This is amazing. Thank you!
2,243 karma · joined April 22, 2008
FWIW I've been admiring your HN posts for many years. When I see your handle in an HN discussion, my learned expectation is of knowledge and insight.
I don't know why you keep repeating that him deleting his blog ensures the lack of anonymity (in this sense) where there'll be a NYT article or not. I suspect he'd disagree, and I think that if this story ends with no NYT article that outs his name, all the discussions and media articles about this story that happened so far won't change the desirable status quo re: his patients not connecting his name to Online Controversies through cursory Googling.
That many people especially admire SSC for even-handedness, which they see as false, is especially infuriating to them.
If I understand correctly, antigens identify the virus by its S proteins. Would we be able to use the same methods <a href='https://www.nature.com/articles/nm.3985'>as in this paper</a> to replace the S protein in a harmless cold-like coronavirus with the S protein of SARS-CoV-2, and would then the resulting immunity defend against SARS-CoV-2?
And yes, Stallman wrote twice in the past that he thought consensual pedophilia was not particularly harmful - about three lines worth among many thousands of entries in his inane "political blog" that approximately nobody reads or cares about (and on which of course nobody commented or cared about at the time they were made). These lines were lovingly collected and placed, with some other examples of his wrongthink, e.g. on the geek feminism wiki years ago. They did not play a significant role in creating the moral panic - the initial Medium article with the lie about him saying Epstein's victims were entirely willing didn't have them. The articles in the media with headlines like "Star MIT Scientist Calls Epstein Victim Virginia Giuffre "Entirely Willing"" mostly didn't include them, though a few did when the journalists went to dig for more fodder.
Thomas, I've never met you, but admired your writings over the years. Seeing you defend the mob against Stallman with well-worn platitudes like "He is welcome to his thoughts and opinions, and everyone else is entitled to their freedom not to associate with him as a result" is particularly saddening.
OpenAudible looks like a nice tool that'll automate all of these steps for you.
I think "singing along" is not a sharp enough distinction - the Netflix guy, in a meeting about sensitive words, was presumably saying something like "The words that are awful and completely out of bounds and should never be used, like c-, f- or n-..." Such a use is firmly on the other side of a use-mention boundary in the way that singing along some lyrics is not. So it seems striking (at least to this non-American) that it should be considered heavily tabooed in this way.
In a discussion on reddit, someone said
"Within the last fifteen years, I've had readings in university English classes that contained said word and we're discussed and quoted in class with no one batting an eye. This was also the case in highschool before this. There's a big difference between referring to a Twain or Goines novel and using the word maliciously or with any intent beyond dispassionate quotations."
So are they wrong, or was there a huge shift in the social norm of using the word in the last 15 years (discuss in class vs being ostracized and fired, both for mentioning the word with full awareness of its awful context) or is the consensus not really quite a consensus yet, and asserting that it is, like in the Netflix CEO's letter or your comment, are (well-meant) attempts at driving it further along?
>Also, it's not like this is unique to black people and the N-word. LGBT people have similar words and we have reached similar detentes about them.
That sounds strange to me. Are you sure? Presumably you mean the word f-, but I think I see or hear the word being discussed and quoted in the context and with the awareness of its awfulness quite often without heavy, or any, social consequences.
It's said to have been based on DragonFly BSD, with new components written in Ada (!), but also supporting Cyclone.
In fact, I want to start the year with a MOOC or a video course on deep learning. I went over the current offerings, and am hesitating between Andrew Ng's new Deep Learning course at Coursera in 5 parts, and self-studying through the cs231n Stanford course (it has all the videos, assignments etc. posted). Can anyone help me decide between the two, or offer something else? I'm looking for mathematical understanding as well as hands-on experience, and am not afraid of math.
In the Soviet Union, things were typically inexpensive but unobtainable. So for example if you wanted a fully automatic washing machine that didn't break down after a few months, the problem wasn't that you couldn't afford it, the problem was that too few of them were imported or produced compared to the number of people who wanted them. In theory, these fancy washing machines went straight from producer/importer to distribution centers and then to regular stores to be bought by regular people. In practice, every chain in the distribution link was highly corrupt, and desirable items never reached regular stores (or, in some cases they did, but the knowledge of when exactly and in what store they'll be giving out something "good" was by itself desirable). Everybody worked intensely at cultivating connections with the right kind of people who were gatekeepers in some distribution chains, or officials in charge of those chains. Now repeat that for every kind of expensive item: furniture, cars, good clothes, shoes, some kinds of high-quality food (from caviar down to sausage), getting your kids into a good school, summer camps, vacation packages, a work trip abroad... all could be and were obtained through nepotism.
