But Moore's Law and its other psuedoscientific corrolaries say that CPU horsepower and network bandwidth keep getting cheaper, so it may become cost-effective for them (or some competitor) to bring it back in a year or two, especially if the economy improves in a way that makes advertising revenue valuable again.
Just think what a Peer-to-Peer Doubleclick Reimplementation could get away with - if they could include their evil warez into a popular home server application, like a game system or IM/ICQ client or music-piracy\\\\\\\distribution system or whatever, they might be able to cut the costs of distributing and summarizing their advertising information.
It's easy to make doubleclick appear to go away - on Unix systems, I alias it to "127.0.0.2", which is the machine next to mine, instead of 127.0.0.1, but that only works for things named doubleclick.com. If they start naming their domain names things like adserver.customer1.com and innocuous.customer2.com, it becomes much harder to block, and if they do a different software version that runs a CGI on the same server that their customer uses to serve images, then you wouldn't be able to block it without blocking the interesting site, unless you examine more detail than just the domain name.
Ad services are already going beyond the banner model, with many major sites putting up bigger ads in the middle of their content (e.g. ZDnet) as well as developing annoying popups, popunders, using garish blinking, etc. As costs come down, building effective tracking tools will become easier, while developing better ways to make you interested in the advertiser's content will continue to be a hard problem. That suggests that better-tracked ads should have a growing value relative to less-tracked ads, though both may still decline in absolute value, cost, and price.
The first two lessons we learned were "here's what to do with a keypunch" and "if you don't comment your code we'll give you a bad mark even if it appears to work fine", but the first *real* lesson we learned was "Your program can *never* *ever* trust its input."
And to make sure we got the point, they'd make us run our programs on their input decks, which often had maliciously designed explorations of the limits of programs - what if the input field is missing, or too short, or too short by 1, or precisely as long as the maximum, or maximum+1, or way too long, or not a number, or a negative number, or had spaces in it, or had magic-looking values like 999 or 32767, or duplicated things that were supposed to be unique, or used values that weren't on the list of the-only-values-the-user-can-input. This was on Evil Mainframes with EBCDIC, so there are some modern forms of Bad Input that didn't exist (like backspaces or carriage returns in alphabetic fields ) but there were other evil things that could be done, like bogus punchcards, or characters that weren't from the 48-character character set the old printer supported or the 64-character set that the new one supported, or had data that ran into columns 73-80 which are only for sequence numbers. One of many annoying things about punchcard-oriented systems was that the edit-compile-run cycle was very slow, but it forced you to think very carefully about what you were doing. On the other hand, there are kinds of Bad Input that come from lots of experiments of throwing Nasty Looking Stuff into a program to see what it does that you wouldn't bother with on a punchcard system.
It's not really even 25 cities - it's really half the Canadian population living in half a dozen big cities, and 2/3 of the population / top 25 cities gets you down to places that are really Not Very Big, and it's all downhill from there. I've also heard that something like 90% of the population lives within 50 miles of the US border (not really true, but if you exclude Calgary and Edmonton, it's pretty close.)
Sure, the parts of Canada that not many people live in, like the parts of Australia that not many people live in, or of Alaska, are much less dense than a similar fraction of the land-area than in the US, and much much less than China, but most people live in a city big enough for broadband to be cost-effective, and near enough to the US border that they could get US TV, at least if cities like Buffalo and Detroit had much TV:-)
In most of the country, there are only a few DSL layer-2 providers (at most the telco, Covad, and the late lamented Northpoint, Rhythms, Jato, etc., and maybe a few upstarts), and often only the telco, but they'll provide PVCs from your house to any ISP that wants to buy a T3 line (or sometimes a T1 line) to deliver the PVCs to them, and those ISPs offer a range of services from simple IP packet forwarding to email to web hosting to shell accounts to whatever. Some DSL providers may do IP routing down at the DSLAM, or with a small amount of ATM-based regional concentration, though many of them use evil technologies like PPPoE or PPPoATM to retain lots of control over the packets they deliver to the ISPs. But for regulatory reasons, most telcos do pretend to offer service to multiple ISPs, and Covad and the other CLECs did it for business reasons.
Part of the problem is that the Bell telcos and their competitors talked the government into giving them monopolies on voice telephony early in the last century. Part of it is that the cable TV companies talked lots of small towns into giving them "franchise" monopolies in the 60s-80s (usually selected not by who had the most forward-looking telecom policies, but by whose brother-in-law got the street paving contract), and then took advantage of their small-region monopolies into talking pro-regulation Congresscritters like Markey into giving them more power.
