Piqua, of course, is the literary (?) home of the Jerome Hurwitz Elementary School and its Principal Krupp. Never heard of them? Ask any third grader! Dav Pilkey's Captain Underpants books are set there. Kids love this stuff, and it puts Piqua onto the same map as, say, Bedrock.
I assume that their Net Nanny would censor Pilkey's site
The bill calls for 255 MHz of contiguous spectrum for what it calls U-NII services. The present U-NII allocation includes 200 MHz of contiguous spectrum in the 5.7-5.8 GHz range, plus another noncontiguous 100 MHz down around 5.2.
The lowest 100 MHz is very restricted, since satellites use it too, so it's indoor only, very low power. Some 802.11a devices hang out there. The power limits on the upper two sub-bands are higher. The 5.85-5.925 band (just above U-NII) is used for vehicle-oriented services -- see the FCC web site http://www.fcc.gov/ for a current proceeding concerning that spectrum. Not to mention a slick spectrum chart covering 300-3000 MHz.
So really this bill isn't asking for all that much, probably just an extension of the 5.7 GHz band. The problem has been lower volume and higher cost for that equipment; if it catches on, prices will fall. But the rules may need some tweaking.
Voice doesn't use much bandwidth anyway. So banning it from 2.4 GHz wouldn't do much good.
The 2.4 GHz band is used by microwave ovens, TV cable box extenders, and all sorts of other unlicensed devices. Phones don't belong there only because they can get better performance and less interference at 900 MHz. The recent move of cordless phones to 2.4 is a victory of stupid imagemongering over technical reality. (The first 2.4 GHz phones were priced at a premium, so everyone mistakenly thought it was better.)
I'm probably late to post in this thread, but if anybody knowledgeable reads this...
I tried Knoppix to see if Linux would work on an old laptop that I recently picked up. Very nice, though it didn't ge the sound going. But the hard drive install is intriguing. It's based on Debian, which is famously hard to install. Once Debian's in place, though, it's said to be very easy to maintain, using apt-get. I use Mandrake now and while urpmi is supposed to help, it's no apt-get! Indeed it still leaves me in dependency hell more often than not, when trying to install a package not included in the base distro.
So would Knoppix on HD be a nice shortcut to a working Debian system? If so, it could be a killer tool for the Debian world, and I could try it on my "spare" partition (where I last had Gentoo, a nice hack in is own right but talk about hard to install...).
Note the phrase "Covered OEMs". Not "all OEMs". Just the "covered" ones. I believe, from previous articles on the topic, that the DoJ settlement only covers the top 20 OEMs. HP, Dell, IBM, Gateway, Toshiba, Sony, NEC, and other top brand names. Everybody else is not covered. The 40%+ of the market that's "white box" is not covered.
No. "Reasonable and non-discriminatory" is standard language for how patents are licensed for a fee. Remember the stink at the W3C when IBM wanted its patented technologies to be added to the web standards on a "RaND" basis? All it meant was that everybody could pay them the same fee for licensing, and it might have to be less than, say, a million dollars down and a hundred bucks a browser.
So MS is prohibited from charging different licensing fees to each of its Top 20 oems (the only ones, I think, "protected" by this settlement). But they can certainly impose charges that would be incompatible with GPL. They'd only be valid if covered by valid patents, of course; SAMBA is built without licenses, and reverse engineering is a way around non-patent licensing.
Well, no. I don't expect everybody to work for free.
However, when the FCC licensed the AM and FM bands, they created multiple channels, for separate ownership. For satellite radio, they allowed a technology to be developed that limited the service to only two providers. Thus I *cannot*, even if I had a billion or two to spend, put up my own radiosat. It's a duopoly.
That's a nice way for Saddam and his kid Uday to each have their own radio network, but it hardly seem to suit the American tradition. But of course the old free press tradition is giving way to extreme media concentration, and in this case, the new medium is legally exclusive.
I suggest that if there is a valid technical reason to have only two transmitters, that the satellite operators have an obligation, as cable operators do already, to allow some number of channels to be leased to third parties, on a nondiscriminatory basis.
The FCC adopted the wrong model for satellite radio. The pigopolists pretty much got what they wanted, and are suffering for it.
For technical reasons, there are only two satellite radio networks, Sirius and XM. Both have capacity for a lot of channels. The FCC decided to use a "market" approach and allow each company to choose details of its own technology, so their radios are incompatible. Imagine how well TV or FM radio might have done if different stations required different receivers! Consumers are locked in. Sure, it's nice work if you can get it, but consumers aren't quite as dumb as the companies wanted them to be.
