srn

Yes, the first link addresses a strawman but I bring it up because other people still bring it up.

The second link discusses interests and influences for female undergrads majoring in some kind of engineering, not CS, though I still think it is somewhat relevant. I find the influence on family members to be the most interesting.

The third link has papers specifically related to interest. One is titled "The Anatomy of Interest." It discusses why women who started college enthusiastic about CS ended up leaving.

My personal observation is that last night I stayed up too late reading the linux kernel code. Probably won't help anyone that much.

I like this article because it points out women are also geeky and that it's OK to be geeky. Women just may tend to have a slightly different way of expressing that geekiness.

You want girls to feel comfortable because you want the girls who are good at CS to go into CS. This is independent of how many women go in total. But since we're at it, let's talk about the biology arguments:

http://geekfeminism.org/2009/10/17/how-does-biology-explain-...

http://www.mun.ca/cwse/Cannon,Elizabeth.pdf

according to the above study, undergraduate women in engineering showed comparatively a higher interest in math than men though slightly lower interest in engineering and yet lower interest in physics.

http://www.cs.cmu.edu/afs/cs/project/gendergap/www/papers/

Through additional work CMU increased their percentage of women in CS from 8% to 1995 to 42% in 2000. One thing I recall is that women were not as experienced when they first got into CS and had to work harder. However they believed they had caught up by junior year.

Check out http://www.opencores.org/

I play tabletop RPGs which could be considered fake achievement. The types I play however reward creativity and problem solving so I think they have merit.

Quake gave me a sense of direction when I was a teenager. Before that I had either walked very short distances or been driven everywhere. My sense of direction meant I could figure out that this street was parallel to that street etc. but it was not based on cardinal directions. When I started using a map during college was when my sense of north aligned with north.

In silicon valley it's interesting because people sometimes define north/south as parallel to the freeway north/south 101. It actually runs ~east/west where I am.

Carriers and HW companies may not like making side-by-side comparisons so easy. Alternately there should be names other than A,B,C which are not inherently ranked.

Someone (nokia?) has/had farms of different devices which you could test on remotely. However this is unlikely to be available to run-of-the-mill developers.

Agreed on the numbers.

Almost every web browser on the planet can run javascript, independent of operating system or underlying hardware. I don't think you can get more accessible than that

The hacker dojo may meet some of your needs.

Good compilers will warn you.

Extend this to functions as well.

I had a bug like

if (foo(b == A))

instead of

if(foo(b) == A)

No love for nokia n900 :( Though it is not out yet

http://maemo.nokia.com/n900/specifications/

Only comes unlocked in US, 3G only works on tmobile.

According to their research, women started out with much less confidence but by the junior year it has started to level out when the classes become more difficult for both genders. See section "Gap Between Perceived and Actual Ability" and "Confidence Gap Narrows"

Also see "Geek Mythology: Lore about Being in CS" specifically "It is important to note that most of the CS students (both male and female) we interviewed feel they do not match the stereotype: their interests are varied (including sports, theater, poetry) and not isolated to computer science. "

http://www.cs.cmu.edu/afs/cs/project/gendergap/www/papers/si....

This paper expounds more on the view that a student must be absorbed in computers 24/7 and that women's perception is that they must be this way to succeed: http://www.cs.cmu.edu/afs/cs/project/gendergap/www/papers/an... men who are not like that still feel like they can be successful in computer science under "The Non-Hacker Male" from that paper.

Based on this information, I believe the stereotypes are self perpetuating and breaking the feedback loop will help a lot. I ask that you personally help by keeping an open mind.

Well, obviously Dilbert is an extreme caricature of what happens in most offices, but the comic in and of itself is not a good emissary for the software engineering profession. It unfortunately may have unconsciously biased many people away from jobs in industry.

It's about Dilbert. If I have the skills to go into a profession where I'll be respected or into a profession where I'll be treated, well, shall we say not so well, which will I pick?

I would consider "bugs opened per month" to be a good metric, assuming someone uses the software.

Isn't apple just as bad with itunes and the palm pre?

http://www.technewsworld.com/story/67615.html?wlc=1256230659

Google isn't immune either

http://phandroid.com/2009/09/25/cyanogen-gets-cd-from-google...

None of these companies are golden. Buy what works for you.

Also from what I know, which is not much, you should be requesting the ports via request_region. If that fails you should probably not be writing to the ports and should also mean you're sharing the port with another module.

kernel newbies is theoretically a good place to go with these kinds of questions too.

More explicit handling of error conditions is also good FWIW - don't see you doing it here. The kernel is not forgiving of using resources your don't have or failing to free resources acquired.

This may not be likely, but - you don't specify a level for your printk statement. It could be your message is just not getting printed to console. Check out info for "/proc/sys/kernel/printk" as well as information about syslogd and klogd as the message may be logged to somewhere else.

I also would like to mention that in general memory barriers can be useful for working with i/o irrespective of your particular situation.

Second on unnoticed exclusionary behavior. Making sexist assumptions which are "reasonable" because there are not many women in the field may very well be self perpetuating.

Going back to the original topic, I think that being active in FOSS, even at low levels @ a couple hours a week, is a good way to keep evidence of some skillset. This should hopefully help getting back into the workforce.

Female oriented apps with what, pink unicorns and fluffy bunnies?

That's a pretty silly statement no matter how you take it. Remember the myth (or the myth of the myth) about nobody needing more than 640K of ram? There will always be opportunities for novel applications, more user friendly applications, applications which take advantage of new technologies, etc.

However if you do market research or usability testing, dependent on the use case please do with both genders and perhaps different ages - for example, soft thumbboards may be perfectly fine for someone with one size hands to use and ridiculously hard to use with another. That's an obvious example, and there are other things not so obvious that may come out from that.

Listen to music. Get the headphones that encompass your entire ear so you don't hear other people talk. I've done this for 10 years now. Also possibly get people to write you emails instead of dropping by the office even if you're 5 feet away. Work from home or from somewhere else occasionally, and work hours when other people aren't there.

I'll have to try applying some of these study techniques, thanks.

Wouldn't it be interesting if we taught people how to learn instead of just teaching them the material?

Sponsored locations.

I've seriously been thinking about such issues; there is a definitely a point of diminishing returns. Deciding when to cut your losses on maintaining code that is on the limits of your ability to follow is difficult.

What other unusual languages do people use for linux kernel modules?

This project is interesting, given that it's open source and you can dynamically add or remove units. As far as power goes, crank down the clock speeds and any processor is going to run cooler. arm is also well known for taking less power.

But about the whole changing your basic architecture, consider Tilera:

http://www.itjungle.com/tlb/tlb082107-story02.html http://www.theregister.co.uk/2008/09/23/tilera_cpu_upgrade/

And it runs linux! ;)Granted it's closed and probably costs a bazillion, but the big guys haven't exactly missed out on the memory bus being a bottleneck.