But it seems that the SDMI-crack-discovered-it notch occurs only in some *segments* of the music. Not all the time. So it isn't nearly as obvious.
Of course in some kinds of music, notably classical again, it is still likely to be quite intrusive. I think SDMI only intended to be useful with things like Spice Girls and 2Pac. (Not that I want to associate those with each other.) Classical is small business, so if they screw it up, they don't really care.
The principal Verance watermark is similar. According to the paper, it's basically a periodic up or down amplitude tweak (each baud interval being 16/50 of a second?) to a set of narrow ranges in the top octave. Sort of like typing onto a graphic equalizer. Using hashed track data to create a weak checksum. Again, it's hard to tell perceptually, especially on pop/rock, but might intrude onto some classical recordings.
No, Canada now permits local telephone competition, so Telus' monopoly is historic rather than de jure. But due to the economics (e.g., residential service from Telus is priced below the cost of raw wire from them to a competitor), it's very hard for a competitor to sell residential service profitably. Unless it's the CATV, who's already there.
Trees don't help, but they don't absolutely block the signal. MMDS is licensed with more power than the 2.4 GHz unlicensed radios, so it has more fade margin to begin with. In some areas, trees could be a problem. But it's not like optical, where they cut you right off, or LMDS which is close.
Multipath's a different story. Different radios survive it differently. Cisco, for instance, is quite vocal about how their OFDM radios for MMDS are multipath resistant (they work okay even if the signal is bounced off of a few buildings). Some others aren't. I don't know what radio Spring is using here but I suspect it's one of the newer, more multipath-resistant ones. That, btw, is what vendors are talking about when they say they support non-line-of-sight. Hills are a different story -- that's no path!
You get an earful of lies, is what you get from your friends at the Bells!
The law is pretty clear. If Bell runs a line, a CLEC can demand Bell provide him with access at COST PLUS. Not minus. Maintenance is a cost that is included in the rent. Thus, my CLEC can lease raw copper wire analog/ADSL loops from VZ/New York Telephone in Ballston for, oh, $19/month or so, while VZ's local flat rate telephone service, built out of similar wire plus the whole network, costs less than that. Bells are notorious for their creative accounting.
Sprint is already doing this in a few other cities, such as Phoenix, where they have the MMDS license. Worldcom is doing it in Jackson, MS and a few other cities, and will be expanding it too (they have the NY and Boston licenses).
Note that "line of sight" for MMDS is much better than optical; it means "not over the horizon". Since Chicago is basically flatland, hills aren't the problem they would be in, say, New England. Which is why this Chicago rollout is so important; it could give the technology a real boost. MMDS operates around 2.5 GHz. It is not subject to significant rain fade, and passes easily enough through trees. (Contrast this to LMDS at 29 GHz, which has a typical reliable range of around 2 miles, because of rain fade, though it goes much farther on dry days.)
Each market has one MMDS licensee. This was the FCC's last pre-auction lottery, nicknamed "wireless cable". It was intended for pay-TV broadcast distribution. A bunch of shady operators took fees to enter people into the license lottery ca. 1993. The MMDS companies who bought up the licenses from the lottery winners discovered that there wasn't much of a market, so they went bankrupt or sold out to Sprint and Worldcom (who between them have most of the country's population covered by their licenses, but are just starting to offer service). Now it's viewed as a DSL alternative. Some other operators are also in business; Oxford Telecom, for instance, does MMDS data in Portland, Maine.
This is mostly two-way radio, something the FCC authorized a couple of years ago. (Early systems were dial-up return.) I don't really think there's enough bandwidth there to replace DSL or cable modems in urban areas, but it's a good alternative for people who are out of range of those services. Alas, with only one license per city (spectrum being a scarce resource), it's not totally competitive.
No, salami. The Bells are ALLOWED already to do almost whatever they want locally, provided they're not blatantly anticompetitive. They are ALLOWED to run fiber to the home, and used to promise it. They're ALLOWED to run DSL to the home. They just are trying to get around some laws that require them to let others in the business too. So they hold back things they could be doing, as well as interfere with competition, and demand absolute monopoly control. (Specifically, Ed Whitacre of SBC has recently bitched to the Illinois Commerce Commission that he can't provide DSL to most homes there because the current rules require him to let competitors, like Covad, lease copper from him. He wants to cut their copper.)
If they had the monopolies they want, do you think ISPs would get a fair deal? Fat chance. Note too that VZ DSL here has a 90k byte upstream speed, much lower than competitors. It's just TV to them. Your keyboard is a remote control.
Precisely. The phrase "Computer Science" refers to a math/logic subject which was given that name. But computer can describe science, the same way any noun can be used as an adjective. Such "computer science" isn't the same thing, so the headline is bad work.