But a major problem is that Roosevelt's New Deal FCC quasi-nationalized the radio spectrum, giving broadcasting rights to big broadcasting businesses (most of the Roosevelt anti-trust activity was really trading around which big powers got market control, unlike their rhetoric about supporting the little guys) to prevent competition in return for regulation of content, and limiting most other spectrum use to business-speech-verboten applications such as "ham radio", which also wasn't allowed to use privacy protections and limited access to highly technically skilled people. By limiting the number of stations allowed to broadcast, there's an inherent push toward central powers buying them up. But also, by preventing widespread free-speech radio deployment except for the ubergeeks of the day, they prevented the use of radio as a common home telephony substitute, which slowed the development of radio systems as well as limiting telephony to capital-intensive wired systems which are often a bad economic choice for rural areas. There were exceptions - CB Radio was the AOL of the 70s, especially as truckers discovered that ham radio equipment could be used to give them kilowatt CB instead of the legal-maximum 5 watts - but most of them weren't generally usable.
Right of way is one of the obvious critical issues for running fiber. Most long-haul connections are run along railroads, or highways, or gas pipelines, and all the telcos buy lots of right of way along them. Sprint was originally the Southern Pacific Railroad's internal telecom business, before they spun it off into a company. Williams was in the gas pipeline business, and started several telecom businesses (including Wiltel, which was bought by Worldcom) initially using pipeline right-of-way and in some cases, installing inside unused/spare/obsolete pipelines, which is MUCH easier than trenching. Also, Williams had the advantage that their "Don't Dig Here" signs didn't say "Fragile, Expensive Glass - please be careful", which attracts backhoe drivers - they said "Gas Pipeline - If you dig here, it'll blow up and you'll die!", which most people can figure out not to dig near:-)
Go walk along a railroad track sometime, especially near bridges over rivers or the crossings with big highways. You'll often see multiple "Telco A - Don't dig here" "Telco B - Don't Dig Here" signs, which should give you some idea of how much redundancy you actually get by buying service from multiple companies...
Unlike you commies in the north, who have a small enough population and large enough tradition of mercantilism to have one big railroad company, down here in the warmer states, we had a number of different railroad companies competing with each other. (No, no, not Purist Randian Free-Market competition - they were competing for who could out-bribe which politicians to get the best land deals for themselves and get whatever kinds of monopoly they could to keep other competitors out, or at least get contiguous enough chunks of land to block the other companies from crossing their tracks. I said we're not THAT different....)
The US also had other issues - much more of the land that "Manifest Destiny" said must surely belong to Americans was occupied by inconvenient other people who thought they already lived there, like the Indians, or who thought they'd already stolen it from the Indians, like the Mexicans, so there were a few decades of expensive war to clear them out, plus there was the bit of unpleasantness that the losers referred to as the "War Between The States" that took a while to put down, and also provided a diversion because lots of people wanted to rip off whatever they could in the South. And there was that minor "54-40 or Fight" attempt to steal British Columbia from the Brits.
I'd guess that Canada had less trouble from Indians, but also had a narrower area that was good for building railroads on?
Typical speeds for cable modem were about 3Mbps downstream, except that AT&T throttled the users they took over from @Home to 1.5Mbps downstream, and either 128kbps or 768kbps upstream depending on whether your cable modem company has new enough equipment to enforce the lower bandwidth.
DSL speeds are dependent on distance from your telco office - Maximum is usually 1.5Mbps or 1.1Mbps, 384kbps is common, 128 kbps is available farther away (it uses ISDN physical layer with DSL applications instead of telephone switches), and many people have asynchronous service that's either 1.5Mbps/384kbps or 608/128, where your downstream bandwidth may be lower than maximum depending on distance and conditions.
I'm under the impression that Canadian cable modem systems are usually faster than US systems, particularly upstream bandwidths (unless they're measuring in Canadian bits, which are half the size of US bits:-)
I strongly agree with bricriu'scomment. But I'd extend it beyond that - the TOS-don't-provide-any-services rules prevent development of other interesting services as well as preventing development/delivery of interesting content. Some applications get developed anyway, particularly at universities that have ethernet-wired dorms, such as Napster and its clones and followons. Some get developed even in the dialup world, like Cu-SeeMe and ICQ. Some of them have been around since the beginning - it's silly to go through the complication of uploading your vacation pictures or pictures of your kids or cats to an advertising-run web server offering 20 meg of free space when you've got 20 GIG of space on your disk drive. Sure, the commercial service may have better bandwidth and reliability, and may be a good place for a front page pointing to your home system, but most pictures you'd want to serve will work just fine in a low-throughput environment, because most people's home sites don't get heavy use. (Obvious exceptions are music-sharing, but the systems that have sufficiently scalable indexing can put up with slow uploads.)
