Respectfully, I think you may have missed the point of the article. In the committee hiring process that Facebook uses, it would seem that every candidate was actually not being judged purely on their capabilities. So telling folks to "distinguish yourself on your capabilities," while generally good advice, doesn't actually help if the process doesn't reward that.
If instead, as the recruiters in the article suggest, pedigree of college and other factors are being prioritized, then that's a problem. Criteria like this disproportionately disadvantages URM candidates in particular, which is how Facebook ends up with only 1% black and 3% latino tech workers, even after prioritizing diversity at the recruiter level.
> I don't think the decision makers are racist or sexist.
I recommend you read up a bit on [individual vs. systemic racism](1). An important point to note is that even if individual actors don't exhibit what you consider to be racism, the system that they make up can itself be racist. This is a really important point that often gets lost in discussions like this on HN, especially since the largely-white largely-male population here is unlikely to ever directly feel the effects of systemic racism.
> Trying to make it fair for one demographic will likely end up making things unfair for another.
Just to be clear, I think what you are trying to say is that adjusting for bias against URM candidates will inherently lead to bias or disadvantages to white folks. I would question any notion of "fairness" that relies on another group being treated unfairly. To borrow a common expression, when you’re accustomed to privilege, equality feels like oppression.
Genes may play a factor, but they are certainly not the most important one, or even a significant one in this context. If you grow up in a poor family / poor neighborhood, you won't have access to the same education as those who grow up in middle-class areas. Education is far more important to what we consider "intelligence" in this context. Basic math, financial literacy, or even reading comprehension is far more a result of your education than it is of your genetic material.
It's easy, as a product of good education and a safe environment, to look at people in poverty and say "I guess they're not as smart as me." This hides a far more uncomfortable truth, which is that everyone who doesn't grow up in poverty has access to better education, better opportunities, and a better environment than those in poverty. Try as you might to claim that people in poverty are just "stupider", it's simply impossible to make this comparison given that people on opposite ends of the financial spectrum grow up in totally different environments. It's short-sighted to attribute this to genetics and nothing else.
I'm just going to hazard a guess that you didn't grow up in poverty, am I right? If you had, I'm sure you would have a very different idea about how difficult it is to grow up without all the advantages of a middle-class upbringing.
I don't think it's correct to say one way or the other what the critical driver is, as to my knowledge we don't have the necessary data to make such statements. Both genetics and environment likely play a role. We as a society have decided that it is unethical / not in our societal best interest to perform the experiments necessary to further investigate the mechanism, and so we should focus our efforts on alleviating environmental drivers of the problem. With that being said, to write off theoretical heritable causes of poverty or low intelligence just because they makes us uncomfortable would be bad science. We should be transparent about the limitations of our data and admit that as a society we are content with not knowing the answer to this question.
The other issue is that society has become so detached from nature that a lot of people are completely oblivious. If you breed animals, questions like this are pretty obvious and easy to answer.
I grew up in a reasonably poor household. We certainly weren't wealthy. I couldn't care less about poverty or not poverty, it's irrelevant. I was incredibly lucky to have good parents, with 'smart' genes.
If anything, I could see Uber being one of the first adopters of self-driving cars when they're ready. They're already investing in it heavily, hiring away 40 robotics researchers from CMU to focus on it. [1] They want to corner the market, get everyone using Uber, and then eventually make the change to self-driving vehicles when they're ready.
(Full disclosure: I'm the developer for this site).
You make a really interesting point, and I think the reason for this is that each college's solution is different in tiny ways. In my opinion this is not an insurmountable problem, as seen with Facebook's success. Clearly not every single person has the same exact needs with a social network, but the key is 1. Facebook is typically "good enough" for most people and 2. Everyone uses Facebook, so people look over tiny inconsistencies with their own usage to be on the same network as their friends.
However the difference, as I see it, between Facebook and something like a Craigslist for campus, a campus chat board, or campus event system is that there is only pressure to use it within distinct communities. For instance, if our site grows to dominate this space at Penn, then only students at Penn feel the pressure to use the site. A student at college X will have different classes, different events, and different needs, so even if all his/her friends at Penn use it, they would get no utility out of hopping onto the College X version of our site, if no one at College X used it.
What we have there is a situation where instead of Facebook's virus-like growth, you have distinct bubbles that need extensive, personal attention to break into. You can't just launch to a new college and immediately have users rolling in, you need a personal "in" for each college, someone (ideally multiple someones) to get real students to use it. This, naturally, requires much more effort and a lot of connections, and it's to this that I'd attribute your observation that no one product has grown to dominate this space.
For now, we're considering our site a simple experiment for Penn, and if we do end up expanding, it would only be if we could justify-—with profit from the site—-the effort needed to expand to other colleges.
Yeah, this is something we're concerned with as well. The reason we have it the way it is now is because we're concerned about having too many category-specific criteria. For instance, price would be nice to show on the front page, but we're hesitant to put a price input on the new post form, because it wouldn't be relevant to many posts (for instance posts about events or groups on campus).
For now, we're focused on keeping the posts as general as possible with only Title, Text, and Category options. In future I could see introducing different parameters for individual categories (price for selling something, location for events, etc.) but for now our priority is in making the new post process as frictionless as possible.
Thanks for the input though! I really appreciate another pair of eyes on a new product.
I'm the developer for this site, and we're using Node.js w/Express, MongoDB, and deployed on Heroku. Mongo might not scale super well for this kind of thing, but for ease of development you really can't beat it.
Yeah we found that library posted on HN a few days ago, and figured since we have very low traffic right now we'd have nothing to lose with a library like this (even with the additional requests it sometimes incurs when a user hovers but doesn't click). It's so fast now I'm not sure I could ever bear to build a site without it ever again.
Just testing that site (thanks to maxucho for providing the example :) ). Interestingly, even if I hold my mouse button down for a second or two before releasing it, there's still a perceptible loading time of maybe 500ms - 1s.
Not sure if that's expected as part of the design?
Hmm that could be just an issue on my site, not necessarily with the plugin. I've been experiencing some strange load times with it even before using InstantClick.
If instead, as the recruiters in the article suggest, pedigree of college and other factors are being prioritized, then that's a problem. Criteria like this disproportionately disadvantages URM candidates in particular, which is how Facebook ends up with only 1% black and 3% latino tech workers, even after prioritizing diversity at the recruiter level.
> I don't think the decision makers are racist or sexist.
I recommend you read up a bit on [individual vs. systemic racism](1). An important point to note is that even if individual actors don't exhibit what you consider to be racism, the system that they make up can itself be racist. This is a really important point that often gets lost in discussions like this on HN, especially since the largely-white largely-male population here is unlikely to ever directly feel the effects of systemic racism.
> Trying to make it fair for one demographic will likely end up making things unfair for another.
Just to be clear, I think what you are trying to say is that adjusting for bias against URM candidates will inherently lead to bias or disadvantages to white folks. I would question any notion of "fairness" that relies on another group being treated unfairly. To borrow a common expression, when you’re accustomed to privilege, equality feels like oppression.
(1): http://www.ucalgary.ca/cared/formsofracism