As many people have stated already, DSL is a copper technology -- it uses higher frequencies on the same twisted pair as your analog home phone line to handle the data side of things. In fact, some DSL installations require putting a filter on all the phones in your house to keep you from hearing those frequencies.
Ignoring the Fiber/Copper thing, DSL has always had problems with remotes. Remotes are devices that are not co-located with the CO switch, but sit closer to customers -- they're often in large green-ish boxes, typically one per neighborhood or so. (Remotes can actually get quite large...) Anyway, a Remote takes a large number of phone lines and relies on the fact that not all of them are going to be on the phone, to concentrate down to a smaller number of lines that go to the Central Office (CO). Typically, these conversations are sampled and sent digitally to the CO.
So, stepping back, DSL works by having your twisted pair connected up to a DSLAM (AM = Access Module, I believe), which strips out those higher frequencies and forwards your audio onto the CO switch.
Any remote will have an issue with DSL, because the DSLAM is typically located at the CO. Digital remotes don't even transfer the higher-frequency signals to the CO, and analog remotes (much less common) are set up so your twisted pair isn't always connected to the CO.
Now, phone companies will occasionally run a few twisted pairs straight from the CO to the site of the remote, often for future expansion. (A T1 trunk is physically just 2 twisted pairs.) It's possible for the CO to use those pairs to handle individual lines instead of configuring them as trunks.
More modern remotes use fiber to talk to the CO -- fiber carries many more conversations and doesn't have to be regenerated as often. So, if there's only fiber going to the CO, then it isn't possible to use DSLAMs at the CO.
Now, recently, it's become possible to install DSLAMs at the remote, extending the range of DSL. Not too common yet, but will be the way that many phone companies do it.
For fiber-to-the-curb scenarios (this is what I have), my local phone company (Sprint) is offering me what they call DSL, but which is actually ethernet straight from the ONU. (Optical Network Unit -- it's a smaller box that sits on our curb and translates from fiber to copper. We only have copper for the last 500 ft or so.) The ONU has a a router card in it that forwards my IP traffic into the rest of their network via fiber.
So, in short...
1. It is possible to get DSL if you're on a remote. Either use a pair straight to the CO, or use a DSLAM at the Remote.
2. With fiber, it's harder, but IP can be carried over fiber if the phone company wants to make the investment.
So, about 12 years ago, I was the head grader for a CS course at CMU, and I devised a similar approach to catching cheaters on computer programs. The class was around 100 students, so comparing each pair of programs was ruled out. Instead, I just ran the programs through a script that counted things like semi-colons, braces, parentheses, number of lines and so on. Then, it was just a matter of looking at the ones that were "close" based on those numbers.
It all works because cheating is, almost by definition, lazy -- if you're going to go through the effort to change the original sufficiently to make it hard to tell that you cheated, then why not just do the work yourself?
As it happened, among the people we caught were 2 people who not only happened to be roommates, but also happened to be alphabetically next to each other in the class, and who can been caught cheating once before.
The "CD Protectors" readily admit that they can't be 100% successful, just because of the wide variety of CDROMs, rippers, &c out there. So, a few things are going to happen:
1. People will discover the technology that allows ripping, and it will actually be a feature of that technology. Knowledge of what technology is necessary will spread around and more people will use it.
2. Those who don't possess the technology will sometimes buy a "copy-protected CD," and try to convert it to mp3s for use in their portable player. Unable to do so, they'll go looking out on the 'net for mp3 versions of these songs, and will find them because some people will possess technology that allows these CDs to be ripped.
3. Once person X discovers how easy it is to get copies of mp3s, he may not see any reason to buy the original CDs to begin with -- after all, if all he wants to do is play it on his mp3 player, the original CD doesn't do him any good at all.
So, I expect that this will actually result in more piracy, rather than less.
Two other predictions:
1. Out of spite, a lot of people will buy the Charlie Pride CD and then return it to the store because "it didn't work in my car" or whatever.
