I admin a small research lab, and I'm most useful if people can talk to me without having to go to another room.
If I _really_ need screen privacy and people are there (like, say, I'm editing/etc/ppp/chap-secrets... I've really gotta get that whole ldap/radius thing working and kill that file), I grab my laptop and head for a private place.
Have to agree with everything I'm reading here though... modular office furniture with lockable cabinets/drawers, hardware locks, keeping more things serverside (via roaming profiles and samba shares in windows or any of a couple dozen possible implementations in *nix), a little bit of sane best-practices living (don't write passwords down! if you MUST, don't leave them lying around), and most complaints about it are pretty groundless.
your "weakest link" argument makes an interesting point...
however, I'd like to point out two things:
(1) Attacks get better. The Sha-1 attacks are improving. They're at least 64 times faster now than the initial publication, which is itself about 2000 times better than a brute force attack. This is drawing near the range of computability now. Sure, not like WEP or LanMan computability, but it's still broken.
(2) Which is better? Continuing to support and enforce weak, fragile, or downright broken standards? Or moving to stronger, standardized algorithms? Even if other parts of the chain may still weak, that doesn't mean that working on strengthening the chain is a waste of time, and the view you're suggesting is saying that they shouldn't bother trying to fix it at all, just let it break.
Further along the same line, isn't it better to prepare for the future, when the warning signs are all there pointing at practical exploits for algorithms we're using now, maybe as soon as 5 years from now? Especially given that Microsoft's policies on product lifecycle promise at least 5 years of support on whatever platform they're looking at.
Who knows what Vista will bring to the security table? But if it's the same sort of improvement over XP-SP2 that SP2 was over a fresh install, I'll be more than glad to see it. And it sounds like they're learning the crypto lessons that they were ignoring before, which is really the key to that.
I'm no economist, but... aren't most of the things that war consumes made in the USA? The tanks, the planes, the bullets, etc. Not to mention the troops, who would otherwise be in the work force competing for jobs, driving unemployment higher and salaries down (as more people would be competing for the same jobs).
Since that spending is pretty much internal to the economy, it has NOTHING to do with trade deficits. As far as I can see, the only things that war really negatively impacts in our economy are overall government spending (budget deficits != trade deficits) and negative public opinion.
Hard work and increased production do NOT a healthy economy make. Overproduction was a major driving force in the depression, about a century ago. Increased production has to match increased consumption, and war is GREAT at consuming a LOT of things from our own economy, ostensibly dumping the money back into American hands.
If you want to address the trade deficit, you need to reduce imports and improve exports. Things like reducing dependency on foreign oil, manufacturing more high-tech and end-user goods here instead of overseas, counteracting the outsourcing trend, and getting the steel industry's head out of it's ass and back into competitive production would go pretty far in that direction. Not that all of that is exactly trivial...
Now, budget deficit on the other hand... war definitely hits that pretty hard.
If you have the technology to steal software to begin with (computer, network, modern non-free operating system), you don't count as "the poor". Sorry. You just don't.
"The poor" end up without computers, without decent clothes, without houses that keep out the cold of winter. They end up without basic needs.
If your basic needs are met, and you have a computer, you should be subject to the same regulations and the same choices that everyone else who does is. Saying otherwise is like saying "people with incomes under $X should be exempt from speed limits, highway tolls, and emissions standards".
And, as this IS a FOSS-themed medium, I'd like to point out that there ARE free (as in beer, if not as in freedom) alternatives to any software you could possibly consider "necessary" for personal use. Just that they're arguably not as good as commercial alternatives that you're unwilling to pay for but want anyway.
Warez is a domain strictly limited to things that people want, but don't need, and don't want to pay for.
That said, I think it would be a misappropriation for the CIA, NSA, DHS to take part in these things (The DHS recently did, which was stupid). I would think the FBI would have better things to do, but copyright investigation IS their jurisdiction.
And in a lot of cases, "the bought version is better" is a fallacy. A lot of warez games are iso rips, and the only thing about them that's broken is the copy protection. A lot of software piracy comes in the form of broken copy protection or simply shared serial numbers and activation codes, or simply binaries that have the copy-protection parts of them torn out. In those cases, there's functionally no difference between the legitimate version and the illegitimate one, except that legit users have to jump through more hoops to use the software.
