Having read the paper (https://gofile.io/d/WfcxoF), I can't take the idea that Google wanted to suppress it seriously (as the article argues).
In the paper, another paper is summarized describing the environmental costs of training large models. It is then argued that global warming will disproportionately affect marginalized people. This constitutes "environmental racism" because the main beneficiaries of these models are rich white people whereas the people bearing the costs are poor people who live near the equator. In my mind, this is two-points-make-a-line thinking.
The rest of the paper is more interesting. It argues that machine learning is conservative in that it "reinforces hegemonic language". I think this is true. Models are trained using data, data is what exists rather than what we would like to exist. If you're unhappy with what exists (as Gebru is, terribly), this represents a problem. Two of her solutions are "curating" data (censoring data) and "working with panels of experiential experts through the Diverse Voices methodology" (constituting powerful groups made up of her ideological allies). I think these solutions are dead-ends. It seems to me that we have to make peace with the conservatism inherent in these models (and maybe all models).
The paper also argues (I think rightly) that problems will materialize "should humans encounter seemingly coherent language model output and take it for the words of some person or organization who have accountability for what is said". In other words, we grant other humans responsibility for what they say and do but what about models that can mimic humans very accurately? A similar problem exists today when a Google or Youtube account is erroneously locked by some algorithm. There's a sense that, really, no one is responsible for this outcome because an algorithm did it. Even if the owner manages to get his account restored, there's no clear way to assign responsibility for the mistake. Perhaps this problem will become even worse if we start to see algorithms powered by models that are near-indistinguishable from people.
Anyway, the paper doesn't seem controversial to me. It seems clear that Gebru's strident tone and personal style were the cause of her firing/resignation. Jeff Dean's claim that the paper was rejected for not citing other papers may be false. The truth seems much simpler: based on her own public communications, Gebru felt she was the subject of "harassment" and "dehumanization" but those grievances seem to be hyperbole, and the personal injury that she felt became an excuse for her to treat her coworkers poorly. In other words, she didn't drop her twitter persona at work.
Coal mines should be unionized because the miners have their backs against the wall.
If I felt I was being under-compensated, working in poor conditions, and had my back against the wall, I'd want a union too. But describing my job like that is hysterically funny, much less Google programmers.
What's really going on here is simple. If you work at Google, your job is too good. There's no struggle. People want a struggle and so they invent one by LARPing as Marxists.
The idea of Google exploiting its programmers is an insult to actual exploited labor. These aren't coal miners. There's no factories.
The programmers where I work have too much power. We're exploiting the company. If I don't feel like writing code, I just do github stuff and answer questions. In fact that's what I did yesterday and the day before. No one notices or cares because, among other things, they couldn't replace me because there aren't thousands of hyper-educated new college grads after my job. I think many of the Google employees who are unhappy at Google due to "exploitation" would be happier working at a smaller, less famous company. Of course there's a pay cut and you lose that special feeling when everyone instantly recognizes you employer and assumes you're a genius.
Yes, the parent comment was about "lack of administrative rigor". The cheerful acceptance of that situation was the topic of my post.
There is no reason why someone can't support the welfare state and also demand efficiency. That demand represents an ideal; there will always be errors, but the insistence that all errors are unavoidable and that (as the parent suggested) we shouldn't even talk about them is extremely irritating. It provides the enemies of the welfare state with the best possible arguments ("you don't even care if the money is going to people who need it").
Because, as I said in my original post, "[shoddiness, etc] represents the best objections to the welfare state".
In other words, blithe dismissals like "I accept some inefficienies and waste" provide the other side with the best possible ammunition against the idea of a welfare state.
It's laughable to see the number of people here who, straight-faced, say nonsense like "Nitpicking efficiency is for the capitalists".
Ironically, the idea you're pushing, something like "this doesn't really matter, the important thing is the government giving people money, the more the better, who even cares if they're American," precisely represents the best objections to the welfare state. Which are its shoddiness, its lack of administrative rigor, and so on. These objections raise the question: can a welfare state be effective if it's administered so poorly? If the American government is sending money to Swedes, can we trust it to send money to "people who need it"?
Does Sweden send welfare checks to Americans who aren't Swedish citizens? If not, perhaps that partly explains why the Swedes are more comfortable with an expansive welfare state than Americans.
Just because you view all history as the result of a "secret cabal of evildoers" (in this case, plotting to preserve slavery) doesn't mean I have to.
As you let slip here, Horne's writings aren't about history: they're a form public of agitation so that "the USA will do less racist shit". Which would be fine if he was a social critic or a politician. But he's been passing himself off as a historian.
Horne is a clumsy propagandist and you've been thoroughly propagandized.
In this kind of history, all nuance (like Adams' views on slavery) is lost and all historical events are made to fit a predetermined narrative. In addition to being wrong, it's incredibly boring stuff.
The DHS was established in the panic following 9/11. If they're knocking on doors trying to intimidate dissidents, it seems clear that it has outlived its usefulness (if it ever had any) and should be dismantled.
