Live data from Hacker News

The Kolmogorov option

scottaaronson.com

401–410 of 476 posts

Re: The Kolmogorov option

#401
post #33

Let's say this is about the Google memo. And let's say, for the sake of argument, you're a person who thinks Damore had some good points and some bad points but you think the hysterically censorious response to him was way over the line. But you don't want to become a pariah yourself, so you stay quiet about it. The argument Damore was making was fiddly, kind of subtle and takes a long time to explain, it's not worth…

Let's say this is about the Google memo.

Let's just say this - whatever one thinks about Google's snap decision to fire that guy, he ain't no Kolmogorov. Or even a "little Kolmogorov". And as to the suggestion that there might be some thematic overlap between his "struggle for the truth" and the struggles against Stalin or the Nazis... let's please not go there.

No one stopped and offered a third option: actually discussing his argument, acknowledging where he was right, and discussing what he got wrong.

Probably because it wasn't cogently presented, and has some quite glaring gaps in the basic logic of what it does present. And in its tone, was more than a bit inflammatory.

Re: The Kolmogorov option

#402
When I reached the comment field at the end of page, to my big surprise the name and email have been already filled in. In fact, it seems like data of the person last to comment have been present there. Something along the way have gone quite wrong.

Re: The Kolmogorov option

#403
post #34

Tell me, truly, is this a view held by many here? That, as the church denounced Galileo's observation that we turn around the sun as heretical, so too our modern PC culture suppresses legitimate inquiry into wether gender is a determinant of programming aptitude? This is a comparison worth drawing? Galileo was a scientist. He published papers, books, treatises. He devoted his life to the pursuit of the truth, found a…

First of all he never ever said women were inferior. He said they were statistically less likely to be interested in computer science. That's not an objectionable claim at all. It's certainly not a "hostile work environment". Firing people for having political opinions sure is though.

The "manifesto" wasn't the best argument ever written on the subject. But as far as I can tell its basically correct. There are many well documented biological differences between men and women. Not just minor ones, but some of the largest effect sizes I've ever seen in a social science. I think it's absurd to expect exactly equal gender ratios given such enormous differences.

The NPR article is interesting. It was brought up before and there was some debate about how accurate it was. It doesn't make much sense given the gender ratio in other similar areas like engineering is similar and hasn't changed that much.

But even if its correct, the theory they present is basically compatible with the Google memo. That parents are less likely to buy girls computers and so they don't go into comp sci. If that's the case, Googles sexist programs to discriminate against men over women isn't going to change anything. All the statistics show the "problem" isn't in the tech industry and is happening by high school or before.

Re: The Kolmogorov option

#404

Earlier quoted context omitted.

What would you need to do to limit doxxing potential?

Most of all, consistently apply the maximum level opsec right from the beginning. Even if the first hundred posts on your new blog are inoccuous, you can't let them be connected to you when five years from now you find out that you actually want to write about something controversial for once. The hardest part is planning ahead and investing a lot of time and effort before the point where you actually publish the dan…

Is there some kind of guide about this?

How can you beat a linguistic analysis? If you publish elsewhere and someone guesses to compare your work, are you screwed? Are there any programs that scan writing to determine if the writer's english is Canadian or American or British etc? Or maybe your gender? Could you use that to weed out any regional phrases, or use regional phrases from other places to confuse the text? How do you make sure you don't sound the same in your real life, using similar phrases (For example, if Scott Alexander from Slate Star Codex had another blog that was not anonymous, would it be nessecary to not use expressions like 'Steelman' or refer to effective altruism?

Should you look in the academic literature about language, and try to make it so your style can't be detected by theoretical methods of linguistic analysis that haven't yet been implimented computationally?

How do you deal with private communication? Does it make sense to simply have no possible way of privately emailing you, making all communication public (thus giving you plausible deniability if you click any links phishing for your identity). Should you not even interact with public comments?

What about any information you might giveaway even when you are being a VPN or something (browser info? Some kind of computer associated seriel number? internet cookies?). Is it overkill to simply have one device dedicated to researching/blogging, and restricting yourself from doing normal day to day work on that computer? What about a virtual machine?

