It's been a few years since I looked into the hemp paper issues, so things may have changed, but the main problem with hemp paper is that it tends to curl a bit when heated, so it's not very useful in laser printers or copiers - it'll usually work ok for one-sided printing, but isn't well-behaved for two-sided, because it doesn't like to stack flat enough and tends to jam. (I was trying to do brochures that needed to be two-sided, so I went with recycled-dead-tree paper instead.)
This was before ink-jet printers totally took over the small-volume printer market, and I don't know how it is for ink absorbency.
And of course smoking isn't the best use for marijuana - you can make brownies with it, and be kinder to your lungs.
You can set up your basic Delaware corporation for $100 or so. If it gets busted, bummer, you've burned that $100 bill, but as long as you sold two bottles of Herbal Fake Viagra Substitute pills, you've made enough to pay for the next shell corporation. I've tracked one spammer to an address in Greenville Delaware, which was the same street address as The Company Corporation (which has been the canonical place to get your Delaware corporation set up for the last 100-plus years), so they weren't even trying to hide it with a mailbox company.
If you structure things right, the corporation that sends the spam is renting all their hardware from your virtual hosting company and hiring snailmail fulfillment services from your shipping company, so if they get busted you express that you're shocked, shocked to find AUP violations being perpetrated by *your* customers and send them an angry note cancelling their services, while running the perl script that creates a clone with a different name to sell the same worthless junk.
Following the money trail requires there to be enough money that's easily confiscated and no jurisdictional borders in the way. Usually the trail leads to some loser in a trailer park, or some server in China rented by some bogus entity. If it does lead to one of the ROKSO Top 200 spammers, it probably ends up with some disposable corporate shell that can go bankrupt rather than getting the real spammer.
If you do your own VOIP and your own IPSEC, yes, you can prevent wiretapping. If you use a VOIP-to-Telco carrier like Vonage or AT&T, there might or might not be encryption on the IP segment (and you've definitely got no control over it), but they have to support wiretapping at the head end of the connection. If you do Skype, there's definitely encryption available, but the protocols are proprietary and closed so you've got no way to be sure, and SkypeOut is in a fairly similar position to other carriers. If you're using other types of emerging VOIP service providers, anything that the FBI can get its nasty little claws into has wiretapping features (hint - what's the difference between a wiretap and a 3-way call to a voicemail system? Encryption isn't going to help you there.
Raw telephone bandwidth is 64kbps (that's 8 KBytes/sec.) The most common VOIP compression rate in the US is 8kbps, though there's also some 6.5kbps, 13kbps, 32kbps, etc., plus IP overhead. There are a bunch of sources of latency, some of which are easier to fix than others.
PC interrupt-handling latency - dedicated IP phones usually don't have this problem.
Sampling rate latencies - most VOIP algorithms do their calculations on 10ms, 20ms, or 30ms samples, depending partly on the amount of data the algorithm needs to get its compression and partly on tuning parameters to minimize IP overhead (if you send half as many packets per second, you cut the overhead in half, and the overhead's a lot larger than the real packets.)
Silence Detection Timing - You can save a lot of bandwidth by not transmitting in a given direction when nobody's talking, but this balances out against latency if you don't want to be too aggressive about clipping. (This is an easy thing to tune badly, and it's really annoying if done wrong.)
Jitter buffers - Jitter is short-time-period variation in packet transfer latency, which is largely caused by variations in packet sizes, queuing, and variations in arrival times of packets from different packet streams. To prevent calls from sounding like M-m_M-Maxxxx-Headddr-r-r-r-ooo-mmm, the receiving side of a voice call typically has a buffer of some number of milliseconds deep to smooth out jitter problems, so there's always some data queued up to decompress and play. The more jitter your network has, the bigger the buffer you need, and that can be a lot of the latency, especially for untuned environments like the open internet between multiple carriers. A single carrier can usually do a lot better, and fat access lines help a lot. (Also, jitter is _really_ hard to measure well enough to write a service level agreement about.)
Typical targets are more like 150ms, and some people forget that India and Singapore are far away from the US East Coast no matter how good a network you've got underneath.