2. Within a day of being released, mp3s will be available in 100 places on the net.
If the music companies want to stem piracy, they need to meet their customers needs: Distribute their own digital music on-line, Sell individual songs instead of entire CDs with 2 good songs and 8 crappy ones, and make the price reasonable enough and the process easy enough that buying from them just makes the most sense.
In short, their problem is that their customers are sick of spending $15 for a CD with 2 good songs on it.
So, how does one force an international loosly-organized volunteer organization to do anything?
Exactly what punishment would a US court enact? A fine? Who do you send the bill to? What if they just issue an order that the IETF do something, and then at the next IETF meeting (in London), they just ignore the order? Will the judge hold the IETF in contempt and order it arrested? Maybe its assets should be seized. (What assets does the IETF have? The only one I can think of, ironically, is the 'IETF' name.)
I understand that the US Congress authorized ICANN, but I don't recall that they ever passed anything that said everybody needed to pay attention to them.
The internet has always worked under rough consensus, where people generally agree on how things are going to work. The ICANN changes that model, really without consent.
1. A 'billion million years' is effectively permanent, and would, I suggest be unconstitutional.
2. The dissenting judge has a point -- the Constitution grants the power to (A) "Promote the progress of science and useful arts", and (ONLY as a method of promoting science and useful arts) (B) to secure "for limited times to authors and inventors the exclusive right to their respective writings and discoveries." So, if Congress does (B) above in a manner inconsistant with (A) above, then it has acted unconstitutionally.
It's like I said: "I'm going to let you live in my house for the next year and I'll let you take responsibility for upkeep, by giving you access to my bank account." I come back a year later, and my house is in decent shape, but you've emptied my bank account and bought yourself a land rover. Under the majority's iterpretation, you'd say "Well, yeah, you gave me access to your bank account," but I gave you that access to take care of the house, not to buy yourself a car.
The Constitution allows the Congress to set laws instituting Patents and Copyrights "To promote the Progress of Science and useful Arts" "for limited times." How long do you believe copyright should reasonably exist on a work and what "fair-use" of copyrighted materials do you believe should be allowed? In addition, are there any types of work that are currently copyrightable which you either do not believe are worthy of copyright or believe should be treated differently in copyright law?
Obviously not, otherwise I could send you a copy of Microsoft office without any prior agreement between us and you would then be able to copy and distribute it at will.
The key point there is "obligation whatsoever *to the sender*." If Microsoft were to send me a copy of Office, then this regulation would allow me to do whatever I wanted to. However, there are other federal laws and regulations (ie copyright law) which prevent you from doing just that. I suspect that any reasonable court would find that copyright is still protected even in light of this postal regulation.
So, it seems to me that if the government wants to change how the US is defined, they need to submit a draft in the appropriate format to the appropriate IETF working group for consideration at the next meeting. The US government can't just bypass the procedures that have been set up for managing the internet. (Don't get me started about ICANN.)
Not saying that.us doesn't need some changes, just that there's an established procedure for change and the commerce department can't just run rough-shod over that procedure.
As for the last question ("How did the fans find out [about the music]"), by-and-large, the answer is that record companies pay radio stations and MTV to play their music when it first comes out. In effect, they plant a seed of interest in the minds of consumers. Then, after a while, the music companies stop paying for the music to be played, instead demanding payment when it is. But, at that point, the consumer interest is already there. Then, the music companies use their distribution systems to get the same music out to the stores.
This method of advertising and distribution has been next to impossible for unsigned acts to accomplish. However, the internet & MP3s change all that -- suddenly, you don't need a massive distribution network, because the king of networks (the internet) does it for you. And, in a environment where people are well-connected, word-of-mouth becomes a much more effective method of getting known.
So, you can see why the major labels are worried: their entire raison d'etre is being challenged by the internet: If you don't need a record company to do your advertising, and you don't need a record company to get your music to consumers, what's left for them to do?
So, it seems to me that there are two ways of dealing with this:
1. Prevent your browser from accessing the offending site (doubleclick in this case...)
and...
2. Sending so many bogus requests to the site that any real data they collect would be totally obscured by the gazillions of bogus html requests they get in. (Hmm... an IP address associated with Senator so-n-so's office has visited the "Curing Impotence" site 5 million times in the past hour.)
