Computer people are just like that. They type all day, keep an eye on the developments around them, and have good
memories, so it's no big deal for them to sit down one day and type up the highlights of all the developments they
have seen (within a narrow focus) in the last 20 years of their career.
That's exactly the reason for historians. Because all the things that those people remember will be wrong. Okay, that was hyperbolic, but I'll bet that if you do this with six people, you'll get six different versions of what happened, even with projects that are only ten years in the past, much less fifty (and that's the time scale we're already talking about).
The job of a historian is to take all those primary and secondary sources, grind them up, and tell us what he, she, or it thinks really happened. Because they'll never agree.
There's a very very steep learning
curve for emacs, especially if you don't want to use the (admittedly slightly better) Xemacs [....]
It's been fifteen years since I started using EMACS. I worked through the tutorial (C-H t) and just started working with it. I used "apropos" a lot. I still do. It took me two or three days to become as productive as I was with vi.
Menus? Don't use 'em. Why bother?--it's more work to reach for the mouse than to use the command, even if I have to type a prefix of the name.
As far as I'm concerned, if I can't run a program in it, it isn't a programmer's editor. Check out Microsoft Tu^H^HWord if you want a fine piece of editing machinery for beginners....
First of all, they granted the NSA the patent, not this guy.
I thought "this guy" was William Friedman. That's kind of like saying "some dude" in an article about cars when you meant "Henry Ford." Friedman was quite the crypto boy--in fact, many of his books are still available from Aegean Park Press . I highly recommend ELEMENTS OF CRYPTANALYSIS--it will really help you wreck the Cryptoquips....
This is just another example of that trite but true observation that whenever you read a news article about anything you know about, it's wrong. At least they give his credentials at the end of the article--he's Bill Gates' hairdresser or something like that....
I'm having a hard time buying that it "accidentally" looks like a minor Star Wars character. First of all, the likeness is nearly exact. Second, half the young geeks I've met are either Star Wars freaks or have reached their post-Star Wars period.
Their managers probably don't want to admit it was intentional because they're worried about a lawsuit and/or getting their monkeys spanked. Remember "project Sagan," later known as "Butt-head Astronomer?"
Imagine this scenario [someone copies your fingerprint...]
You mention that SecureID, S/Key, etc. are better. This sounds like pretty much the same thing with the addition that the token authenitcates the user to itself before authenticating to the remote system. If someone steals one and "copies your fingerprints(TM)," you just key up another one and revoke the stolen item. As I understand it, you can require a password for authentication in addition to the prints, too.
So how is this weaker than the systems that do exactly the same thing minus the fingerprint id?
Also, remember one or two scientists on the first A-bomb project were worried that the explosion would ignite all the oxygen in the atmosphere?
I believe the concern was starting a fusion reaction, not "burning all the oxygen."
You're right about the cloud--I thought the same thing when I read it. I assume that's an example of the reporter making up something that he thought sounded plausible (I don't know what the approved reportorial terminology for this is...). If not, then the article is probably a complete fabrication.
Incidentally, someone I don't remember once had a science fiction short story where a Coca Cola analog puts their logo on the face of the moon by spreading lamp black ballistically over the entire visible face. Cooler than doing it with a laser....
I've seen a lot of complaints like a) the source is printed in it, and I already run that on my Java keychain b) it doesn't cover XYZZY module (for instance, the file system) or c) it's out of date. These are all valid assessments, but I think there is another target audience this book suits perfectly.
This is the ideal book for someone who's studied operating systems in school but has never seen real source code. This seems to be the case for most folks I see that are new holders of a B.S. (or sometimes M.S.).
Theoretical descriptions in texts frequently skip the messy parts (otherwise, they might as well print the source code!). Once you've seen an example of how it's done with a few helpful annotations, you're better able to deal with other systems that are implemented differently.
I've seen a lot of online kernel documentation, but there isn't any of it that I'd point a new team member at in preference to this book, unless the thirty bucks was an issue. On the other hand, if said new team member is already a BSD kernel hacker, there will be much less benefit to reading it.
I did find a few clangers while I was browsing through it. For example, the author seems to think that the top half of a driver is the same as the hardware interrupt handler. Even so, it's a lot better than Inside Windows NT 2nd Edition, and several of the Linux kernel books I've read.
No, you don't understand what you're talking about. Both UNIX and NT are C1 systems.
If what you mean is, "both UNIX and NT possess the features needed to get a C1 rating," you're right. What you said is wrong, though. UNIX can't get a C1 because there isn't any single "UNIX". NT 3.51 with no networking was granted a C-something rating. This has nothing to do with reality, but doesn't prevent marketing using the rating.