QoS can help with latency if you're using a network that has significant queuing delay, but won't help the speed of light any. It can also help with jitter - but you also need fat enough pipes and/or really small MTU sizes. For instance, a 1500 byte packet on a 56kbps line takes about 220ms, so even with QoS, if some big FTP packet finally got its turn on the wire, you've got to wait for it to clear out. On a T1 line (1536kbps), it's only about 10ms, so it's less of a problem, but you don't want to get stuck behind a bunch of them.
You're mixing up a bunch of different issues here. First of all, international dialup used to be expensive partly because undersea cables were expensive and partly because monopoly telcos could charge whatever outrageous price they wanted because they were the monopoly. Telecom liberalization has gotten to most of the world, as people realize that low-cost communications leads to increased business and job creation, and as competition keeps getting its foot in the door with technologies like call-back services. Moore's Law has long since caught up with the cost per bit of transmission, and telecom company bankruptcies have meant that you can buy useful undersea cables for pennies on the original construction dollar (e.g. Global Crossing Asia selling for 1% of its original cost.)
Digital and VOIP are much different issues - traditional circuit-switched communications depends on relatively centralized intelligence on relatively old-technology switches, and while phone calls _do_ turn into digital when they hit the first telephone exchange, that's 64kbps digital, and it mostly stays that speed except for international circuits where voice compression pays off. There are some VOIP services that use 64kbps, but it's pretty common to compress to 8kbps or so. Depending on how you handle IP headers, this can inflate to ~25kbps without header compression or ~11kbps with header compression. Compressed voice doesn't sound quite as good as telco-quality voice, but one thing cell phones have taught the market is that people will put up with that, and your VOIP phone can sound just fine since it has a decent microphone and you're not using it in a car with traffic noise in the background.
The costs of switching equipment are substantially different - for VOIP, you do a database lookup at the beginning to find your destination, but after that, it's all IP routing, and routers have not only become dirt-cheap and scale very far, but all the heavy lifting really gets done by the CPUs at the caller's phone or PC, and the prices of wholesale internet transmission have been in total free-fall for a couple of years, especially including international transmission.
Another major change is VOIP-based PBXs, which have taken advantage of the PC hardware commoditization curve better than traditional PBXs have - if you're building a new office, there's usually no reason to use a non-VOIP PBX. It might or might not pay to spend the capital costs to rip out your old PBX - that depends a lot on features, which depends a lot on whether your old PBX was made by a stuffy clueless telco or by somebody who understood that open systems are critically important for their own developers even if they're not giving out the source code.
Here in the US, the big pricing anomaly for VOIP is that traditional wireline telcos charge about 2 cents/minute for delivering calls to a user, so a VOIP company has to either do per-minute pricing or else charge a high monthly flat rate and hope they win statistically, even though their costs of the long-distance part of the connection may amortize to 0.1 cents/minute. Also, a huge amount of the current long-distance switching infrastructure's complexity is to support call-center features for toll-free calling, which is something that people cared a lot more about when phone calls were 25 cents/minute than when they become 0.1 cents/minute.
Moving phones from one office to another, or changing the phone numbers for phones without moving the phones, has been a standard feature on many PBXs for a long time. If you're paying $100 to phone service guys to do it, then you've either got a really dumb PBX or a really dumb contract with your PBX guys (though such contracts are often part of hiring somebody else to manage your PBX instead of training your folks to manage a PBX yourself, which can be a lot of work for a small company.) New PBXs have been smart enough to do that since about the mid-1980s when digital PBXs came out, and certainly anything built in the 90s was that smart because the software really ran on a PC-equivalent even if it was hidden in the box with the switching fabric.
If your PBX appears too dumb to do that, it's probably because that's an "extra value-added application feature" that's not in the application package you bought, though real difference is probably whether the user interface menu items are enabled or disabled rather than whether the API behind it works.
The FBI are trying to extend their tentacles into the VOIP world, because people can make phone calls they can't easily tap with their existing methods, and it's a good excuse to get Congress to let them tap more of the Internet so they can stop the EEEEVILLLL TERRORISTS AND DRUG DEALERS. It's really annoying that they're trying to interfere with technology because technology's getting away from them.