What kinds of applications would people develop if it only took creativity and technical skill and wasn't forbidden by usage policies? Most interesting applications include at least some kind of server somewhere - even an ICQ or IM "client" is technically a server, because it's sitting there on your system waiting for people to connect to it, and it's often advertising its presence using some kind of presence server (the ICQ login stuff or Napster index servers or whatever.) Some successful applications were carefully planned by a small or large group of people, but many of them just happened - somebody tried it, and a few people liked it, and it caught on. And the more opportunity you have for people to develop things that probably won't catch on, the more chance that somebody will develop things that DO succeed. Maybe it'll be a "neighborhood watch" or "home traffic/weather cam" application, or maybe cheap cameras and better PC audio will allow the ICQ-phone to replace large chunks of the phone company (so duhh, either team with a gateway company like Net2Phone or a long-distance phone company to profit from professionalizing it), or maybe simply getting $40/month instead of $80/month from people working at home over VPNs is enough to be happy with, or maybe you can provide a $5/month IP relay service an 802.11 client software so that wireless users will become paying customers instead of service-stealing evil leeches. Or maybe it won't come from home developers, it'll come from game developers, like the integration of networking, Dancepads, and Quake into Combat Aerobics, or the World Wide Rave Network, burning its 15 minutes of fame before something else takes over. Whatever. More likely, it'll be something I haven't thought of, and much more likely, it'll be something the cable companies haven't thought of, because it's a decentralized decision-making process, not central planning.
But if you're a cable modem company desperately needing enough customers to sign on to pay for growing your infrastructure, decreasing the chances of potential customers finding the killer-for-them app that makes *them* want to buy service from you is really, Darwinianly stupid.
Cablecos do have things they're legitimately afraid of, though it was worse in the past than today. Upstream bandwidth is still limited, and people running popular amateur porn or warez websites on their cable modems could dog down performance for their neighborhoods (unlike commercial sites, which need better performance than the typical 128kbps upstream of current cable modem.) And that gives them a bad reputation for performance, and encourages the local phone company to run "Web Hog" ads taunting them. And Napster and Movie-ripoff-ster and other copyright-violation-promoting services directly hurt the business of their major business partners, so they need some way to discourage them. And the band on "email servers" is partly driven by fear of spammers, though it's largely driven by the sheer corporate greed assumption that if it's a mail server, it's either a business that you'd be willing to pay more money for or that you're taking away potential cablemodem customers instead of encourage more people to get cable. But blanket "can't serve anything because we don't want to monitor your content or upgrade our hardware to meter" policies are just stupid.
Moore's commentary on Sturgeon's law says that the 90% of stuff out there that's crap keeps doubling every 18 months, and typical Freshmeat experience says that lots of projects will die out before they reach usable stages. But that's ok, and if we're lucky many of them would be in the 90% and not the 10%, or that the ideas in the good ones will get recycled by somebody else.
OK, so it's intended as humor, but it really *is* relevant to deciding what kinds of messages to send to Extra Terrestrials. Maybe they're not made of meat.
Just by coincidence, I was hearing radio ads today for Lily Tomlin's play "Search for Signs of Intelligent Life in the Universe" which she's doing here in San Francisco. Terrestrial intelligence being in such short supply, perhaps she can help...
But it's also *about* personalization
on
Making It Personal
·
· Score: 1
The most effective way to make money fast by spamming, other than credit card fraud (:-), is to take the lists of suckers who have responded to spam and sell them to other spammers.
