Dude! Are you still using Lisp Machines? This development just proves they're dying!
I was originally going to title this "Symbolics is Dying", but it turns out Symbolics is still around, selling things like Macsyma, just not making Lisp Machines any more.
C is still my favorite practical programming language. It's lightweight, elegant, terse, does exactly what you tell it, doesn't hide the machine from you, and people should stop using it for significant projects because too many people can't use it safely 100% of the time. Not only do they shoot themselves in the foot, but they're getting blood all over the floor and other people are slipping on it.
All of these mechanisms are providing steel-toed boots so that usually when you shoot yourself in the foot, the bullets ricochet harmlessly, making enough noise to wake you up, though sometimes the bullet goes through and nails your boot to the floor, but in neither case did you hit the target you actually wanted. It's there to reduce collateral damage, and to make programs crash loudly enough that you can debug them.
There are a variety of Emacs versions and imitators out there, and most of them are either written in some version of LISP or else interpret some version of LISP for many of their library functions. Usually it's Emacs's ELISP as opposed to Common Lisp or other major dialects.
Perhaps there aren't other popular counterexamples, but there's at least that one. I do remember the old Winterp XLISP-based window package - one of my friends used it for fast prototyping, and found that the prototypes usually had better performance than the evenual production code, as well as often having more features...
Replacing your SUV with a fuel-efficient car might be the "right thing" to do, but it might not, and that doesn't mean a Hybrid is the right choice of fuel-efficient cars. Any new car requires a lot of natural resources to build, and current battery technologies have toxic-waste issues, though hybrids require less battery capacity than full-electrics.
There have been lots of cars built that do better than 30MPG, some better than 40MPG. Most of them are light-weight uncool econo-boxes, and Fancy New Technology isn't necessarily better than old-fashioned cheapskate technology. My 1985 Toyota wagon got 27mpg or better for its first decade, and was still getting 25mpg (plus oil) when it was an old beater; it's annoying that my 2001 PT Cruiser only gets 22mpg on the highway. A friend of mine had a 45mpg car back in the 70s when he was stationed in Japan, though it probably wouldn't have been street legal in the US.
It might be that the "Right Thing" car for you to use most of the time is a motorcyle, whether that's a big bike for freeway commutes or a scooter for most city driving. Unfortunately, most US state insurance regulations mean that insurance companies are going to charge you lots more money if you've got more vehicles, even if you're not driving them much more, so having a bike that you ride on dry days and a less efficient car that you ride on rainy days may not be cost-effective, even though it's resource-efficient.
Miles Per Gallon* matters - it affects how much fuel you need to buy, as well as how much greenhouse-inducing CO2 plant food your car emits. (If you don't like the antiquated US measurement systems, kilometers-per-liter matters, but you've been paying extortion-level-taxed petrol prices for long enough that you already own a small car.)
The price you pay for your car also matters, because you're amortizing that as well as buying gas, so if it costs you an extra $10K to buy the car, that's about $1000-3333/year extra, depending on how long you're planning to keep the car, your cost of money, and future value of the potential added resale value if you eventually sell the car. So if buying a hybrid saves you $1000-3333/year in gasoline costs, it was worth spending the money, otherwise not. On the other hand, if you're replacing your leased-during-the-internet-boom BMW with a hybrid instead of another BMW, justifying it because the hybrid is worth as many Coolness Points as another luxury car, you've probably saved money so the lower gas prices are gravy.
I'm a relatively light driver - about 10,000 miles/year on my 1987 Chevy Van, which is about 600 gallons/year, plus my wife drives about 10,000 miles/year on her PT Cruiser, about 450 gallons/year. Replacing the van would saved me about 300-400 gallons/year, which would have been $500/year at last year's gas prices or $1000 at this year's prices, so it'd break even if I could do that for $10K + $2000-3000 tradein value, i.e. not too likely. Instead, since it's an old vehicle, I'm replacing it piece by piece:-) (This year's current pieces are a water pump and however much steering-column it takes to get the electrical system working again....)
One of my friends is on the iTunes staff, responsible for keeping the iTunes download system running. She spent New Years Eve party glued to her pager/palmtop watching the server stats instead of paying attention to the cute guy she's been seeing, because it was her turn to make sure everybody who got an iPod for Christmas could keep buying tunes. Shutting down iTunes for a week would mean that she and her co-workers could go on vacation....