Can you buy and pay for a domain anonymously?

Should you make a list of things you are willing to reveal about yourself, and stick to it? For example, A/S/L and then make sure never to reveal other details (former locations, trips with dates, schooling, etc) Should you change details of anecdotes if you share them?

If you trust someone, perhaps a girlfriend, or wife, or really good friend, is it too risky to share with them your identity, even if you agree to never discuss any of it digitally? Assuming they also keep a wall between themselves and that identity (not sharing posts, not telling friends, etc) is that safe? If you do break up, should you create a new blog, and if so, is it worth it to make the writing style clearly different from the old blog? Are there any high profile, clearly psuedonymous people who have remained so for long periods of time?

By making this post, should I now do none of these things?

Re: The Kolmogorov option

#405
post #399
post #350

Earlier quoted context omitted.

I saw the links. No expert on the actual matter at hand (women participation in tech) would consider them relevant at all to the issue.

> Did he research social structures or even the psychological differences between the sexes? > No expert on the actual matter at hand (women participation in tech) would consider them relevant at all to the issue. Your inability to maintain a frame of reference for your objections is disheartening. It's clear you're just entrenched in a win/lose mentality and have no interest in the topic at all.

On the contrary. This is a subject I studied (although indirectly) at university, and that's why I could easily tell that the author of the manifesto is both ignorant and uninterested in the subject, which is why he quotes studies that are irrelevant even if their findings are representative. In other words, that he confuses the question of "natural inclination" with the problem of inclusion of women in tech shows that he has done no research on the subject at all.

Re: The Kolmogorov option

#406

Earlier quoted context omitted.

OK, well I'm glad you're just inquiring genuinely. It's hard not to get defensive when you trot out some pretty horrible characters. I'm in strong agreement that discrimination against gays, minorities and women is bad. I think most reasonable people are. My issue with the progressive movement is how it doesn't seem to want to stop in the middle by just stopping racism, but instead wants to flip it over by making dis…

> It's hard not to get defensive when... I thought classical liberals were incapable of emotional involvement in a rational discussion.

I've seen what happens to people who are falsely tagged with certain labels. Id like to discuss politics calmly but when the big words get trotted out I feel the need to defend myself.

Re: The Kolmogorov option

#407

Earlier quoted context omitted.

I actually think Galileo vs. religion is a quite fitting analogy for these kinds of issues today. There is a prevailing ideology that, if you are seen as disagreeing with it, and if you doing so is visible enough, can get you ousted from polite society. This ideology has a set of dogmatic views about the world which may not be challenged, some of which are seen as axioms of morality. Even though a good argument can b…

Fox News is a highly rated TV channel. Right-wing talk radio has tens of millions of listeners. Our national government is currently dominated by people who disagree with this supposed "prevailing ideology." I am baffled by the idea that liberals dominate discourse and conservatives have to hide in the shadows to avoid persecution. I see it in a lot of comments here, and half of Damore's manifesto revolved around it.…

> half of Damore's manifesto revolved around it

And, Damore is now interviewing with YouTube personalities claiming Google silenced him.

Meanwhile, every day are 4 new articles about his manifesto, which everyone can find fully linked up and formatted. There's nothing left to complain about. He's got the debate he wanted.

I feel like the strategy of denouncing media began with Trump. It's earned him tons of attention, and others have begun doing the same because they think it's true. They think their ideas are being drowned out.

I don't see that. I see tons of online discussions. From time to time, someone asks me to frame my point in a way that is more to their liking, and I'm baffled. That isn't how open debate works.

Re: The Kolmogorov option

#408
post #390

Earlier quoted context omitted.

