departure from the prior princess formula that Disney used. Frozen was partially a deconstruction of their prior trope.

Not really. There's a thin patina of modernness, but no more so than in Little Mermaid and Beauty and the Beast.

Let's give credit where credit is due -- Frozen features Elsa and Anna as sisters who genuinely care deeply about each other, and that is not part of any formula in Disney's princess movies, and is, as Bechdel could point out, somewhat rare in Hollywood media at all. And it's a nice message for the little girls who are the primary audiences for the movie.

The rest of Frozen is fairly conventional albeit well-executed Disney animation feature, and the Anna/Elsa relationship, while driving a couple of important scenes, is not what I'd call the core of the movie. But it is there, and it is more than simply a patina.

You're right that the sister relationship is the one story/character element that is special about Frozen. Almost everything else in the film is derivative and unconvincing. But a touching, realistic sister relationship was already done by Disney in the successful Lilo and Stitch. Among the princess movies, Little Mermaid was the first (and really only) one to feature an intense and somewhat real relationship between father and daughter (quick, name 15 Disney movies with a doddering, buffoonish father figure! It's easy, isn't it?). Beauty and the Beast featured a prince who wasn't the stereotype, and a 'princess' who wasn't all about looks. Mulan was a 'princess' who took a completely unique path of cross-dressing and going to war. And so on. All of these films have their little hook, some more successful than others. But ultimately it was the kick-ass songs that elevated Frozen into the stratosphere.

And the snowman.

Fair point about Lilo & Stitch, though Lilo's sister was in loco parentis, while Elsa and Anna are more conventionally sororal.

This is another prong of the Disney PR attack

Trivia: the creators of one or more of the songs in Frozen are Filipinos, or so I understand.

Frozen would not have happened had Disney not bought Pixar. John Lasseter went back to Disney Animation and brought back a studio that was almost shut down. Note that Big Hero 6 (this year's surprise hit) was another Disney Animation (not Pixar) movie.

Meanwhile, Pixar is now making the sequels it would have rejected outright in the non-Disney days.

Doug, Disney Animation wasn't about to be shut down when Iger bought Pixar. It WAS close to being shut down after Sleeping Beauty, and after Walt died, and after Eisner and Katzenberg came in. But not in 2006, when Tangled was already deep in development (as Rapunzel). Lasseter's big contribution when he came in at Disney was to unsuccessfully retool A Day with Wilbur Robinson and American Dog (which became Meet the Robinsons and Bolt), push The Frog Princess forward (became Princess and the Frog), and lay off most of the remaining traditional animators.

Iger gambled big by overpaying for Pixar, and for that reason the Disney machine has every reason to credit buying Pixar with rejuvenating Disney animation. There is some truth to this, but the bigger picture is that animation studios go through cycles of creativity/innovation and then stagnation. And if they last long enough, the come full circle. By the time Disney bought Pixar it was already clear to many people that Pixar was becoming the stagnant studio, which has been born out by the list of sequels and second rate efforts (Cars being the prime example). Meanwhile, Disney was ripe for rebirth, after a period of epic mismanagement under the later years of Eisner and empty suits like David Stainton. At one point there were literally 47 vice-presidents in charge of the animation department, to give you an idea of the degree of mismanagement. Mickey Mouse could have come in and ended up looking like a hero.

Out of curiousity, how different would the princess movies have to be, for you to regard them as examples of innovation and variety?
And re-writing other culture's stories is an ancient tradition in Western creative culture, think for example of Shakespeare. Or James Joyce's Ulyssess.

Everything I've ever seen from Disney I've found revolting, with two exceptions. I've seen bits of Mary Poppins on the telly, and it seemed OK in an anodyne way. I have enjoyed The Jungle Book, though somewhat amazed that nobody ever remarks on its racist aspects.

My impression of Disney stuff is treacle smeared over tween porn.

The New Zealand sheep butchers remind me of a wool dealer I met in Rio Gallegos, Argentina (near the Magellan Strait). He'd grown up on a Patagonia sheep ranch and spoke Spanish with a Scottish accent though he didn't speak English. He had been to New Zealand and South Africa, a world traveler who kept far south of the equator.

I thought it was inspired to use the chap from Dannevirke as the illustration.

