I read Cryptonomicon a year or so ago and loved it, and also really liked Snow Crash. So I went to check out one of William Gibson's books and found that Idoru was the only one at my local library. I checked it out, but after a couple of chapters I gave up. It just didn't impress me.
My question is, is Idoru considered to be among Gibson's best work? If not, what's the best introduction to his style?
As a mathematician-turned-programmer who's also been programming since the age of 12 (which adds up to a few more years total in my case) and as somebody who's seen a LOT of badly written code in my time, I can tell you that a couple of years' worth of college-level mathematics courses will make you a much better programmer. Here's a few courses you should definitely have under your belt:
* Calculus (usually 3 semesters)
* Linear Algebra
* Discrete Mathematics (generally a mixed bag with some combinatorics and graph theory)
and of course any prerequisites to the courses above that you didn't get in high school.
In addition, here's some courses that would be useful, some in particular areas:
* Abstract algebra
* Number theory
* Geometry (not high-school geometry)
* Real analysis (sometimes called advanced calculus)
All of the changes being suggested and argued about have one thing in common--they provide a cheap alternative to actually solving the problem.
It would be possible to solve the problem of security in our skies without taking away any of our liberties. Make sure there are (frequently) law enforcement officers on board flights. Yes, this is being considered, but it is expensive. It's cheaper to build a database and track each person flying. It is an invasion of privacy, but it is cheaper.
It would be possible to solve the problem of Middle Eastern terrorism, but it is expensive. It would require assistance to the desperately poor parts of the region, to build schools, hospitals, and the other things they need to support a decent life. It would require us to be willing to pay a higher price to get oil that is not purchased from tyrants. It would require us to give up our notion of "client states" and recognize that the people who are considered too poor and powerless to worry about today will be desperate enough to follow a madman tomorrow. But it's cheaper to try and spy and assassinate our way out of the problem. It won't work, of course, and will create bigger problems down the line, but it is cheaper than solving the problem.
The world didn't change last week, really. Many innocent people lost their lives in a senseless tragedy. The tragedy will be compounded if we don't start educating ourselves about the world we live in, and if we don't realize that there is no person on this planet too poor, too different, or too desperate to be important.
I worked at the NSA for three years, and left recently enough to have
been there for some of the events discussed in the "60 Minutes"
story. Disclaimer: When I left the agency, I was
disgusted at a personal (they screwed me over), ethical (I have this
funny thing about hypocrisy) and professional (mathematics & computer
science) level, so take this for what it's worth.
I also signed an agreement when I left saying I would submit anything
I wrote about NSA for publication to their public relations office for
approval before submitting it, but I seem to have lost their address.
Oh well.
One thing I've noticed in a lot of the other responses is a reaction
to the apparent contradiction between "they can read everything,
everywhere" and "our systems are broken, we need more money or we'll
fall apart." In a sense, both things are true, and it illustrates
what I think is one of NSA's biggest problems.
Yes, of course, NSA has some amazing technology. But these gee-whiz
supercomputers and super-secret devices are like little islands in a
sea of technological muck. When you hear about secretaries doing word
processing on Crays at NSA, it is not because they have so much excess
capacity just lying around, it is because the secretary's desktop unit
probably really sucks and she has no other choice. Did the entire
network shut down for nearly a week? Yes, I recall going to work one
morning and seeing a sign posted on the turnstiles, "Don't log in when
you get to your office." While all those brilliant minds were busy
with gee-whiz projects (that is, after all, how you get the cash
awards and the promotions), the infrastructure was being allowed to
rot. After all, what looks better at evaluation time: "I played a
small, seemingly insignificant part in making sure that NSA's
wide-area network stayed up" or "I created a new system using
insert hot technology here that resulted in a insert big
percentage here increase in processed traffic against insert
country name here, a high priority target"?
That's the big problem I was talking about. NSA does have some
really smart people, but their management stinks. I mean, really
stinks. It's been referred to as the "Glen Burnie full employment
project" (a Baltimore suburb near Ft. Meade). After all, there might
be some incentive to go through all the security nonsense (an
essentially random process which can't be proven to prevent anything) to
get a job there, if you're a techie and you think you might get to
play with some neat toys, or if you're a mathematician in a bad
academic job market. But if you're a manager, the only way you'll be
interested in NSA is if you are really not that talented, but heck,
you have an uncle who can probably get you into a pretty good position and you
won't have to worry about getting laid off, like you would at the
phone company.
So when you hear that the agency's response to some new technology is
a hamhanded effort to make it illegal, or at least unexportable, and
you ask yourself, "What could they be thinking? They can't be that
stupid!" think again.