But of course most of the business is non-personalized, such as the lists for 9 billion validated email addresses. It would be nice if they at least ran some basic coordination, so I wouldn't be getting ads for both penis enlargement AND breast enlargement though:-)
Back when I used to do lots of business travel, I found this extremely helpful. If I got an aisle seat and a vegetarian meal, I knew they had my frequent flyer number in the registration, whereas if I got a window seat and no veg. meal, I knew there was a high chance that corporate travel had screwed up again, and I should give the airline my frequent flyer number and start haggling with Travel and Amex to make sure they weren't charging my flight to some *other* w.stewart in the company, who would probably soon be wondering who'd flown to Australia on his credit card:-)
Those pigeons aren't in the Eighth Dimension - they're somewhere over New Jersey
There is a way to make compression like this work - for each string you want to compress, there's a compression program that losslessly compresses it to an arbitrarily short output string (one bit is fine...), but if the output string is N bits long, the program only works for 2**N input strings, and in general requires SIZE(INPUT) bits of program per input string (though for non-random strings, or for related strings, you can do better.) In other words, it's not useful for general-purpose compression, but you can use it for special-purpose compression - you can't design a small compression program to perfectly describe "Alice"'s or "Bob"'s appearance, but you can design a small program that outputs "Alice", "Bob", or "Somebody else".
Similarly, with pigeons, you can play Hundred-Pigeon Monte, and attract investors to your company, or use this to attract customers for your other products, or have a big crowd on the street intently watching you play hundred-pigeon monte with your shill while a pickpocket walks around behind the crowd.
Pons and Fleischman, as near as I can tell, believed they had some interesting physics going on, though they were mistaken about quite what, and jumped to publication way prematurely. (As somebody said about their work "If it's not real, they've still invented the world's most interesting battery".)
This is more like Usenet Crank Robert E. McElwaine who published lots of articles with his (capital-preserving) tagline "UN-altered REPRODUCTION and DISSEMINATION of this IMPORTANT Information is ENCOURAGED."
And that may be giving them more credit than they deserve - it looks like a compression algorithm designed for use on digital wallets....
That's the real secret of this device, is that it was designed with mounting brackets to attach on top of the Ginger aka Segway scooter. Just think about what you want to compute, and it'll go do that, as well as displaying the output of your digital video camera so you can see where you're going on the screen.
If you want a low-end PC, you're looking at more like $400-500, not $1000 - that's retail with a cheapo monitor. Spending $1000 should get you some significant upgrades. Typical retail prices for a motherboard and 900MHz AMD are about $99, 40-60GB disk is about $100. $50 for 512MB RAM. $20 for ether. If you're not a gamer, spend $29 on video card and use the built-in audio. If you *are* a gamer, spend an extra $150 for video and $50 for audio and upgrade the power supply and add joysticks, heavy speakers, chair-shaker...:-) CD-burners are commonly $79, as are DVD/CD players.
If you do want that gorgeous 22-inch Apple flat-panel monitor, now _that_ will set you back actual cash, and even 17-19" flat panels are in the $500 range.
Of course they'd be Mac users - most news companies are probably willing to spend a bit more on hardware in return for reducing the time they spend on user education, broken machines holding up people on deadlines, and sysadmin labor, and they're willing to ignore the occasional smart-quote turned into a Q,R,or S in return for getting their work done. Also, Mac layout software is relatively common, so it's a compatibility thing.
And besides, before they got Macs, lots of people in that business used to use things like XYWrite and Wang Word Processors, and after using the Mac they've got no desire to go back to the old days, or even halfway back:-)
It's true, the seek time is better on faster-RPM drives. But fundamentally, for an MP3 player, you don't care, because the throughput of the drive is much much faster than the 128kbps you're playing tunes at - so other than the user interface hit of a less than ten milliseconds the first time you hit the "Play" button and it reads the first track into cache, the drive's got plenty of time to seek before you need to read the next one, and your file system is probably designed to keep most tunes in mostly sequential files anyway.
That's the problem with having too much karma -- You forget to turn off the "give yourself an extra point" on articles that frankly don't deserve it:-)
Sigh. It's getting to be like that old joke about the Soviet digital watch (from back when digital watches were still cool) - solid, brightly lit, but carrying around the car battery to keep it running took half the fun out of it.
