4 ms·

It occurred to me recently that if the goal is to eventually get to gender equality, female focused tech groups should set the proportion of their tickets avail

It occurred to me recently that if the goal is to eventually get to gender equality, female focused tech groups should set the proportion of their tickets available at x% female, y% male where x and y are the inverse of their proportions at some equivalent male dominated group. Over time, this policy should pull men and women together instead of apart. What good is it if women only know other women and men only know other men?

soneca 13y ago
I do think that there is good in a only women conference. But i also do think that your suggestion is incredibly smart. Minimize criticism of those who think this is sexism, while keeping its goal of both making a statement that there is inequality and providing inspiration to women - and still giving a chance to a few men understand how difficult can be to be on a environment dominated by the other gender.
001sky 13y ago
Nothing is wrong with your point, but you are applying a 1->N scaling strategy to a 0->1 problem. That is to say, this type of event is brings people together because n=small, and the act of concentration brings more scaling effects that a larger act of dispersion. Over time, this will shift...and your preferred methods will make more sense in a context where dispersion is more important than concentration. But it is a good point to make that neither one is perfect or viable as a standalone method.
- wellboy 13y ago
  Very technical analysis, but it actually makes sense! :)
- thatthatis 13y ago
  I disagree slightly with your categorization of the problem. I think your characterization, while accurate in itself, is missing the "otherizing" and "in-grouping" that happens if we only foster no-boys-allowed special interest groups. Lets say a given women in tech event is phenomenally successful, that means that men are now precluded from attending a key event. Whether consciously or unconsciously men understand this and have a discentive to making the group phenomenally successful. The more successful it is, the worse the world becomes for the average man's career. When it is x%/y% it puts things into a different category. In x%/y% all success is shared. 1) Men interested in attending these female focused events would have a vested interest in increasing both the size of the female event (-> more male tickets) and increasing the % of women at other events (-> more male tickets). If I, as a man, wanted to attend the YC female founders conference as it was just held, there is no way I can - so it isn't worth me spending any time on. If however, the event had an x%/y% structure, one of the ways I could gain admission would be to actively bring women into tech (I assume this would be a large part of the selection criteria by which the small number of men get admitted). Everyone's interests are now aligned. 2) The connections formed between the small number of men and the large number of women at x%/y% female events are connections that will act as a bridge to getting more women at the regular events. It's well demonstrated in organizational behavior that even small amounts of conscious cross department interaction leads to greater sharing of ideas and follow on connections in both groups. 2b) I'm concerned that most female focused groups are only creating in-group connections. Meaning the approach they're taking leads to B0->B1 and eventually B1->BN (where the existing tech scene is already at AN as a boy's club). Which is to say, they're creating a silo of connections. We shouldn't be looking at the problem as "How do we create the best shadow tech scene for women." But rather, "how do we make the tech scene gender neutral." 3) By making these groups x%/y% you put men in the tech scene in a more acutely aware state. "You can't go in there, it's women only" is easy to parse as a group of 'others.' Human psyches are pretty good at dealing with blanked exclusions -- its a rival tribe, which is fine, I have my own tribe. The situation of "You could go in there, but it's very hard to get in because they only let in x% men" makes the problem tangible, evaluative, and solvable. Long story short on this point, I think with x% men, men feel the exclusion in a stronger and more productive way. Men would feel excluded in a way that is more productive for society if events were x%/y%. In the current state of affairs, nobody can write a blog post titled "How I hacked my way into a men's ticket to the YC women's conference." 