The rest of the article is good, though, in pointing out how compute-intensive much modern science is. Still, it's not CS; computers are just a powerful tool.
Idealab this time seems again to be Cluelesslab. DNS is a hierarchical system. Most retail end users now get their DNS service from their ISPs; commercial end users (leased lines, etc.) may have an in-house DNS but that too is hierarchical under their ISP's.
Now anybody can point to any other DNS, but it's not trivial for Joe User, so it's really best if the ISPs do the fix in their DNS servers. They can simply add these alternative roots next to.com et al. Poof, it's done, for all applications.
The tricky part is Worldcom's UUNET, the largest backbone ISP (upstream from a lot of retail services). Vint, the Elizabeth Taylor of the Internet (famous because he's famous, and didn't he once make a movie about a horse when he was a kid?), works for them. So Roberts gave him the hot title at ICANN, so he won't defy them. But UUNET's customers (who mostly run DNS servers) can still do the fix.
But plug-ins? Gross needs a cranial plug in.
(BTW, Idealab has put up for sublease its fancy Boston digs, occupied for less than a year. They're shrinking away with their stock portfolio)
Re:Cable/DSL isn't real world, but 9600bps is?
on
The Modem Lives On
·
· Score: 2
Major parts of The City of Boston (47 sq. miles, 580k pop.) are out of DSL range. The COs just aren't put close together, so urban neighborhoods (row houses, etc.) are not loop qualified. We're not talking the woods. It's just a crap shoot if you qualify -- the BEST RBOC loop quals are a bit over 50%, while some cities (not woods) are below 20%.
9600 is not quite real world, but 24k (top speed of a modem on a phone line served by Universal DLC, which is common) is quite real, so if you can work at 9600 at all, then at 24k it's probably decent. Maybe today they should test at 24k. No faster.
Well put, but also good advice for web designers!
on
The Modem Lives On
·
· Score: 4
Network games and web sites both need to be designed by people who pay attention to low-rate connections. Besides the fact that the Internet path itself gets rather slow quite often, modem-speed access indeed will remain the norm for some years, as many people have no choice. And there's nothing that the ISPs can do about it, because it's the telecomm carriers and cable companies who have the wires.
Why some folks just can't get broadband:
* To get a cable modem, the cable company has to have the upgraded Hybrid Fiber-Coax (HFC) plant, as well as a cable modem terminating system (CMTS) and the rest of the needed gear. This is becoming more common but lots of cable systems aren't there yet; for instance, the old TCI systems were real fixer-uppers.
* To get DSL, the subscriber must be within 15,000 wire feet or so of the ILEC central office. This rules out a lot of homes, even in cities, because COs are spaced wider than that in all but the densest places.
* DSL won't work if the subscriber is served by a Digital loop carrier (DLC) system, which is the norm for new installations more than 12,000 feet or so from a CO (and sometimes much closer). Somebody could theoretically put a DSLAM at the DLC location, but the economics and practicality are often poor; it's very very rare at present.
* DSL won't work if the wire is not in good shape. That's often the case. (Especially in former-NYNEX territory!)
* There still has to be a DSLAM; this is hard to justify in smaller COs. DSL's basically an urban service.
* Wireless bandwidth is expensive. Even if you could go faster than 9600 bps on cellular, you wouldn't like the price. Unlicensed wireless is "free" bandwidth, but the range is short, so again most people don't live in range of a provider (wireless ISP). And that requires an antenna location, decent near-line-of-sight path, etc.
So I have advised my consulting clients to design their web sites using a 9600 bps link! If it's usable at 9600, then it'll be grand for most folks. I really hate sites that are slow to load even on a cable modem or T1 link. And those are too too common -- the developers aren't designing for the real world, but for an indoor demo.
Covad bought Bluestar last year. Bad move. They're now mostly closing down much of what they bought from Bluestar, and considering a write-down of the whole purchase. Bluestar did retail DSL+ISP while Covad was primarily wholesale DSL to ISPs. The combination made Covad a competitor of their customers. Bad move. Plus their customers (ISPs) have been, uh, slow to pay.
Covad doesn't compete much with cable modems or cheap consumer DSL. Their main business is SDSL to business. Rather oversaturated; several CLECs (not usually ILECs) sell in that space, and there's not enough business to go around. Covad might end up the survivor though.
"At the moment, many users are technical, and not consumer"? That's true, but the point of Maximum Linux was to help spread the good news of Linux to the consumer community. Not the lusers who thought Packard Bell was cool, but the non-programmers who wanted an alternative to Windows, and who appreciated what Linux had to offer -- even potentially offer to non-programmers. The trouble with Linux (and most Unix in general) is that it's built by programmers for programmers. That runs out of users pretty fast. ML understood that.