Will variable pricing happen? Probably, though right now the two main prices in the market are 99 cents at iTunes and 0 cents at many other fine distributors. 99 cents per tune is typically $10-15 per album, which is about the same as the current prices from the big labels, i.e. way too high. On the other hand, many music consumers don't download all the tunes from an album, so overpriced single tunes, while a threat to the Album-Oriented Business Model, may not be as overpriced as they look.
Will some of the indie labels let Steve buy their tunes for less and sell them for less than 99 cents? It could happen, especially if the major labels conspire to raise their prices in an attempt to deal with the fact that they're doomed anyway.
I've got an iPod Shuffle. I've downloaded some 99-cent songs from Apple, and some free MP3s from legitimate sources (e.g. bands that let you download some songs from their websites and hope you'll download more, and amateur singers putting their music out for fun.) I would have ripped some CDs I already own, but WindowsXP doesn't seem to know how to get the audio track off my CD-burner, but you can either blame Bill Gates or Doctor Polaroid (burnmax48 drive) or the CD standards bureaucrats for that; it ain't Steve Jobs's fault.
As I said, MP3s play just fine on my iPod. If one major music distributor channel doesn't like Steve selling their tunes when the public would have been happy to buy them, there are other major labels and lots of indies who will let Steve sell their music. If all the major labels gang up to illegally price-fix their tunes by not selling to Steve or by raising their prices, well, they know that MP3s of their music plays just fine and there are lots of places to download them, Yarrrr!!!!, and meanwhile we can still get smaller labels' music from iTunes and lots of jam-band stuff off eTree.
These aren't the Internet Fractures you're looking for - you can move along.
DNS isn't "The Internet" - it's just DNS. Fracturing the root would be stupid, especially when it can be easily glued back together by sticking ".newTLD" under an existing TLD or 2LD, e.g. ".newTLD.net" or ".newtld.neustarsucks.com". But it wouldn't break The Internet - IP would still work, and protocols that embed DNS names into their interactions such as HTTP and SMTP can all support multiple names, e.g. the people who own foo.newtld might need to accept connections for www.foo.newtld.net and user@www.foo.newtld.net or they'd find that lots of people can't access their sites, but that's really no different than the problems faced by people who used the earlier Alternate Root projects such as Orange Root or ORSC or whatever.
ICANN is badly broken - it's not responsive to the user community, and the only "IP" it cares about is "Intellectual Property", not "Internet Protocol". That's why ICANN has insisted that everybody who registers domains in TLDs controlled by ICANN provide True Names and ICBM-addresses to facilitate trademark lawsuits, in spite of the major privacy problems with that change in whois semantics, and why it took them many years to add any additional TLDs, after taking over from the IETF Ad-Hoc Committee that had already developed a plan to do so.
However, most of the proposals for "Internet Governance" that the WSIS gang have come up with have been evil, clueless, or both.
ICANN doesn't control the Internet, only DNS policies and IP address assignments, and expanding that scope would be Bad.
China wants to "govern" the Internet by getting the rest of the world to enforce their censorship policies, which are currently too easy for Chinese citizens to evade by using non-China-based websites, email, and IM servers. A few other governments also want to use "governance" to censor pornography, free speech that criticizes them politics, and pornography. (Really, it's just about pr0n and evil nasty terrorists, pay no attention to that press censor behind the curtain.) ICANN currently has no control over this except perhaps blocking registry of Fulan-Gong.com
Some third-world countries want "Internet Governance" to tax rich Internet users to subsidize internet connectivity into their countres. Not only do they fundamentally misunderstand how the Internet works, the major problem in many of those countries is telecom monopolies that provide overpriced inadequate service, and the first step in getting their citizens decent internet access is to get the telco monopolies out of the way. That doesn't mean there aren't also infrastructure problems, or that an infusion of cash couldn't be useful, but in general they'd be giving more money and power to their PTT monopolies, which is mostly counterproductive.
I really hate treating ICANN as the Good Guys here, so I won't - this is a conflict between the Bad Guys and the Worse Guys.
DNS isn't The Internet - splitting DNS would be ugly, stupid, and easily repaired, e.g. by creating records like [newTLD].[existingTLD] or [newTLD].[NewTLDowner].net.