The big problem for the Feds is that VOIP is structured fundamentally differently than traditional phone switching. Call setup uses servers, which can be run by anybody in the world (but are usually run by a company's internal telecom department for IP PBXs, by a VOIP-to-telco gateway company for many commercial VOIP services, or by software companies with ill-defined business models or popular Instant Messaging services with proprietary protocols), and the voice channel of the call is carried separately on whatever IP networks connect the callers, which may not be under the control of the gatekeeper (call setup server) providers. There are two fun places to hack this kind of communications - VOIP-to-telco gateways (where you can tap the telco side using traditional methods, but you have to bully the VOIP provider into giving you call detail records for the IP side, which is much tougher in the decentralized VOIP world than in traditional telcos), and in the call setup gatekeepers themselves, which can tell the callers to set up the call through an unexpected conference bridge service instead of directly (since the caller's VOIP software is designed to believe that sort of request and wouldn't have an easy way to check it if didn't trust the gatekeeper.) The latter obviously requires hacking the gatekeeper, which is even harder to do in some environments, since it's possible to use a service with distributed gatekeepers that might not even be in the jurisdiction of the cops who want to wiretap you.
Disclaimer time: I work for a large telecom company, and the following is strictly my personal opinion, not theirs: I work for a large telecom company, so the rules about obeying wiretap orders appear to apply when we're providing managed VOIP services for customers. But if a router or PBX vendor wants to sell hardware to a customer, that hardware doesn't need to have wiretapping service built in, and if the hardware vendor wants to sell management services for the hardware, they're not a telecom company so wiretapping rules don't appear to apply to them (unless the government thinks up some new rules.) From a competitive standpoint, this is really really annoying, especially because as a civil libertarian, I think wiretapping is wrong and rude and often done illegally, and a huge amount of CALEA and followon laws and regulations seems to be designed to make massive-scale illegal or at least unsupervised wiretapping easy.
AT&T and the LECs have been selling that for years also. The service name is often SINA (Static Integrated Network Access), and you can connect some channels to voice and others to data applications, which originally tended to be private line or frame relay but these days is often Internet. Buying two T1 lines isn't guaranteed to get you any useful diversity - it protects you against many kinds of equipment failures, but both lines are almost guaranteed to take the same physical route back to the nearest telco office, so one backhoe or water leak can take out both of them as easily as one.
The savings depends a lot on your calling patterns - are you making enough IP calls to justify the cost of the hardware you need? You can save money if you're making lots of calls between your locations (especially international locations), or if you're using a VOIP phone company that lets you deliver calls to a VOIP-to-Telco gateway, but if your calling patterns aren't right, you might not be able to justify the hardware cost unless you needed to buy a new PBX anyway (in which case it'll probably be VOIP.) Alternatively, you can save money on long distance calls by telling your long distance company sales rep that the X cents/minute rate they're charging you is laughably high in today's market and you can get a better price buying phone cards at the 7-11, and have them grumble and cut your price yet again.
Disclaimer: I work for a large telecommunications company which can sell you any of these services or hardware, so we make depressingly thin margins regardless of which option you take, and no, this isn't an official company position, it's just my opinion.
Here in Collie-Fornia, laptops in the front seats of cars have been illegal since January, as are other digital devices with display screens that aren't built-in and are forward of the driver's seat-back. I'm not sure if that includes DVD players in the front seat section (or if that's already illegal), and the law doesn't appear to have been intended to ban cellphones, but it's definitely not legal for the driver or even the passenger to operate laptops in the front seat. The motivation is to prevent accidents caused by distracted drivers, but wardriving gets hit as a side effect, unless you do it from the back seat while your chauffeur drives.
If you're writing a law that can throw people in jail or fine them huge amounts of money, you really have the responsibility to get the details right, and I don't think Congress has the skill set to do it without doing lots of collateral damage.
For instance, one common piece of Evil Spyware offers to be a helpful toolbar for your browser that'll give you more information, but really offers you occasional advertising and transmits your browsing preferences back to their company.
Yet Google Toolbar for IE does much the same thing, it's just more upfront about it and less obnoxious about ads and presumably better behaved with the data it collects.
Many spammers include web bugs in their HTML email to track whether you've read their material, either as 1x1 nearly-invisible GIFs or as big jumpy advertising GIFs/JPGs.
When _you_ send somebody HTML with URLs for your vacation pictures on your web site, you can also check your web server logs to see if they've accessed the pictures.
It's mostly about _intent_, and Congress is sufficiently clueless technically that they're not likely to be able to write a clearly defined law that covers either the intent or the technology.