Another big problem with the NSA, CIA, etc. is an inherent
contradiction at the heart of what they do. In the middle of a
(supposedly) free society, that is made up of a mixture of cultures
from all over the world, you have a bunch of people who do all of
their work in secret and steal information from other countries.
Okay, maybe they call it "maintaining information on a need-to-know
basis" and "intelligence gathering" but we know what the point is
here, right? When we're not at war with a country, it's pretty hard
to justify doing things to them as an organization that, on a personal
level, would be wrong and just creepy. Especially when you might have
good neighbors who were born in that country. Or even relatives.
Back in the 1960's, when the average engineer was kind of a WASPy
dweeb, the contradiction wasn't so apparent. But take a look at the
population of any engineering class now. It's way more diverse, and
you're simply going to have a harder time justifying the "we're in a
constant state of war" line with these people. It just doesn't make
sense anymore. And yes, the terrorism that NSA talking heads mention
when they're begging for money has something to do with the crap we've
dealt people around the world for years. After all, what are Bin
Laden and his followers upset about? The US presence in Muslim holy
lands. Why are we there? Leftovers from the Gulf War against Saddam
Hussein. How did he get so powerful? The US backed him against
Iran. Why was Iran a big problem? The people who overthrew the Shah
hated the US. Why? Because the Shah was a tyrant backed by--who?
That's right, the CIA.
This is already too long, but one final thing about NSA listening to
your phone conversations. If you're not a US citizen and you're not
currently in the US, you're fair game to these people, but you're also
probably not very interesting (see the comment about getting cash
awards and promotions above). If you are interesting to them for some
(possibly nonsensical) reason, that is if listening to you can get
some analyst a promotion, there you go. If you are a US citizen, or
you are currently residing in the US, then the NSA cannot legally spy
on you, and nobody gets promoted if the lawyers aren't happy. But if
the FBI develops a (perhaps nonsensical) interest in you, it is not
hard to get a warrant for whatever kind of surveillance they want to
do, and guess where they get their technology?
I think this was a great speech by Stallman--interesting timing, too. Today a delegation from the EU that came to Washington with some questions about just how much the US spies on its European "allies" was basically snubbed by the CIA and the NSA. They eventually gave up on getting any straight answers and went home.
The article I read about this contains the following:
Coelho and his team also met with former CIA director James Woolsey, as well as civil liberties groups.
In interviews last year, Woolsey acknowledged the United States collects information on European companies. But he insisted it did so only in cases where companies were suspected of violating U.N. or U.S. sanctions, of offering bribes to win contracts, or to keep tabs on technologies that have civilian and military applications.
Does this information collection include looking for "intellectual property" violations? Imagine if, in addition to the MPAA, you had the NSA on your back!
...we just got rid of legal restraints on the use of so-called "intellectual property"? If these innovations that Microsoft keeps talking about are really so much in demand, simple economics says that someone would find a way to meet that demand.
I agree that people who do good work have a right to be compensated for it, but isn't there a way to fund the work other than grants and other than guaranteeing some corporation exclusive rights to use the result?
I think the answer to that question is important not only for the software industry, but also for other industries that depend heavily on R&D, like pharmaceuticals and biotech. Look at what happened in South Africa a couple of weeks ago. If it's true that the only way to get new AIDS drugs is to guarantee some company exclusive rights to the result, we're in big trouble. There must be another way.
I also think a lot of the recent controversies over the marketing of drugs and genetically engineered seeds stem from the belief on the part of these corporations that "We've spent so many millions of dollars on research for this thing, now we have to create a market for it." As a result, there's no way to tell the difference between products that can really help people and overengineered, dangerous junk.
If anyone is going to be careful not to run afoul of Supreme Court case precedent, it's the feds.
WRONG. That's a common fallacy--they're the government, they must know what they're doing. Here's the truth--people do what they're paid to do. The DEA gets its funds from Congress, and Congress wants to appear tough on drugs, so the DEA will do whatever it takes to give them what they're paying for--whether it is constitutional or not.
The profile they get of an Amtrak passenger is not fundamentally different from what they could get standing around the station observing people. The only difference is they are looking at data on a screen rather than faces and clothes.
WRONG. Ever heard of "expectation of privacy"? When I walk through an Amtrak terminal, I don't have any expectation that I won't be seen. When I buy a ticket, I have every expectation that my name and travel times are not publicly available. That makes (or should make) a BIG difference when it comes to law enforcement. Of course, the bigger issue is the transition from "identify the possible bad guys based on probable cause and track them" to "track everybody and see if anything interesting turns up."
In fact, this form of profiling has the potential to be less racially biased than face-to-face observation. When you get over the knee-jerk reaction and think about it, it really is preferable to the current system. The current system being basically "stop people with brown skin".