Even worse, the duopolists were not charged as common carriers, but as programmers. So XM and Sirius determine what they will carry, and if they don't want something run, it won't run. Sure, they've figured out that they have to offer some kind of musical variety, so they have country & western streams, '70s rock streams, '80s rock streams, sports streams, etc. But the plain fact remains that they control the horizontal, they control the vertical, and a Sirius or XM subscriber won't be exposed to anything that the suits at Sirius or XM don't want them to hear. I guess to them, a stream playing Wilco and an NPR stream are radical enough.
So if this turkey fails, maybe somebody else will try again. If an operator were less greedy, and leased enough channels to independent programmers, then a workable business might be found.
A 5ESS uses proprietary custom software. It's actually a very primitive machine -- the 5E was designed in the late 1970s, though it didn't get widespread deployment until 1982 or so. You can tell by its forward-looking design: Analog lines normally attach through an analog crossbar stage! It's implemented using high voltage diodes ("gated diode crosspoint") which the Bell boys were very proud of back in 1980. After all, it saved on codec chips (which today cost a few cents).
Still, the 5E's code is super-reliable. It takes a couple of years for "hello world" to go through the validation processes that telcos want. And the 5E has a great feature set. It would be nice if they ported it to modern hardware.
Some parts of the code do run under Solaris, but not the critical real time stuff.
Some new switches, however, do use Solaris, which has a great reputation for reliability. I've recently been invovled with cutting one over whose main real time CPU is a Sun blade. (Its primary management interface is a Windows app, but if that crashes, the switch just carries on.) Telco gear is designed to last for years with minimal maintenance. Linux is a lot of fun but frankly it's not optimized for that application.
I've never installed Debian, though these reviews gave me a good idea of the flavor. (And I did install Yggdrasil on a 386SX back in, oh, 1983 or so. I know my own hardware pretty well.) Debian's sounds tricky and confusing.
But at least it's an installer. I installed Gentoo 1.1a a few months ago to see what the buzz was about. It doesn't even have an installer! Gentoo's installation is, I suppose, based most closely on Linux From Scratch. It was a bunch of instructions about how to fetch this package and have portage install it, then use nano to write this or that configuration file, etc. So by contrast, Debian 3.0 sounds pretty easy.
But in the meantime I'm using Mandrake 9.0, which may mark me as Fat, Dumb and Lazy but it does get an awful lot of software up and running very easily.
>Now if I could pay for 3G access on my monthly phone bill, and be able to use it on my laptop at no extra charge from anywhere in the country.
Yes, you could pay on your phone bill, but the charge on the phone bill would likely be *more* than the extra charge at the airports.
3G bandwidth is not cheap. Figure a good deal is one that charges regular voice-level minutes. VZW, for instance, offers plan that for $30/month. Another plan doesn't cap minutes but bytes instead, for around $100/month.
These developments (3G, BLAST) are a way to widen the firehose, but they don't reduce the price of the Perrier water coming through it.
You're right -- they overbid at auction. Winning was losing.
It didn't have to be that way. Yes, it would be bad to not win a license at auction, but if the auction price goes to high, the "loser" can just wait and pick up the asset later, at a bankruptcy sale or similarly depressed price. I expect that to happen.
I was working at the time for a consulting firm who did a lot of wireless carrier work. I worked for one client that was thinking of bidding at an auction. So we figured out how much it would cost to build the network, how much they could charge customer to use it, and what kind of profit was left. We told them, and also noted that given the bidding that had been going on around the world, it was unlikely that the bidding would stay that low.
Another group was working for some of the European cellphone companies, doing what I understand to be glorified secretarial work (Eurocrats love paperwork) and claiming "success" every time their client "won". My cow-orkers gagged.
We're now seeing the Euro vendors ask for permission to share networks. It costs a lot to build W-CDMA, and they don't see the need any more for competitive facilities. If it catches on, then they might have enough demand to use up their licensed bandwidth. But for now there's not much demand for "high speed data" in the Euro/minute price range.
You are correct. ISPs are not common carriers. There is specific law about who is and is not a common carrier. ISPs are "information service" providers. Under the Telecommunications Act, common carriers are subject to Title II regulation, while ISPs are unregulated though to some extent technically subject to Title I.
Some laws do specify what an ISP can and cannot do, or must do when told. It's not quite the same as common carrier rules, but has a lot in common.