Just because the display is damaged doesn't mean that the laptop can't work just fine with a monitor, often at higher resolution than with the original screen. You _can_ use it as a server with no display, but you're not limited to that.
One of the three PCs on my desk is a laptop with a dead screen that works just fine with the monitor when I switch the KVM switch over to it, and keeps a couple of old Win95 apps alive; when we're done with them I'll probably turn it into a Linux server of some sort, but it's happy for now, or I may just use it as a printer server.
I remember Rob and Ken ranting against microkernels back in the day, but Plan 9 was smaller and cleaner than most of its competition, and things that were viewed as "hopelessly bloated" a decade ago are now comparatively small and under-featured. How do we get back to relatively small kernels and fast reliable operating systems without ditching all the useful things that have been added since then? Hardware support for PCs and connected devices including accelerated video cards and USB plugins are critical issues.
Rob gave a talk at ?Nashville Usenix showing off the 8.5 Windowing System. One of his comments was that "Ken and I spent 10 years learning what things windowing system shouldn't do and we've written one that doesn't do them." 8.5 (the ".5" is properly the Unicode "1/2" glyph...) is small and blazingly fast, firing up a windowing system in about the time you'd expect a normal carriage return to give you a shell $ prompt back on a 680x0 machine, and the Acme windowing user interface on it was also lean, mean, and very fast.
He's not an African-American if he hasn't lived here and wasn't born here. But I don't care if his papers are in order - if he lives here for a while and wants to stay, regardless of whether he's a citizen, non-citizen, Green Card, long-term-tourist, or whatever, then, yeah, I'd call him an African-American even if it tweaks the politically-correct definition.
Back when I was in college and the term was newer, I had an Afrikaner friend who referred to himself and his wife as African-Americans, and they were, just as you can be an Italian-American regardless of what your skin colour is.
One of my current friends was from Rhodesia, and became a Zimbabwean (the name changed in 1980 after the British Empire recognized the independence they'd declared for themselves in 1965) before he moved here to the US as a teenager, and yeah, he's African-American in spite of being a light-skinned blondie. (I'm pretty sure he's both young enough and old enough that he was born Rhodesian, but I don't know if he's also a UK subject from the disputed-ownership period...)
Somebody else mentioned a Moroccan friend calling himself African-American - that really doesn't tell you whether he looks Negro, because Morocco is an extremely diverse country. I'm guessing from context that he's probably Arab, but I've met Moroccans ranging from Berbers in the south (black, but mostly looking quite different from Sub-Saharan West Africans or East Africans) to Arabs to Sephardic Jews to leftover French bureaucrats.
Sure, it'd be cool to hop out with your surfboard and oxygen tanks and try it, but it's not likely that SpaceShip One's aerodynamics are designed to have the door opened at full speed. They might be able to get away with it because there's very little atmosphere up there and they're going relatively slowly at the top of the curve. But also, the lack of atmosphere means you'd probably need a pressure suit and not just oxygen tanks, so your maneuverability once you're down to low altitude will be pretty limited.
Sure, people who want to actually work in space would be willing to live in a tin can - but if you want a lot of people up there, either as tourists or as workers, you need a multi-user tin can, with enough resources to handle a bunch of people and enough redundancy that equipment failures don't become emergencies. So you might as well call it a hotel, and have somebody working up there as concierge / desk clerk / janitor / cook / bottlewasher.
It's easy to find lots of this - check out your local railroad tracks and look for the "Telecom Cable - Don't Dig Here" signs, especially if you see more than one of them (e.g. Local Telco, AT&T, Sprint, MCI all along the same railroad). Call up the "Don't Dig Here" center and say that you're Bubba from Bubba's Backhoe Company and you want to make sure you don't do something stupid, and they'll be real happy to talk to you, because too many guys like Bubba don't even bother calling unless the sign says "Gas Pipeline - Don't Dig Here or It'll Explode and You'll Die".
At least one major US telecom company used to regularly fly planes along major cable routes looking for construction activities - not sure if they still do that. There were usually over a thousand backhoes within a quarter mile of their east coast cable routes, but they'd still get surprises like that farmer burying a dead cow with his backhoe in the early 90s.