3b) Right now 90% of the tech scene has no skin in the game for getting in more women. Some may have philosophical beliefs or take a very long term view and thus work towards female events happening anyways. But, x%/y% would mean we can a much larger portion of the 90% working on the problem in a way that has tangible near term benefits for everyone. Finally, we as a society only really permit discriminatory subgroups like this when the size of the subgroup is very very small. "Men in teaching," "Women in tech." By setting up "women in tech" groups as permitted discriminatory subgroups we are betting against these groups ever achieving parity. Because, when they do achieve anything even approaching parity they'll have to reconstitute. Think League of Women voters -- they're now a league of concerned voters of all genders, but their historical permitted discriminatory subgroup focus hampers their activities to this day. x%/y% means the group will never have to go through an identity crisis or a period where it is in the grey area of whether it is still acceptable to keep their historical membership barriers.
  - 001sky 13y ago
    Lets actually look at a more realistic example. Women in CS and Engineering today might be ~10-20pct of FTEs. That means, in a medium sized shop, where there might be 20-30 engineers, you have a situation whee how many are women? 2-6? That sounds about right. Now, if you are in this position...the issue is that the guys have 18-24 FTEs amongst their co-hort. In order for the girls to get the same type of network-density you need to bring together the equivalent of 5 to 10x shops together. That is just simpe seeding of network effects. If you worked in a shop with only 2 or 3 other people who did your job, the "law of small numbers" means your chances of leveraging this social infrastructure are "hit or miss". There are too many simple variables (personality type, work habits, social background, etc) for an effective "matching problem" to be solved. Or use another example: would you go to a school with only 2 or 3 students in your major? Most people would see that as a red-flag. Social isolation is at odds with the type of network effects most people seek to develop at school. MBAs for example typically invest $250k to buy a commodity education, paired with a very exclusive rolodex. The value of this degree would almost be worthless in a school with 3 other students (regardless of the teaching). This is quite a different argument from lowering standards or affirmative action or whatever you want to call it. In fact, it is precisely the strategy YC used to bring its co-horts together. Back in 2005, VCs did not back many founders without an advanced degree (MA/PhD or MBA). As a result, it made sense to bring together (concentrate) a cohort, because the general peer group of VC backed CEO's would not have much in common with a 22 year old with no job experience. By bringing folks together that face similar challenges, you are not (per-se) giving these people special treatment or lowering standards; you're just providing a level os support infractructure that lets them "self-help" there way through many problems. It is actually critical that you don't "dumb down" standards, for this to work.
    - thatthatis 13y ago
      I don't see how you can suggest that the scenarios I'm discussing are unrealistic. This thread is in response to a concrete female only conference of exactly the type I'm discussing. They don't fit the model(s) you're proposing as well, but that makes them different cases, not unrealistic ones. I think we are talking about a fairly different types of things. I'm talking about low-touch high n networking as you find in events, conferences, and meetups; I am not talking about high touch networking like accelerators and degree programs. Whether my suggestions are valid or not in the high touch scenario you're discussing, I don't know, as I haven't considered that case in sufficient depth. Your suggestion of there being value in a separate network hinges on, I think, there being a fundamental incompatibility between the groups. 22yo vs 32yo MBAs have real structural differences in their lives that make divided self help a plausible solution. I just don't buy that the differences we're talking about are significant enough to create fundamental incompatibilities. "The female tech network" vs "The tech network" smacks of separate but equal to me. I have no problem with female focused accelerators, but remember that YC doesn't systemically exclude MBAs to create a network of 22y.o.s. I'm enjoying the discussion (minus the part where you try to dismiss my cases as unrealistic) as you're bringing up models and cases that I hadn't yet considered. Is the goal to have two networks, or to achieve gender neutrality in the existing meta-network? I think our goal is gender neutrality.