Good comment. Easy is perceived as bad. Yes, some easy things (Microsoft Bob) are really bad, and some difficult things are good, but Linux, like Unix in general (sorry to generalize but it's a cultural inheritance), suffers from an "old boys' club" mentality in which earned knowledge of arcana is viewed as an entry point. Sort of like morse code has been to ham radio -- there aren't rational reasons to require it any more (though some of us actually like to use it, sportingly), but removing the requirement brings out all sorts of anger from old timers who had to learn it.
Where it currently bugs me the most: Gentoo looks like a swell distro. Installing from source ends dependency hell and optimizes performance; I can buy it. But the setup is dreadful, basically more Linux From Scratch than anything else. The topic of an installer came up on Gentoo Forums. The "consensus" of the Gentoo user base is that "Gentoo is a hard distribution, and so the installation should be hard too." What rot! Once installed, no distribution should be gratuitously hard to live with. And while Gentoo lacks some of the GUI tools of say Mandrake or Red Hat, it's basically a clean system that shouldn't be that hard to manage. But the install procedure basically consists of printing out a lengthy set of instructions and doing a lot of hand edits of files, step by step, and hoping your system is enough like the developers' to work right.
Personally I don't find the current kernel config (make xconfig) to be that hard, just a little nerve-racking where some new options show up that I don't understand. Which is what Bruce set out to fix. We can quibble about implementation details but his heart's in the right place. Linux won't prosper so long as it lives with the old boys' mentality. If I want to join the Freemasons, I will.
The original author failed to research how vinyl records work, something that "everybody" knew 20 years ago, before CDs.
Now to see if my memory still works. Mono LPs used horizontal modulation; the needle moved back and forth within the groove. Stereo can be viewed two ways. Vertical is difference (L-R), horizontal is sum of the L+R. Viewed differently, the two diagonal walls of the groove are the two channels.
A flatbed scanner can only see the horizontal, so it might work a bit with mono, but it won't work too well! However do note that some very, very expensive ($10k+?) new turntables actually do use optical "needles" to track the groove without touching it. Talk about low tracking force!
Funny that the site should come from TRAC. This is a classic example of Washington "astroturf", phoney grass roots organizations. TRAC is run by Sam Simon of Issue Dynamics, a P.R. firm that works for incumbent telephone monopolists like Verizon. Their main activity is publicizing anyone but AT&T as being a better long-distance alternative to AT&T. There's a long and bitter blood feud between AT&T and Verizon (fka Bell Atlantic). Sam's a hired gun.
In this case, it's likely that Verizon is smarting over the cost of spam to its Verizon Online operation. Verizon is trying to get the FCC and/or Congress to give them an absolute monopoly over DSL ISP service. The FCC has a proposal on the table now (FCC NPRM 02-33) to remove common carriage rules from ILEC DSL, so that Verizon can say that if you use their DSL telecom service, you must use their ISP. Bye-bye to all the local ISPs who buy Verizon DSL to reach their subs. There's also the Dingell-Tauzin bill in Congress which effectively repeals the Telcom Act, and bans competitive LECs from accessing ILEC raw wire in order to provision their own DSL. Also note that Verizon Online forces you to put @verizon.net in the From: field of mail sent through their SMTP relay; you can't be From: your choice of mail hosts! Forced advertising, no less.
So when they come out with an anti-spam site (and Sam doesn't brush his teeth without a Verizon top exec's permission), I take it with a boulder of salt.
I use an AT&T-Broadband cable modem. They do let me use a web server.
Of course the server is theirs, on their site, not mine. This makes perfect sense, because bandwidth on the cable is very asymmetrical. A typical cable modem segment, shared among dozens of users, has 2.5 Mbps upstream and 27 Mbps downstream. (These numbers can vary somewhat but that's in the range.) So if users put servers on their sites, the upstream bandwidth could congest pretty quickly.
This isn't some nefarious plot. CATV networks begin their downstream at Channel 2 (in the USA, 54-60 MHz), and need a guard band between that and the upstream. So upstream is below 40 MHz, and is shared between modems, telephony, cable box response, and a whole lot of noise in that part of the spectrum. Upstream bandwidth on a cable is just naturally scarce. To have more, the cable would need a "high split", say at 100 MHz, which would reduce the number of TV channels, especially the valuable "cable ready" analog ones (VHF 2-6).
Cable operators make mistakes, and don't always "get it", but they are stuck dealing with reality.
Paperless good. Paperless plus good. Paper can stick around, carrying un-information. Paperless makes it easier for the Ministry of Truth to remove references to unpersons, and to change unfacts when they are no longer true.