#5 There is a radio interview with Gunnar Halldorsson, the manager of the Icelandic chain, here

#5 illustrates the advantages of the Dreamliner over the A380. In the future these guys will be flying south to London and then getting a connecting flight to Reykjavik. (A connecting flight to Blonduos will be a bit of a stretch). A little bit of global warming will of course help.

#3 was excellent.

#3: "Alexander Grothendieck, the greatest of all modern mathematicians ..."

I wonder how many mathematicians would agree with Landsburg on this.

Maybe these guys are also worthy of that title?

Terry Tao:

Paul Erdos:

You can divide mathematicians into problem solvers and theory builders. Grothendieck was the greatest of the theory builders. Most theory builders look down upon the problem solvers, and consider what they do to be barely worthwhile mathematical work. They are confident that he would indeed be the greatest mathematician of this century. Most problem solvers would say many great things about Erdos, but are likely declare they are incompetent to say anything about Grothendieck, while possibly making a comment that he once mistook 57 for a prime number. (Terry Tao can be crudely put in the problem solver side of the story, but he's not as far of that scale as Erdos).

@math, Al: are you guys talking about faded stars? Erdos is what, 70? Googling now... no, he's dead. I vote the greatest mathematician is the one that popularizes math. So it would be Martin Gardner (dead), John Allen Paulos (a Greek), Keith Devlin, and, frankly, myself in my own humble way.

Re # 3: Hearsay only, but there can't be more than a handful of people in the world qualified to state that the popular algebraic geometer who recently passed away - may he be in a better world now - is a better mathematician than Kolmogorov, Neumann, Ramanujan, Penrose, Coxeter, Weil, Connes (to throw in another humble laborer in one of the less venerable branches of number sciences), and Tao. I would be more than astounded if some such person wrote for, or commented on, a typical American libertarian weblog. But I have been wrong before. Al- Wasn't Erdos a recreational mathematician? Not the same as a real mathematician, surely ?

Do you know these guys 'indirectly' too?

Brian Donohue - Are you angry? Why? Was it the mild prayer for the eternal soul of Monsieur G. ? Or did you think that the comment was, perhaps, uninformed or humble or anti-libertarian or arrogant in a way that offended you? Or are the math reputations of middling geniuses and over-middling geniuses your special field, on which somebody directly encroached? Or is there something else going on that I can't figure out? Personally, I have known a real genius or two in my day, but none of the eight I mentioned. But I am not sure your question was directed to me.

I don't think it was directed at you, Housman. I think it was directed at Ray Lopez.

Read Ray's comments in this blog post to get the background:

Al (at 11:12 PM), Thanks for letting me know that. Still I will never use the word "hearsay" on the internet again...

Yikes! I shall weigh in here with those who think that Landsburg has way overstated the importance of Grothendieck and his work, creative and innovative as it was. Landsburg even goes so far as to suggest that Grothendieck "might" be the greatest mathematician of all time, not just of the 20th century, which is really letting his mouth pull his brain out of his head and dump ii on the sidewalk in full view of the audience.

Some others in the 20th century besides those already mentioned might be Hilbert and Goedel.

Heck, the paper by McLarty that Landsburg links to makes it clear how much of the stuff Grothendieck did was founded on important works by other people within his field, such as Serrre, Eilenberg, and MacLane, as well as Weil mentioned above. One could say that the really big idea was Weil's and that Grothendieck was the "problem solver:" who figured out how to rigorously establish the Weil conjectures. A crutical tool in all that was category theory, a very important invention, which was done by MacLane and Eilenberg, not Grothendieck. Arguably Grothendieck was the greatest of the Bourbaki gang, but he is nto all that far ahead of Serre, who influenced him mightily, not to mention the leader of the group, Cartan, who was also the main math teacher of Gerard Debreu, who was the most important figure introducing Bourbakism into economics.

I find it amusing all this talk of Grothendieck building buildings and being outside of universes or mathematical mansions, something he himself cllaiimed for himself. But one can easily argue that all he did was to clearly show how certain passageways identified by others, see Weil especially, actually went between romms that had previously been viewed as disconnected. If a new building was built, others did it, with Grothendieck providing a lot of important finishings to it. How hilarious this all is can be seen by the bizarre remark by McLarty that Grothendieck took an 80 page paper by Serre and "simplified" it into a 1000 page work. Uh hun. Some simplification there.