Of course the whole case is bogus, and allowing cops to exercise warrants on people in secret rather than to their face are bogus, and laws against gambling are bogus, especially in states that run lotteries themselves, and racketeering laws that make conspiracy to repeatedly run gambling games into Federal crimes are bogus (yer winnings, governor!), and bogus laws like that encourage gambling to be run by thugs like Scarfo, and the idea that Feds should be able to call technology like this "classified information" when you can buy products that do this on the street and when they're lobbying Congress to let them develop better ones is bogus, but leaving all of that aside.... There's a difference between the Feds sniffing the passphrase, which is indirect evidence, and sniffing the contents of the file as he typed it, which would have been more direct evidence had they done that. The Feds are trying to hide how they stole the passphrase, and they're arguing about exactly what kind of warrant is needed for stealing it (wiretap vs. search warrant), but once they've stolen the passphrase and legally obtained the encrypted file, they can use it to show a jury that the passphrase they stole decrypts the file into the text they're alleging that Scarfo typed which allegedly shows that he's a mobster. And if they'd simply guessed the passphrase (hint, don't use simple words or your father's prison ID # as your passphrase) they could have done the same. By contrast, if they'd used the SEEKRIT keyboardsniffer to snarf up the file itself, they'd have to tell the jury "Nicky really typed this incriminating letter, trust us, we can't tell you how we know that, cuz it's RILLY SEEKRIT, but we're the FBI and we'd never lie to you, so he's GUILTY GUILTY GUILTY", they'd have a much weaker case. (Any self-respecting jury would throw them out on their expletive-deleted for even trying that, but American juries often fall for that sort of thing, and judges fall for it even more often.)
US rules of evidence, since the early-1960s Supreme Court decisions which promulgated the "Exclusionary Rule", say that you can't use illegally obtained evidence, and there's a doctrine called "Fruit of the Poisoned Tree" which says that if you illegally obtain information that you use to obtain other information, you can't use that as evidence either. So if they'd beaten or tortured the information out of Scarfo, or if they hadn't had a warrant when they first searched his computer, they'd be unable to use it legally, which is part of why Scarfo's lawyers were arguing about the precise type of warrant they needed before stealing his passphrase.
On the other hand, if they'd gone asking around the mobster social club if anybody wanted to call in an anonymous tip with Nicky's usual passwords or offering get-out-of-jail-free cards to temporarily-retired mobsters in return for the passphrase, that'd be legal, and unlike the cases where stool pigeons give false testimony about people in return for reduced jail time, a passphrase is demonstrably either correct or incorrect. (And of course, an "anonymous tip" is often nearly indistinguishable from illegally gathered evidence used to obtain a search warrant.)
It's not so much like the bag of dope on your car seat - it's more like them sneaking in your house, copying your car keys, opening your trunk, and finding the bag of dope there.
The economics keeps changing, political clout keeps changing, the price of congresscritters keeps changing, the needs and desires of the public keeps changing. The 1992 law wasn't written in stone or amended into the Constitution - it's just a law, and a newer law can supersede it, change it, discard it, replace it with something better or worse or different, depending on whatever 51% of both houses of Congress can be talked into voting for this week. It's not a Supreme Court decision about Constitutionality that has some sticking power, and even those evolve.
So when the act came out, did you believe it was the record industry's Final Answer, or did you believe it was the stinkin' camel's nose getting into the tent, or just another milepost along the way?
It's one of those "eternal vigilance" things. If you care about it, keep watching them, keep lobbying your Congresscritters, support groups like the EFF and ACLU and NRA and keep watching them too.
Slashdot Mirror
Just think what a Peer-to-Peer Doubleclick Reimplementation could get away with - if they could include their evil warez into a popular home server application, like a game system or IM/ICQ client or music-piracy\\\\\\\distribution system or whatever, they might be able to cut the costs of distributing and summarizing their advertising information.
It's easy to make doubleclick appear to go away - on Unix systems, I alias it to "127.0.0.2", which is the machine next to mine, instead of 127.0.0.1, but that only works for things named doubleclick.com. If they start naming their domain names things like adserver.customer1.com and innocuous.customer2.com, it becomes much harder to block, and if they do a different software version that runs a CGI on the same server that their customer uses to serve images, then you wouldn't be able to block it without blocking the interesting site, unless you examine more detail than just the domain name.
Ad services are already going beyond the banner model, with many major sites putting up bigger ads in the middle of their content (e.g. ZDnet) as well as developing annoying popups, popunders, using garish blinking, etc. As costs come down, building effective tracking tools will become easier, while developing better ways to make you interested in the advertiser's content will continue to be a hard problem. That suggests that better-tracked ads should have a growing value relative to less-tracked ads, though both may still decline in absolute value, cost, and price.