With new paperless media like DVD-ROM and Ebook and the DMCA, it is already a federal crime to circumvent the Ministry's control over information. Once the DVD player expires, or the rights-managed usage grant expires, then it will be a crime to circumvent protections in order to access un-information. Downloaded music from PressPlay will all die when PressPlay does. It becomes un-music! You'll have to buy the next big star they want to sell you!
Answering the originator's question about how much power is consumed by computers and the like, we did a study for DoE last year. Note that it excludes home computers, focusing on business use, but this picks up the bulk of power consumption. NTIS.gov will sell you a CDR of report PB2002-101438 for $47. A PDF slide set summary is available at:
http://www.tiax.biz/pdf/EIA-OffTelecom-TIAX.pdf
One interesting conclusion: Commercial Office and Telecommunications equipment electricity consumption represents just under 3% of national electricity consumption, and a little over 1% of national energy consumption.
Note that the report was prepared by Arthur D Little Inc.; TIAX acquired the group that did this study.
"Real" PBX systems have very complex software. Basic call control is easy. Features, especially keyset emulation, take a lot of code. This isn't something to code yourself in your spare time. You might have fun adapting what's out there, but don't confuse "IP telephony" with a real PBX. IP phones are expensive and don't sound as good as circuit phones.
Standard PBX systems are designed for the famous "five nines" reliability. You don't get that from a regular PC -- for instance, PC hardware can't do hot swap, which any PBX worth its salt can.
If you're adventurous, you can cobble together "carrier grade" hardware nowadays using off-the-shelf cards in the Compact PCI (which is more accurately "collosal PCI") form factor plus the H.110 bus, which supports 4K time slots of TDM voice. Of course that's overkill for a home system, but some serious phone gear is built that way, using off-the-shelf Sparc or PowerPC CPUs.
The article makes it sound as if the USPS wanted a "private express" type monopoly in email. I have a long memory of these things, and very seriously doubt it! Remember, email as we know it began over the ARPAnet in 1972; single-computer email goes back farther. Lots of people (myself included) were on the ARPAnet in the late '70s, using email galore. There was no thought of shutting us down. In the 1980s, there was also a lot of uucp mail, fido, DECnet, BITnet, and other types of email besides the venerable SMTP. These just could not be banned or shut down.
And don't forget X.400, the 1980s idiot bastard child of the ITU itself, an email protocol so baroque that only a Lotus Notes developer could love it. X.400 was a bad implementation of a good idea, that being to have a multivendor standard. They just ignored SMTP's existence, even as millions used it. Right into the early 1990s there were people arguing that X.400's supposedly greater capabilities were necessary.
Various worldwide postal agencies did build something called IntElPost (sp?) in the late '70s and early '80s; basically, it was international Group IV fax service between post offices. The USPS was not allowed to participate; it still operates in some countries.
Somebody else has noted how the USPS introduced a truly awful RJE-printer papermail service, ECOM, which flopped big time. I note that MCI Mail, a 1981-ish consumer/business email service, had a paper-output option too; I occasionally used it to send paper mail.
The USPS could potentially play a role in a future e-post system; that might be one way to cut spam. I'd be happy to pay, oh, a penny or so per email, provided that spammers did too. More likely, it would have to be some kind of micropayment scheme wherein my inbox would block something without an e-stamp, which would cost too much for a spammer. Of course that doesn't need the USPS, but they could be a player if they got their act together.
I assume that the common version of this works in Windoze, since that's the mass market fave. But what do Linux and other Unix fans do? Clearly, this is one debugger that needs an open source version!
Leesse... Gnu Debugger, new switches to be added: repel mouse repel mosquito repel cowboy neal... no, he's okay.
Who's responsible? The article points out that the decisions are being made by none other than Theresa LaPore. She's the genius behind 2000's Butterfly Ballot. No, it didn't conform to Florida law, but Jeb was willing to let it slide so long as it benefitted his family. Now there's a voting machine with no real recount possible? Sounds like Jeb must have recommended it.
Smaller isn't always better. In this case, much of the subscriber module is an antenna. Smaller antennas have less gain, so the range would be worse. The actual electronics probably fit into a small box inside the subscriber module.
We're obviously dealing with a reading comprehension problem here. I didn't say that the beams were built in order to block RF, but that was a side effect which proved unfortunate in an emergency.
And btw, movie theatres don't usually have assigned seats and ushers to lead you to them.