ICANN isn't controlled by the citizens of the US; it's basically autonomous, and went to a lot of work to free itself of the shackles of public participation in its board, especially any subset of the public including Karl Auerbach. It's occasionally influenced by the US Commerce Department, which is occasionally influenced by the Bush Administration, who are occasionally influenced by right-wingers, rich corporations, or cosmic rays, but AFAICT, the only effect the Bushies have had on it is to suggest that they'd like ICANN to create.xxx and then insist that ICANN delay.xxx because some other right-wingers didn't like it.
The "IP" that ICANN cares about isn't the "Internet Protocol" - it's "Intellectual Property". The real influence behind ICANN is the WIPO-mongers, whoever they are, that have gotten ICANN to insist that anybody who registers a domain name anywhere under their control provide enough information to serve them with a subpoena for a trademark lawsuit (and possibly for a website-content lawsuit from RIAA, etc., but it's really trademarks that matter.)
I've used Exchange/Outlook/MSMail for over a decade, and it really has improved a lot, but it's still a fundamentally broken system - in spite of how addictive the calendar system is once you start using it. Because the protocols aren't an open standard, you're either stuck using Outlook as a client if you want the full feature set, or using POP3 clients with limited features, and that means that all the things wrong with Outlook are your problems also.
The original product was designed to work in a NetBIOS LAN-only file-server environment, with a proprietary dialup interface hacked on later, and SMTP client support grudgingly and unreliably added even later. Back in 1994, it was the third-worst mail system I'd ever used, and I'd been online over a decade using and managing a wide variety of mail systems. Many BandAids, kilometers of duct tape, and spools of baling wire later, it's had some LDAP-like stuff added underneath, and it's possible to use on a laptop that's sometimes attached to your work LAN, sometimes disconnected, sometimes on random wireless or wired internet or private-net connections, and it usually doesn't get hosed up on me more than once a week, refusing to let me queue email or refusing to transmit the queued stuff or access my calendar or whatever, but it's really still made for people whose computers are tethered to their desktops 7x24x365 and have a big air-conditioned server farm for Exchange to live on. And don't even get me started about what helpful header-munging does to spam-filtering, or why I can't run my own black/white/keyword/filter lists without getting disconnected from the corporate filters our IT droids run.
The calendar's integration into Exchange's proprietary protocols is a major reason for its success and continued purchases of upgrades - but a Calendar program could just as easily have been built around HTTP/CGI, which would allow most email clients that support clicking on URLs to access it, and allow a much wider variety of client programs and clientless browser interfaces, so you wouldn't have to go to the overhead of firing up Outlook just to check your calendar - a major issue in a laptop environment.
US trials generally can't be photographed either. One of the networks did a re-enactment using actors, but I preferred Keith Olbermann's "Michael Jackson Puppet Theatre" on his Countdown news show on MSNBC. It had tacky little stick puppets, and entirely no respect for Jackson, his supporters, or his lawyers.
My undergrad was also at Cornell, many years ago, and sleep *is* for the weak. Most of the teachers were excellent (with a few exceptions, but fortunately not in critical foundation classes.) I did grad school at Berkeley, and the teaching was probably even better. The styles of the institutions were much different, though part of that was because it was grad school and not mass-quantity undergrad courses - Cornell expected lots of students would blow off some lectures and make it up by reading, problem sets, and lab sessions. Berkeley expected you to show up for everything, and expected the professors to make it worth your while. And I was married, rather than living in a fraternity, so my life was a bit calmer and I got to bed much earlier and more consistently, except on Thursday nights when I had to stay up late doing time series problem sets due the next morning (professor didn't think we should waste scarce computer time doing graphs, which goes to show what life was like before PCs, so I had to copy them all by hand after doing them on a computer....)
Inadequate teaching in fundamental courses like calculus is inexcusable, and any college that's failing its students like that needs a major wake-up call.
If it costs the content distributors less to distribute via BitTorrent, there's likely to be more stuff you want to download available. Some small indie garage studio may only be able to afford a T1 line, but BT lets them compete with the big studios.
You'll also get material you want faster and more reliably, because BitTorrent scales its capacity to meet the demand. Without BT, if the studio distributing an interesting movie trailer has to buy a gigabit ethernet connection, and ten thousand people with cable modems jump on it, you're still only going to get 10 kB/s, and you're fairly likely to get an incomplete copy. On the other hand, with BT, they need to have enough bandwidth to keep the tracker working and get the first copy out to the world, but it's a lot less critical.