And to make sure we got the point, they'd make us run our programs on their input decks, which often had maliciously designed explorations of the limits of programs - what if the input field is missing, or too short, or too short by 1, or precisely as long as the maximum, or maximum+1, or way too long, or not a number, or a negative number, or had spaces in it, or had magic-looking values like 999 or 32767, or duplicated things that were supposed to be unique, or used values that weren't on the list of the-only-values-the-user-can-input. This was on Evil Mainframes with EBCDIC, so there are some modern forms of Bad Input that didn't exist (like backspaces or carriage returns in alphabetic fields ) but there were other evil things that could be done, like bogus punchcards, or characters that weren't from the 48-character character set the old printer supported or the 64-character set that the new one supported, or had data that ran into columns 73-80 which are only for sequence numbers. One of many annoying things about punchcard-oriented systems was that the edit-compile-run cycle was very slow, but it forced you to think very carefully about what you were doing. On the other hand, there are kinds of Bad Input that come from lots of experiments of throwing Nasty Looking Stuff into a program to see what it does that you wouldn't bother with on a punchcard system.
Sure, the parts of Canada that not many people live in, like the parts of Australia that not many people live in, or of Alaska, are much less dense than a similar fraction of the land-area than in the US, and much much less than China, but most people live in a city big enough for broadband to be cost-effective, and near enough to the US border that they could get US TV, at least if cities like Buffalo and Detroit had much TV :-)
In most of the country, there are only a few DSL layer-2 providers (at most the telco, Covad, and the late lamented Northpoint, Rhythms, Jato, etc., and maybe a few upstarts), and often only the telco, but they'll provide PVCs from your house to any ISP that wants to buy a T3 line (or sometimes a T1 line) to deliver the PVCs to them, and those ISPs offer a range of services from simple IP packet forwarding to email to web hosting to shell accounts to whatever. Some DSL providers may do IP routing down at the DSLAM, or with a small amount of ATM-based regional concentration, though many of them use evil technologies like PPPoE or PPPoATM to retain lots of control over the packets they deliver to the ISPs. But for regulatory reasons, most telcos do pretend to offer service to multiple ISPs, and Covad and the other CLECs did it for business reasons.
But a major problem is that Roosevelt's New Deal FCC quasi-nationalized the radio spectrum, giving broadcasting rights to big broadcasting businesses (most of the Roosevelt anti-trust activity was really trading around which big powers got market control, unlike their rhetoric about supporting the little guys) to prevent competition in return for regulation of content, and limiting most other spectrum use to business-speech-verboten applications such as "ham radio", which also wasn't allowed to use privacy protections and limited access to highly technically skilled people. By limiting the number of stations allowed to broadcast, there's an inherent push toward central powers buying them up. But also, by preventing widespread free-speech radio deployment except for the ubergeeks of the day, they prevented the use of radio as a common home telephony substitute, which slowed the development of radio systems as well as limiting telephony to capital-intensive wired systems which are often a bad economic choice for rural areas. There were exceptions - CB Radio was the AOL of the 70s, especially as truckers discovered that ham radio equipment could be used to give them kilowatt CB instead of the legal-maximum 5 watts - but most of them weren't generally usable.
Go walk along a railroad track sometime, especially near bridges over rivers or the crossings with big highways. You'll often see multiple "Telco A - Don't dig here" "Telco B - Don't Dig Here" signs, which should give you some idea of how much redundancy you actually get by buying service from multiple companies...
The US also had other issues - much more of the land that "Manifest Destiny" said must surely belong to Americans was occupied by inconvenient other people who thought they already lived there, like the Indians, or who thought they'd already stolen it from the Indians, like the Mexicans, so there were a few decades of expensive war to clear them out, plus there was the bit of unpleasantness that the losers referred to as the "War Between The States" that took a while to put down, and also provided a diversion because lots of people wanted to rip off whatever they could in the South. And there was that minor "54-40 or Fight" attempt to steal British Columbia from the Brits.
I'd guess that Canada had less trouble from Indians, but also had a narrower area that was good for building railroads on?
DSL speeds are dependent on distance from your telco office - Maximum is usually 1.5Mbps or 1.1Mbps, 384kbps is common, 128 kbps is available farther away (it uses ISDN physical layer with DSL applications instead of telephone switches), and many people have asynchronous service that's either 1.5Mbps/384kbps or 608/128, where your downstream bandwidth may be lower than maximum depending on distance and conditions.