It wasn't perfect. Some of their "attitude" was tiresomely manufactured. But they had great CDs (I got a few distros off of them, among other things) and some really helpful articles. I'll miss 'em. I hope somebody picks up the slack.
Compulsory licensing has plenty of precedent. The one that strikes me as most relevant is radio broadcasting. A radio station may play a record on the air without negotiating a royalty rate with the record company. They also pay a clearance fee, which allows the owners of the music to be paid at the fixed compulsory license rate.
Before 1943, radio stations were not copyright-licensed to play recorded songs! They played mainly live music; networks had orchestras, paid musicians to show up, etc. Those were the "radio days", before there were disk jockeys. The law was changed in 1942 and the musicians' union revolted. They refused to make records for over a year! Radio stations could play them but there was no new material.
The story tends to get publicized every Christamstime, because there was only one hit Christmas record released in 1943. Spike Jones recorded "All I Want For Christmas Is My Two Front Teeth" at a New Years' Eve party in 1942, minutes before the strike. Come late 1943, it was the only new record out there! Naturally, it was a hit. In 1944 they settled, and while the musicians lost some radio work, the record business prospered, and radio play became an important source of record sales. The world adapted.
Now what's music on the web? I think a case can be made that it is functionally a substitute for radio. I can turn on the radio, or I can select tunes off of a web site or Internet music service (like, say, Napster). As it stands, web radio stations are not granted compulsory license. But that is the sort of controversial call that Congress needs to look into. What is the correct balancing of interests?
And I point out that "compulsory license" is NOT a "big government" compulsion. As others have noted, Intellectual Property, and private property in general including real estate, are creations of government. As a society we agree to restrict our rights to do or take certain things for the common good. Government steps in to enforce those rights. Compulsory licensing is actually a smaller-government approach, because it tends to simplify or reduce the amount of contract enforcement (a key function of a government or, in societies that don't have a working government, a mafia) by the government. I doubt they have a lot of "intellectual property" rights in Somalia (a country not ruled by any government today).
Yes, this is an old trick. Whenever there's a meteor shower (a few days of increased activity, when Earth passes near the residue of a comet; this happens a few times a year on schedule) a lot of ham operators point their VHF (2 meter or 70 cm CW, typically) antennas at it and bounce signals off of the tails. Not exactly the meat of long conversations, but a nice way to get some well-beyond-the-horizon contacts into the logbook (think "radiosport"; we do it for the challenge, often competitively).
I suppose if you throw enough at it, you can find enough tiny meteors to make it a fairly regular means of communications. But it's rarely the medium of choice, what with dirt-cheap geostationary satellite bandwidth for the sites that can't get to the fiber networks.
Public Utilities Fortnightly, which is perhaps a bit less well-read than Sports Illustrated, had a good article in its January 1 issue. Basically, it noted that last summer's electricity crunch (which was not quite as acute as this winter's) was apparently caused by the Independent System Operator (ISO), the organization that intermediates between generators and distributors (PG&E, etc.).
Last summer ISO set very high margins for spare capacity before declaring a shortage. WHenever they declared a shortage, spot market prices skyrocketed. Even if the supply wasn't any tighter than what used to be considered normal. Shortages are in the generators' interests.
It could be massive incompetence, but it's potentially much worse, a cartel among producers. ISO allows the generators to have the kind of cartel that OPEC couldn't create. OPEC countries frequently cheat on their quotas. ISO monitors production and reports what each generator puts out. So if a generating company (Enron, Duke, whatever) happens to be holding back in order to raise the price, and another one boosts production to make some quick bucks, the holding-back generators know it. That prevents cheating, and keeps the supply down and the price high.
The article at pur.com is not available online to nonsubscribers, alas.
The FCC does not apparently want to get too involved in enforcing this Bad Law, which specifies that they have to enforce it. So (if you read the NPRM it's clear) what the FCC wants the schools and libraries to do, when filling out the form acknowleging receipt of the federal money, is say "we comply" or "we do not need to comply".
Beyond that "self-certification", the FCC says nothing about filtering itself. Clearly they are not enthusiastic.
And don't blame Clinton: McCain pushed this one through as a Rider to a big appropriations bill. Vetoing that bill would have been very hard. Riders are a trick for passing bad law without going through the constitutional approval process.
This is probably the key issue for fighting this action! The first amendment supposedly guarantees freedom of the press. That includes the right to set up a press with no licensure. Broadcast airwaves, on the other hand, have always been subject to a scarcity doctine. They are the press, so they are somewhat free, but that freedom is circumscribed because there isn't enough spectrum to go around. That is a technical matter and it is recognized legally.