I'm assuming that "U of Washington", like most state universities, charges higher tuition for out-of-state students than for state residents, typically substantially higher. So follow the money - they're really happy to have foreigners and any other out-of-state students who are forking over big bucks than to have in-state students getting in cheap. Did anybody see if the population of state residents was representative of the population of the state, or only the whole population?
Also, Asian immigrants, like Jewish immigrants in the 19th and 20th centuries, heavily value education, and make sure their kids work hard and get it. Most of the Asians who've moved here in the last few decades are educated as well, so they want their kids to get educated, but even the ones who weren't educated had the initiative to get off the farm or out of the slum and go for the opportunities, and just because they're working in restaurants doesn't mean they want their kids doing that when they grow up. Here in the San Francisco Bay Area, there are lots of after-school/weekend-tutoring businesses to help people make sure their kids get ahead, and when I drive by them at the times the parents are picking up their kids, it's about 90% Asians (mainly Chinese), a few Caucasians, and a few Mexicans. My town's probably about 1/4 Chinese and 1/3 Mexican; we don't seem to have a lot of Indians compared to some of the other towns in the area.
Ok, then, so let's go rerun the experiment with a Slashdot-sized set of subjects. If your Slashdot ID is even, go have some pot. If it's odd, go send some text messages to other odd-numbered Slashdotters tonight. End of the evening, we'll see who's got better karma, who had a better evening, and who just stayed home and ordered pizza...
Once you've gotten past the "It's not a skirt, it's a kilt", nobody's going to give you grief about your sporran resembling a purse. Of course, that depends on whether it's an appropriate shape and pattern to be a sporran as opposed to an obvious handbag.
Back when people carried big clunky calculators around, belt packs were fairly common on engineers, and nobody gave them any crap about it.
The verb tenses I learned in my school books in the Northeastern US didn't *always* match the way the kids on the street or the people on TV spoke, and most people didn't use the more complex Latin-like forms (subjunctives and optatives and the like) very much. But Southerners have all sorts of different verb forms, especially for future or potential future events. I'm not just talking about uneducated-white-boy Ebonics-equivalent or "ain't" or the assertion that a Southern accent is like losing 20 IQ points (which I've mainly heard from Southerners:-). It's forms like "I might could do that" or "I might coulda done that" or "I'd been fixin' to get around to that." Some of this is because of insular communities that have been around from various sets early-colonial British-Isles immigrants, and some probably has African influences, and some just kinda happened.
The Constitution forbids taking private property for public use without just compensation. If the Feds want to take the "intellectual property" for "National Security" or whatever reasons, they are required to compensate them - assuming the patent owner files a lawsuit in the right form asking the right questions. Doesn't sound like they've done that yet.
The benefit is that the key you generate using DH is disposable - you don't keep the keyparts after the session is over, so nobody can recover it. With RSA, you've encrypted the session key with the recipient's public key, so if somebody records the conversation and sometime later gets the private keys of the recipient, they can decode the key exchange and therefore the whole conversation
The way to avoid MITM attacks with DH is to digitally sign the parts of the key exchange, which you typically do with RSA or the ElGamal signature algorithms. There are other ways to do it - have each side read the other a fingerprint of the key they're using, but you can't automate that (or a MITM could fake it), so it's not really practical for everyday conversations, as opposed to using it for military security or something. Diffie likes the signed DH approach.
In RSA, the way you avoid MITM attacks is to make sure you've really got the other person's public key, which you typically do by using a Public Key Infrastructure or else by handing them the key on some other channel, such as printing a hash of the key on your business card. Skype does some keyserver stuff to automate this. But you're still vulnerable to somebody getting the private key later and being able to read all conversations they've wiretapped in the past.
That's $1.1M per crack, not 1.1M per machine, and no, the NSA is unlikely to have a better machine than that, because the academic crypto world has pretty much lef them in the dust the last decade or so. If you've pissed off the Feds enough that they're willing to spend $1M to crack one of your phone calls, you've got much more serious problems. It's still a lot cheaper for them to black-bag your laptop and install a key-logger to steal your passwords and email your microphone's signals to kgbvax.
But from a theoretical security perspective, even 512-bit RSA wouldn't be the weak link - it's likely to be key-handling or other protocol issues. How's the information handled, how's it stored, how's it refreshed, how are the random numbers generated, what leaks? Can it make a three-way conference call without telling you? Why isn't it using Diffie-Hellman for key exchange, with the RSA just for authentication? Without Adult Supervision by credible crypto professionals, you simply can't trust the stuff. The source doesn't need to be Free-As-In-Beer open, or Free-As-In-Speech open, or Accept-Patches-From-The-Public open, but it does need to have the design docs and the code reviewable by the community.