That is assuming the "1337 hax0rs" don't get hold of the algorithms. I can just imagine people messing around with p2p networks just for fun.
early in the lives of gotwoot and scarywater (large, fairly well known fansub bittorrent tracker sites), they encountered ddos issues...
people were using botnets and what amounts to trivial network code to send false complete requests to the trackers, and volunteering as seeds. So, in a field of maybe 100-200 legitimate seeds, there would be ~30,000 fakes poisoning the tracker. The tracker couldn't tell they were fakes, so was redirecting 99% of requests for blocks to the fakes advertising themselves as seeds (And eventually running out of memory as more bots were activated and the server broke under the load).
The recent weaknesses found in md5 and sha1 also make block poisoning a possibility. Which opens the door to download pool poisoning. If an attacker can generate a block that checksums to a known good block, then the downloader will only be able to detect that poisoned block in a many-blocks hash, not in individual block hashes. This means that the bad block would be propagated before it was detected, and poison the whole larger block (chunk).
Even further, clients would have no way of determining exactly which block is bad, so would have to discard the entire chunk and start again... and again, may very well end up with the poisoned data.
That's assuming that the app is still using a broken hash though. This becoming a problem would probably force the application into a better hashing algorithm (the yet-unbroken sha256 over sha1 or md5, for example), or into complete unusability, assuming the attackers were determined enough to poison every file and to do so intently enough to make an impact.
nah, it's not that it's already happened irrevocably.
It's that it's already happened insidiously, in the hearts and minds of the vast majority of people.
By claiming we're powerless, we just make excuses to justify our own complacence at the progressive removal of our rights and privileges as private citizens.
if the drive is one that has that executables and swap space and so forth. Which isn't to say that it does, or that it should.
All my data is on non-system disks (and in fact, not on a desktop system at all... it's on my fileserver in the corner).
I've got 2 hard drives (120 giggers) in my fileserver that have generated SMART errors in the past. Both of those drives have rendered individual, discrete files on them unrecoverable, without impacting the rest of the system, or even the rest of the filesystem, in any discernable way. Both are still functional, though I make it a point not to put anything important on them.
Further, a friend of mine was storing a large amount of manga scan rar archives on a hard drive of his fileserver, and discovered that the drive was going bad when I was trying to extract the files and got checksum errors. Without causing any other problems, the disk was randomly losing bits.
Almost nobody I know among my technically literate friends (with enough money to avoid doing so, that is) uses a single disk for their data storage needs, because everyone knows that the next time the OS dies or the partition table corrupts itself in a power outage or the windows install degrades to unusability, the quickest and cleanest way back up is usually wiping the system disk and starting over.
I'm in the US, I perceive flicker on 70hz and below refresh rate monitors, and on some old fluorescent lighting (but I've gotten used to it and can deal with it). But the thing is, a properly ballasted fluorescent lamp doesn't flicker at 50 or 60 hz. It flickers at 100 or 120 -- the ballast doubles the frequency from the mains frequency. Which is faster than most people perceive. However, solid state ballasts go WAY faster than that... Wikipedia's entry on ballasts is pretty informative.
So, pretty much, newer better lamps shouldn't flicker perceptibly. I know my CFL's don't, and ever since we got the ballasts replaced the tubes at work don't either. But I guess YMMV.
bare incandescents are just as ugly as fluorescents. Maybe uglier! Those filaments are such a small point, and such an intense light source that it's pretty unpleasant to look at them. I'd rather have one of my 14 watt fluorescents shined in my eyes than the 60 watt incandescents they replaced.
I think the main reason there's not widespread adoption of those fluorescent bulbs is that people don't think in the long-term. In the short term, a pack of walmart brand bulbs costs 75 cents and there's 4 of them, great! Or hell, splurge and get the $1.50 GE bulbs or whatever. Those crazy fluorescents are like $7.00-$12.00 for a two-pack.
A quick cocktail-napkin calculation based on my last power bill ($80 for the month) and usage level (980kwh), and estimating bulbs that are on for about 6 hours a day (which is probably an underestimate in my case, since I'm pretty much only at home at night, and usually spend more like 8-10 hours with at least some light on)... those fluorescents save me about $1.00/month each in comparison with the bulbs they replaced. Meaning in the course of 5 months, instead of having to discard 2-4 burned out incandescents and spend another 75 cents to get another pack, I've saved the cost of the fluorescents, and they've effectively cost me nothing. Now they continue to save me a few bucks a month on the bills, which is nothing spectacular, but it's pretty nice, and it adds up. In the short term, my checkout cost could have been 9 bucks lower than it was, but in the longer term, I'm pretty much winning out over everyone who didn't pay that up-front cost.
so instead you opt to be a receptionist, which requires you sitting at a desk 8 hours a day with your mind in neutral and getting half the salary you'd be making in IT?