soneca 13y ago

I do think that there is good in a only women conference. But i also do think that your suggestion is incredibly smart. Minimize criticism of those who think this is sexism, while keeping its goal of both making a statement that there is inequality and providing inspiration to women - and still giving a chance to a few men understand how difficult can be to be on a environment dominated by the other gender.

001sky 13y ago

Nothing is wrong with your point, but you are applying a 1->N scaling strategy to a 0->1 problem. That is to say, this type of event is brings people together because n=small, and the act of concentration brings more scaling effects that a larger act of dispersion. Over time, this will shift...and your preferred methods will make more sense in a context where dispersion is more important than concentration. But it is a good point to make that neither one is perfect or viable as a standalone method.

wellboy 13y ago
Very technical analysis, but it actually makes sense! :)
thatthatis 13y ago
I disagree slightly with your categorization of the problem. I think your characterization, while accurate in itself, is missing the "otherizing" and "in-grouping" that happens if we only foster no-boys-allowed special interest groups. Lets say a given women in tech event is phenomenally successful, that means that men are now precluded from attending a key event. Whether consciously or unconsciously men understand this and have a discentive to making the group phenomenally successful. The more successful it is, the worse the world becomes for the average man's career. When it is x%/y% it puts things into a different category. In x%/y% all success is shared. 1) Men interested in attending these female focused events would have a vested interest in increasing both the size of the female event (-> more male tickets) and increasing the % of women at other events (-> more male tickets). If I, as a man, wanted to attend the YC female founders conference as it was just held, there is no way I can - so it isn't worth me spending any time on. If however, the event had an x%/y% structure, one of the ways I could gain admission would be to actively bring women into tech (I assume this would be a large part of the selection criteria by which the small number of men get admitted). Everyone's interests are now aligned. 2) The connections formed between the small number of men and the large number of women at x%/y% female events are connections that will act as a bridge to getting more women at the regular events. It's well demonstrated in organizational behavior that even small amounts of conscious cross department interaction leads to greater sharing of ideas and follow on connections in both groups. 2b) I'm concerned that most female focused groups are only creating in-group connections. Meaning the approach they're taking leads to B0->B1 and eventually B1->BN (where the existing tech scene is already at AN as a boy's club). Which is to say, they're creating a silo of connections. We shouldn't be looking at the problem as "How do we create the best shadow tech scene for women." But rather, "how do we make the tech scene gender neutral." 3) By making these groups x%/y% you put men in the tech scene in a more acutely aware state. "You can't go in there, it's women only" is easy to parse as a group of 'others.' Human psyches are pretty good at dealing with blanked exclusions -- its a rival tribe, which is fine, I have my own tribe. The situation of "You could go in there, but it's very hard to get in because they only let in x% men" makes the problem tangible, evaluative, and solvable. Long story short on this point, I think with x% men, men feel the exclusion in a stronger and more productive way. Men would feel excluded in a way that is more productive for society if events were x%/y%. In the current state of affairs, nobody can write a blog post titled "How I hacked my way into a men's ticket to the YC women's conference." 3b) Right now 90% of the tech scene has no skin in the game for getting in more women. Some may have philosophical beliefs or take a very long term view and thus work towards female events happening anyways. But, x%/y% would mean we can a much larger portion of the 90% working on the problem in a way that has tangible near term benefits for everyone. Finally, we as a society only really permit discriminatory subgroups like this when the size of the subgroup is very very small. "Men in teaching," "Women in tech." By setting up "women in tech" groups as permitted discriminatory subgroups we are betting against these groups ever achieving parity. Because, when they do achieve anything even approaching parity they'll have to reconstitute. Think League of Women voters -- they're now a league of concerned voters of all genders, but their historical permitted discriminatory subgroup focus hampers their activities to this day. x%/y% means the group will never have to go through an identity crisis or a period where it is in the grey area of whether it is still acceptable to keep their historical membership barriers.
- 001sky 13y ago
  Lets actually look at a more realistic example. Women in CS and Engineering today might be ~10-20pct of FTEs. That means, in a medium sized shop, where there might be 20-30 engineers, you have a situation whee how many are women? 2-6? That sounds about right. Now, if you are in this position...the issue is that the guys have 18-24 FTEs amongst their co-hort. In order for the girls to get the same type of network-density you need to bring together the equivalent of 5 to 10x shops together. That is just simpe seeding of network effects. If you worked in a shop with only 2 or 3 other people who did your job, the "law of small numbers" means your chances of leveraging this social infrastructure are "hit or miss". There are too many simple variables (personality type, work habits, social background, etc) for an effective "matching problem" to be solved. Or use another example: would you go to a school with only 2 or 3 students in your major? Most people would see that as a red-flag. Social isolation is at odds with the type of network effects most people seek to develop at school. MBAs for example typically invest $250k to buy a commodity education, paired with a very exclusive rolodex. The value of this degree would almost be worthless in a school with 3 other students (regardless of the teaching). This is quite a different argument from lowering standards or affirmative action or whatever you want to call it. In fact, it is precisely the strategy YC used to bring its co-horts together. Back in 2005, VCs did not back many founders without an advanced degree (MA/PhD or MBA). As a result, it made sense to bring together (concentrate) a cohort, because the general peer group of VC backed CEO's would not have much in common with a 22 year old with no job experience. By bringing folks together that face similar challenges, you are not (per-se) giving these people special treatment or lowering standards; you're just providing a level os support infractructure that lets them "self-help" there way through many problems. It is actually critical that you don't "dumb down" standards, for this to work.
  - thatthatis 13y ago
    I don't see how you can suggest that the scenarios I'm discussing are unrealistic. This thread is in response to a concrete female only conference of exactly the type I'm discussing. They don't fit the model(s) you're proposing as well, but that makes them different cases, not unrealistic ones. I think we are talking about a fairly different types of things. I'm talking about low-touch high n networking as you find in events, conferences, and meetups; I am not talking about high touch networking like accelerators and degree programs. Whether my suggestions are valid or not in the high touch scenario you're discussing, I don't know, as I haven't considered that case in sufficient depth. Your suggestion of there being value in a separate network hinges on, I think, there being a fundamental incompatibility between the groups. 22yo vs 32yo MBAs have real structural differences in their lives that make divided self help a plausible solution. I just don't buy that the differences we're talking about are significant enough to create fundamental incompatibilities. "The female tech network" vs "The tech network" smacks of separate but equal to me. I have no problem with female focused accelerators, but remember that YC doesn't systemically exclude MBAs to create a network of 22y.o.s. I'm enjoying the discussion (minus the part where you try to dismiss my cases as unrealistic) as you're bringing up models and cases that I hadn't yet considered. Is the goal to have two networks, or to achieve gender neutrality in the existing meta-network? I think our goal is gender neutrality.