Can you give more details (accessible to programmers rather than social scientists) how that'd work, what do you actually model and aim to find out? Sounds pretty interesting.
Technically it is self-contained and nothing beyond high school knowledge is needed to understand it; but while technically true, in reality it's a cruel deceit. If you try to learn it without advanced training in physics and math, you'll run into limitations related to mathematical maturity (https://en.wikipedia.org/wiki/Mathematical_maturity) very quickly, within the space of the first few chapters.
The idea of this book seems noble and uplifting, but it cannot be done in this fashion. It doesn't matter that every theorem and assumption are clearly stated and follow each other in an orderly sequence. There's a huge number of new ideas, mathematical and physical, which require proper acquaintance to settle in your mind, not the 1-2 pages Penrose is able to give each of them. If someone is serious about learning physics to that level "properly" and is willing to invest the time and effort, they're much, much better off working through conventional college-level (and later grad-level) textbooks in particular subjects, taking care of math prerequisites before and during such study. That's very very hard to do, but is in fact possible, whereas studying solely from Penrose's book, I think, is not. It functions, sadly, as a sort of trap for impressionable bright people very eager for advanced knowledge - they will try to struggle through it, inevitably give up rather early on, and blame themselves.
So things like:
- viewing primitive objects, structs and arrays as blobs of bytes arranged between address X and address Y
- a pointer as an address of that blob. Understanding what it means to copy that blob vs. to pass around the pointer: calling by value/reference not in terms of the semantic effect, but in terms of what happens underneath.
- the stack and how it's typically used, esp. in recursion. Local vars vs. heap-allocated vars. How control flow leaves its trace on the stack. How arguments are passed and return values are returned.
- complex data objects viewed as blobs of memory pointing to each other vs. pointers. Concept of "owning" such a blob and how ownership is passed. Understanding how your language's runtime keeps track of the blobs if you don't have to, and what are common pitfalls.
- clear understanding of the difference between int8, int32, int64. Strings as null-terminated or counted arrays. Bitops.
"Learn C" is just a useful way to force you to internalize all of the above, because you can't properly "learn C" without doing that. But it's the above that helps you back in your favorite language. That, and perhaps the fact that C gives you a feeling what it's like when you can look at a line of source code and understand immediately what happens in the machine (broadly) when executing it. No hidden effects. C++ doesn't have that (constructors you don't know about when looking at the line, exceptions etc.) That "local clarity" isn't the most important thing in the world, but if you feel and appreciate it, perhaps you'll strive for local clarity back in your favorite language, too.
Long and short vowels must obviously be distinguished. I think either yōko or yooko are fine. Either Tōkyō or Tookyoo will let a beginner pronounce the word correctly. Some prefer youko and it works fine in almost all cases too, but perhaps there's no real value in clinging to the accidental way with which hiragana happens to represent long vowels; it isn't inherent to the language. Besides, there's a small added benefit in disambiguating real [ou] pronunciations like omou.
Your point about visiting Japan is well-put. There's no arguing with the fact that anyone who wishes to understand Japanese should learn kana and kanji. But I think that there's a real value in putting emphasis on the spoken language first, and a well-developed minority tradition of teaching Japanese like this in a serious manner, not as a crutch. E.g. Eleanor Jorden's JSL textbook follows this principle, as do a number of other textbooks and guides.
"To get started in learning to read modern Japanese along the correct lines, it is essential that first of all you have some knowledge of the modern spoken language. This is so important that it is well to emphasize it strongly from the beginning - contrary to what you may think, you undoubtedly will not make the best progress in learning to read Japanese by starting in your study of the language directly with reading and writing in the Japanese script. Some previous work with spoken Japanese is essential, and the better your grasp of the patterns and the forms of the spoken language, the faster and surer will be your progress with the reading materials in this book" (Roy Andrew Miller, _A Japanese Reader_).