WRONG (at least as far as I'm concerned). I've never been a fan of "the lesser of two evils" and I'm especially not a fan of it when it comes to law enforcement.
Slashdot Mirror
My question is, is Idoru considered to be among Gibson's best work? If not, what's the best introduction to his style?
As a mathematician-turned-programmer who's also been programming since the age of 12 (which adds up to a few more years total in my case) and as somebody who's seen a LOT of badly written code in my time, I can tell you that a couple of years' worth of college-level mathematics courses will make you a much better programmer. Here's a few courses you should definitely have under your belt:
* Calculus (usually 3 semesters)
* Linear Algebra
* Discrete Mathematics (generally a mixed bag with some combinatorics and graph theory)
and of course any prerequisites to the courses above that you didn't get in high school.
In addition, here's some courses that would be useful, some in particular areas:
* Abstract algebra
* Number theory
* Geometry (not high-school geometry)
* Real analysis (sometimes called advanced calculus)
Good luck--it'll be worth it.
All of the changes being suggested and argued about have one thing in common--they provide a cheap alternative to actually solving the problem.
It would be possible to solve the problem of security in our skies without taking away any of our liberties. Make sure there are (frequently) law enforcement officers on board flights. Yes, this is being considered, but it is expensive. It's cheaper to build a database and track each person flying. It is an invasion of privacy, but it is cheaper.
It would be possible to solve the problem of Middle Eastern terrorism, but it is expensive. It would require assistance to the desperately poor parts of the region, to build schools, hospitals, and the other things they need to support a decent life. It would require us to be willing to pay a higher price to get oil that is not purchased from tyrants. It would require us to give up our notion of "client states" and recognize that the people who are considered too poor and powerless to worry about today will be desperate enough to follow a madman tomorrow. But it's cheaper to try and spy and assassinate our way out of the problem. It won't work, of course, and will create bigger problems down the line, but it is cheaper than solving the problem.
The world didn't change last week, really. Many innocent people lost their lives in a senseless tragedy. The tragedy will be compounded if we don't start educating ourselves about the world we live in, and if we don't realize that there is no person on this planet too poor, too different, or too desperate to be important.
I worked at the NSA for three years, and left recently enough to have been there for some of the events discussed in the "60 Minutes" story. Disclaimer: When I left the agency, I was disgusted at a personal (they screwed me over), ethical (I have this funny thing about hypocrisy) and professional (mathematics & computer science) level, so take this for what it's worth.
I also signed an agreement when I left saying I would submit anything I wrote about NSA for publication to their public relations office for approval before submitting it, but I seem to have lost their address. Oh well.
One thing I've noticed in a lot of the other responses is a reaction to the apparent contradiction between "they can read everything, everywhere" and "our systems are broken, we need more money or we'll fall apart." In a sense, both things are true, and it illustrates what I think is one of NSA's biggest problems.
Yes, of course, NSA has some amazing technology. But these gee-whiz supercomputers and super-secret devices are like little islands in a sea of technological muck. When you hear about secretaries doing word processing on Crays at NSA, it is not because they have so much excess capacity just lying around, it is because the secretary's desktop unit probably really sucks and she has no other choice. Did the entire network shut down for nearly a week? Yes, I recall going to work one morning and seeing a sign posted on the turnstiles, "Don't log in when you get to your office." While all those brilliant minds were busy with gee-whiz projects (that is, after all, how you get the cash awards and the promotions), the infrastructure was being allowed to rot. After all, what looks better at evaluation time: "I played a small, seemingly insignificant part in making sure that NSA's wide-area network stayed up" or "I created a new system using insert hot technology here that resulted in a insert big percentage here increase in processed traffic against insert country name here, a high priority target"?
That's the big problem I was talking about. NSA does have some really smart people, but their management stinks. I mean, really stinks. It's been referred to as the "Glen Burnie full employment project" (a Baltimore suburb near Ft. Meade). After all, there might be some incentive to go through all the security nonsense (an essentially random process which can't be proven to prevent anything) to get a job there, if you're a techie and you think you might get to play with some neat toys, or if you're a mathematician in a bad academic job market. But if you're a manager, the only way you'll be interested in NSA is if you are really not that talented, but heck, you have an uncle who can probably get you into a pretty good position and you won't have to worry about getting laid off, like you would at the phone company.
So when you hear that the agency's response to some new technology is a hamhanded effort to make it illegal, or at least unexportable, and you ask yourself, "What could they be thinking? They can't be that stupid!" think again.