Slashdot Mirror
I was originally going to title this "Symbolics is Dying", but it turns out Symbolics is still around, selling things like Macsyma, just not making Lisp Machines any more.
Rob Pike (or maybe Ken Thompson} once gave a Plan 9 talk where he said that threads are for operating systems that can't do fork() very well...
C is still my favorite practical programming language. It's lightweight, elegant, terse, does exactly what you tell it, doesn't hide the machine from you, and people should stop using it for significant projects because too many people can't use it safely 100% of the time. Not only do they shoot themselves in the foot, but they're getting blood all over the floor and other people are slipping on it.
All of these mechanisms are providing steel-toed boots so that usually when you shoot yourself in the foot, the bullets ricochet harmlessly, making enough noise to wake you up, though sometimes the bullet goes through and nails your boot to the floor, but in neither case did you hit the target you actually wanted. It's there to reduce collateral damage, and to make programs crash loudly enough that you can debug them.
Perhaps there aren't other popular counterexamples, but there's at least that one. I do remember the old Winterp XLISP-based window package - one of my friends used it for fast prototyping, and found that the prototypes usually had better performance than the evenual production code, as well as often having more features...
There have been lots of cars built that do better than 30MPG, some better than 40MPG. Most of them are light-weight uncool econo-boxes, and Fancy New Technology isn't necessarily better than old-fashioned cheapskate technology. My 1985 Toyota wagon got 27mpg or better for its first decade, and was still getting 25mpg (plus oil) when it was an old beater; it's annoying that my 2001 PT Cruiser only gets 22mpg on the highway. A friend of mine had a 45mpg car back in the 70s when he was stationed in Japan, though it probably wouldn't have been street legal in the US.
It might be that the "Right Thing" car for you to use most of the time is a motorcyle, whether that's a big bike for freeway commutes or a scooter for most city driving. Unfortunately, most US state insurance regulations mean that insurance companies are going to charge you lots more money if you've got more vehicles, even if you're not driving them much more, so having a bike that you ride on dry days and a less efficient car that you ride on rainy days may not be cost-effective, even though it's resource-efficient.
The price you pay for your car also matters, because you're amortizing that as well as buying gas, so if it costs you an extra $10K to buy the car, that's about $1000-3333/year extra, depending on how long you're planning to keep the car, your cost of money, and future value of the potential added resale value if you eventually sell the car. So if buying a hybrid saves you $1000-3333/year in gasoline costs, it was worth spending the money, otherwise not. On the other hand, if you're replacing your leased-during-the-internet-boom BMW with a hybrid instead of another BMW, justifying it because the hybrid is worth as many Coolness Points as another luxury car, you've probably saved money so the lower gas prices are gravy.
I'm a relatively light driver - about 10,000 miles/year on my 1987 Chevy Van, which is about 600 gallons/year, plus my wife drives about 10,000 miles/year on her PT Cruiser, about 450 gallons/year. Replacing the van would saved me about 300-400 gallons/year, which would have been $500/year at last year's gas prices or $1000 at this year's prices, so it'd break even if I could do that for $10K + $2000-3000 tradein value, i.e. not too likely. Instead, since it's an old vehicle, I'm replacing it piece by piece :-) (This year's current pieces are a water pump and however much steering-column it takes to get the electrical system working again....)
One of my friends is on the iTunes staff, responsible for keeping the iTunes download system running. She spent New Years Eve party glued to her pager/palmtop watching the server stats instead of paying attention to the cute guy she's been seeing, because it was her turn to make sure everybody who got an iPod for Christmas could keep buying tunes. Shutting down iTunes for a week would mean that she and her co-workers could go on vacation....
Will some of the indie labels let Steve buy their tunes for less and sell them for less than 99 cents? It could happen, especially if the major labels conspire to raise their prices in an attempt to deal with the fact that they're doomed anyway.
As I said, MP3s play just fine on my iPod. If one major music distributor channel doesn't like Steve selling their tunes when the public would have been happy to buy them, there are other major labels and lots of indies who will let Steve sell their music. If all the major labels gang up to illegally price-fix their tunes by not selling to Steve or by raising their prices, well, they know that MP3s of their music plays just fine and there are lots of places to download them, Yarrrr!!!!, and meanwhile we can still get smaller labels' music from iTunes and lots of jam-band stuff off eTree.