It's one thing to be unable to get information about your phone company or their outages, and it's annoying as a consumer, but it's not life-threatening (disclaimer - I work for a large telecomm company, and own stock in several others, and this is just my opinion, not the official opinion of any of them.)
But Terrorists-Under-The-Bed have been used as an excuse for blocking public access to lots of critical safety information, particularly in industries like oil refining, chemical manufacturing, and anything nuclear, where there can be serious risks of toxicity, chemical spills, and even major explosions (I'm not ranting about nukes here - fertilizer plant explosions are much more likely.) The Feds, who used to force public disclosure of lots of this information are now banning it, and databases that used to be accessible are being closed to the public, because Terrorists and other enemies of American Industry (like anti-pollution activists and various other NIMBYs) might abuse it.
Yeah, I'm glad somebody else thought of it and posted it - I realized the joke a few minutes after my original posting, but following up my own post would have been tacky:-)
If they've implemented it right, even if the Trust-Us-chip gets activated, that shouldn't bother your Linux any. It might prevent you from running Windows, or applications that use the chip, unless you've got the right licenses.
The main feature of the "trusted computing" platforms is that you can write software that will only run natively on the platform and won't run in conditions it doesn't want to run under, such as running on a computer that wasn't the one it was licensed for. So you _could_ write "Trusted Malware" that only runs on trusted platforms and refuses to infect untrusted platforms, but that kind of misses the point of most malware..... I suppose you could use it to make a virus that was harder to dissect than regular viruses, though it might not have enough potential victims to propagate efficiently unless "Trusted Computing" platforms really take off.
The trusted computing platform can prevent some kinds of malware from running, by limiting what you can do to the operating system without the platform's consent. Depending on how it's implemented, that could limit the spread of the malware, or could just turn it into a denial-of-service attack against the operating system, or could turn it into a denial-or-service against other Trusted-Platform-Only software (because the platform looks like it's been tampered with, the copy-protection in "Example Trusted FPS Game" decides not to let you play.) But more likely, malware vendors will concentrate on writing applications like IE Plugins that'll run fine on trusted platforms.
"Trusted Computing" doesn't mean that you trust your computer. It means that other people can write software that can trust your computer not to run in ways that they don't want (e.g. without paying them per view, or without making sure you've licensed the software on that hardware, etc.)
What "Trusted Computing" gains you as a user is the ability to buy and run software written for the Trusted-By-Them environment, which the vendors might not sell for non-trusted platforms. So if you want to play "Example FPS Game" or play music CDs using the "Evil Record Label's Protected Music Player", using a trusted-by-them platform lets you do that. If you don't run a Trusted-By-Them platform, they might not sell you their game/music at all, or they might sell you a CD that plays the low-res version of the music on untrusted platforms, or sell you the crippleware version of the game that only has levels 1-3, or lets you play the whole game but doesn't include the Surround Sound or graphics accelerator support or something.
The argument that the space race or various military arms race programs were worthwhile because they brought us all these cool non-space spinoff things like communications satellites, velcro, and Tang is bogus. Yes, those things were potentially valuable, but they not only diverted money from the applications that the private sector could have used for other applications, they diverted a lot of really talented engineers and scientists from working on private-sector applications. What kinds of things would they have built if they hadn't been making rockets? Would they have improved the energy efficiency of cars, airplanes, and houses by 10%? Would they have built the Franistan that would have revolutionized the industry? (You've never seen or used a franistan, but that's because the team that would have invented it were improving military helicopter gunnery systems instead.)
Slashdot Mirror
This was before ink-jet printers totally took over the small-volume printer market, and I don't know how it is for ink absorbency.
And of course smoking isn't the best use for marijuana - you can make brownies with it, and be kinder to your lungs.
If you structure things right, the corporation that sends the spam is renting all their hardware from your virtual hosting company and hiring snailmail fulfillment services from your shipping company, so if they get busted you express that you're shocked, shocked to find AUP violations being perpetrated by *your* customers and send them an angry note cancelling their services, while running the perl script that creates a clone with a different name to sell the same worthless junk.
Following the money trail requires there to be enough money that's easily confiscated and no jurisdictional borders in the way. Usually the trail leads to some loser in a trailer park, or some server in China rented by some bogus entity. If it does lead to one of the ROKSO Top 200 spammers, it probably ends up with some disposable corporate shell that can go bankrupt rather than getting the real spammer.