Another big problem with the NSA, CIA, etc. is an inherent contradiction at the heart of what they do. In the middle of a (supposedly) free society, that is made up of a mixture of cultures from all over the world, you have a bunch of people who do all of their work in secret and steal information from other countries. Okay, maybe they call it "maintaining information on a need-to-know basis" and "intelligence gathering" but we know what the point is here, right? When we're not at war with a country, it's pretty hard to justify doing things to them as an organization that, on a personal level, would be wrong and just creepy. Especially when you might have good neighbors who were born in that country. Or even relatives. Back in the 1960's, when the average engineer was kind of a WASPy dweeb, the contradiction wasn't so apparent. But take a look at the population of any engineering class now. It's way more diverse, and you're simply going to have a harder time justifying the "we're in a constant state of war" line with these people. It just doesn't make sense anymore. And yes, the terrorism that NSA talking heads mention when they're begging for money has something to do with the crap we've dealt people around the world for years. After all, what are Bin Laden and his followers upset about? The US presence in Muslim holy lands. Why are we there? Leftovers from the Gulf War against Saddam Hussein. How did he get so powerful? The US backed him against Iran. Why was Iran a big problem? The people who overthrew the Shah hated the US. Why? Because the Shah was a tyrant backed by--who? That's right, the CIA.
This is already too long, but one final thing about NSA listening to your phone conversations. If you're not a US citizen and you're not currently in the US, you're fair game to these people, but you're also probably not very interesting (see the comment about getting cash awards and promotions above). If you are interesting to them for some (possibly nonsensical) reason, that is if listening to you can get some analyst a promotion, there you go. If you are a US citizen, or you are currently residing in the US, then the NSA cannot legally spy on you, and nobody gets promoted if the lawyers aren't happy. But if the FBI develops a (perhaps nonsensical) interest in you, it is not hard to get a warrant for whatever kind of surveillance they want to do, and guess where they get their technology?
European Panel Probing NSA Departs Abruptly
I think this was a great speech by Stallman--interesting timing, too. Today a delegation from the EU that came to Washington with some questions about just how much the US spies on its European "allies" was basically snubbed by the CIA and the NSA. They eventually gave up on getting any straight answers and went home.
The article I read about this contains the following:
Coelho and his team also met with former CIA director James Woolsey, as well as civil liberties groups.
In interviews last year, Woolsey acknowledged the United States collects information on European companies. But he insisted it did so only in cases where companies were suspected of violating U.N. or U.S. sanctions, of offering bribes to win contracts, or to keep tabs on technologies that have civilian and military applications.
Does this information collection include looking for "intellectual property" violations? Imagine if, in addition to the MPAA, you had the NSA on your back!
...we just got rid of legal restraints on the use of so-called "intellectual property"? If these innovations that Microsoft keeps talking about are really so much in demand, simple economics says that someone would find a way to meet that demand.
I agree that people who do good work have a right to be compensated for it, but isn't there a way to fund the work other than grants and other than guaranteeing some corporation exclusive rights to use the result?
I think the answer to that question is important not only for the software industry, but also for other industries that depend heavily on R&D, like pharmaceuticals and biotech. Look at what happened in South Africa a couple of weeks ago. If it's true that the only way to get new AIDS drugs is to guarantee some company exclusive rights to the result, we're in big trouble. There must be another way.
I also think a lot of the recent controversies over the marketing of drugs and genetically engineered seeds stem from the belief on the part of these corporations that "We've spent so many millions of dollars on research for this thing, now we have to create a market for it." As a result, there's no way to tell the difference between products that can really help people and overengineered, dangerous junk.
If anyone is going to be careful not to run afoul of Supreme Court case precedent, it's the feds.
WRONG. That's a common fallacy--they're the government, they must know what they're doing. Here's the truth--people do what they're paid to do. The DEA gets its funds from Congress, and Congress wants to appear tough on drugs, so the DEA will do whatever it takes to give them what they're paying for--whether it is constitutional or not.
The profile they get of an Amtrak passenger is not fundamentally different from what they could get standing around the station observing people. The only difference is they are looking at data on a screen rather than faces and clothes.
WRONG. Ever heard of "expectation of privacy"? When I walk through an Amtrak terminal, I don't have any expectation that I won't be seen. When I buy a ticket, I have every expectation that my name and travel times are not publicly available. That makes (or should make) a BIG difference when it comes to law enforcement. Of course, the bigger issue is the transition from "identify the possible bad guys based on probable cause and track them" to "track everybody and see if anything interesting turns up."
In fact, this form of profiling has the potential to be less racially biased than face-to-face observation. When you get over the knee-jerk reaction and think about it, it really is preferable to the current system. The current system being basically "stop people with brown skin".
WRONG (at least as far as I'm concerned). I've never been a fan of "the lesser of two evils" and I'm especially not a fan of it when it comes to law enforcement.