DNS isn't "The Internet" - it's just DNS. Fracturing the root would be stupid, especially when it can be easily glued back together by sticking ".newTLD" under an existing TLD or 2LD, e.g. ".newTLD.net" or ".newtld.neustarsucks.com". But it wouldn't break The Internet - IP would still work, and protocols that embed DNS names into their interactions such as HTTP and SMTP can all support multiple names, e.g. the people who own foo.newtld might need to accept connections for www.foo.newtld.net and user@www.foo.newtld.net or they'd find that lots of people can't access their sites, but that's really no different than the problems faced by people who used the earlier Alternate Root projects such as Orange Root or ORSC or whatever.
However, most of the proposals for "Internet Governance" that the WSIS gang have come up with have been evil, clueless, or both.
- ICANN doesn't control the Internet, only DNS policies and IP address assignments, and expanding that scope would be Bad.
- China wants to "govern" the Internet by getting the rest of the world to enforce their censorship policies, which are currently too easy for Chinese citizens to evade by using non-China-based websites, email, and IM servers. A few other governments also want to use "governance" to censor pornography, free speech that criticizes them politics, and pornography. (Really, it's just about pr0n and evil nasty terrorists, pay no attention to that press censor behind the curtain.) ICANN currently has no control over this except perhaps blocking registry of Fulan-Gong.com
- Some third-world countries want "Internet Governance" to tax rich Internet users to subsidize internet connectivity into their countres. Not only do they fundamentally misunderstand how the Internet works, the major problem in many of those countries is telecom monopolies that provide overpriced inadequate service, and the first step in getting their citizens decent internet access is to get the telco monopolies out of the way. That doesn't mean there aren't also infrastructure problems, or that an infusion of cash couldn't be useful, but in general they'd be giving more money and power to their PTT monopolies, which is mostly counterproductive.
- I really hate treating ICANN as the Good Guys here, so I won't - this is a conflict between the Bad Guys and the Worse Guys.
DNS isn't The Internet - splitting DNS would be ugly, stupid, and easily repaired, e.g. by creating records like [newTLD].[existingTLD] or [newTLD].[NewTLDowner].net.It's occasionally influenced by the US Commerce Department, which is occasionally influenced by the Bush Administration, who are occasionally influenced by right-wingers, rich corporations, or cosmic rays, but AFAICT, the only effect the Bushies have had on it is to suggest that they'd like ICANN to create
The "IP" that ICANN cares about isn't the "Internet Protocol" - it's "Intellectual Property". The real influence behind ICANN is the WIPO-mongers, whoever they are, that have gotten ICANN to insist that anybody who registers a domain name anywhere under their control provide enough information to serve them with a subpoena for a trademark lawsuit (and possibly for a website-content lawsuit from RIAA, etc., but it's really trademarks that matter.)
The original product was designed to work in a NetBIOS LAN-only file-server environment, with a proprietary dialup interface hacked on later, and SMTP client support grudgingly and unreliably added even later. Back in 1994, it was the third-worst mail system I'd ever used, and I'd been online over a decade using and managing a wide variety of mail systems. Many BandAids, kilometers of duct tape, and spools of baling wire later, it's had some LDAP-like stuff added underneath, and it's possible to use on a laptop that's sometimes attached to your work LAN, sometimes disconnected, sometimes on random wireless or wired internet or private-net connections, and it usually doesn't get hosed up on me more than once a week, refusing to let me queue email or refusing to transmit the queued stuff or access my calendar or whatever, but it's really still made for people whose computers are tethered to their desktops 7x24x365 and have a big air-conditioned server farm for Exchange to live on. And don't even get me started about what helpful header-munging does to spam-filtering, or why I can't run my own black/white/keyword/filter lists without getting disconnected from the corporate filters our IT droids run.
The calendar's integration into Exchange's proprietary protocols is a major reason for its success and continued purchases of upgrades - but a Calendar program could just as easily have been built around HTTP/CGI, which would allow most email clients that support clicking on URLs to access it, and allow a much wider variety of client programs and clientless browser interfaces, so you wouldn't have to go to the overhead of firing up Outlook just to check your calendar - a major issue in a laptop environment.
Already spent too much on dental work this summer - sigh... Root canals are really annoying.