Now, the FCC is the agency charged with determining technical matters on the airwaves. They determined that the old FM interference rules, promulgated in the 1950s during an era of vacuum-tube radios, are obsolete. With considerable study, they loosened up the requirements just a smidge, but enough to let in a lot of <= 100 watt stations. (They explictly rejected their original proposal of allowing LPFMs to have 1000 watts, and rejected 2-channel spacing, but allow 3-channel spacing instead of the old rule's 4-channel requirement.)
Now the NAB et al are concerned about competition. They are making noises about interference, but there's a strong record at the FCC to show that those concerns are not realistic. So they're trying to change the law to override the FCC. Trouble is, the FCC's law interfered with freedom of press because of interference concerns. The FCC's record should be a good weapon in court to prove that the new law is a violation of the free press -- it goes beyond what has been demonstrated technically necessary.
And if that law falls, then the whole issue of broadcast licensure may even be subject to reopening. After all, Congress has demonstrated that it's based on anticompetitive theory, and thus has made the whole thing suspect. Wouldn't that be grand!
Yes, Prodigy's still around. Its two main owners are a) Telmex, or at least its controlling owner Carlos Slim, and b) SBC, the American mega-Bell telco.
So there are Prodigy posters on every pay phone in Mexico. And Prodigy picked up all of SBC's DSL and retail customers, including the PacBell area. Thus it's not the same company that Sears and IBM owned, and it's acting like any other retail ISP, but it's still pretty big.
That's just ridiculous chauvinism. Unix isn't so great. It has obvious advantages, in some areas, over certain well-known straw horses which Unix devotees like to poke fun at. But it has many flaws too. The problem is that these flaws become accepted as part of the religion: Since they're part of holy Unix, they must be right. But they're not.
And when it comes to quality, Unix gives you all the tools you could possibly want to make software buggy.
No, I said that because AT&T provides ISP service over cable, they should be allowed to be the only company offering ISP service *over their own cable*, if they're so foolish as to exclude other ISPs. (They're currently under a contractual obligation to exclude all but @Home.)
If you were an ISP and pulled your own cable, or stuck up your own spread-spectrum radios, would you want other ISPs to have access to it at state-set rates? Would you want the state to set your rates?
The Forced Access movement is orchestrated by the (ILEC) telcos as a way to distract users from the telcos own failure to meet their own legal obligations.
Well, no. Competitors are startups, and they have to pick potentially-profitable businesses. As it stands, few CLECs break even, so if you required CLECs to go everywhere at once, they'd
a) need infinite capital to start,
b) have no chance at making a profit,
c) removing their access to capital.
There are specialist CLECs focusing on smaller towns. But they were not the first ones out there, and it'll take a while before the boonies get competition. The big telcos generally charge CLECs more for access to wires in non-urban areas, too, which doesn't help.
Well no, AT&T's right and you're wrong, but it's not entirely obvious why, so I'll explain.
The ILECs (incumbent telcos) used to have a legally-sanctioned monopoly on phone service. Cable companies did only broadcast TV. The cable companies generally did NOT have legal monopolies, but it's economically hard for a second cable operator to enter a town that already has one. RCN, Knology and a few others are trying.
Now, cable companies are allowed to enter the phone and cable-modem businesses. But in those fields, they are the *second* provider, the competitor, since the ILEC had the wire first and had it on a monopoly basis. The rule in this country is that competitive entrants can charge whatever they please, because they're not monopolies -- they're challenging them and need to establish a business model.
So when a cable company provides ISP service over its plant, it is acting purely as a new competitor -- five years ago, only the phone company was there. If you change the rules and regulate the cable companies' provision of ISP service, then you're a) entering the scary domain of regulating ISPs, and/or b) applying rate regulation to a competitive new entrant, which in so doing dries up capital for new entrants faster than an Arizona summer day.
The cable companies should learn that the ISPs are their friends, and they should be competing with the phone companies for the wire business. They've not figured it all out yet, but AT&T is going down that road. They are not going to renew existing exclusivity deals with @Home. Regulating it, hwoever, is the ILECs trick for keeping the cable companies to the TV business and out of their hair. Legally, DSL providers (LECs) are regulated common carriers open to all ISPs at a posted price; cable companies are not common carriers.
AT&T lost the use of the "Bell" name in 1984. Today's AT&T is very different from the old Ma Bell. Most of the monopolist culture stuck with the RBOCs, who got the shared Bell trademark (though it's fading in some areas).
