Hacker News Reader

Comment by UnGravitas | Hacker News Reader

I'm not sure if we read the same article, there seem to be several critiques:

1. Silicon valley imagines itself to be a meritocracy which is a harmful belief as this is not actually the case.

2. The pipeline problem is not the entirety of the problem and overstating it risks ignoring problems which people in tech can solve more easily.

3. Power and money are disproportionately in the hands of white men

Engaging with her points rather than ad hominem attacks and responding to tone might result in an interesting discussion.

View on HN

My point is that she doesn't go beyond stating her views (and it's not like she has done anything of value that would make me respect her views on their own, sans proof).

Let's say I believe something that most people do not. If I want them to give me a chance, perhaps I should start with demonstrating why my theories are valid. If you read 'The Structure of Scientific Revolutions', Kuhn describes the process by which a scientific paradigm changes. And that starts with observations that do not fit with currently established models (as opposed to upstart scientists showing up and labeling everybody else a racist idiot).

Ie for every point you raised, where is the data that demonstrates that commonly held beliefs (that she disagrees with) are wrong?

Linked to directly from the article:

1. http://www.google.com/diversity/at-google.html#overall

2. http://www.ncwit.org/blog/did-you-know-demographics-technica...

Unless one assumes that industry staffing should closely correlate with overall country demographics, I don't see its relevance.

Ie that's like starting a scientific paper that aims to demonstrate that g=5 (and not 9.81) with the opening 'Let's assume that g is equal to 5'.

I'll give you an idea of what I am talking about. Let's say we had a few cases of women / minority staffed/founded companies that just blew it out of the park. So one could point at them and say - 'Look at how these guys did when not encumbered with white patriarchy'. That would be example of data that would challenge the notion that SV is meritocratic .

Or another example - showing how top CS programs have 50% graduates who are women / minorities but Google only having 10% as developers. That would also mean that something is amiss - why isn't Google hiring them?

> Unless one assumes that industry staffing should closely correlate with overall country demographics, I don't see its relevance.

Well, all that takes is 2 assumptions:

(1) natural talent is evenly spread throughout the population.

(2) SV is a meritocracy where advancement is based on talent.

That would naturally lead to a SV population that mirrors the population as a whole. Since it very much does not, one of those assumptions must be false. There's no real evidence that (1) is false, and so thus (2) must be false. It's pretty simple.

"There's no real evidence that (1) is false"

Since there is ample difference in results, you actually have to demonstrate that natural talent is evenly spread throughout the population, as opposed to waving the magic wand 'I want to believe it, so it must be true'.

Let's step away from SV and look at academia - who gets CS PhD's?

http://cra.org/uploads/documents/resources/taulbee/CS_Degree... - page 5. 86% male, 80% White or Asian (and 8% foreigners - who are probably mostly Asian). So roughly 87% for both.

There is also research (sadly paywalled) that considers whether PhD attainment can be explained by factors other than natural ability (it can't)

http://www.sciencedirect.com/science/article/pii/S0272775797...

I think one of the problems reading the critique is that her points are buried in too much ranting and wild exaggerations, so it's off-putting.

The approach is good for politicking (soundbites and getting the choir feverish) but it takes from presenting reasonable ideas for regular people to consider and process and say, yeah, that makes sense, or no, that's not quite the way i see it.

That said, one approach garners thought conversation, the other garners lots more clicks.

Reply on news.ycombinator.com