As other have noted, ferrite would block all public safety frequencies; it's not selective to cell phones. And there are public safety frequencies in the 700-900 MHz spectrum, near cellular.
Slashdot Mirror
http://www.davpilkey.com/ too.
The bill calls for 255 MHz of contiguous spectrum for what it calls U-NII services. The present U-NII allocation includes 200 MHz of contiguous spectrum in the 5.7-5.8 GHz range, plus another noncontiguous 100 MHz down around 5.2.
The lowest 100 MHz is very restricted, since satellites use it too, so it's indoor only, very low power. Some 802.11a devices hang out there. The power limits on the upper two sub-bands are higher. The 5.85-5.925 band (just above U-NII) is used for vehicle-oriented services -- see the FCC web site http://www.fcc.gov/ for a current proceeding concerning that spectrum. Not to mention a slick spectrum chart covering 300-3000 MHz.
So really this bill isn't asking for all that much, probably just an extension of the 5.7 GHz band. The problem has been lower volume and higher cost for that equipment; if it catches on, prices will fall. But the rules may need some tweaking.
Voice doesn't use much bandwidth anyway. So banning it from 2.4 GHz wouldn't do much good.
The 2.4 GHz band is used by microwave ovens, TV cable box extenders, and all sorts of other unlicensed devices. Phones don't belong there only because they can get better performance and less interference at 900 MHz. The recent move of cordless phones to 2.4 is a victory of stupid imagemongering over technical reality. (The first 2.4 GHz phones were priced at a premium, so everyone mistakenly thought it was better.)
I'm probably late to post in this thread, but if anybody knowledgeable reads this...
I tried Knoppix to see if Linux would work on an old laptop that I recently picked up. Very nice, though it didn't ge the sound going. But the hard drive install is intriguing. It's based on Debian, which is famously hard to install. Once Debian's in place, though, it's said to be very easy to maintain, using apt-get. I use Mandrake now and while urpmi is supposed to help, it's no apt-get! Indeed it still leaves me in dependency hell more often than not, when trying to install a package not included in the base distro.
So would Knoppix on HD be a nice shortcut to a working Debian system? If so, it could be a killer tool for the Debian world, and I could try it on my "spare" partition (where I last had Gentoo, a nice hack in is own right but talk about hard to install...).
Note the phrase "Covered OEMs". Not "all OEMs". Just the "covered" ones. I believe, from previous articles on the topic, that the DoJ settlement only covers the top 20 OEMs. HP, Dell, IBM, Gateway, Toshiba, Sony, NEC, and other top brand names. Everybody else is not covered. The 40%+ of the market that's "white box" is not covered.
No. "Reasonable and non-discriminatory" is standard language for how patents are licensed for a fee. Remember the stink at the W3C when IBM wanted its patented technologies to be added to the web standards on a "RaND" basis? All it meant was that everybody could pay them the same fee for licensing, and it might have to be less than, say, a million dollars down and a hundred bucks a browser.
So MS is prohibited from charging different licensing fees to each of its Top 20 oems (the only ones, I think, "protected" by this settlement). But they can certainly impose charges that would be incompatible with GPL. They'd only be valid if covered by valid patents, of course; SAMBA is built without licenses, and reverse engineering is a way around non-patent licensing.
Well, no. I don't expect everybody to work for free.
However, when the FCC licensed the AM and FM bands, they created multiple channels, for separate ownership. For satellite radio, they allowed a technology to be developed that limited the service to only two providers. Thus I *cannot*, even if I had a billion or two to spend, put up my own radiosat. It's a duopoly.
That's a nice way for Saddam and his kid Uday to each have their own radio network, but it hardly seem to suit the American tradition. But of course the old free press tradition is giving way to extreme media concentration, and in this case, the new medium is legally exclusive.
I suggest that if there is a valid technical reason to have only two transmitters, that the satellite operators have an obligation, as cable operators do already, to allow some number of channels to be leased to third parties, on a nondiscriminatory basis.
The FCC adopted the wrong model for satellite radio. The pigopolists pretty much got what they wanted, and are suffering for it.
For technical reasons, there are only two satellite radio networks, Sirius and XM. Both have capacity for a lot of channels. The FCC decided to use a "market" approach and allow each company to choose details of its own technology, so their radios are incompatible. Imagine how well TV or FM radio might have done if different stations required different receivers! Consumers are locked in. Sure, it's nice work if you can get it, but consumers aren't quite as dumb as the companies wanted them to be.
