That's right, unionize. So instead of overworked Silicon Valley, we end up with collapsed Detroit. That's best case; the more likely case is that the companies will find no shortage of non-union labor.
A meeting at the middle is eventually going to happen. Sucks for comparatively wealthy westerners, but it sure is great for all those poor people around the world being uplifted from poverty.
Unfortunately, wrong. The actual middle is still abject poverty. No sooner will those poor people be uplifted than the next billion will drag them down.
But here's an issue. What if we do all this, what if we add gender sensitivity to the qualifications for men, and start making programming a mandatory class for women, and not allow men to take the classes, and we still don't have gender equity?
My guess: If you were looking for a course of action to make programming even less attractive to women, you'd be hard-pressed to beat that one. But then, as I said on the other thread, I've been seeing women-only sections of such courses fail for a long time; the women who actually take the sections find them useful, but they can't get enough women to fill them.
I know this flies in the face of all allowed opinion, but given some of the feedback I've received, the young ladies do not particularly like the idea of the late hours, the living on takeout food, and the lack of a regular schedule.
I think this one may be spurious; few people, men or women, like that sort of thing. Yet as you point out, women go into veterinary science, and nurses are overwhelmingly women. Both fields involve late and irregular hours (much worse than software in many cases). In other words, it might be more informative to ask why women go into other difficult fields rather than software, instead of asking why they don't go into software -- any signal in the latter question might be drowned out by the noise of people who just want an easy job.
Anyhow, after all those years of trying, and actually seeing less female participation, while the workplace had been scrubbed of sex, and the only people talking about sex were the women, I have to say that despite the willingness of some to point out dongle jokes, this one might not be all men's fault. And when you don't get to the real problems, you don't fix them.
Indeed. I suspect the current "men in tech are pigs and that's the issue" doctrine is actually driving more women away from the field than any actual man-pigs in tech. If the #1 message young women hear about tech, from other women, is that it's filled with misogynistic pigs (whether this is true or not), why would they even consider the field?
But you do need some calculus for computer science. Not all those integration techniques (thankfully I've never had to do an integration by parts in my professional career), but things like limits and rates of change and logarithms and the relation between polynomial powers.
BTW, integral of sqrt(1/csc(x)) = integral of sqrt(sin x), which isn't expressible in terms of elementary functions.
Remember, arbitrage is only legal when dealing with intangible financial instruments. Arbitrage with actual products is gauche and therefore punishable.
After all, a sufficiently smart compiler, given the assumption that integer overflows are impossible, should optimize out those pre-tests anyway, by virtue of the fact that you're about to add them, and integer overflows are impossible.
Fortunately, that particular optimization is not allowed.
Maybe we need a construct which catches overflows, something like
ifoverflow(arith_expr) {...} else {...}
If any overflow occurs within the protected expression, any assignments made are set to implementation-defined (not undefined) values.
ifoverflow(c = a+b) {
Put this one with practical commercial fusion, economical solar energy, and flying cars. It wasn't that long ago Slashdot was all excited about "anything to oil"... went nowhere, of course.
It has been often pointed out that the top North American and European female athletes in many sports currently have better performance statistics than the top males in the same sport 50 or so years ago. This supports the claim that the differences are primarily of social origin, not genetic.
ROTFL. That's one of the dumbest things I've read on slashdot. Well, dumbest that actually formed a coherent thought, anyway.
wait to see if this increases the number of women taking these courses and going into CS.
Why wait? I've seen these sorts of initiatives -- including women-only sections of classes -- for almost 30 years. They never work. This won't either. Percent of women in CS degrees peaked in 1984; everyone keeps trying to look for the reason for the downslide, but IMO what they really should be doing is looking for the reason for the increase in the first place.
There IS a big push, especially in nursing and education
There's a push for men in nursing. However, it doesn't seem to include blaming the situation on female nurses driving men out of the field, nor does it go so far as to attempt to exclude women from any nursing programs.
Education? Ha. Token. It's still assumed that any guy interested in elementary education is a pedophile.
Perhaps the problem is that standard C isn't expressive enough to express some operations that some programs require, especially with respect to detection of integer arithmetic overflows.