There is a lot of extra overhead to do MAC checks and MAC intentionally sort of puts people in a prison.
It is true there is additional work necessary to check additional policies. Traditional DOD-style MAC checks are fast -- one comparison and a few ands and equality checks. This doesn't have to be significant. As for "putting people in a prison," you're actually more interested in putting Trojan Horse programs in prison--and if you don't need the feature, just give everything the same sensitivity label.
Microsoft won't comment on where Whistler is in the development process. But sources close to the company say the latest "stable" internal developers build is numbered 2207. The most recent internal test build is 2214, sources add.
I have confidential internal information that the release numbers correspond directly to the projected year of release....
[....] that HR ends up being the department that "champions" cutting the really interesting, set-your-company-apart-from-the-crowd type of benefits... and that only seems to happen when the company reaches some critical mass (200+ people).
Interesting observation. Perhaps what you've really discovered is the point at which a company usually hires some high-powered H.R. manager type.
My vast experience (okay, two companies in ten years) has been that you only find people that are members of the solution set in H.R. if they've recently entered the field.
My theory is that people above a certain level in H.R. are chiefly evaluated based on perceived "savings" they generate. I can't explain the higher "break point," but I'm heartened by the possibility of its existance.
Seems like something people in the cushy part of the world should learn to endure, some amount of minor suffering.
The heart of civilization is the ability to specialize. Speaking for everyone else, we volunteer to let you get all the colds, including the ones we aren't going to be getting any more. As "designated sufferer," you will have a ready market for screeds on your agonizing sinuses and post-nasal drip.
Till then, it's nice to know that female genital mutilation, land mines, child slavery, and crushing poverty are pretty much like the common cold....
When I was confined to a Winblows box for a while, I swore I wouldn't use InYerNeck Exploder, and I got fed up with Nutscrape when it insisted on bringing along his pal, the Little Yellow Man.
I know, I know, I should just have nuked LYM, but I got kind of scared to modify anything on NT because a) it takes all day to figure out how to do it and b) you usually break something else.
Anyway, Hot Java was interesting. My 350 Mhz PII took about three minutes to launch it. Then, I looked at a page that was supposed to be specially intended for HotJava. It said something like "if you're looking at this with a lesser browser, you won't see Little Jeb (or whatever his name is--the guy that looks like Tux run over with a steam roller) waving."
Needless to say, li'l Jeb was stationary. I decided I didn't want to have to get a four-processor PIII box just to browse the web, so I nuked it. Too bad, I kind of like programming in Java.....
How do you libel the government? I thought that political free speech was the most strongly protected of all. You can't even libel an office-holder unless you mention them by name....
Since most uSoft software has serious security flaws that are caused by applying power to the system, maybe there's a single point at which these problems could be fixed....
Someone submitting invalid data can ruin this entire project for everyone.
I'm feeling simple-minded today. Suppose I claim that I detected some spectacular message in the data I downloaded. How does the "ruin the entire project for everyone?" Does my message just get forwarded direct to a headline in the New York Times? If so, I think their protocol could use a little work....
SCI AM admittedly tends to be a bit more serious than, say, Popular Science
Most people don't know, but the official name of the magazine was changed to Popular Scientific American a number of years ago. The same time that they started printing profiles of scientists with their photos. (I understand next year they'll be requiring them to pose in swimming suits). For something much like the old Scientific American, I suggest you check out American Scientist. I've heard they got some of the SA staff, but can't verify that myself.
Seems like I recall some law or something that is supposed to keep people in the dark about comparative wages in an industry.
Don't know where you're posting from, but in the United States we have this Bill of Rights thing. Any law such as you describe would violate its free speech provisions.
Sounds like the kind of FUD that an H.R. drone would like to disperse, though.
Correct me if I'm wrong here....If you were going to purchase something online, and agree to sign a contract digitally, then wouldn't you already be able to access all the fine print (via secured web sight, physically printing the contract, secure e-mail, etc.)?
Okay, you're wrong.
Here's what I'd do if I was a company. I'd say "gee, here's a way to avoid notification costs, and also to screw people by not notifying them!" I'd have the legal department add a notice in the standard contract agreeing to digital notification. Most customers would just sign it without question, because no one reads those contracts anyway (at least people are always amazed when I do). The one percent that complained would be told that "it's just a formality--they don't notify anyone by email." I'd then provide a really inconvenient method for customers to receive email (i.e., an email-only Decwriter in one location in the metro area). Whatever would just barely meet whatever standard is mentioned in the law. And finally, I could do whatever I wanted so long as I sent email to myself first that would never be read.
Of course, I'm sure no business would ever try to screw their customers....