That seems like a bad choice.
Moreover, this isn't just limited to IT. A lot of technical fields are this way.
I guess instead of "gender-typical" I should say "gender-acceptable".
There's a lot of both sexes in english and literature majors. But not a lot of women in engineering.
But if you're, as you say, pursuing your MRS... you don't want to spend all your time studying to try to pass your classes. You want to spend all your time at the bars, at parties, at social events, places where you've actually got a chance of picking someone up.
yeah, that's why I broke out the caplan and caplan book, and alluded to other gender studies about educational differences. And why I pointed out that there was an anecdotal part of the post.
Thanks for noticing though. Tell that to the other posters.
there's no question that the interest simply isn't there.
The question is... WHY is the interest not there?
Now, a "sexist pig" (or Harvard president) would suggest that this is strictly an innate difference. Someone a little more educated in the field of psychology (specifically gender studies) would be more prone to say that this is a socialized difference.
Women aren't electing to be programmers (or any of the numerous other IT positions out there), just like they aren't swarming to engineering and physics and chemistry. Nobody's saying they should be forced into jobs they don't want. But there's an indication of a problem when women as a whole are being indoctrinated with the idea that they CAN'T pursue these jobs, when they are capable of doing them.
More to the point, this isn't a gender difference that's always around. Women don't abandon their interests in science and technology until they're in their teens, as a rule. Ask 20 fourth grader girls, they all want to be scientists and doctors and executives and astronauts. Ask 20 9th grade girls, 18 of them will want to be thinner and more attractive, and have substantially no long-term goals beyond their appearances.
The problem is "why is the interest disappearing when these girls start puberty?".
that's not just a "French" viewpoint you were taking. And that's a lot of the source of the problem.
Those things that many men hold true, many women also come to hold true, at least on some level.
Your belief in "what women were for" came from somewhere, and more likely than not what indoctrinated you indoctrinated someone of the opposite sex just as deeply.
And more likely than not, the women who were raised believing that are living out those beliefs, hunting for husbands, working dead-end jobs, and trying to look good, instead of trying to build careers.
social desirability theory says that, in general, women percieve themselves as less desirable if they're good at math, or involved in the sciences. If they're not 'normal' they're different.
Women in science aren't in science to "hook a man". They're there to study science.
The women going to college hoping to get married along the way and be a dependent for life are the ones that go into gender-typical classes (ie: elementary education, liberal arts, to a lesser extent management or nursing).
Re:You're modded as +3 funny but...
on
Women Leaving I.T.
·
· Score: 5, Insightful
anecdotal, but...
3 of the best programmers I know are women. That includes my boss, and 2 people that went through the CS curriculum with me.
Now... I wouldn't trust any of them to do the job I do (mixed environment system administration), because it's not what they know. But in their fields, they're significantly better equipped than most of the men they graduated with.
There's a gender difference in teaching though. Men tend to get called on more than women in classes, and also tend to get taken more seriously than women, all the way back into elementary schools, by both male and female teachers.
Caplan and Caplan's "Thinking Critically about Women and Gender" has a good chapter on educational differences.
Ultimately, the women in IT are just as good as the men, but they're a far smaller sample. There's a lot of piss-poor programmers and sysadmins and support people who are men, and a smaller number in the same positions who are women. If a man screws up, it's more likely to be blamed on his incredible incompetence, where if a woman screws up, you're more likely to draw the attribution that it's because she's a woman.
People always bring up the issue of what the female sex can and can't do, well IMHO it's all BS, it's all about what they want to do.
And what they're socialized into doing.
Women feel social pressure NOT to be in science and technology. They're not supposed to be smart, they're supposed to look good. At least, that's how it is after junior high, for a vast majority of girls.
Men, on the other hand, have no associated stigma with being smart. In fact, we're pressed to be intelligent and successful, where they're pressed not to be.
Some gender psychologists tack a lot of the blame for the low turnout of women in science and technology quite firmly on that, and there's a lot of very good research to back that view up.
Re:Looking at the distribution ...
on
Women Leaving I.T.
·
· Score: 3, Interesting
And to put things in more perspective: I prefer Female managers over Male ones. I am very sexist at that because I think women have generaly more empathy and people skills, things a good manager needs.