How hard would it be to write a process that spoofed IP addresses on HTTP requests? You could even make it part of a virus -- imagine that, a virus that actually did something good.
I work for a large telecom company, and we typically sell networks consisting of many products that were developed independently, possibly including 3rd party equipment and/or equipment from companies where we just bought the company out-right.
So, we have the same problem on a smaller scale -- we may have 10 different implementations of the same protocol -- some in firmware, some in (ack) hardware, some in one language, some in another, some off-the-shelf, etc.... (We try to avoid having multiple people doing the same thing, but sometimes it just can't be helped.)
To enforce adherence to the standards: we write that requirement into contracts with third-party vendors; we have testers who pound on the different protocol implementations and generate a lot of negative press when something isn't up to standards; and we do a lot of interoperability testing -- if an implementation isn't doing what it's supposed to do, then the people who did it get a *lot* of heat put upon them to fix it.
1. I'm a software guy making more than twice the starting salary our company offers to kids right out of school.
2. A big chunk of the value of experience depends on the *industry* you're in. For example, I'm in the telecom industry, and it takes a good couple of years just to learn all the terminology and special requirements in a telecom environment. Companies like Cisco, Lucent & Nortel are continually raiding each other for experienced people.
3. Who really wants to be programming in 20 years? I'm 32 years old, and if I haven't advanced beyond programming, I'm going to be very frustrated.
4. Your value goes down once you graduate, not because of your salary, but because the industry is moving so quick -- if you aren't continually updating your skills, next years college graduate is going to know more than you do. They say that the half-life of an education is 5 years.
Re:This is the wrong question
on
Linux Failover?
·
· Score: 2
So, in the telephony industry, this is a big requirement... The problem is that you don't want the outage of a single network link to take your machine out of service, or to interrupt any existing transactions.
So, for example, if you have a TCP stream going to a specific NIC and the link between the NIC and the switch gets cut, or the NIC fails or something, then you need to be able to continue the same TCP stream on a second interface.
You end up with several issues: On a lot of NICs, it's not that easy to figure out when it's having problems. Secondly, the second NIC is typically at a different hardware address, so you need to update the ARP cache of any machine sending to you. And, you have to figure out how to tell when the first NIC is working again.
So... It seems to me that it's only illegal to distribute copies of a work to people who are not licensed for that work: I can take my favorite book to the photocopier, but I can't give the photocopy of the book to somebody who doesn't already own a copy of the book. (unless I also give that person the original and every other copy I've made.)
My understanding is that the my.mp3.com service verifies that the clients actually own the CD, so I don't see where the problem is. It sounds like the recording industry has successfully convinced a judge that fair use should be banned.
So, it seems to me that there's a actually a good argument in the text of the constitution to say that the Congress' retroactive extending of copyright is unconstitutional. Last year, in the "Sonny Bono Copyright Act," Congress extended copyright to 95(!!) years, primarily in response to lobbying from Disney and other media companies.
The problem is that extending copyright retroactively doesn't encourage the arts -- the work has already been produced, quite often by somebody who is already dead! And, since the constitution allows Congress to institute copyright to encourage the arts, retroactive copyright extension doesn't do it.
I wonder what their response is going to be when doing a web-search for "Leonardo" comes up with a bunch of "Leonardo Finance is a bunch of fascist bastards" websites.
Article 1, Section 9, Clause 5 of the US Constitution:
"No Tax or Duty shall be laid on Articles exported from any State."
The Internet Tax Freedom Act, sect 1101(a):
"No State or poilitical subdivison thereof shall impose any of the following taxes during the period beginning on October 1, 1998 and ending 3 years after the date of the anactment of this Act -- [..] (2) multiple or discriminatory taxes on electronic commerce"
Internet Tax Freedom Act, section 1104 ("Definitions"), section 2:
Discriminatory tax -- The term "Discriminatory tax" means --
(A) Any tax imposed by a State or political subdivision thereof on electronic commerce that --
(i) is not generally imposed and legally collectible by [...] on transactions [...] accomplished through other means;
[...]