I too have AT&T Broadband/formerly-MediaOne phone service and it's quite good. AT&T still has management problems but in point of fact they're sworn blood enemies of the Bell companies, especially Bell Awful/Bell Titanic/VeriZontal. That's reason alone to wish AT&T well.
Slashdot Mirror
Plain old spectrum notches are really obvious.
But it seems that the SDMI-crack-discovered-it notch occurs only in some *segments* of the music. Not all the time. So it isn't nearly as obvious.
Of course in some kinds of music, notably classical again, it is still likely to be quite intrusive. I think SDMI only intended to be useful with things like Spice Girls and 2Pac. (Not that I want to associate those with each other.) Classical is small business, so if they screw it up, they don't really care.
The principal Verance watermark is similar. According to the paper, it's basically a periodic up or down amplitude tweak (each baud interval being 16/50 of a second?) to a set of narrow ranges in the top octave. Sort of like typing onto a graphic equalizer. Using hashed track data to create a weak checksum. Again, it's hard to tell perceptually, especially on pop/rock, but might intrude onto some classical recordings.
No, Canada now permits local telephone competition, so Telus' monopoly is historic rather than de jure. But due to the economics (e.g., residential service from Telus is priced below the cost of raw wire from them to a competitor), it's very hard for a competitor to sell residential service profitably. Unless it's the CATV, who's already there.
Multipath's a different story. Different radios survive it differently. Cisco, for instance, is quite vocal about how their OFDM radios for MMDS are multipath resistant (they work okay even if the signal is bounced off of a few buildings). Some others aren't. I don't know what radio Spring is using here but I suspect it's one of the newer, more multipath-resistant ones. That, btw, is what vendors are talking about when they say they support non-line-of-sight. Hills are a different story -- that's no path!
You get an earful of lies, is what you get from your friends at the Bells!
The law is pretty clear. If Bell runs a line, a CLEC can demand Bell provide him with access at COST PLUS. Not minus. Maintenance is a cost that is included in the rent. Thus, my CLEC can lease raw copper wire analog/ADSL loops from VZ/New York Telephone in Ballston for, oh, $19/month or so, while VZ's local flat rate telephone service, built out of similar wire plus the whole network, costs less than that. Bells are notorious for their creative accounting.
Sprint is already doing this in a few other cities, such as Phoenix, where they have the MMDS license. Worldcom is doing it in Jackson, MS and a few other cities, and will be expanding it too (they have the NY and Boston licenses).
Note that "line of sight" for MMDS is much better than optical; it means "not over the horizon". Since Chicago is basically flatland, hills aren't the problem they would be in, say, New England. Which is why this Chicago rollout is so important; it could give the technology a real boost. MMDS operates around 2.5 GHz. It is not subject to significant rain fade, and passes easily enough through trees. (Contrast this to LMDS at 29 GHz, which has a typical reliable range of around 2 miles, because of rain fade, though it goes much farther on dry days.)
Each market has one MMDS licensee. This was the FCC's last pre-auction lottery, nicknamed "wireless cable". It was intended for pay-TV broadcast distribution. A bunch of shady operators took fees to enter people into the license lottery ca. 1993. The MMDS companies who bought up the licenses from the lottery winners discovered that there wasn't much of a market, so they went bankrupt or sold out to Sprint and Worldcom (who between them have most of the country's population covered by their licenses, but are just starting to offer service). Now it's viewed as a DSL alternative. Some other operators are also in business; Oxford Telecom, for instance, does MMDS data in Portland, Maine.
This is mostly two-way radio, something the FCC authorized a couple of years ago. (Early systems were dial-up return.) I don't really think there's enough bandwidth there to replace DSL or cable modems in urban areas, but it's a good alternative for people who are out of range of those services. Alas, with only one license per city (spectrum being a scarce resource), it's not totally competitive.
No, salami. The Bells are ALLOWED already to do almost whatever they want locally, provided they're not blatantly anticompetitive. They are ALLOWED to run fiber to the home, and used to promise it. They're ALLOWED to run DSL to the home. They just are trying to get around some laws that require them to let others in the business too. So they hold back things they could be doing, as well as interfere with competition, and demand absolute monopoly control. (Specifically, Ed Whitacre of SBC has recently bitched to the Illinois Commerce Commission that he can't provide DSL to most homes there because the current rules require him to let competitors, like Covad, lease copper from him. He wants to cut their copper.)
If they had the monopolies they want, do you think ISPs would get a fair deal? Fat chance. Note too that VZ DSL here has a 90k byte upstream speed, much lower than competitors. It's just TV to them. Your keyboard is a remote control.