Indeed; the compiler's even allowed to assume signed integer overflow doesn't happen, which is where you get into trouble. Yet we have this perfectly good mechanism for detecting integer overflow (condition codes) and no way to reach them from high level languages (C isn't unique in this respect)
This should not be about a blame game, but instead about trying to discover the causes and trying to fix them.
Discover the causes, sure. Try to fix them? Might want to know what they are first. But that's not what this is about. It's already been decided that any disparity is due to bias, and therefore when disparity is found either one must find the male chauvinist pigs and root them out (basically Politically Correct Reign of Terror), or one must apply sufficient bias in the other direction to cancel the bias towards evil White Males (oh yeah and Asian mlaes too). In the US, the second tends to fall afoul of the same laws which prevent discrimination against minorities, so there's a lot more push for the first.
And the figures seem to show that this bias has been changing over time, we have far fewer women in computing jobs than we had a few decades ago.
You see, you've already assumed that the disparity is due to bias.
Most of them just go off and get other jobs. They realize that they don't want to work in that kind of environment and find a better one.
Right. Maybe they go into nursing, where they have to deal with nasty bodily fluids and asshole doctors (of both sexes) all day. Or perhaps into marketing or advertising, where the men are always perfect angels. Or even finance, which is completely free of douchebag assholes making sexist remarks.
The narrative that the tech workplace is an especially hostile to women just doesn't hold up. As for slashdot comments, trolls will be trolls, and Slashdot is not the workplace.
That's a nice theory, but you haven't addressed the elephant in the room: virtually all resumes for tech jobs are from white men. I don't even know how they came up with this 35% number.
No great mystery there; Slashdot got it wrong. Yahoo's Tech is only 15% female (globally), they didn't release US-only number.
Imagine not being able to communicate your fears, frustrations, desires, etc to anybody... ever. This drug might offer a path to reduce their symptoms to the point that they could communicate and possibly function in neurotypical society.
Joke's on them. Once they have the ability, they'll find out that no one cares about their fears, frustrations, or desires anyway.
AP tests are made to get you college credit, but many CS programs won't accept AP credit to fulfill requirements in CS. So there's not a lot of point for a student wanting to become a CS major to take the AP CS test.
Also note it is (or was, it's been a while) possible to take the AP test without taking the AP class.
Ugh, noticed a problem right after posting. I should be comparing P(sick | immunized) to P(sick | !immunized), not to P(sick). P(sick | !immunized) = (0.15 / 0.07) * P(sick) P(sick) = (0.07 / 0.15) * P(sick | !immunized) P(sick | immunized) = (0.85 / 0.93) * (0.07 / 0.15) * P(sick | !immunized)
= ~43% * P(sick | !immunized) or about 57% better chance of not getting sick if you're immunized. That's much better but still not great. It's supposed to be over 80% effective.
With the opposite assumption of unknown status, it's (89/93) * (7/11) = 61%, reducing your chance of getting sick by 39%, which is quite poor.
Testing the sensitivity to assumption of those immunized in the population, if 95% are immunized you get (85 / 95) * (5 / 15) = 30%, a 70% effective vaccine (still not great).
Only if the percentage immunized in the population is 97% or better does the vaccine meet the 80% standard.
So the vaccine is probably crap, but not as crappy as I thought.
15% wasn't the un-immunized number for the population. It was the percentage of those who got sick who either weren't up to date or whose immunization status was unknown. If immunization works, this should be considerably greater than the percentage of those in the population who aren't up-to-date on immunizations.
DTP (3 doses) coverage rate for children was 93% (+/- 4%) in California as of the most recent survey on the CDC web site. If we assume this holds for the population (probably not true, but this is a slashdot post and not an academic paper) and assume all unknowns are unimmunized, we can use Bayes Theorem to determine the efficacy of the vaccine.
P(Immunized | sick) = 0.85 P(Immunized) = 0.93 P(sick | immunized) = (P(Immunized | sick) * P(sick)) / P(immunized)
= (0.85 / 0.93) * P(sick)
= 0.91 * P(sick) Or in other words, immunization reduces your chances of getting sick by less than 10%. That's a crappy vaccine.
If all those of unknown status were in fact vaccinated, it's 89/93 = 96% -- that is, the vaccine reduces your chance of getting sick by 4%. A vaccine which only reduces your chance of illness between 4% and 10% is almost completely ineffective.
