4 ms·
*backed up with heavily contested social science references
by cookie_monsta 7y ago
*backed up with heavily contested social science references
- tenpies 7y agoAnd refuted not with "uncontested" social science (an oxymoron if there was ever one), but with job termination. We had someone clearly trying to have a constructive discussion on a topic and he was treated as if he had written the manifesto for the Fourth Reich and was demanding all women be sent back to the kitchen.
- krapp 7y agoPlenty of science-based refutations of James Damore's manifesto can be found in HN discussions and throughout the web if you care to look. No, I am not going to provide a canonical list here.
- krageon 7y agoThere is a time and a place for "constructive discussion" of certain topics and that place is apparently not Google. It might be regrettable that he (supposedly) does not understand it, but really it is not a surprise to me and should not be a surprise to any working individual that your workplace is not a space in which you discuss things of a sensitive nature (especially not if it is very visible). This is the case in almost every large corporation, so the fact that this specific incident is so widely discussed is very strange to me.
- xyzzyz 7y agoIt is definitely very much fine to discuss things of sensitive nature at Google, and Google has been explicitly very much encouraging of that. The issue was that only some of the opinions on the issues of sensitive nature were allowed, but it wasn't made clear by written company policy, only by enforcement.
- krageon 7y agoCorporations frequently embellish the truth. Let us not pretend we are so naive that we do not understand that this is the case. Care should be taken when talking about things that people feel strongly about, especially if those people are your boss. These are hard lessons that everyone who works at a big corporation at a level above janitor has learned within the first year of working there, so to me arguments along the lines of "but they said it would be different this time, they promise" just seem ill-advised to me. As if you are sticking your head in the sand and proceeding as if you can't depend on your own good thinking.
- dragonsngoblins 7y ago> There is a time and a place for "constructive discussion" of certain topics and that place is apparently not Google I'd agree with you if he wasn't discussing it in direct reference to how Google approached things like hiring from a strategic perspective. However what he discussed was in the context of how google approaches something. Secondly from what I've heard Google provided a space for precisely these kinds of discussions, so suggesting he should have known better than to discuss seems pretty unfair to me. Even if what he said was wrong, wrong, wrong, it really seems to me like he was trying to engage in good faith and people didn't respond to it in good faith at all
- krageon 7y ago> hiring from a strategic perspective There is a reason people who make actual strategic decisions speak as if they are eels covered in oil and it is to avoid situations like the one this person found themselves in. > Google provided a space for precisely these kinds of discussions, so suggesting he should have known better than to discuss seems pretty unfair to me. Yes, corporations embellish the truth (or lie, if you want to be that way). At this point, you need to understand this - especially if you work for a corporation that does as many questionable things as Google does. Everyone should be free to make their own decisions of course, this person decided to make an ill-advised argument at work and the consequence was that he cannot work there anymore. > it really seems to me like he was trying to engage in good faith Regardless of whether or not I believe this is true, "engaging in good faith" is not a shield against poor decisions.
- lordCarbonFiber 7y agoNot to mention math so bad he wouldn't pass a high school stats class. Sampling from a highly cross correlated population expecting to see population-wide gausians; amateur hour before you even got to the phrenology and the misapplication of personality theory. I feel like 90% of people didn't read the drivel he wrote; I'd have fired him purely for the fact anyone that bad at stats (and proud of it) shouldn't be anywhere near a machine learning team.
- dragonsngoblins 7y agoOr they also aren't great at stats. Stats is surprisingly difficult to get right