That would probably take nearly as long to figure out as the Cryptoquip in the morning paper. If someone is going to the trouble to sniff your keyboard, assuming they can't solve a monoalphabetic substitution is wishful thinking.
Now, if you put a one-time pad in the keyboard driver, you could fool them. Of course, it would slow your typing down a little, but you'd probably get good at it eventually. Might even break 1wpm....
Is it just me, or does anyone else find the concept of a product called "... for dummies" a slight on the customer?
I've refused to buy any "For Dummies" or "For Idiots" or "For Microcephalics" or whatever for a long time. My theory is that you either need a much weaker or a much stronger self image than I have to purchase one, though I imagine they make excellent gifts for that hard-to-please in-law.
[....] "SHEVA" is a obvious anagram of "Shiva" [....]
SHEVA = AEHSV
SHIVA = AHISV
AEHSV != AHISV
Or does "anagram" also mean "variant spelling" in some world?
I agree with the hundreds of other posters who noted that people love to give cheesy allusive names to things -- that's certainly the way it works in the computer world.
Slashdot Mirror
That's exactly the reason for historians. Because all the things that those people remember will be wrong. Okay, that was hyperbolic, but I'll bet that if you do this with six people, you'll get six different versions of what happened, even with projects that are only ten years in the past, much less fifty (and that's the time scale we're already talking about).
The job of a historian is to take all those primary and secondary sources, grind them up, and tell us what he, she, or it thinks really happened. Because they'll never agree.
Good luck to the original poster.
It's been fifteen years since I started using EMACS. I worked through the tutorial (C-H t) and just started working with it. I used "apropos" a lot. I still do. It took me two or three days to become as productive as I was with vi.
Menus? Don't use 'em. Why bother?--it's more work to reach for the mouse than to use the command, even if I have to type a prefix of the name.
As far as I'm concerned, if I can't run a program in it, it isn't a programmer's editor. Check out Microsoft Tu^H^HWord if you want a fine piece of editing machinery for beginners....
I thought "this guy" was William Friedman. That's kind of like saying "some dude" in an article about cars when you meant "Henry Ford." Friedman was quite the crypto boy--in fact, many of his books are still available from Aegean Park Press . I highly recommend ELEMENTS OF CRYPTANALYSIS--it will really help you wreck the Cryptoquips....
This is just another example of that trite but true observation that whenever you read a news article about anything you know about, it's wrong. At least they give his credentials at the end of the article--he's Bill Gates' hairdresser or something like that....
Their managers probably don't want to admit it was intentional because they're worried about a lawsuit and/or getting their monkeys spanked. Remember "project Sagan," later known as "Butt-head Astronomer?"
You mention that SecureID, S/Key, etc. are better. This sounds like pretty much the same thing with the addition that the token authenitcates the user to itself before authenticating to the remote system. If someone steals one and "copies your fingerprints(TM)," you just key up another one and revoke the stolen item. As I understand it, you can require a password for authentication in addition to the prints, too.
So how is this weaker than the systems that do exactly the same thing minus the fingerprint id?
I believe the concern was starting a fusion reaction, not "burning all the oxygen."
You're right about the cloud--I thought the same thing when I read it. I assume that's an example of the reporter making up something that he thought sounded plausible (I don't know what the approved reportorial terminology for this is...). If not, then the article is probably a complete fabrication.
Incidentally, someone I don't remember once had a science fiction short story where a Coca Cola analog puts their logo on the face of the moon by spreading lamp black ballistically over the entire visible face. Cooler than doing it with a laser....
This is the ideal book for someone who's studied operating systems in school but has never seen real source code. This seems to be the case for most folks I see that are new holders of a B.S. (or sometimes M.S.).
Theoretical descriptions in texts frequently skip the messy parts (otherwise, they might as well print the source code!). Once you've seen an example of how it's done with a few helpful annotations, you're better able to deal with other systems that are implemented differently.
I've seen a lot of online kernel documentation, but there isn't any of it that I'd point a new team member at in preference to this book, unless the thirty bucks was an issue. On the other hand, if said new team member is already a BSD kernel hacker, there will be much less benefit to reading it.
I did find a few clangers while I was browsing through it. For example, the author seems to think that the top half of a driver is the same as the hardware interrupt handler. Even so, it's a lot better than Inside Windows NT 2nd Edition, and several of the Linux kernel books I've read.
If what you mean is, "both UNIX and NT possess the features needed to get a C1 rating," you're right. What you said is wrong, though. UNIX can't get a C1 because there isn't any single "UNIX". NT 3.51 with no networking was granted a C-something rating. This has nothing to do with reality, but doesn't prevent marketing using the rating.