If you do your own VOIP and your own IPSEC, yes, you can prevent wiretapping. If you use a VOIP-to-Telco carrier like Vonage or AT&T, there might or might not be encryption on the IP segment (and you've definitely got no control over it), but they have to support wiretapping at the head end of the connection. If you do Skype, there's definitely encryption available, but the protocols are proprietary and closed so you've got no way to be sure, and SkypeOut is in a fairly similar position to other carriers. If you're using other types of emerging VOIP service providers, anything that the FBI can get its nasty little claws into has wiretapping features (hint - what's the difference between a wiretap and a 3-way call to a voicemail system? Encryption isn't going to help you there.
Digital and VOIP are much different issues - traditional circuit-switched communications depends on relatively centralized intelligence on relatively old-technology switches, and while phone calls _do_ turn into digital when they hit the first telephone exchange, that's 64kbps digital, and it mostly stays that speed except for international circuits where voice compression pays off. There are some VOIP services that use 64kbps, but it's pretty common to compress to 8kbps or so. Depending on how you handle IP headers, this can inflate to ~25kbps without header compression or ~11kbps with header compression. Compressed voice doesn't sound quite as good as telco-quality voice, but one thing cell phones have taught the market is that people will put up with that, and your VOIP phone can sound just fine since it has a decent microphone and you're not using it in a car with traffic noise in the background.
The costs of switching equipment are substantially different - for VOIP, you do a database lookup at the beginning to find your destination, but after that, it's all IP routing, and routers have not only become dirt-cheap and scale very far, but all the heavy lifting really gets done by the CPUs at the caller's phone or PC, and the prices of wholesale internet transmission have been in total free-fall for a couple of years, especially including international transmission.
Another major change is VOIP-based PBXs, which have taken advantage of the PC hardware commoditization curve better than traditional PBXs have - if you're building a new office, there's usually no reason to use a non-VOIP PBX. It might or might not pay to spend the capital costs to rip out your old PBX - that depends a lot on features, which depends a lot on whether your old PBX was made by a stuffy clueless telco or by somebody who understood that open systems are critically important for their own developers even if they're not giving out the source code.
Here in the US, the big pricing anomaly for VOIP is that traditional wireline telcos charge about 2 cents/minute for delivering calls to a user, so a VOIP company has to either do per-minute pricing or else charge a high monthly flat rate and hope they win statistically, even though their costs of the long-distance part of the connection may amortize to 0.1 cents/minute. Also, a huge amount of the current long-distance switching infrastructure's complexity is to support call-center features for toll-free calling, which is something that people cared a lot more about when phone calls were 25 cents/minute than when they become 0.1 cents/minute.
If your PBX appears too dumb to do that, it's probably because that's an "extra value-added application feature" that's not in the application package you bought, though real difference is probably whether the user interface menu items are enabled or disabled rather than whether the API behind it works.
The big problem for the Feds is that VOIP is structured fundamentally differently than traditional phone switching. Call setup uses servers, which can be run by anybody in the world (but are usually run by a company's internal telecom department for IP PBXs, by a VOIP-to-telco gateway company for many commercial VOIP services, or by software companies with ill-defined business models or popular Instant Messaging services with proprietary protocols), and the voice channel of the call is carried separately on whatever IP networks connect the callers, which may not be under the control of the gatekeeper (call setup server) providers. There are two fun places to hack this kind of communications - VOIP-to-telco gateways (where you can tap the telco side using traditional methods, but you have to bully the VOIP provider into giving you call detail records for the IP side, which is much tougher in the decentralized VOIP world than in traditional telcos), and in the call setup gatekeepers themselves, which can tell the callers to set up the call through an unexpected conference bridge service instead of directly (since the caller's VOIP software is designed to believe that sort of request and wouldn't have an easy way to check it if didn't trust the gatekeeper.) The latter obviously requires hacking the gatekeeper, which is even harder to do in some environments, since it's possible to use a service with distributed gatekeepers that might not even be in the jurisdiction of the cops who want to wiretap you.