Even worse, the duopolists were not charged as common carriers, but as programmers. So XM and Sirius determine what they will carry, and if they don't want something run, it won't run. Sure, they've figured out that they have to offer some kind of musical variety, so they have country & western streams, '70s rock streams, '80s rock streams, sports streams, etc. But the plain fact remains that they control the horizontal, they control the vertical, and a Sirius or XM subscriber won't be exposed to anything that the suits at Sirius or XM don't want them to hear. I guess to them, a stream playing Wilco and an NPR stream are radical enough.
So if this turkey fails, maybe somebody else will try again. If an operator were less greedy, and leased enough channels to independent programmers, then a workable business might be found.
A 5ESS uses proprietary custom software. It's actually a very primitive machine -- the 5E was designed in the late 1970s, though it didn't get widespread deployment until 1982 or so. You can tell by its forward-looking design: Analog lines normally attach through an analog crossbar stage! It's implemented using high voltage diodes ("gated diode crosspoint") which the Bell boys were very proud of back in 1980. After all, it saved on codec chips (which today cost a few cents).
Still, the 5E's code is super-reliable. It takes a couple of years for "hello world" to go through the validation processes that telcos want. And the 5E has a great feature set. It would be nice if they ported it to modern hardware.
Some parts of the code do run under Solaris, but not the critical real time stuff.
Some new switches, however, do use Solaris, which has a great reputation for reliability. I've recently been invovled with cutting one over whose main real time CPU is a Sun blade. (Its primary management interface is a Windows app, but if that crashes, the switch just carries on.) Telco gear is designed to last for years with minimal maintenance. Linux is a lot of fun but frankly it's not optimized for that application.
I've never installed Debian, though these reviews gave me a good idea of the flavor. (And I did install Yggdrasil on a 386SX back in, oh, 1983 or so. I know my own hardware pretty well.) Debian's sounds tricky and confusing.
But at least it's an installer. I installed Gentoo 1.1a a few months ago to see what the buzz was about. It doesn't even have an installer! Gentoo's installation is, I suppose, based most closely on Linux From Scratch. It was a bunch of instructions about how to fetch this package and have portage install it, then use nano to write this or that configuration file, etc. So by contrast, Debian 3.0 sounds pretty easy.
But in the meantime I'm using Mandrake 9.0, which may mark me as Fat, Dumb and Lazy but it does get an awful lot of software up and running very easily.
>Now if I could pay for 3G access on my monthly phone bill, and be able to use it on my laptop at no extra charge from anywhere in the country.
Yes, you could pay on your phone bill, but the charge on the phone bill would likely be *more* than the extra charge at the airports.
3G bandwidth is not cheap. Figure a good deal is one that charges regular voice-level minutes. VZW, for instance, offers plan that for $30/month. Another plan doesn't cap minutes but bytes instead, for around $100/month.
These developments (3G, BLAST) are a way to widen the firehose, but they don't reduce the price of the Perrier water coming through it.
You're right -- they overbid at auction. Winning was losing.
It didn't have to be that way. Yes, it would be bad to not win a license at auction, but if the auction price goes to high, the "loser" can just wait and pick up the asset later, at a bankruptcy sale or similarly depressed price. I expect that to happen.
I was working at the time for a consulting firm who did a lot of wireless carrier work. I worked for one client that was thinking of bidding at an auction. So we figured out how much it would cost to build the network, how much they could charge customer to use it, and what kind of profit was left. We told them, and also noted that given the bidding that had been going on around the world, it was unlikely that the bidding would stay that low.
Another group was working for some of the European cellphone companies, doing what I understand to be glorified secretarial work (Eurocrats love paperwork) and claiming "success" every time their client "won". My cow-orkers gagged.
We're now seeing the Euro vendors ask for permission to share networks. It costs a lot to build W-CDMA, and they don't see the need any more for competitive facilities. If it catches on, then they might have enough demand to use up their licensed bandwidth. But for now there's not much demand for "high speed data" in the Euro/minute price range.
You are correct. ISPs are not common carriers. There is specific law about who is and is not a common carrier. ISPs are "information service" providers. Under the Telecommunications Act, common carriers are subject to Title II regulation, while ISPs are unregulated though to some extent technically subject to Title I.
Some laws do specify what an ISP can and cannot do, or must do when told. It's not quite the same as common carrier rules, but has a lot in common.