Sure, I agree that he consciously talks about both preference and ability. Why is he talking about preference? Does that have to do with his thesis? I think it's because he wants to write sentences like that one about "preferences and abilities" in an attempt to pretend that arguments about preferences translate to arguments about abilities. That is, I would have agreed with a claim that he has two distinct theses (p…

> Why is he talking about preference? Does that have to do with his thesis? If preferences partially explain why psychology is dominated by women and why computer science is dominated by men, then preferences can partially explain part of why psychologists tend to be women and tech employees tend to be men. And that's a big part of the memo: why is there a gender gap, what causes it, etc. So it's not surprising that…

Preferences are important. I'm not disputing that. And there are definitely gender correlations in preferences (I agree with at least three of his points here, namely that women seek better work-life balance, that men seek high-status position, and that there's more work to be done in dismantling the patriarchy's oppression of men who don't conform to gender roles).

I am disputing that preferences are relevant to the business purpose of this memo, which was to claim that what Google does with applicants once they have applied is both unlikely to work and illegal.

I am also disputing, more directly, that he or anyone has shown that preferences are relevant to the gender gap for engineering roles at Google; I'd think that there exist both more qualified men and more qualified who are interested in working at Google than Google has positions for, and the difficulty is in identifying these people. This is supported by how the specific diversity initiatives that the author calls out are all about properly evaluating applicants, and none are about encouraging people to apply or have an interest in the field. (And anecdotally, as someone who received a job offer from Google in May and spent three months going through team selection and ended up accepting another offer, both my own experience and the stories of others I've talked to both inside and outside Google is that headcount is sparse.)

These two claims I'm making (that preferences are irrelevant, and that the memo does not argue that preferences are relevant) are definitely falsifiable, so I'm interested in evidence to the contrary.

Note that this is a different point from the gender gap in tech as a whole. I'd definitely believe that preferences are much more relevant there, especially if we count things like wanting better work/life balance (or wanting better maternity leave, etc.) as preferences.

Re: The Kolmogorov option

#409
post #347

Earlier quoted context omitted.

Nope, a "civil debate" about PHP vs. ASP is a waste of time. No reasonable employer will give you space for debate between the two. If there were a company-wide week-long debate about PHP vs. ASP at my employer, I would seriously question whether I should be at a company where these sorts of questions are considered debate-worthy, whether I have to justify every decision I make about common-sense things (why am I wri…

There's an important difference between silencing opinions that (you think) are very stupid for practical engineering reasons, and not wanting to work with the fools who hold them; and between silencing opinions because you think they are morally abhorrent and harmful just by being expressed, and punishing people for holding or expressing them. Morality is a psychologically distinct category. Being morally wrong is n…

I think that argues in favor of a swifter response to someone who is objectively wrong on a moral matter that's relevant to the company's business objectives than someone who's objectively wrong on a merely technical matter that's relevant to the company's business objectives, no? Maybe I'm misunderstanding you.

I mean, we could argue that this isn't a moral matter, but the fact is that a huge number of employees at the company do see it as a moral matter, and it affects everything from morale to recruiting to trust to retention. A warning is a luxury we have when this isn't true, but it's not one that's affordable here. To be clear, I am absolutely on board with giving him a generous severance (and a sinecure or PIP if needed to maintain immigration status or something), because I care very much about the wellbeing of all humans regardless of whether I find them to act morally. I'm not here to punish him for his immoral behavior. But the important thing is that he needs to be removed from day-to-day activities and decisionmaking at least as much as someone who seriously and earnestly advocated ASP would.

(In fact, one of the few scientifically-cited claims in the document is "moralizing things that shouldn't be moralized is bad," but there's no strong argument, scientific or otherwise, for "... and this is one of those things that shouldn't be moralized.")

Re: The Kolmogorov option

#410

Earlier quoted context omitted.

It's easy to identify a disaster with the safety and comfort of decades of hindsight, but rather less easy to see today's contemporary equivalents.

We're not talking about hindsight. The people who got excited in the 30s, then backed away, are easy to forgive. We're talking about people supporting soviet communism literally for decades after the evidence for the brutality required to sustain it was incontrovertible.

It's about the hindsight from your and my perspective, today.

Think about it: in another few decades people may (or may not) look back and wonder how intellectuals could support the system we have today despite of all the bad things happening.

Post reply on HN