Disclaimer time: I work for a large telecom company, and the following is strictly my personal opinion, not theirs: I work for a large telecom company, so the rules about obeying wiretap orders appear to apply when we're providing managed VOIP services for customers. But if a router or PBX vendor wants to sell hardware to a customer, that hardware doesn't need to have wiretapping service built in, and if the hardware vendor wants to sell management services for the hardware, they're not a telecom company so wiretapping rules don't appear to apply to them (unless the government thinks up some new rules.) From a competitive standpoint, this is really really annoying, especially because as a civil libertarian, I think wiretapping is wrong and rude and often done illegally, and a huge amount of CALEA and followon laws and regulations seems to be designed to make massive-scale illegal or at least unsupervised wiretapping easy.
The savings depends a lot on your calling patterns - are you making enough IP calls to justify the cost of the hardware you need? You can save money if you're making lots of calls between your locations (especially international locations), or if you're using a VOIP phone company that lets you deliver calls to a VOIP-to-Telco gateway, but if your calling patterns aren't right, you might not be able to justify the hardware cost unless you needed to buy a new PBX anyway (in which case it'll probably be VOIP.) Alternatively, you can save money on long distance calls by telling your long distance company sales rep that the X cents/minute rate they're charging you is laughably high in today's market and you can get a better price buying phone cards at the 7-11, and have them grumble and cut your price yet again.
Disclaimer: I work for a large telecommunications company which can sell you any of these services or hardware, so we make depressingly thin margins regardless of which option you take, and no, this isn't an official company position, it's just my opinion.
Here in Collie-Fornia, laptops in the front seats of cars have been illegal since January, as are other digital devices with display screens that aren't built-in and are forward of the driver's seat-back. I'm not sure if that includes DVD players in the front seat section (or if that's already illegal), and the law doesn't appear to have been intended to ban cellphones, but it's definitely not legal for the driver or even the passenger to operate laptops in the front seat. The motivation is to prevent accidents caused by distracted drivers, but wardriving gets hit as a side effect, unless you do it from the back seat while your chauffeur drives.
- For instance, one common piece of Evil Spyware offers to be a helpful toolbar for your browser that'll give you more information, but really offers you occasional advertising and transmits your browsing preferences back to their company.
- Yet Google Toolbar for IE does much the same thing, it's just more upfront about it and less obnoxious about ads and presumably better behaved with the data it collects.
- Many spammers include web bugs in their HTML email to track whether you've read their material, either as 1x1 nearly-invisible GIFs or as big jumpy advertising GIFs/JPGs.
- When _you_ send somebody HTML with URLs for your vacation pictures on your web site, you can also check your web server logs to see if they've accessed the pictures.
It's mostly about _intent_, and Congress is sufficiently clueless technically that they're not likely to be able to write a clearly defined law that covers either the intent or the technology.One of the three PCs on my desk is a laptop with a dead screen that works just fine with the monitor when I switch the KVM switch over to it, and keeps a couple of old Win95 apps alive; when we're done with them I'll probably turn it into a Linux server of some sort, but it's happy for now, or I may just use it as a printer server.
I remember Rob and Ken ranting against microkernels back in the day, but Plan 9 was smaller and cleaner than most of its competition, and things that were viewed as "hopelessly bloated" a decade ago are now comparatively small and under-featured. How do we get back to relatively small kernels and fast reliable operating systems without ditching all the useful things that have been added since then? Hardware support for PCs and connected devices including accelerated video cards and USB plugins are critical issues.
Rob gave a talk at ?Nashville Usenix showing off the 8.5 Windowing System. One of his comments was that "Ken and I spent 10 years learning what things windowing system shouldn't do and we've written one that doesn't do them." 8.5 (the ".5" is properly the Unicode "1/2" glyph...) is small and blazingly fast, firing up a windowing system in about the time you'd expect a normal carriage return to give you a shell $ prompt back on a 680x0 machine, and the Acme windowing user interface on it was also lean, mean, and very fast.
Back when I was in college and the term was newer, I had an Afrikaner friend who referred to himself and his wife as African-Americans, and they were, just as you can be an Italian-American regardless of what your skin colour is.
One of my current friends was from Rhodesia, and became a Zimbabwean (the name changed in 1980 after the British Empire recognized the independence they'd declared for themselves in 1965) before he moved here to the US as a teenager, and yeah, he's African-American in spite of being a light-skinned blondie. (I'm pretty sure he's both young enough and old enough that he was born Rhodesian, but I don't know if he's also a UK subject from the disputed-ownership period...)