Good comment. Easy is perceived as bad. Yes, some easy things (Microsoft Bob) are really bad, and some difficult things are good, but Linux, like Unix in general (sorry to generalize but it's a cultural inheritance), suffers from an "old boys' club" mentality in which earned knowledge of arcana is viewed as an entry point. Sort of like morse code has been to ham radio -- there aren't rational reasons to require it any more (though some of us actually like to use it, sportingly), but removing the requirement brings out all sorts of anger from old timers who had to learn it.
Where it currently bugs me the most: Gentoo looks like a swell distro. Installing from source ends dependency hell and optimizes performance; I can buy it. But the setup is dreadful, basically more Linux From Scratch than anything else. The topic of an installer came up on Gentoo Forums. The "consensus" of the Gentoo user base is that "Gentoo is a hard distribution, and so the installation should be hard too." What rot! Once installed, no distribution should be gratuitously hard to live with. And while Gentoo lacks some of the GUI tools of say Mandrake or Red Hat, it's basically a clean system that shouldn't be that hard to manage. But the install procedure basically consists of printing out a lengthy set of instructions and doing a lot of hand edits of files, step by step, and hoping your system is enough like the developers' to work right.
Personally I don't find the current kernel config (make xconfig) to be that hard, just a little nerve-racking where some new options show up that I don't understand. Which is what Bruce set out to fix. We can quibble about implementation details but his heart's in the right place. Linux won't prosper so long as it lives with the old boys' mentality. If I want to join the Freemasons, I will.
The original author failed to research how vinyl records work, something that "everybody" knew 20 years ago, before CDs.
Now to see if my memory still works. Mono LPs used horizontal modulation; the needle moved back and forth within the groove. Stereo can be viewed two ways. Vertical is difference (L-R), horizontal is sum of the L+R. Viewed differently, the two diagonal walls of the groove are the two channels.
A flatbed scanner can only see the horizontal, so it might work a bit with mono, but it won't work too well! However do note that some very, very expensive ($10k+?) new turntables actually do use optical "needles" to track the groove without touching it. Talk about low tracking force!
Funny that the site should come from TRAC. This is a classic example of Washington "astroturf", phoney grass roots organizations. TRAC is run by Sam Simon of Issue Dynamics, a P.R. firm that works for incumbent telephone monopolists like Verizon. Their main activity is publicizing anyone but AT&T as being a better long-distance alternative to AT&T. There's a long and bitter blood feud between AT&T and Verizon (fka Bell Atlantic). Sam's a hired gun.
In this case, it's likely that Verizon is smarting over the cost of spam to its Verizon Online operation. Verizon is trying to get the FCC and/or Congress to give them an absolute monopoly over DSL ISP service. The FCC has a proposal on the table now (FCC NPRM 02-33) to remove common carriage rules from ILEC DSL, so that Verizon can say that if you use their DSL telecom service, you must use their ISP. Bye-bye to all the local ISPs who buy Verizon DSL to reach their subs. There's also the Dingell-Tauzin bill in Congress which effectively repeals the Telcom Act, and bans competitive LECs from accessing ILEC raw wire in order to provision their own DSL. Also note that Verizon Online forces you to put @verizon.net in the From: field of mail sent through their SMTP relay; you can't be From: your choice of mail hosts! Forced advertising, no less.
So when they come out with an anti-spam site (and Sam doesn't brush his teeth without a Verizon top exec's permission), I take it with a boulder of salt.
I use an AT&T-Broadband cable modem. They do let me use a web server.
Of course the server is theirs, on their site, not mine. This makes perfect sense, because bandwidth on the cable is very asymmetrical. A typical cable modem segment, shared among dozens of users, has 2.5 Mbps upstream and 27 Mbps downstream. (These numbers can vary somewhat but that's in the range.) So if users put servers on their sites, the upstream bandwidth could congest pretty quickly.
This isn't some nefarious plot. CATV networks begin their downstream at Channel 2 (in the USA, 54-60 MHz), and need a guard band between that and the upstream. So upstream is below 40 MHz, and is shared between modems, telephony, cable box response, and a whole lot of noise in that part of the spectrum. Upstream bandwidth on a cable is just naturally scarce. To have more, the cable would need a "high split", say at 100 MHz, which would reduce the number of TV channels, especially the valuable "cable ready" analog ones (VHF 2-6).
Cable operators make mistakes, and don't always "get it", but they are stuck dealing with reality.
Paperless good. Paperless plus good. Paper can stick around, carrying un-information. Paperless makes it easier for the Ministry of Truth to remove references to unpersons, and to change unfacts when they are no longer true.