Note that my assumption about vaccination rate can't affect things much. If nearly all the population was immunized and 85% of the sick people were immunized, the vaccine reduced the probability of getting sick by a maximum of 15%, which is still piss-poor. If less of the population was immunized, it's even worse -- if 85% of the population were immunized and 85% of the sick people were immunized, the vaccine is completely ineffective.
Seems to me the data points to the current pertussis vaccine being crap.
The reason for combining opoids and acetaminophen rather than ibuprofen or even prescribing straight opoids has nothing to do with J&J lobbying and a lot to do with its hepatotoxicity. The drug warriors -- who are in fact scum of the earth -- would prefer that someone deliberately abusing pain pills die a painful death from liver failure than remain a live addict.
Slashdot Mirror
That's right, unionize. So instead of overworked Silicon Valley, we end up with collapsed Detroit. That's best case; the more likely case is that the companies will find no shortage of non-union labor.
Unfortunately, wrong. The actual middle is still abject poverty. No sooner will those poor people be uplifted than the next billion will drag them down.
Yeah, that's what the number theory weenies thought, until Diffie and Helman went and invented their key exchange.
My guess: If you were looking for a course of action to make programming even less attractive to women, you'd be hard-pressed to beat that one. But then, as I said on the other thread, I've been seeing women-only sections of such courses fail for a long time; the women who actually take the sections find them useful, but they can't get enough women to fill them.
I think this one may be spurious; few people, men or women, like that sort of thing. Yet as you point out, women go into veterinary science, and nurses are overwhelmingly women. Both fields involve late and irregular hours (much worse than software in many cases). In other words, it might be more informative to ask why women go into other difficult fields rather than software, instead of asking why they don't go into software -- any signal in the latter question might be drowned out by the noise of people who just want an easy job.
Indeed. I suspect the current "men in tech are pigs and that's the issue" doctrine is actually driving more women away from the field than any actual man-pigs in tech. If the #1 message young women hear about tech, from other women, is that it's filled with misogynistic pigs (whether this is true or not), why would they even consider the field?
Unfortunately, I believe that current educational "thinking" is much closer to the "talent is a myth" theory than yours.
But you do need some calculus for computer science. Not all those integration techniques (thankfully I've never had to do an integration by parts in my professional career), but things like limits and rates of change and logarithms and the relation between polynomial powers.
BTW, integral of sqrt(1/csc(x)) = integral of sqrt(sin x), which isn't expressible in terms of elementary functions.
Remember, arbitrage is only legal when dealing with intangible financial instruments. Arbitrage with actual products is gauche and therefore punishable.
It's because colleges and universities are natural collectors of Element 0 -- Administratium
Fortunately, that particular optimization is not allowed. Maybe we need a construct which catches overflows, something like ifoverflow(arith_expr) {...} else {...} If any overflow occurs within the protected expression, any assignments made are set to implementation-defined (not undefined) values. ifoverflow(c = a+b) {
Put this one with practical commercial fusion, economical solar energy, and flying cars. It wasn't that long ago Slashdot was all excited about "anything to oil"... went nowhere, of course.
ROTFL. That's one of the dumbest things I've read on slashdot. Well, dumbest that actually formed a coherent thought, anyway.
Why wait? I've seen these sorts of initiatives -- including women-only sections of classes -- for almost 30 years. They never work. This won't either. Percent of women in CS degrees peaked in 1984; everyone keeps trying to look for the reason for the downslide, but IMO what they really should be doing is looking for the reason for the increase in the first place.
There's a push for men in nursing. However, it doesn't seem to include blaming the situation on female nurses driving men out of the field, nor does it go so far as to attempt to exclude women from any nursing programs.
Education? Ha. Token. It's still assumed that any guy interested in elementary education is a pedophile.
Indeed; the compiler's even allowed to assume signed integer overflow doesn't happen, which is where you get into trouble. Yet we have this perfectly good mechanism for detecting integer overflow (condition codes) and no way to reach them from high level languages (C isn't unique in this respect)
Discover the causes, sure. Try to fix them? Might want to know what they are first. But that's not what this is about. It's already been decided that any disparity is due to bias, and therefore when disparity is found either one must find the male chauvinist pigs and root them out (basically Politically Correct Reign of Terror), or one must apply sufficient bias in the other direction to cancel the bias towards evil White Males (oh yeah and Asian mlaes too). In the US, the second tends to fall afoul of the same laws which prevent discrimination against minorities, so there's a lot more push for the first.