There's sexism, and there's realism. The reality is that there is a significant gender difference in leadership styles. Men tend to be authoritarian leaders, women tend to be more democratic. There's a time and a place for both, and one's not universally better than the other.
Your preference in leadership probably reflects the way you work best. Sexism would be "I prefer female managers because they're more fun to look at".
Re:Looking at the distribution ...
on
Women Leaving I.T.
·
· Score: 5, Interesting
here's some figures for you to dispute.
I'm a CS undergrad at purdue. Our CS undergrad program, as of the start of last semester had 40 women in it. 24 of them are graduating. it's estimated that 6-10 at most are coming in, by figures I've heard. This is down from 10-15% of the department 4 years ago.
This is in a curriculum which has 800 or so undergrads, if I remember correctly.
I'm currently in a 300-level class (a major requirement, no less) that has 80 students, none of whom are female. Last semester I was in a database class that had 50 students, with a single woman in it. The semester before, I was in a class with 150 people, and a grand total of 4 women, and I know that after that class one of them changed majors out of computer science.
As of the end of this semester, 20-26 out of 800+. Those are very discouraging numbers, for women in CS. And the IT curricula in the school of technology aren't faring much better, I'm told.
python, like perl, is something I'd relegate to the "sysadmin/netadmin toolkit" over the "programmer toolkit".
Good stuff if you're doing things to make your own life easier. Great for writing internal apps and utils. Not so great for production software, just because of acceptance and familiarity levels (and some other things too).
Really depends on where you want to go with it, but having more tools to choose from is very seldom a bad thing.
In security, there's the concept of "least privelege". This means: for a given task you employ just enough access to accomplish the task. Many administrators subscribe to this exclusively, saying "in order to maintain security, you can only use the computer in ways I've predefined" -- a rather restrictive belief.
I don't entirely buy into least privelege, though. As a philosophy, it ends up being draconian and ultimately breeds the sort of animousity that the parent demonstrates. That user is one who's not going to bother with security regardless of what they have, out of sheer spite. Sure, I don't run as root or administrator unless I need root or administrator privs, and nobody else gets full privs on my network... just, there's a very fine line between removing the ability to break the system and removing the ability to use the system freely. I'm definitely more a proponent of "most allowable access" -- deny the users the access they shouldn't have, allow them everything else.
But before I worked out exactly what to deny users, I had one critical workstation taken out for about 2 days, crippled by about 30 spyware titles that invited each other in over the course of 2 days in the hands of a power user. 2 days of lost productivity for that user because I let him have default "power user" access, so the user could run max and ICQ... but anyway...
I think the big key to user happiness in a network environmeent is administrator flexibility. Sometimes IT/IS gets so caught up in its own little world it forgets that it has 2 roles: protect (the company resources) and serve (the users), and it has to balance those. When you forget one of those roles, you make bad policies that ultimately hurt more than they help.
As for the "old fart" teachers... I found that a lot of the best teachers in my old high school were the ones that weren't worrying about whether their contract would be renewed. The single best teacher I had had been teaching for around 40 years at the same school, and the year he taught me chemistry was the last year he was allowed to teach (because of his age). He was the best goddamned teacher I've ever had, too.
Rating teachers by "gpa" is a bad thing though. It encourages teachers to artificially inflate student's marks. GPA is an extremely subjective thing -- a B at harvard says about the same thing as a D at MIT, for example.
If you want to rate teachers, schools, etc... you need to do it on an objective basis. That's what Bush's No Child Left Behind tries to do, but ultimately the emphasis of the whole education process becomes the test. "Teaching the Test" is a huge injustice, it narrows the students' experience to the test's subject matter.
There ARE valid reasons for lower income children to have problems though. Last year my mom's 5th grade class had 5 students (out of 19) who were either Fetal Alcohol Syndrome babies or were born cocaine-addicted. Prevalence of substance abuse during pregnancy is a LOT lower in middle and upper class families.
A lot of the problem IS the teachers. Way too many of them are clueless. The GOOD side of NCLB is the teacher accreditation it's imposed -- weeding out some of the incredibly stupid teachers around who lack basic skills in english and math.