(iii) imposed an obligation to collect or pay the tax on a different person or tentity than in the case of transactions [...] accomplished through other means."
There are other problems: Purchasing items sent as gifts to people in other states, purchasing items that are not subject to state sales tax.
So, toward the beginning of the letter, the authors make the statement that having an "end-to-end" architecture (ie intelligence on the edge), with generic protocols in the network and applications running on clients is one of the strengths of the internet. Then, it goes on to say that ISPs can provide intelligence in the network, through things like web-page cacheing, and that you should be able to choose an ISP that does that.
This entire argument seems hokey to me. At most, it seems to me that AT&T should allow users to say "I don't want to use AT&T's web servers or usenet servers or mail servers. Please take $15 off my monthly bill." The user could then configure their software to point to somebody else's servers.
This repeatedly goes back to my point (made in this space before) that ISPs are an anachronism caused by trying to route a packet-based network over a switched network: You have to have gateways that convert from one to the other. This, currently, is the main function of ISPs. In a world where you don't need the gateway -- ie where the packet-based network extends all the way to the machines that want to see it (as is the case with packet-over-cable), then the main purpose of the ISP goes away.
Now, I would argue that if Mindspring & AOL want to stay relevant, they need to transform themselves into "ASPs" -- Application Service Providers. At the low end, this means running the same servers that the ISPs ran before -- mail, usenet, web, &c. But, there's a lot of opportunity beyond that -- PC backup services, video servers, network file storage and so on. These are the logical place for the ISPs to move into. But, they're so entrenched in what they've been doing that they can't see to move beyond it.
1. (*main purpose*) Provides the other end for your modem to connect to, establishing IP connectivity to your house, effectively extending the internet to the home.
2. Mail Server 3. Web Server space 4. Usenet setver etc...
I would argue that cable modems, by themselves, do the job of #1, above -- they effectively extend the internet into the home. With cable, there is no set of corresponding cable modems on the other end that you're switched to, depending on what phone number you dialed. By attempting to force cable systems into the ISP model, you enormously drive up the cost. because the cable system would need to know "Ok. This house is a mindspring customer. Route his packets to the mindspring router." Just plain silly.
As for the other 3 things I listed above, I see no reason why they should be grouped together -- why can't I use Hotmail as my mail server, some other usenet server and buy server space from a third place.
So, I argue that if the government is going to mess around with the broadband market, it should prevent forced bundling of services -- The company proving access via cable modem should not be able to force people to also pay to use its mail servers, for example. (This is similar to how Local & Long Distance services are set up -- Sprint is my local carrier, but they don't force me to use Sprint as my LD carrier.)
Let's not confuse the "Red Hat Linux Distribution v6.0" (ie an ordered collection of 1's and 0's) with "Red Hat Linux v6.0," the product. Red Hat Linux, the product includes a CD, manual, box, assorted packing material and the right to obtain support under certain conditions. People who take a CD and burn the 1's and 0's onto it are not creating "Red Hat Linux" -- they're creating a CD containing the "Red Hat Linux Distribution." The two are *not* the same thing.
So, I actually went and *looked* at the mediametrix web site, and saw their list of the top-50 web sites. Note the following observations:
1. Sites #2, #3, & #4 are AOL, MSN and Geocities -- sites where individuals can put up their own web pages.
2. Out of the top 10 sites, 5 are primarily search engines (yahoo, go, lycos, excite, angelfire).
3. Another 2 of the top sites, netscape.com and microsoft.com are (I believe) the default start page for two very popular browsers.
4. The first e-commerce site is amazon.com, at #11 with barnesandnoble.com at #35. The second most-popular commerce site is ebay.com
5. Free-email sites, such as hotmail.com (and yahoo.com) are prominent in the top 50.
6. The first 'news' site is msnbc, at #24, followed by zdnet at #26, pathfinder at #28, wather.com at #30, cnn.com at #31.
7. There are a few unexpected entries -- passport.com (????), ivillage.com ? Lends to questioning how broad of a sample the study used.