Precisely. The phrase "Computer Science" refers to a math/logic subject which was given that name. But computer can describe science, the same way any noun can be used as an adjective. Such "computer science" isn't the same thing, so the headline is bad work.
The rest of the article is good, though, in pointing out how compute-intensive much modern science is. Still, it's not CS; computers are just a powerful tool.
Yes, that would be a better way.
.com et al. Poof, it's done, for all applications.
Idealab this time seems again to be Cluelesslab. DNS is a hierarchical system. Most retail end users now get their DNS service from their ISPs; commercial end users (leased lines, etc.) may have an in-house DNS but that too is hierarchical under their ISP's.
Now anybody can point to any other DNS, but it's not trivial for Joe User, so it's really best if the ISPs do the fix in their DNS servers. They can simply add these alternative roots next to
The tricky part is Worldcom's UUNET, the largest backbone ISP (upstream from a lot of retail services). Vint, the Elizabeth Taylor of the Internet (famous because he's famous, and didn't he once make a movie about a horse when he was a kid?), works for them. So Roberts gave him the hot title at ICANN, so he won't defy them. But UUNET's customers (who mostly run DNS servers) can still do the fix.
But plug-ins? Gross needs a cranial plug in.
(BTW, Idealab has put up for sublease its fancy Boston digs, occupied for less than a year. They're shrinking away with their stock portfolio)
Major parts of The City of Boston (47 sq. miles, 580k pop.) are out of DSL range. The COs just aren't put close together, so urban neighborhoods (row houses, etc.) are not loop qualified. We're not talking the woods. It's just a crap shoot if you qualify -- the BEST RBOC loop quals are a bit over 50%, while some cities (not woods) are below 20%.
9600 is not quite real world, but 24k (top speed of a modem on a phone line served by Universal DLC, which is common) is quite real, so if you can work at 9600 at all, then at 24k it's probably decent. Maybe today they should test at 24k. No faster.
Network games and web sites both need to be designed by people who pay attention to low-rate connections. Besides the fact that the Internet path itself gets rather slow quite often, modem-speed access indeed will remain the norm for some years, as many people have no choice. And there's nothing that the ISPs can do about it, because it's the telecomm carriers and cable companies who have the wires.
Why some folks just can't get broadband:
* To get a cable modem, the cable company has to have the upgraded Hybrid Fiber-Coax (HFC) plant, as well as a cable modem terminating system (CMTS) and the rest of the needed gear. This is becoming more common but lots of cable systems aren't there yet; for instance, the old TCI systems were real fixer-uppers.
* To get DSL, the subscriber must be within 15,000 wire feet or so of the ILEC central office. This rules out a lot of homes, even in cities, because COs are spaced wider than that in all but the densest places.
* DSL won't work if the subscriber is served by a Digital loop carrier (DLC) system, which is the norm for new installations more than 12,000 feet or so from a CO (and sometimes much closer). Somebody could theoretically put a DSLAM at the DLC location, but the economics and practicality are often poor; it's very very rare at present.
* DSL won't work if the wire is not in good shape. That's often the case. (Especially in former-NYNEX territory!)
* There still has to be a DSLAM; this is hard to justify in smaller COs. DSL's basically an urban service.
* Wireless bandwidth is expensive. Even if you could go faster than 9600 bps on cellular, you wouldn't like the price. Unlicensed wireless is "free" bandwidth, but the range is short, so again most people don't live in range of a provider (wireless ISP). And that requires an antenna location, decent near-line-of-sight path, etc.
So I have advised my consulting clients to design their web sites using a 9600 bps link! If it's usable at 9600, then it'll be grand for most folks. I really hate sites that are slow to load even on a cable modem or T1 link. And those are too too common -- the developers aren't designing for the real world, but for an indoor demo.
Covad doesn't compete much with cable modems or cheap consumer DSL. Their main business is SDSL to business. Rather oversaturated; several CLECs (not usually ILECs) sell in that space, and there's not enough business to go around. Covad might end up the survivor though.
"At the moment, many users are technical, and not consumer"? That's true, but the point of Maximum Linux was to help spread the good news of Linux to the consumer community. Not the lusers who thought Packard Bell was cool, but the non-programmers who wanted an alternative to Windows, and who appreciated what Linux had to offer -- even potentially offer to non-programmers. The trouble with Linux (and most Unix in general) is that it's built by programmers for programmers. That runs out of users pretty fast. ML understood that.
It wasn't perfect. Some of their "attitude" was tiresomely manufactured. But they had great CDs (I got a few distros off of them, among other things) and some really helpful articles. I'll miss 'em. I hope somebody picks up the slack.
Welcome the the recession.