The biggest blame ultimately lies on families though. A teacher can only teach maybe 7 hours a day 5 days a week, 8 or 9 months a year. The family gets 9 or 10 hours a day 7 days a week, all year, and often the family doesn't bother to get involved in their children's education. If you aren't willing to know what your kid is learning in school, and you don't teach them values at home, school isn't going to be terribly useful to them no matter how good the teachers are. That's probably your rural connection, too.
Slashdot Mirror
I'm in the same boat.
/etc/ppp/chap-secrets ... I've really gotta get that whole ldap/radius thing working and kill that file), I grab my laptop and head for a private place.
I admin a small research lab, and I'm most useful if people can talk to me without having to go to another room.
If I _really_ need screen privacy and people are there (like, say, I'm editing
Have to agree with everything I'm reading here though... modular office furniture with lockable cabinets/drawers, hardware locks, keeping more things serverside (via roaming profiles and samba shares in windows or any of a couple dozen possible implementations in *nix), a little bit of sane best-practices living (don't write passwords down! if you MUST, don't leave them lying around), and most complaints about it are pretty groundless.
your "weakest link" argument makes an interesting point ...
however, I'd like to point out two things:
(1) Attacks get better. The Sha-1 attacks are improving. They're at least 64 times faster now than the initial publication, which is itself about 2000 times better than a brute force attack. This is drawing near the range of computability now. Sure, not like WEP or LanMan computability, but it's still broken.
(2) Which is better? Continuing to support and enforce weak, fragile, or downright broken standards? Or moving to stronger, standardized algorithms? Even if other parts of the chain may still weak, that doesn't mean that working on strengthening the chain is a waste of time, and the view you're suggesting is saying that they shouldn't bother trying to fix it at all, just let it break.
Further along the same line, isn't it better to prepare for the future, when the warning signs are all there pointing at practical exploits for algorithms we're using now, maybe as soon as 5 years from now? Especially given that Microsoft's policies on product lifecycle promise at least 5 years of support on whatever platform they're looking at.
Who knows what Vista will bring to the security table? But if it's the same sort of improvement over XP-SP2 that SP2 was over a fresh install, I'll be more than glad to see it. And it sounds like they're learning the crypto lessons that they were ignoring before, which is really the key to that.
I'm no economist, but ... aren't most of the things that war consumes made in the USA? The tanks, the planes, the bullets, etc. Not to mention the troops, who would otherwise be in the work force competing for jobs, driving unemployment higher and salaries down (as more people would be competing for the same jobs).
...
... war definitely hits that pretty hard.
Since that spending is pretty much internal to the economy, it has NOTHING to do with trade deficits. As far as I can see, the only things that war really negatively impacts in our economy are overall government spending (budget deficits != trade deficits) and negative public opinion.
Hard work and increased production do NOT a healthy economy make. Overproduction was a major driving force in the depression, about a century ago. Increased production has to match increased consumption, and war is GREAT at consuming a LOT of things from our own economy, ostensibly dumping the money back into American hands.
If you want to address the trade deficit, you need to reduce imports and improve exports. Things like reducing dependency on foreign oil, manufacturing more high-tech and end-user goods here instead of overseas, counteracting the outsourcing trend, and getting the steel industry's head out of it's ass and back into competitive production would go pretty far in that direction. Not that all of that is exactly trivial
Now, budget deficit on the other hand
doubtful.
That would require IT security people being cooperative instead of adversarial...
If you have the technology to steal software to begin with (computer, network, modern non-free operating system), you don't count as "the poor". Sorry. You just don't.
"The poor" end up without computers, without decent clothes, without houses that keep out the cold of winter. They end up without basic needs.
If your basic needs are met, and you have a computer, you should be subject to the same regulations and the same choices that everyone else who does is. Saying otherwise is like saying "people with incomes under $X should be exempt from speed limits, highway tolls, and emissions standards".
And, as this IS a FOSS-themed medium, I'd like to point out that there ARE free (as in beer, if not as in freedom) alternatives to any software you could possibly consider "necessary" for personal use. Just that they're arguably not as good as commercial alternatives that you're unwilling to pay for but want anyway.
Warez is a domain strictly limited to things that people want, but don't need, and don't want to pay for.
That said, I think it would be a misappropriation for the CIA, NSA, DHS to take part in these things (The DHS recently did, which was stupid). I would think the FBI would have better things to do, but copyright investigation IS their jurisdiction.