I'm under the impression that Canadian cable modem systems are usually faster than US systems, particularly upstream bandwidths (unless they're measuring in Canadian bits, which are half the size of US bits :-)
What kinds of applications would people develop if it only took creativity and technical skill and wasn't forbidden by usage policies? Most interesting applications include at least some kind of server somewhere - even an ICQ or IM "client" is technically a server, because it's sitting there on your system waiting for people to connect to it, and it's often advertising its presence using some kind of presence server (the ICQ login stuff or Napster index servers or whatever.) Some successful applications were carefully planned by a small or large group of people, but many of them just happened - somebody tried it, and a few people liked it, and it caught on. And the more opportunity you have for people to develop things that probably won't catch on, the more chance that somebody will develop things that DO succeed. Maybe it'll be a "neighborhood watch" or "home traffic/weather cam" application, or maybe cheap cameras and better PC audio will allow the ICQ-phone to replace large chunks of the phone company (so duhh, either team with a gateway company like Net2Phone or a long-distance phone company to profit from professionalizing it), or maybe simply getting $40/month instead of $80/month from people working at home over VPNs is enough to be happy with, or maybe you can provide a $5/month IP relay service an 802.11 client software so that wireless users will become paying customers instead of service-stealing evil leeches. Or maybe it won't come from home developers, it'll come from game developers, like the integration of networking, Dancepads, and Quake into Combat Aerobics, or the World Wide Rave Network, burning its 15 minutes of fame before something else takes over. Whatever. More likely, it'll be something I haven't thought of, and much more likely, it'll be something the cable companies haven't thought of, because it's a decentralized decision-making process, not central planning.
But if you're a cable modem company desperately needing enough customers to sign on to pay for growing your infrastructure, decreasing the chances of potential customers finding the killer-for-them app that makes *them* want to buy service from you is really, Darwinianly stupid.
Cablecos do have things they're legitimately afraid of, though it was worse in the past than today. Upstream bandwidth is still limited, and people running popular amateur porn or warez websites on their cable modems could dog down performance for their neighborhoods (unlike commercial sites, which need better performance than the typical 128kbps upstream of current cable modem.) And that gives them a bad reputation for performance, and encourages the local phone company to run "Web Hog" ads taunting them. And Napster and Movie-ripoff-ster and other copyright-violation-promoting services directly hurt the business of their major business partners, so they need some way to discourage them. And the band on "email servers" is partly driven by fear of spammers, though it's largely driven by the sheer corporate greed assumption that if it's a mail server, it's either a business that you'd be willing to pay more money for or that you're taking away potential cablemodem customers instead of encourage more people to get cable. But blanket "can't serve anything because we don't want to monitor your content or upgrade our hardware to meter" policies are just stupid.
Moore's commentary on Sturgeon's law says that the 90% of stuff out there that's crap keeps doubling every 18 months, and typical Freshmeat experience says that lots of projects will die out before they reach usable stages. But that's ok, and if we're lucky many of them would be in the 90% and not the 10%, or that the ideas in the good ones will get recycled by somebody else.
OK, so it's intended as humor, but it really *is* relevant to deciding what kinds of messages to send to Extra Terrestrials. Maybe they're not made of meat.
Just by coincidence, I was hearing radio ads today for Lily Tomlin's play "Search for Signs of Intelligent Life in the Universe" which she's doing here in San Francisco. Terrestrial intelligence being in such short supply, perhaps she can help...
But of course most of the business is non-personalized, such as the lists for 9 billion validated email addresses. It would be nice if they at least ran some basic coordination, so I wouldn't be getting ads for both penis enlargement AND breast enlargement though
Back when I used to do lots of business travel, I found this extremely helpful. If I got an aisle seat and a vegetarian meal, I knew they had my frequent flyer number in the registration, whereas if I got a window seat and no veg. meal, I knew there was a high chance that corporate travel had screwed up again, and I should give the airline my frequent flyer number and start haggling with Travel and Amex to make sure they weren't charging my flight to some *other* w.stewart in the company, who would probably soon be wondering who'd flown to Australia on his credit card :-)
There is a way to make compression like this work - for each string you want to compress, there's a compression program that losslessly compresses it to an arbitrarily short output string (one bit is fine...), but if the output string is N bits long, the program only works for 2**N input strings, and in general requires SIZE(INPUT) bits of program per input string (though for non-random strings, or for related strings, you can do better.) In other words, it's not useful for general-purpose compression, but you can use it for special-purpose compression - you can't design a small compression program to perfectly describe "Alice"'s or "Bob"'s appearance, but you can design a small program that outputs "Alice", "Bob", or "Somebody else".