Just to hammer this home a bit more, at the end of his article, McLarty admits that some of Grothendiecks's more famous inventinos, notably toposes, are not used by many researchers. In short, if Grothendieck really did build brand new mansions "from scratch," as his loudest fans proclaim, some of those mansions are eerily empty of inhabitants.

The 20th century mathematical portraits that hang on my office wall --- along with the pre-20th century portraits of Archimedes, Fermat, Newton, Euler, Gauss and Hilbert --- are of Noether, Cartan, Weil, MacLane, Eilenberg, Serre, Grothendieck and Quillen, so clearly there's a lot of overlap between your mathematical heroes and mine. And of coures we're all entitled to revere our heroes in the order we choose.

But I do maintain that by some reasonably objective criteria, Grothendieck towers over even this field of titans. What other project can compare with the tranformative power of SGA? You suggest that some of Grothendieck's ideas (e.g. toposes) are not widely influential. I'd disagree with that, but no matter. We are judged not by our least influential work, but by our most. Surely you can't say that schemes are not widely influential --- indeed, modern algebraic geometry is the theory of schemes (except, perhaps, when it's the theory of stacks or motives, which also came from Grothendieck). I think it's a fair statement that virtually every single paper now written in algebraic geometry stems from Grothendieck, that every single paper written in K-theory stems from Grothendieck, that a majority of the papers written in commutative algebra and a substantial chunk of those written in number theory stem from Grothendieck, and that these ideas spill over importantly into topology, theoretical physics, and on and on.

You're absolutely right that Grothendieck stood on the broad shoulders of Eilenberg and MacLane (two of my other great heroes) and that's absolutely true. It's also true that Einstein stood on the shoulders of Maxwell, but we don't ordinarily think of that as diminishing Einstein's achievement. Eilenberg and MacLane invented categories; Grothendieck, beginning with Tohoku, made them more powerful than E&M could ever have dreamed.

Serre, too, was clearly indispensable, and made vast contributions to nearly every field in which Grothendieck worked (plus others!), either directly through his own research or indirectly by pointing Grothendieck in the right directions.

But I propose the following semi-objective test: Pick five random pages from the 10,000 pages of SGA and EGA. Measure their depth, in terms of how long it takes to fully digest them, how much new insight they convey, how important they've been for further developments. Now tell me who else has produced 10,000 such pages. Let alone the other thousands of pages in Tohoku, GRR, FGA, the Brauer Group papers.....

Or another test: Ask active mathematicians. I've got no hard data but I do have several decades of casual observations, and it seems pretty clear that Grothendieck's name commands a unique level of respect.

If you're unmoved by those tests, that's your right. But I do object to the (perhaps unintentional) suggestion that I just randomly elevated Grothendieck to the top of my pantheon. There are good and reasonably objective criteria by which he belongs there. If you have other criteria that point elsewhere, that's fine too.

reply to Steve Landsburg - For what it is worth, after reading your comment on this blog, I am now convinced - because I will never be able to refute paragraphs 5 and 6 of your comment on this blog (the paragraph referring to examining the ratio between interesting and non-interesting publications, and the paragraph on your decades or casual interaction)- that you did not "randomly elevate Grothendieck". So, again for what it is worth, I retract the part of my comment that might be taken to mean that your assessment was random. And, on further thought, after reading your explanation of your assessment techniques, I should not have said I would be "astounded" at a certain number of individuals being able to assess the mathematical progress in question. That word - astounded - was more a reaction to the comments of the commenters on your blog on this subject, than a reaction to to any real facts of this world. That being said, from what I understand of him, Kolmogorov, for one, is really really hard to put in second place to anyone whose life overlapped his.

I think there is no objective way to resolve this. Such rankings and judgments are clearly subjective and depend on which parts of math people consider most important, not to mention judgments regarding particular works. Clearly AG was brilliant and enormously influential. But ahead of those folks hanging on your wall? Well, we do not agree on that, and this will not be resolved. Anyway, let me make just a few points, recognizing the fundamental indeterminacy involved here.