Compulsory licensing has plenty of precedent. The one that strikes me as most relevant is radio broadcasting. A radio station may play a record on the air without negotiating a royalty rate with the record company. They also pay a clearance fee, which allows the owners of the music to be paid at the fixed compulsory license rate.
Before 1943, radio stations were not copyright-licensed to play recorded songs! They played mainly live music; networks had orchestras, paid musicians to show up, etc. Those were the "radio days", before there were disk jockeys. The law was changed in 1942 and the musicians' union revolted. They refused to make records for over a year! Radio stations could play them but there was no new material.
The story tends to get publicized every Christamstime, because there was only one hit Christmas record released in 1943. Spike Jones recorded "All I Want For Christmas Is My Two Front Teeth" at a New Years' Eve party in 1942, minutes before the strike. Come late 1943, it was the only new record out there! Naturally, it was a hit. In 1944 they settled, and while the musicians lost some radio work, the record business prospered, and radio play became an important source of record sales. The world adapted.
Now what's music on the web? I think a case can be made that it is functionally a substitute for radio. I can turn on the radio, or I can select tunes off of a web site or Internet music service (like, say, Napster). As it stands, web radio stations are not granted compulsory license. But that is the sort of controversial call that Congress needs to look into. What is the correct balancing of interests?
And I point out that "compulsory license" is NOT a "big government" compulsion. As others have noted, Intellectual Property, and private property in general including real estate, are creations of government. As a society we agree to restrict our rights to do or take certain things for the common good. Government steps in to enforce those rights. Compulsory licensing is actually a smaller-government approach, because it tends to simplify or reduce the amount of contract enforcement (a key function of a government or, in societies that don't have a working government, a mafia) by the government. I doubt they have a lot of "intellectual property" rights in Somalia (a country not ruled by any government today).
Sort of brings new meaning to the word "phlegm", no?
Yes, this is an old trick. Whenever there's a meteor shower (a few days of increased activity, when Earth passes near the residue of a comet; this happens a few times a year on schedule) a lot of ham operators point their VHF (2 meter or 70 cm CW, typically) antennas at it and bounce signals off of the tails. Not exactly the meat of long conversations, but a nice way to get some well-beyond-the-horizon contacts into the logbook (think "radiosport"; we do it for the challenge, often competitively).
I suppose if you throw enough at it, you can find enough tiny meteors to make it a fairly regular means of communications. But it's rarely the medium of choice, what with dirt-cheap geostationary satellite bandwidth for the sites that can't get to the fiber networks.
Public Utilities Fortnightly, which is perhaps a bit less well-read than Sports Illustrated, had a good article in its January 1 issue. Basically, it noted that last summer's electricity crunch (which was not quite as acute as this winter's) was apparently caused by the Independent System Operator (ISO), the organization that intermediates between generators and distributors (PG&E, etc.).
Last summer ISO set very high margins for spare capacity before declaring a shortage. WHenever they declared a shortage, spot market prices skyrocketed. Even if the supply wasn't any tighter than what used to be considered normal. Shortages are in the generators' interests.
It could be massive incompetence, but it's potentially much worse, a cartel among producers. ISO allows the generators to have the kind of cartel that OPEC couldn't create. OPEC countries frequently cheat on their quotas. ISO monitors production and reports what each generator puts out. So if a generating company (Enron, Duke, whatever) happens to be holding back in order to raise the price, and another one boosts production to make some quick bucks, the holding-back generators know it. That prevents cheating, and keeps the supply down and the price high.
The article at pur.com is not available online to nonsubscribers, alas.
The FCC does not apparently want to get too involved in enforcing this Bad Law, which specifies that they have to enforce it. So (if you read the NPRM it's clear) what the FCC wants the schools and libraries to do, when filling out the form acknowleging receipt of the federal money, is say "we comply" or "we do not need to comply".
Beyond that "self-certification", the FCC says nothing about filtering itself. Clearly they are not enthusiastic.
And don't blame Clinton: McCain pushed this one through as a Rider to a big appropriations bill. Vetoing that bill would have been very hard. Riders are a trick for passing bad law without going through the constitutional approval process.
TENEX, and its twin TOPS-20, were way cool. Compared to Unix, it's like the difference between driving a Mercedes and driving a Willys Jeep.
This is probably the key issue for fighting this action! The first amendment supposedly guarantees freedom of the press. That includes the right to set up a press with no licensure. Broadcast airwaves, on the other hand, have always been subject to a scarcity doctine. They are the press, so they are somewhat free, but that freedom is circumscribed because there isn't enough spectrum to go around. That is a technical matter and it is recognized legally.