Somebody else mentioned a Moroccan friend calling himself African-American - that really doesn't tell you whether he looks Negro, because Morocco is an extremely diverse country. I'm guessing from context that he's probably Arab, but I've met Moroccans ranging from Berbers in the south (black, but mostly looking quite different from Sub-Saharan West Africans or East Africans) to Arabs to Sephardic Jews to leftover French bureaucrats.
Sure, it'd be cool to hop out with your surfboard and oxygen tanks and try it, but it's not likely that SpaceShip One's aerodynamics are designed to have the door opened at full speed. They might be able to get away with it because there's very little atmosphere up there and they're going relatively slowly at the top of the curve. But also, the lack of atmosphere means you'd probably need a pressure suit and not just oxygen tanks, so your maneuverability once you're down to low altitude will be pretty limited.
Sure, people who want to actually work in space would be willing to live in a tin can - but if you want a lot of people up there, either as tourists or as workers, you need a multi-user tin can, with enough resources to handle a bunch of people and enough redundancy that equipment failures don't become emergencies. So you might as well call it a hotel, and have somebody working up there as concierge / desk clerk / janitor / cook / bottlewasher.
At least one major US telecom company used to regularly fly planes along major cable routes looking for construction activities - not sure if they still do that. There were usually over a thousand backhoes within a quarter mile of their east coast cable routes, but they'd still get surprises like that farmer burying a dead cow with his backhoe in the early 90s.
But Terrorists-Under-The-Bed have been used as an excuse for blocking public access to lots of critical safety information, particularly in industries like oil refining, chemical manufacturing, and anything nuclear, where there can be serious risks of toxicity, chemical spills, and even major explosions (I'm not ranting about nukes here - fertilizer plant explosions are much more likely.) The Feds, who used to force public disclosure of lots of this information are now banning it, and databases that used to be accessible are being closed to the public, because Terrorists and other enemies of American Industry (like anti-pollution activists and various other NIMBYs) might abuse it.
Yeah, I'm glad somebody else thought of it and posted it - I realized the joke a few minutes after my original posting, but following up my own post would have been tacky :-)
Some of these images have been copied elsewhere. Google Image Search shows some of the pictures, such as rotsnake2.gif
If they've implemented it right, even if the Trust-Us-chip gets activated, that shouldn't bother your Linux any. It might prevent you from running Windows, or applications that use the chip, unless you've got the right licenses.
The trusted computing platform can prevent some kinds of malware from running, by limiting what you can do to the operating system without the platform's consent. Depending on how it's implemented, that could limit the spread of the malware, or could just turn it into a denial-of-service attack against the operating system, or could turn it into a denial-or-service against other Trusted-Platform-Only software (because the platform looks like it's been tampered with, the copy-protection in "Example Trusted FPS Game" decides not to let you play.) But more likely, malware vendors will concentrate on writing applications like IE Plugins that'll run fine on trusted platforms.
What "Trusted Computing" gains you as a user is the ability to buy and run software written for the Trusted-By-Them environment, which the vendors might not sell for non-trusted platforms. So if you want to play "Example FPS Game" or play music CDs using the "Evil Record Label's Protected Music Player", using a trusted-by-them platform lets you do that. If you don't run a Trusted-By-Them platform, they might not sell you their game/music at all, or they might sell you a CD that plays the low-res version of the music on untrusted platforms, or sell you the crippleware version of the game that only has levels 1-3, or lets you play the whole game but doesn't include the Surround Sound or graphics accelerator support or something.
It's so nobody swipes the RAM in the middle of the night to upgrade their own desktops....
The argument that the space race or various military arms race programs were worthwhile because they brought us all these cool non-space spinoff things like communications satellites, velcro, and Tang is bogus. Yes, those things were potentially valuable, but they not only diverted money from the applications that the private sector could have used for other applications, they diverted a lot of really talented engineers and scientists from working on private-sector applications. What kinds of things would they have built if they hadn't been making rockets? Would they have improved the energy efficiency of cars, airplanes, and houses by 10%? Would they have built the Franistan that would have revolutionized the industry? (You've never seen or used a franistan, but that's because the team that would have invented it were improving military helicopter gunnery systems instead.)