So, I may be biased. I personally knew both the late Sammy Eilenberg and Saunders MacLane, having had several fairly in-depth conversations about a variety of mathematical issues with the latter. I also am not a big fan of hard line Bourbakism, which makes me less inclined to AG as a result. So, biases on my part (and I am even more closely personally connected to Godel, and am somewhat of the view that his most famous work is greater than AG's, although this may also be a matter of logic being more important than algebraic geometry, although category theory is the tool that links these up, one of those funky tunnels linking prevously disparate rooms in the mansion of mathematics, and although this was invented by, ahem, Eilenberg and MacLane, not AG).

Regarding category theory, it has clearly had an enormous influence across various parts of math. I would note that while AG may have proposed the most important extensions and apps of it, many others also made such extensions and apps as well, not going to give a list. You might say that AG is Einstein while E and S are merely, well, Lorentz or Planck or... Maybe. But, frankly, I think it is a very far stretch to claim as much influence on math by AG as Einstein had on physics (and beyond). The most cited work of AG, his paper on algebraic homology theory II, has only gotten about 1300 citations, certainly a lot, but not as many as have some other math papers, and certainly not remotely in the league of Einstein. I am perfectly willing to grant that citations are a poor measure, but I think it also indicates that fans of AG are to some extent engaging in a :"He deserves more respect than he has gotten." Maybe, but not settled.

I could go on at much greater length, but this represents certain core views of mine on this.

Barkley Rosser: There is much we can agree to disagree on. But I do just want to say one thing about citations: The reason Grothendieck gets fewer citations than you might expect is that the ideas he introduced are so very central that people routinely use them without citing a source, just as they use, say, the commutativity of addition without citing a source. The material of Tohoku, for example, is used so routinely that one would never think of citing it. Ditto, and even moreso, for scheme theory --- if every paper that used schemes cited Grothendieck, he'd be the most cited author in any subject in all of history. Or etale cohomology. Or K-theory. I've published a dozen papers about K-theory, ALL of which are ultimately commentaries on Grothendieck, but I don't believe I've ever cited him.

Point taken, Steve. But this does not come close to showing he would be the most cited person ever. The same is true of other far more important figures who have been foundational, e.g. Einstein. Lots of people write about ideas drawn from Einstein's work without specifically citing his original work, and Newton, and some of those other mathematicians on your wall.

I suspect your perceptions are influenced by your writing specifically on topics that AG is influential in. The larger question then becomes just how many papers are written on these areas compared to areas where other seminal figures have massive influence so deeply embedded that they no longer get cited? I do not think this is easy to answer, and that is because we are back to the problem of standing on giants. Given the importance of category theory in AG's work, do we count Eilenberg and MacLane whenever something that AG did that depends on category theory is influential, and none of them are cited? This is simply an unsolvable problem.


Final comment on this from me.

The case for AG being the greatest mathematician either ever or at least of the 20th century I think boils down to arguing 1) he is the greatest algebraic geometer ever (or of at least the 20th century), and 2) algebraic geometry is the most important field of mathematics.

I think the case for the first is strong. Sure, I have come on defending Eilenberg and MacLane, but I am willing to grant that AG probably was more important and innovative than them. However, while I respect the case for the second, I do not accept it, and that is a matter that can be debated, but that clearly there will be no resolution regarding.

Interesting. Thanks.

#5. The flight path taken (27 hr 30 min) most probably determined by cost. The flight time the other way round (AKL-LAX-RKV) is only 21 hr 20 min.and may be less waiting time and takeoff and landing time as well.,+New+Zealand/to/Iceland

Yep. Flights via Europe/Asia tend to be much cheaper, because there is a lot more traffic and tons of competition between all the Asian flag carriers on the Europe to Australia/New Zealand routes.

Yeah but if you go AKL-LAX-RKV you have to go through LAX.

Those sheep look so cute!

But I'm sure they're delicious, too.

#1: It's really interesting to see a discussion on Disney's stories evolution when Tyler was pointing at the idea of "affordable luxury" as an emergent consumption pattern powered by children. Priceless =)

Landsburg claims that Grothendieck was the greatest mathematician of the 20 century and possibly of all time. Which is absurd. Weyl, E. Cartan, Siegel, Kolmogorov, Leray, Teichmuller, Connes, Herbrand, Whitney, are just some of the 20th century mathematicians that are more highly regarded than Grothendieck. Only perhaps Teichmuller is comparable to a Galois or Riemann for the claim to greatest mathematician of all time.

Comments for this post are closed