With new paperless media like DVD-ROM and Ebook and the DMCA, it is already a federal crime to circumvent the Ministry's control over information. Once the DVD player expires, or the rights-managed usage grant expires, then it will be a crime to circumvent protections in order to access un-information. Downloaded music from PressPlay will all die when PressPlay does. It becomes un-music! You'll have to buy the next big star they want to sell you!
Paperless good! Long live Big Brother!
Answering the originator's question about how much power is consumed by computers and the like, we did a study for DoE last year. Note that it excludes home computers, focusing on business use, but this picks up the bulk of power consumption. NTIS.gov will sell you a CDR of report PB2002-101438 for $47. A PDF slide set summary is available at:
http://www.tiax.biz/pdf/EIA-OffTelecom-TIAX.pdf
One interesting conclusion:
Commercial Office and Telecommunications equipment electricity consumption represents just under 3% of national electricity consumption, and a little over 1% of national energy consumption.
Note that the report was prepared by Arthur D Little Inc.; TIAX acquired the group that did this study.
"Real" PBX systems have very complex software. Basic call control is easy. Features, especially keyset emulation, take a lot of code. This isn't something to code yourself in your spare time. You might have fun adapting what's out there, but don't confuse "IP telephony" with a real PBX. IP phones are expensive and don't sound as good as circuit phones.
Standard PBX systems are designed for the famous "five nines" reliability. You don't get that from a regular PC -- for instance, PC hardware can't do hot swap, which any PBX worth its salt can.
If you're adventurous, you can cobble together "carrier grade" hardware nowadays using off-the-shelf cards in the Compact PCI (which is more accurately "collosal PCI") form factor plus the H.110 bus, which supports 4K time slots of TDM voice. Of course that's overkill for a home system, but some serious phone gear is built that way, using off-the-shelf Sparc or PowerPC CPUs.
The article makes it sound as if the USPS wanted a "private express" type monopoly in email. I have a long memory of these things, and very seriously doubt it! Remember, email as we know it began over the ARPAnet in 1972; single-computer email goes back farther. Lots of people (myself included) were on the ARPAnet in the late '70s, using email galore. There was no thought of shutting us down. In the 1980s, there was also a lot of uucp mail, fido, DECnet, BITnet, and other types of email besides the venerable SMTP. These just could not be banned or shut down.
And don't forget X.400, the 1980s idiot bastard child of the ITU itself, an email protocol so baroque that only a Lotus Notes developer could love it. X.400 was a bad implementation of a good idea, that being to have a multivendor standard. They just ignored SMTP's existence, even as millions used it. Right into the early 1990s there were people arguing that X.400's supposedly greater capabilities were necessary.
Various worldwide postal agencies did build something called IntElPost (sp?) in the late '70s and early '80s; basically, it was international Group IV fax service between post offices. The USPS was not allowed to participate; it still operates in some countries.
Somebody else has noted how the USPS introduced a truly awful RJE-printer papermail service, ECOM, which flopped big time. I note that MCI Mail, a 1981-ish consumer/business email service, had a paper-output option too; I occasionally used it to send paper mail.
The USPS could potentially play a role in a future e-post system; that might be one way to cut spam. I'd be happy to pay, oh, a penny or so per email, provided that spammers did too. More likely, it would have to be some kind of micropayment scheme wherein my inbox would block something without an e-stamp, which would cost too much for a spammer. Of course that doesn't need the USPS, but they could be a player if they got their act together.
I assume that the common version of this works in Windoze, since that's the mass market fave. But what do Linux and other Unix fans do? Clearly, this is one debugger that needs an open source version!
Leesse... Gnu Debugger, new switches to be added:
repel mouse
repel mosquito
repel cowboy neal... no, he's okay.
Who's responsible? The article points out that the decisions are being made by none other than Theresa LaPore. She's the genius behind 2000's Butterfly Ballot. No, it didn't conform to Florida law, but Jeb was willing to let it slide so long as it benefitted his family. Now there's a voting machine with no real recount possible? Sounds like Jeb must have recommended it.
Smaller isn't always better. In this case, much of the subscriber module is an antenna. Smaller antennas have less gain, so the range would be worse. The actual electronics probably fit into a small box inside the subscriber module.
We're obviously dealing with a reading comprehension problem here. I didn't say that the beams were built in order to block RF, but that was a side effect which proved unfortunate in an emergency.
And btw, movie theatres don't usually have assigned seats and ushers to lead you to them.
As other have noted, ferrite would block all public safety frequencies; it's not selective to cell phones. And there are public safety frequencies in the 700-900 MHz spectrum, near cellular.