US trials generally can't be photographed either. One of the networks did a re-enactment using actors, but I preferred Keith Olbermann's "Michael Jackson Puppet Theatre" on his Countdown news show on MSNBC. It had tacky little stick puppets, and entirely no respect for Jackson, his supporters, or his lawyers.
Squiddy looks a lot like the Flying Spaghetti Monster, as featured in Boing-Boing and elsewhere.
Inadequate teaching in fundamental courses like calculus is inexcusable, and any college that's failing its students like that needs a major wake-up call.
You'll also get material you want faster and more reliably, because BitTorrent scales its capacity to meet the demand. Without BT, if the studio distributing an interesting movie trailer has to buy a gigabit ethernet connection, and ten thousand people with cable modems jump on it, you're still only going to get 10 kB/s, and you're fairly likely to get an incomplete copy. On the other hand, with BT, they need to have enough bandwidth to keep the tracker working and get the first copy out to the world, but it's a lot less critical.
Also, Asian immigrants, like Jewish immigrants in the 19th and 20th centuries, heavily value education, and make sure their kids work hard and get it. Most of the Asians who've moved here in the last few decades are educated as well, so they want their kids to get educated, but even the ones who weren't educated had the initiative to get off the farm or out of the slum and go for the opportunities, and just because they're working in restaurants doesn't mean they want their kids doing that when they grow up. Here in the San Francisco Bay Area, there are lots of after-school/weekend-tutoring businesses to help people make sure their kids get ahead, and when I drive by them at the times the parents are picking up their kids, it's about 90% Asians (mainly Chinese), a few Caucasians, and a few Mexicans. My town's probably about 1/4 Chinese and 1/3 Mexican; we don't seem to have a lot of Indians compared to some of the other towns in the area.
Ok, then, so let's go rerun the experiment with a Slashdot-sized set of subjects. If your Slashdot ID is even, go have some pot. If it's odd, go send some text messages to other odd-numbered Slashdotters tonight. End of the evening, we'll see who's got better karma, who had a better evening, and who just stayed home and ordered pizza...
Back when people carried big clunky calculators around, belt packs were fairly common on engineers, and nobody gave them any crap about it.
The verb tenses I learned in my school books in the Northeastern US didn't *always* match the way the kids on the street or the people on TV spoke, and most people didn't use the more complex Latin-like forms (subjunctives and optatives and the like) very much. But Southerners have all sorts of different verb forms, especially for future or potential future events. I'm not just talking about uneducated-white-boy Ebonics-equivalent or "ain't" or the assertion that a Southern accent is like losing 20 IQ points (which I've mainly heard from Southerners :-). It's forms like "I might could do that" or "I might coulda done that" or "I'd been fixin' to get around to that." Some of this is because of insular communities that have been around from various sets early-colonial British-Isles immigrants, and some probably has African influences, and some just kinda happened.
The Constitution forbids taking private property for public use without just compensation. If the Feds want to take the "intellectual property" for "National Security" or whatever reasons, they are required to compensate them - assuming the patent owner files a lawsuit in the right form asking the right questions. Doesn't sound like they've done that yet.
The way to avoid MITM attacks with DH is to digitally sign the parts of the key exchange, which you typically do with RSA or the ElGamal signature algorithms. There are other ways to do it - have each side read the other a fingerprint of the key they're using, but you can't automate that (or a MITM could fake it), so it's not really practical for everyday conversations, as opposed to using it for military security or something. Diffie likes the signed DH approach.
In RSA, the way you avoid MITM attacks is to make sure you've really got the other person's public key, which you typically do by using a Public Key Infrastructure or else by handing them the key on some other channel, such as printing a hash of the key on your business card. Skype does some keyserver stuff to automate this. But you're still vulnerable to somebody getting the private key later and being able to read all conversations they've wiretapped in the past.
But from a theoretical security perspective, even 512-bit RSA wouldn't be the weak link - it's likely to be key-handling or other protocol issues. How's the information handled, how's it stored, how's it refreshed, how are the random numbers generated, what leaks? Can it make a three-way conference call without telling you? Why isn't it using Diffie-Hellman for key exchange, with the RSA just for authentication? Without Adult Supervision by credible crypto professionals, you simply can't trust the stuff. The source doesn't need to be Free-As-In-Beer open, or Free-As-In-Speech open, or Accept-Patches-From-The-Public open, but it does need to have the design docs and the code reviewable by the community.