Now, the FCC is the agency charged with determining technical matters on the airwaves. They determined that the old FM interference rules, promulgated in the 1950s during an era of vacuum-tube radios, are obsolete. With considerable study, they loosened up the requirements just a smidge, but enough to let in a lot of <= 100 watt stations. (They explictly rejected their original proposal of allowing LPFMs to have 1000 watts, and rejected 2-channel spacing, but allow 3-channel spacing instead of the old rule's 4-channel requirement.)
Now the NAB et al are concerned about competition. They are making noises about interference, but there's a strong record at the FCC to show that those concerns are not realistic. So they're trying to change the law to override the FCC. Trouble is, the FCC's law interfered with freedom of press because of interference concerns. The FCC's record should be a good weapon in court to prove that the new law is a violation of the free press -- it goes beyond what has been demonstrated technically necessary.
And if that law falls, then the whole issue of broadcast licensure may even be subject to reopening. After all, Congress has demonstrated that it's based on anticompetitive theory, and thus has made the whole thing suspect. Wouldn't that be grand!
Yes, Prodigy's still around. Its two main owners are a) Telmex, or at least its controlling owner Carlos Slim, and b) SBC, the American mega-Bell telco.
So there are Prodigy posters on every pay phone in Mexico. And Prodigy picked up all of SBC's DSL and retail customers, including the PacBell area. Thus it's not the same company that Sears and IBM owned, and it's acting like any other retail ISP, but it's still pretty big.
That's just ridiculous chauvinism. Unix isn't so great. It has obvious advantages, in some areas, over certain well-known straw horses which Unix devotees like to poke fun at. But it has many flaws too. The problem is that these flaws become accepted as part of the religion: Since they're part of holy Unix, they must be right. But they're not.
And when it comes to quality, Unix gives you all the tools you could possibly want to make software buggy.
No, I said that because AT&T provides ISP service over cable, they should be allowed to be the only company offering ISP service *over their own cable*, if they're so foolish as to exclude other ISPs. (They're currently under a contractual obligation to exclude all but @Home.)
If you were an ISP and pulled your own cable, or stuck up your own spread-spectrum radios, would you want other ISPs to have access to it at state-set rates? Would you want the state to set your rates?
The Forced Access movement is orchestrated by the (ILEC) telcos as a way to distract users from the telcos own failure to meet their own legal obligations.
Well, no. Competitors are startups, and they have to pick potentially-profitable businesses. As it stands, few CLECs break even, so if you required CLECs to go everywhere at once, they'd
a) need infinite capital to start,
b) have no chance at making a profit,
c) removing their access to capital.
There are specialist CLECs focusing on smaller towns. But they were not the first ones out there, and it'll take a while before the boonies get competition. The big telcos generally charge CLECs more for access to wires in non-urban areas, too, which doesn't help.
Well no, AT&T's right and you're wrong, but it's not entirely obvious why, so I'll explain.
The ILECs (incumbent telcos) used to have a legally-sanctioned monopoly on phone service. Cable companies did only broadcast TV. The cable companies generally did NOT have legal monopolies, but it's economically hard for a second cable operator to enter a town that already has one. RCN, Knology and a few others are trying.
Now, cable companies are allowed to enter the phone and cable-modem businesses. But in those fields, they are the *second* provider, the competitor, since the ILEC had the wire first and had it on a monopoly basis. The rule in this country is that competitive entrants can charge whatever they please, because they're not monopolies -- they're challenging them and need to establish a business model.
So when a cable company provides ISP service over its plant, it is acting purely as a new competitor -- five years ago, only the phone company was there. If you change the rules and regulate the cable companies' provision of ISP service, then you're a) entering the scary domain of regulating ISPs, and/or b) applying rate regulation to a competitive new entrant, which in so doing dries up capital for new entrants faster than an Arizona summer day.
The cable companies should learn that the ISPs are their friends, and they should be competing with the phone companies for the wire business. They've not figured it all out yet, but AT&T is going down that road. They are not going to renew existing exclusivity deals with @Home. Regulating it, hwoever, is the ILECs trick for keeping the cable companies to the TV business and out of their hair. Legally, DSL providers (LECs) are regulated common carriers open to all ISPs at a posted price; cable companies are not common carriers.
AT&T lost the use of the "Bell" name in 1984. Today's AT&T is very different from the old Ma Bell. Most of the monopolist culture stuck with the RBOCs, who got the shared Bell trademark (though it's fading in some areas).
I too have AT&T Broadband/formerly-MediaOne phone service and it's quite good. AT&T still has management problems but in point of fact they're sworn blood enemies of the Bell companies, especially Bell Awful/Bell Titanic/VeriZontal. That's reason alone to wish AT&T well.