Today I watched a talk on skepticism about quantum error correction. Now I don’t agree with the particular criticism’s leveled, but I’m all for people airing their criticisms and, if the majority view is correct, the majority should be able to answer the questions raised. But this isn’t what interested me today. What interested me today was thinking about the following question: when has it been true that a curmudgeon, which I use in the most positive since of the word, been on the winning side? When has it been that a single or very small group of radicals who opposes a majority whose beliefs are set in concrete has actually ended up on the correct side of the argument? But more than this the definition of a curmudgeon indicates an amount of stubbornness which is above and beyond normal stubbornness. Indeed I would say that it even indicates a small amount of being confronted to by contrary evidence, but persisting in spite of this. And maybe not even being confronted not just by a bit of contrary evidence, but maybe a vast encyclopedia of contrary evidence.
I suppose the reason this fascinates me is that I, myself, certainly have some curmudgeonly characteristics. Does this mean I’m doomed to the losing side of the battle? I’m pretty sure I’m doomed, but for other reasons, and only today did it occur to me that I need to add my curmudgeonness to the list of reasons for doom.
O brave new quantum world!
I totally agree that we need outside criticism. Like I said, I’m happy to have Alicki bringing up his criticisms.
I’m just curious about the point of view of the critic. How do you know your criticisms are valid when you get strong pushback? In my experience when someone tells me they don’t believe me, 99.9 percent of the time they are right.
Hey Daniel, aren’t you supposed to be paying attention to the current talk?
Presumably you are talking about Alicki’s talk. Are you? I don’t think that he necessarily believes that fault tolerant quantum computation is impossible. Did he say so outright?
I have always considered the fault-tolerance literature to be more of a dialog, where (please excuse the highly oversimplified characterization by someone who only watches this stuff from the sidelines) someone proves a threshold theorem under plausible assumptions on the nature of the noise, then someone else argues that some source of noise being disregarded would be highly destructive, then another threshold theorem paper is written addressing those issues, etc.
I think that the majority of people in the QC community really do believe that a QC will be built eventually (I think I do too) and celebrate every time the threshold (almost as ridiculous of a number as an IQ) increases. This can potentially be dangerous for our community if we’re all seen as a close-knit group of collaborators constantly patting each other on the back for a job well-done. I think it’s healthy to have some contrarians giving us a real dialog, as opposed to two sides trading insults and yelling at each other in the now infamous string wars.
Dave, it would be very interesting to hear what you disagree with in Alicki’s critique. Maybe we can get a real time online/real life discussion going. There are other QEC07 participants reading this, I’m sure.
I was imagining a scenario from one of the newer Frankenstein movies (Robert de Niro) where the doc interrupts his professor’s lecture in opposition to established ideas on how life might or might not come about. It would have been funny to see someone just get up yell, “SIR, I VEHEMENTLY OBJECT…” and walk out.
“When has it been that a single or very small group of radicals who opposes a majority whose beliefs are set in concrete has actually ended up on the correct side of the argument?”
George Gamow and collaborators, just a handful of people, on the big bang. Mainstream cosmologists stuck with steady state for two decades after it was first proposed. They were wrong, but then of course, cosmologists almost always are.
The history of science is littered with moments when individuals or small groups overcome the status quo. Sometimes it’s almost instantaneous acceptance, sometimes it takes a while. Of course there are far more moments when individuals or small groups have almost no impact, often after working for 40 years on their alternative.
“For every expert, there is an equal and opposite expert.”
— Sir Arthur C. Clarke
Alfred Wegener, the guy who introduced the concept of continental drift and received widespread derision in return, probably qualifies.
Also maybe Georg Cantor.
So far, the curmudgeons have proven mainly right on energy technologies: the disastrous effects of CO2, the expense of solar, the infeasibility of fusion, the dangers of fission, the impracticality of wave and wind.
A good source (including a length essay by John von Neumann) is the 1995 book The Fabulous Future: America in 1980, which holds a prominent place in our QSE Group’s library of subversive literature.
Dave’s topic is interesting IMHO … after all, surely Kurt Godel would qualify as a curmudgeon (albeit an exceptionally young one) for pursuing–and proving–the radically curmudgeonly idea that Hilbert’s consistency-and-completeness program for mathematics might be unachievable?
Surely it is not a logically necessity that curmudgeonly ideas can only be held by the elderly!
Similarly, perhaps Turing’s insight that code-breaking could be done faster by soulless machines than by trained mathematicians, was curmudgeonly?
And didn’t von Neumann have a similarly curmudgeonly faith that fluid dynamics might fruitfully be simulated on machines?
Also, wasn’t Craig Venter’s mathematical faith in shotgun-sequencing assembly algorithms in essence a curmudgeonly faith? Certainly Venter’s criticism of slower sequencing methods was curmudgeonly.
For me, the lesson-learned (on first reflection) is that curmudgeons may be more likely to be right about the value of new scientific methods, than they are likely to be right about the value of new scientific laws.
Many railed the General Theory of Relativity, when first published. However, this did not deter Einstein. When asked if Eddington’s observations of a solar eclipse had produced results which contradicted the predictions of his new theory, he replied, “Then I would have been sorry for the dear Lord-the theory is correct.” This serves to illustrate that consensus can never be trusted to expose truth; it does however often serve to form policy. In the end, one must trust their gut and occasionally at such times it requires that one have some. If you say you have been on the short end at times, in this regard, then you should be confident you possess this prerequisite.
It takes a while to warm up to Dave’s topic, but really, examples are everywhere. A good QIT examples is IBM’s effort to commercialize computers based on Josephson junctions (SQID).
The SQUID visionaries saw the advantages of SQUID computers: (1) speed, and (2) cool “quantumness”. The SQUID curmudgeons saw the disadvantages, which included: (1) poor fan-out, (2) adverse thermodynamic gradient (it’s highly disadvantageous when the computer is colder than the environment), (3) process control issues that impacted device yield, and worst of all (4) SQUID-based computers were chasing a rapidly-improving silicon-based computers.
At the end of the day, the SQUID curmudgeon arguments prevailed … although it seems that no technology is ever utterly abandoned, and SQUIDS may yet return! And very importantly—because it was the main enduring value of the SQUID effort—a tremendous amount of good science was created.
Continuing the techno-curmudgeon thread …
Computation via biased semiconductor junctions is viewed favorably by techno-curmudgeons in consequence of: (1) unlimited informatic fanout consequent to rapid decoherence, (2) highly favorable thermodynamic gradient (the chip substrate is a massive thermal reservoir), (3) excellent process robustness, and (4) favorable “Moore’s Law” scaling of parameters such as device speed and power consumption.
These features are so familiar to us that we tend to overlook how wonderful (and subtle) they are at the quantum level.
The above is a techno-curmudgeon’s explanation of why more DRAM cells are fabricated each year than raindrops fall upon central California.
Just to continue a thread that (most likely) no one but me ever reads …
Doh! How could I have forgotten two classic cases in which the physics-curmudgeons were right: (1) controlled thermonuclear fusion and (2) space-born “Star Wars” defense systems (x-ray lasers and the like). The fundamental physics was marginal … the engineering challenges were severe … and so the programs lost momentum (but never totally died).
Does this occur in biology? You bet. The bio-curmudgeons were proven right when the NIH’s structural genomics programs have largely failed to scale … too many of those darn proteins just don’t want to crystallize. And we can all hope that the bio-curmudgeons will *not* be proven right about the inherent difficulty of developing effective vaccines against rapidly lutating adversaries like the HIV virus and the malaria parasite.