And in a lot of cases, "the bought version is better" is a fallacy. A lot of warez games are iso rips, and the only thing about them that's broken is the copy protection. A lot of software piracy comes in the form of broken copy protection or simply shared serial numbers and activation codes, or simply binaries that have the copy-protection parts of them torn out. In those cases, there's functionally no difference between the legitimate version and the illegitimate one, except that legit users have to jump through more hoops to use the software.
early in the lives of gotwoot and scarywater (large, fairly well known fansub bittorrent tracker sites), they encountered ddos issues...
people were using botnets and what amounts to trivial network code to send false complete requests to the trackers, and volunteering as seeds. So, in a field of maybe 100-200 legitimate seeds, there would be ~30,000 fakes poisoning the tracker. The tracker couldn't tell they were fakes, so was redirecting 99% of requests for blocks to the fakes advertising themselves as seeds (And eventually running out of memory as more bots were activated and the server broke under the load).
The recent weaknesses found in md5 and sha1 also make block poisoning a possibility. Which opens the door to download pool poisoning. If an attacker can generate a block that checksums to a known good block, then the downloader will only be able to detect that poisoned block in a many-blocks hash, not in individual block hashes. This means that the bad block would be propagated before it was detected, and poison the whole larger block (chunk).
Even further, clients would have no way of determining exactly which block is bad, so would have to discard the entire chunk and start again... and again, may very well end up with the poisoned data.
That's assuming that the app is still using a broken hash though. This becoming a problem would probably force the application into a better hashing algorithm (the yet-unbroken sha256 over sha1 or md5, for example), or into complete unusability, assuming the attackers were determined enough to poison every file and to do so intently enough to make an impact.
nah, it's not that it's already happened irrevocably.
It's that it's already happened insidiously, in the hearts and minds of the vast majority of people.
By claiming we're powerless, we just make excuses to justify our own complacence at the progressive removal of our rights and privileges as private citizens.
random crashes maybe....
... it's on my fileserver in the corner).
if the drive is one that has that executables and swap space and so forth. Which isn't to say that it does, or that it should.
All my data is on non-system disks (and in fact, not on a desktop system at all
I've got 2 hard drives (120 giggers) in my fileserver that have generated SMART errors in the past. Both of those drives have rendered individual, discrete files on them unrecoverable, without impacting the rest of the system, or even the rest of the filesystem, in any discernable way. Both are still functional, though I make it a point not to put anything important on them.
Further, a friend of mine was storing a large amount of manga scan rar archives on a hard drive of his fileserver, and discovered that the drive was going bad when I was trying to extract the files and got checksum errors. Without causing any other problems, the disk was randomly losing bits.
Almost nobody I know among my technically literate friends (with enough money to avoid doing so, that is) uses a single disk for their data storage needs, because everyone knows that the next time the OS dies or the partition table corrupts itself in a power outage or the windows install degrades to unusability, the quickest and cleanest way back up is usually wiping the system disk and starting over.
Well, either/or. Maybe both.
I'm in the US, I perceive flicker on 70hz and below refresh rate monitors, and on some old fluorescent lighting (but I've gotten used to it and can deal with it). But the thing is, a properly ballasted fluorescent lamp doesn't flicker at 50 or 60 hz. It flickers at 100 or 120 -- the ballast doubles the frequency from the mains frequency. Which is faster than most people perceive. However, solid state ballasts go WAY faster than that ... Wikipedia's entry on ballasts is pretty informative.
So, pretty much, newer better lamps shouldn't flicker perceptibly. I know my CFL's don't, and ever since we got the ballasts replaced the tubes at work don't either. But I guess YMMV.
bare incandescents are just as ugly as fluorescents. Maybe uglier! Those filaments are such a small point, and such an intense light source that it's pretty unpleasant to look at them. I'd rather have one of my 14 watt fluorescents shined in my eyes than the 60 watt incandescents they replaced.
... those fluorescents save me about $1.00/month each in comparison with the bulbs they replaced. Meaning in the course of 5 months, instead of having to discard 2-4 burned out incandescents and spend another 75 cents to get another pack, I've saved the cost of the fluorescents, and they've effectively cost me nothing. Now they continue to save me a few bucks a month on the bills, which is nothing spectacular, but it's pretty nice, and it adds up. In the short term, my checkout cost could have been 9 bucks lower than it was, but in the longer term, I'm pretty much winning out over everyone who didn't pay that up-front cost.
I think the main reason there's not widespread adoption of those fluorescent bulbs is that people don't think in the long-term. In the short term, a pack of walmart brand bulbs costs 75 cents and there's 4 of them, great! Or hell, splurge and get the $1.50 GE bulbs or whatever. Those crazy fluorescents are like $7.00-$12.00 for a two-pack.