There is a lot of extra overhead to do MAC checks and MAC intentionally sort of puts people in a prison.
It is true there is additional work necessary to check additional policies. Traditional DOD-style MAC checks are fast -- one comparison and a few ands and equality checks. This doesn't have to be significant. As for "putting people in a prison," you're actually more interested in putting Trojan Horse programs in prison--and if you don't need the feature, just give everything the same sensitivity label.
I have confidential internal information that the release numbers correspond directly to the projected year of release....
Interesting observation. Perhaps what you've really discovered is the point at which a company usually hires some high-powered H.R. manager type.
My vast experience (okay, two companies in ten years) has been that you only find people that are members of the solution set in H.R. if they've recently entered the field.
My theory is that people above a certain level in H.R. are chiefly evaluated based on perceived "savings" they generate. I can't explain the higher "break point," but I'm heartened by the possibility of its existance.
The heart of civilization is the ability to specialize. Speaking for everyone else, we volunteer to let you get all the colds, including the ones we aren't going to be getting any more. As "designated sufferer," you will have a ready market for screeds on your agonizing sinuses and post-nasal drip.
Till then, it's nice to know that female genital mutilation, land mines, child slavery, and crushing poverty are pretty much like the common cold....
I know, I know, I should just have nuked LYM, but I got kind of scared to modify anything on NT because a) it takes all day to figure out how to do it and b) you usually break something else.
Anyway, Hot Java was interesting. My 350 Mhz PII took about three minutes to launch it. Then, I looked at a page that was supposed to be specially intended for HotJava. It said something like "if you're looking at this with a lesser browser, you won't see Little Jeb (or whatever his name is--the guy that looks like Tux run over with a steam roller) waving."
Needless to say, li'l Jeb was stationary. I decided I didn't want to have to get a four-processor PIII box just to browse the web, so I nuked it. Too bad, I kind of like programming in Java.....
How do you libel the government? I thought that political free speech was the most strongly protected of all. You can't even libel an office-holder unless you mention them by name....
Since most uSoft software has serious security flaws that are caused by applying power to the system, maybe there's a single point at which these problems could be fixed....
I'm feeling simple-minded today. Suppose I claim that I detected some spectacular message in the data I downloaded. How does the "ruin the entire project for everyone?" Does my message just get forwarded direct to a headline in the New York Times? If so, I think their protocol could use a little work....
Most people don't know, but the official name of the magazine was changed to Popular Scientific American a number of years ago. The same time that they started printing profiles of scientists with their photos. (I understand next year they'll be requiring them to pose in swimming suits). For something much like the old Scientific American, I suggest you check out American Scientist . I've heard they got some of the SA staff, but can't verify that myself.
Don't know where you're posting from, but in the United States we have this Bill of Rights thing. Any law such as you describe would violate its free speech provisions.
Sounds like the kind of FUD that an H.R. drone would like to disperse, though.
Okay, you're wrong.
Here's what I'd do if I was a company. I'd say "gee, here's a way to avoid notification costs, and also to screw people by not notifying them!" I'd have the legal department add a notice in the standard contract agreeing to digital notification. Most customers would just sign it without question, because no one reads those contracts anyway (at least people are always amazed when I do). The one percent that complained would be told that "it's just a formality--they don't notify anyone by email." I'd then provide a really inconvenient method for customers to receive email (i.e., an email-only Decwriter in one location in the metro area). Whatever would just barely meet whatever standard is mentioned in the law. And finally, I could do whatever I wanted so long as I sent email to myself first that would never be read.
Of course, I'm sure no business would ever try to screw their customers....
That would probably take nearly as long to figure out as the Cryptoquip in the morning paper. If someone is going to the trouble to sniff your keyboard, assuming they can't solve a monoalphabetic substitution is wishful thinking.
Now, if you put a one-time pad in the keyboard driver, you could fool them. Of course, it would slow your typing down a little, but you'd probably get good at it eventually. Might even break 1wpm....
I've refused to buy any "For Dummies" or "For Idiots" or "For Microcephalics" or whatever for a long time. My theory is that you either need a much weaker or a much stronger self image than I have to purchase one, though I imagine they make excellent gifts for that hard-to-please in-law.
SHEVA = AEHSV
SHIVA = AHISV
AEHSV != AHISV
Or does "anagram" also mean "variant spelling" in some world?
I agree with the hundreds of other posters who noted that people love to give cheesy allusive names to things -- that's certainly the way it works in the computer world.
Suydam: sendmail [...] has GOT to be the most complete, and configurable piece of software on earth.
Now playing on a computer near you: EMACS: The Extraterrestrial.