Similarly, with pigeons, you can play Hundred-Pigeon Monte, and attract investors to your company, or use this to attract customers for your other products, or have a big crowd on the street intently watching you play hundred-pigeon monte with your shill while a pickpocket walks around behind the crowd.
This is more like Usenet Crank Robert E. McElwaine who published lots of articles with his (capital-preserving) tagline "UN-altered REPRODUCTION and DISSEMINATION of this IMPORTANT Information is ENCOURAGED."
And that may be giving them more credit than they deserve - it looks like a compression algorithm designed for use on digital wallets....
That's the real secret of this device, is that it was designed with mounting brackets to attach on top of the Ginger aka Segway scooter. Just think about what you want to compute, and it'll go do that, as well as displaying the output of your digital video camera so you can see where you're going on the screen.
If you do want that gorgeous 22-inch Apple flat-panel monitor, now _that_ will set you back actual cash, and even 17-19" flat panels are in the $500 range.
At least they're not the old NeXt flopticals...
And besides, before they got Macs, lots of people in that business used to use things like XYWrite and Wang Word Processors, and after using the Mac they've got no desire to go back to the old days, or even halfway back
It's true, the seek time is better on faster-RPM drives. But fundamentally, for an MP3 player, you don't care, because the throughput of the drive is much much faster than the 128kbps you're playing tunes at - so other than the user interface hit of a less than ten milliseconds the first time you hit the "Play" button and it reads the first track into cache, the drive's got plenty of time to seek before you need to read the next one, and your file system is probably designed to keep most tunes in mostly sequential files anyway.
That's the problem with having too much karma -- You forget to turn off the "give yourself an extra point" on articles that frankly don't deserve it :-)
Sigh. It's getting to be like that old joke about the Soviet digital watch (from back when digital watches were still cool) - solid, brightly lit, but carrying around the car battery to keep it running took half the fun out of it.
There's a difference between the Feds sniffing the passphrase, which is indirect evidence, and sniffing the contents of the file as he typed it, which would have been more direct evidence had they done that. The Feds are trying to hide how they stole the passphrase, and they're arguing about exactly what kind of warrant is needed for stealing it (wiretap vs. search warrant), but once they've stolen the passphrase and legally obtained the encrypted file, they can use it to show a jury that the passphrase they stole decrypts the file into the text they're alleging that Scarfo typed which allegedly shows that he's a mobster. And if they'd simply guessed the passphrase (hint, don't use simple words or your father's prison ID # as your passphrase) they could have done the same. By contrast, if they'd used the SEEKRIT keyboardsniffer to snarf up the file itself, they'd have to tell the jury "Nicky really typed this incriminating letter, trust us, we can't tell you how we know that, cuz it's RILLY SEEKRIT, but we're the FBI and we'd never lie to you, so he's GUILTY GUILTY GUILTY", they'd have a much weaker case. (Any self-respecting jury would throw them out on their expletive-deleted for even trying that, but American juries often fall for that sort of thing, and judges fall for it even more often.)
US rules of evidence, since the early-1960s Supreme Court decisions which promulgated the "Exclusionary Rule", say that you can't use illegally obtained evidence, and there's a doctrine called "Fruit of the Poisoned Tree" which says that if you illegally obtain information that you use to obtain other information, you can't use that as evidence either. So if they'd beaten or tortured the information out of Scarfo, or if they hadn't had a warrant when they first searched his computer, they'd be unable to use it legally, which is part of why Scarfo's lawyers were arguing about the precise type of warrant they needed before stealing his passphrase.
On the other hand, if they'd gone asking around the mobster social club if anybody wanted to call in an anonymous tip with Nicky's usual passwords or offering get-out-of-jail-free cards to temporarily-retired mobsters in return for the passphrase, that'd be legal, and unlike the cases where stool pigeons give false testimony about people in return for reduced jail time, a passphrase is demonstrably either correct or incorrect. (And of course, an "anonymous tip" is often nearly indistinguishable from illegally gathered evidence used to obtain a search warrant.)
It's not so much like the bag of dope on your car seat - it's more like them sneaking in your house, copying your car keys, opening your trunk, and finding the bag of dope there.
So when the act came out, did you believe it was the record industry's Final Answer, or did you believe it was the stinkin' camel's nose getting into the tent, or just another milepost along the way?
It's one of those "eternal vigilance" things. If you care about it, keep watching them, keep lobbying your Congresscritters, support groups like the EFF and ACLU and NRA and keep watching them too.