A quick cocktail-napkin calculation based on my last power bill ($80 for the month) and usage level (980kwh), and estimating bulbs that are on for about 6 hours a day (which is probably an underestimate in my case, since I'm pretty much only at home at night, and usually spend more like 8-10 hours with at least some light on)
so instead you opt to be a receptionist, which requires you sitting at a desk 8 hours a day with your mind in neutral and getting half the salary you'd be making in IT?
That seems like a bad choice.
Moreover, this isn't just limited to IT. A lot of technical fields are this way.
t'is true.
I guess instead of "gender-typical" I should say "gender-acceptable".
... you don't want to spend all your time studying to try to pass your classes. You want to spend all your time at the bars, at parties, at social events, places where you've actually got a chance of picking someone up.
:-p
There's a lot of both sexes in english and literature majors. But not a lot of women in engineering.
But if you're, as you say, pursuing your MRS
Besides, who wants to date a geek?
yeah, that's why I broke out the caplan and caplan book, and alluded to other gender studies about educational differences. And why I pointed out that there was an anecdotal part of the post.
Thanks for noticing though. Tell that to the other posters.
there's no question that the interest simply isn't there.
... WHY is the interest not there?
The question is
Now, a "sexist pig" (or Harvard president) would suggest that this is strictly an innate difference. Someone a little more educated in the field of psychology (specifically gender studies) would be more prone to say that this is a socialized difference.
Women aren't electing to be programmers (or any of the numerous other IT positions out there), just like they aren't swarming to engineering and physics and chemistry. Nobody's saying they should be forced into jobs they don't want. But there's an indication of a problem when women as a whole are being indoctrinated with the idea that they CAN'T pursue these jobs, when they are capable of doing them.
More to the point, this isn't a gender difference that's always around. Women don't abandon their interests in science and technology until they're in their teens, as a rule. Ask 20 fourth grader girls, they all want to be scientists and doctors and executives and astronauts. Ask 20 9th grade girls, 18 of them will want to be thinner and more attractive, and have substantially no long-term goals beyond their appearances.
The problem is "why is the interest disappearing when these girls start puberty?".
that's not just a "French" viewpoint you were taking. And that's a lot of the source of the problem.
Those things that many men hold true, many women also come to hold true, at least on some level.
Your belief in "what women were for" came from somewhere, and more likely than not what indoctrinated you indoctrinated someone of the opposite sex just as deeply.
And more likely than not, the women who were raised believing that are living out those beliefs, hunting for husbands, working dead-end jobs, and trying to look good, instead of trying to build careers.
social desirability theory says that, in general, women percieve themselves as less desirable if they're good at math, or involved in the sciences. If they're not 'normal' they're different.
Women in science aren't in science to "hook a man". They're there to study science.
The women going to college hoping to get married along the way and be a dependent for life are the ones that go into gender-typical classes (ie: elementary education, liberal arts, to a lesser extent management or nursing).
anecdotal, but ...
... I wouldn't trust any of them to do the job I do (mixed environment system administration), because it's not what they know. But in their fields, they're significantly better equipped than most of the men they graduated with.
3 of the best programmers I know are women. That includes my boss, and 2 people that went through the CS curriculum with me.
Now
There's a gender difference in teaching though. Men tend to get called on more than women in classes, and also tend to get taken more seriously than women, all the way back into elementary schools, by both male and female teachers.
Caplan and Caplan's "Thinking Critically about Women and Gender" has a good chapter on educational differences.
Ultimately, the women in IT are just as good as the men, but they're a far smaller sample. There's a lot of piss-poor programmers and sysadmins and support people who are men, and a smaller number in the same positions who are women. If a man screws up, it's more likely to be blamed on his incredible incompetence, where if a woman screws up, you're more likely to draw the attribution that it's because she's a woman.
And what they're socialized into doing.
Women feel social pressure NOT to be in science and technology. They're not supposed to be smart, they're supposed to look good. At least, that's how it is after junior high, for a vast majority of girls.
Men, on the other hand, have no associated stigma with being smart. In fact, we're pressed to be intelligent and successful, where they're pressed not to be.
Some gender psychologists tack a lot of the blame for the low turnout of women in science and technology quite firmly on that, and there's a lot of very good research to back that view up.