You'd think that would be something small-government republicans progressive democrats could agree on...
> Do you? Not to over simplify things, but freedom, liberty, and equal treatment were pretty much empty words and applied only to white American men (see: treatment of Native Americans, the Mexican-American war, slaves, and women).
This is nonsense. The US was a left-wing experiment. It was a repudiation of the way things were done in Europe. The fact that it didn't solve every single inequity in one blow is not an interesting observation.
It's fair game to criticize the US, but to misunderstand its history to this extent is just sad. The US was radically egalitarian at its founding and served as the example for the rest of the world for at least a century and a half.
John Adams: "I always consider the settlement of America as the opening of a grand scheme and design in Providence for the illumination and emancipation of the slavish part of mankind all over the earth."
> Don't you see the contradiction there? First he says it isn't helpful if foreigners criticize Russia, because Russians need to learn to criticize themselves. But later he says that his work aimed to lead the Jews to examine themselves, even though he is an outsider to that people and, by his own declared standards, it is none of his business.
This only follows if we accept the axiom "Jews aren't Russian". I'm not convinced that he thinks that (and I certainly don't).
> Also, after the Holocaust it is extremely inappropriate for any non-Jew to suggest that the Jews as a people have done anything bad or negative. The sole appropriate thing Solzhenitsyn could have done with regard to the Jews is either say something nice or not say anything at all.
To me, this is intellectual cowardice.
As far as I know, Solzhenitsyn's controversial view about Jews in the USSR is that they were viewed as less likely to be counterrevolutionary, and therefore quickly climbed the party hierarchy...at least initially. Stalin eventually adopted various anti-semitic measures (and Solzhenitsyn mentions this in The First Circle).
Der Spiegel tries to pin him down as a reactionary Putinist and anti-semite. I think the attempt fails but judge for yourself. I don't think we have a category for people like Solzhenitsyn anymore but "jew-hater, authoritarian, reactionary" doesn't seem accurate to me.
Your post shows how significant events are stripped of their meaning by the dictum that there must be some material end behind every act. But it's a decision to look at history and explain everything in terms of geopolitics. Can't people get together and do something for the glory of doing it?
I think you're partly right but I also think there's something a bit deeper at work here.
These days science is viewed as a means to various ends. These ends are all wonderful...eliminate poverty, curtail climate change, cheaper energy, etc. But what's missing is the idea of doing something for the sake of doing it. It's not totally clear what landing on the moon or maintaining a space station really accomplished in terms of material goals. They're glorious accomplishments because of their difficulty.
I think that attitude is what's missing. Listening to JFK's "we will go to the moon" speech is almost unbelievable today. Politicians of either party absolutely cannot talk like that today.
I understand what you're saying but I disagree: atheism is not a religion. The verb "is" does not mean "fulfills the same function as". Just because atheists have to answer the same questions as theists does not make atheism a religion.
If you want a word to describe what atheism and religion have in common, I'd go with "spirituality". Not in the sense of "burning incense" or "believing in ghosts" but rather in the sense of holding beliefs that are not rational and deciding to care about the world and about humanity in particular.
In the paper, another paper is summarized describing the environmental costs of training large models. It is then argued that global warming will disproportionately affect marginalized people. This constitutes "environmental racism" because the main beneficiaries of these models are rich white people whereas the people bearing the costs are poor people who live near the equator. In my mind, this is two-points-make-a-line thinking.
The rest of the paper is more interesting. It argues that machine learning is conservative in that it "reinforces hegemonic language". I think this is true. Models are trained using data, data is what exists rather than what we would like to exist. If you're unhappy with what exists (as Gebru is, terribly), this represents a problem. Two of her solutions are "curating" data (censoring data) and "working with panels of experiential experts through the Diverse Voices methodology" (constituting powerful groups made up of her ideological allies). I think these solutions are dead-ends. It seems to me that we have to make peace with the conservatism inherent in these models (and maybe all models).
The paper also argues (I think rightly) that problems will materialize "should humans encounter seemingly coherent language model output and take it for the words of some person or organization who have accountability for what is said". In other words, we grant other humans responsibility for what they say and do but what about models that can mimic humans very accurately? A similar problem exists today when a Google or Youtube account is erroneously locked by some algorithm. There's a sense that, really, no one is responsible for this outcome because an algorithm did it. Even if the owner manages to get his account restored, there's no clear way to assign responsibility for the mistake. Perhaps this problem will become even worse if we start to see algorithms powered by models that are near-indistinguishable from people.
Anyway, the paper doesn't seem controversial to me. It seems clear that Gebru's strident tone and personal style were the cause of her firing/resignation. Jeff Dean's claim that the paper was rejected for not citing other papers may be false. The truth seems much simpler: based on her own public communications, Gebru felt she was the subject of "harassment" and "dehumanization" but those grievances seem to be hyperbole, and the personal injury that she felt became an excuse for her to treat her coworkers poorly. In other words, she didn't drop her twitter persona at work.