You see, you've already assumed that the disparity is due to bias.
Right. Maybe they go into nursing, where they have to deal with nasty bodily fluids and asshole doctors (of both sexes) all day. Or perhaps into marketing or advertising, where the men are always perfect angels. Or even finance, which is completely free of douchebag assholes making sexist remarks.
The narrative that the tech workplace is an especially hostile to women just doesn't hold up. As for slashdot comments, trolls will be trolls, and Slashdot is not the workplace.
No great mystery there; Slashdot got it wrong. Yahoo's Tech is only 15% female (globally), they didn't release US-only number.
Joke's on them. Once they have the ability, they'll find out that no one cares about their fears, frustrations, or desires anyway.
They want to regulate mapping apps on smartphones, including those not installed in vehicles? Seems like more than a bit of a stretch.
AP tests are made to get you college credit, but many CS programs won't accept AP credit to fulfill requirements in CS. So there's not a lot of point for a student wanting to become a CS major to take the AP CS test.
Also note it is (or was, it's been a while) possible to take the AP test without taking the AP class.
....3....2...1....GONE!
Ugh, noticed a problem right after posting. I should be comparing P(sick | immunized) to P(sick | !immunized), not to P(sick).
P(sick | !immunized) = (0.15 / 0.07) * P(sick)
P(sick) = (0.07 / 0.15) * P(sick | !immunized)
P(sick | immunized) = (0.85 / 0.93) * (0.07 / 0.15) * P(sick | !immunized)
= ~43% * P(sick | !immunized)
or about 57% better chance of not getting sick if you're immunized. That's much better but still not great. It's supposed to be over 80% effective.
With the opposite assumption of unknown status, it's (89/93) * (7/11) = 61%, reducing your chance of getting sick by 39%, which is quite poor.
Testing the sensitivity to assumption of those immunized in the population, if 95% are immunized you get (85 / 95) * (5 / 15) = 30%, a 70% effective vaccine (still not great).
Only if the percentage immunized in the population is 97% or better does the vaccine meet the 80% standard.
So the vaccine is probably crap, but not as crappy as I thought.
15% wasn't the un-immunized number for the population. It was the percentage of those who got sick who either weren't up to date or whose immunization status was unknown. If immunization works, this should be considerably greater than the percentage of those in the population who aren't up-to-date on immunizations.
DTP (3 doses) coverage rate for children was 93% (+/- 4%) in California as of the most recent survey on the CDC web site. If we assume this holds for the population (probably not true, but this is a slashdot post and not an academic paper) and assume all unknowns are unimmunized, we can use Bayes Theorem to determine the efficacy of the vaccine.
P(Immunized | sick) = 0.85
P(Immunized) = 0.93
P(sick | immunized) = (P(Immunized | sick) * P(sick)) / P(immunized)
= (0.85 / 0.93) * P(sick)
= 0.91 * P(sick)
Or in other words, immunization reduces your chances of getting sick by less than 10%. That's a crappy vaccine.
If all those of unknown status were in fact vaccinated, it's 89/93 = 96% -- that is, the vaccine reduces your chance of getting sick by 4%. A vaccine which only reduces your chance of illness between 4% and 10% is almost completely ineffective.
Note that my assumption about vaccination rate can't affect things much. If nearly all the population was immunized and 85% of the sick people were immunized, the vaccine reduced the probability of getting sick by a maximum of 15%, which is still piss-poor. If less of the population was immunized, it's even worse -- if 85% of the population were immunized and 85% of the sick people were immunized, the vaccine is completely ineffective.
Seems to me the data points to the current pertussis vaccine being crap.
The reason for combining opoids and acetaminophen rather than ibuprofen or even prescribing straight opoids has nothing to do with J&J lobbying and a lot to do with its hepatotoxicity. The drug warriors -- who are in fact scum of the earth -- would prefer that someone deliberately abusing pain pills die a painful death from liver failure than remain a live addict.
Or maybe it's because the current acellular pertussis vaccine just doesn't work all that well.