There's sexism, and there's realism. The reality is that there is a significant gender difference in leadership styles. Men tend to be authoritarian leaders, women tend to be more democratic. There's a time and a place for both, and one's not universally better than the other.
Your preference in leadership probably reflects the way you work best. Sexism would be "I prefer female managers because they're more fun to look at".
here's some figures for you to dispute.
I'm a CS undergrad at purdue. Our CS undergrad program, as of the start of last semester had 40 women in it. 24 of them are graduating. it's estimated that 6-10 at most are coming in, by figures I've heard. This is down from 10-15% of the department 4 years ago.
This is in a curriculum which has 800 or so undergrads, if I remember correctly.
I'm currently in a 300-level class (a major requirement, no less) that has 80 students, none of whom are female. Last semester I was in a database class that had 50 students, with a single woman in it. The semester before, I was in a class with 150 people, and a grand total of 4 women, and I know that after that class one of them changed majors out of computer science.
As of the end of this semester, 20-26 out of 800+. Those are very discouraging numbers, for women in CS. And the IT curricula in the school of technology aren't faring much better, I'm told.
python, like perl, is something I'd relegate to the "sysadmin/netadmin toolkit" over the "programmer toolkit".
Good stuff if you're doing things to make your own life easier. Great for writing internal apps and utils. Not so great for production software, just because of acceptance and familiarity levels (and some other things too).
Really depends on where you want to go with it, but having more tools to choose from is very seldom a bad thing.
In security, there's the concept of "least privelege". This means: for a given task you employ just enough access to accomplish the task. Many administrators subscribe to this exclusively, saying "in order to maintain security, you can only use the computer in ways I've predefined" -- a rather restrictive belief.
...
I don't entirely buy into least privelege, though. As a philosophy, it ends up being draconian and ultimately breeds the sort of animousity that the parent demonstrates. That user is one who's not going to bother with security regardless of what they have, out of sheer spite. Sure, I don't run as root or administrator unless I need root or administrator privs, and nobody else gets full privs on my network... just, there's a very fine line between removing the ability to break the system and removing the ability to use the system freely. I'm definitely more a proponent of "most allowable access" -- deny the users the access they shouldn't have, allow them everything else.
But before I worked out exactly what to deny users, I had one critical workstation taken out for about 2 days, crippled by about 30 spyware titles that invited each other in over the course of 2 days in the hands of a power user. 2 days of lost productivity for that user because I let him have default "power user" access, so the user could run max and ICQ... but anyway
I think the big key to user happiness in a network environmeent is administrator flexibility. Sometimes IT/IS gets so caught up in its own little world it forgets that it has 2 roles: protect (the company resources) and serve (the users), and it has to balance those. When you forget one of those roles, you make bad policies that ultimately hurt more than they help.
As for the "old fart" teachers ... I found that a lot of the best teachers in my old high school were the ones that weren't worrying about whether their contract would be renewed. The single best teacher I had had been teaching for around 40 years at the same school, and the year he taught me chemistry was the last year he was allowed to teach (because of his age). He was the best goddamned teacher I've ever had, too.
... you need to do it on an objective basis. That's what Bush's No Child Left Behind tries to do, but ultimately the emphasis of the whole education process becomes the test. "Teaching the Test" is a huge injustice, it narrows the students' experience to the test's subject matter.
Rating teachers by "gpa" is a bad thing though. It encourages teachers to artificially inflate student's marks. GPA is an extremely subjective thing -- a B at harvard says about the same thing as a D at MIT, for example.
If you want to rate teachers, schools, etc
There ARE valid reasons for lower income children to have problems though. Last year my mom's 5th grade class had 5 students (out of 19) who were either Fetal Alcohol Syndrome babies or were born cocaine-addicted. Prevalence of substance abuse during pregnancy is a LOT lower in middle and upper class families.
A lot of the problem IS the teachers. Way too many of them are clueless. The GOOD side of NCLB is the teacher accreditation it's imposed -- weeding out some of the incredibly stupid teachers around who lack basic skills in english and math.
The biggest blame ultimately lies on families though. A teacher can only teach maybe 7 hours a day 5 days a week, 8 or 9 months a year. The family gets 9 or 10 hours a day 7 days a week, all year, and often the family doesn't bother to get involved in their children's education. If you aren't willing to know what your kid is learning in school, and you don't teach them values at home, school isn't going to be terribly useful to them no matter how good the teachers are. That's probably your rural connection, too.