So, I would argue that the following are more adequate conclusions:
** Many of the most popular web sites are used only to find other websites -- either as search engines or start pages.
** The sites that provide access for user webpages, such as aol, msn and geocities, are very popular -- meaning that lots of people are doing their own web pages.
So, I would argue that rather than a contraction in web sites, we're actually seeing an expansion -- sure the number of companies hosting these sites is pretty compact, but there's a lot of interest in what amateurs are putting on the net! AOL was able to beat the pants off Amazon, BarnesandNoble, ebay and expedia, COMBINED, just by allowing their users to publish their own web pages!
The fact that the search engines are so popular indicates that many people are visiting sites that aren't the big popular ones (those are typically already bookmarked, or obvious: cnn.com).
So, rather than saying that 'the web is contracting', with the connotation of usage on the web shifting toward a few sites, I'd argue that it's expanding because individual people are creating sites and other people are visiting those sites *more* than they are most of the 'mainline' sites.
My first encounter with the internet was in 1986, when I first went to college. And, the university's ISP was... Well, I suppose it was the university itself, but in reality, we had a "direct connection" to the internet. ISPs didn't exist because at the time 2400 baud modems were the fastest available, and 2400 baud just isn't fast enough for IP.
Fast forward a few years.... Modems get faster, and people start using them to make IP connections across phone lines. Still not incredibly fast, but still somewhat convenient. At this point, most IP addresses were still assigned to specific machines.
With invention of the World Wide Web, interest in the internet started to grow, and a cottage industry erupted around being able to provide IP access to people at home. About the same time, protocols became established to provide temporary IP addresses to a dialed-in machine. Voila! The ISP was born. And, as ISPs grew, they began to offer more services to their customers: e-mail, usenet news, web server space, and so on.
So, ISPs began as a solution to a problem: Nobody had the money for a direct internet connection to their home, but a lot of people wanted internet access. As a result, the primary purpose of an ISP is to provide a place where people can dial their modems and connect to the internet. The other big thing that ISPs do is provide a method for client machines to obtain IP addresses. All the other services ISPs provide: Mail server, news server, web space and so on are not intrinsic to the ISP itself -- it's perfectly possible and reasonable for a person to use one company to dial into the 'net, another company as their mail server, and a third company as a usenet server.
So, let's suppose for a second that it was possible and economically feasible for a person to have a "direct connection" to the internet from their home -- what does the ISP provide?? The only thing that I see that they still do is hand out IP addresses. And, for a short time, that would be useful. But, in the next few years, IPv6 is expected to take off, providing enough IP addresses for every man, woman and child to have several dozen. (I'm not sure of the exact ratio, but it's big.) In that case, there's no reason to borrow IP addresses at all, and the only remaining reason for ISPs to exist goes away.
I would argue that 'always on' internet access, via cable modem and DSL is, in fact, a direct connection to the internet. And, with that, nobody needs an "Internet Service Provider," because they already have internet service.
The problem isn't really when telecomm companies buy each other -- in fact, I'd argue that because of economies of scale, that's a good thing: they're able to reduce costs while putting together a large network. The problem comes in when these companies don't have much competition. We still have AT&T, MCI/Worldcom, Level 3, Bell Atlantic/Nynex, Bell South, Ameritec, Sprint, GTE (well... didn't they just merge with somebody?) and a few dozen CLECs.
My guess is that you'll continue to see consolidation along the way, and we'll end up with a handful (3-5) of companies which will compete fiercely with each other.
The thing I think is the most bizarre about this is the sheer difference in size. According to the WSJ, Global Crossing has 148 employees -- they're a *START-UP* company that's trying to buy US West (~54,500 employees) and Frontier ( ~8100 employees).
Slashdot Mirror
As many people have stated already, DSL is a copper technology -- it uses higher frequencies on the same twisted pair as your analog home phone line to handle the data side of things. In fact, some DSL installations require putting a filter on all the phones in your house to keep you from hearing those frequencies.
Ignoring the Fiber/Copper thing, DSL has always had problems with remotes. Remotes are devices that are not co-located with the CO switch, but sit closer to customers -- they're often in large green-ish boxes, typically one per neighborhood or so. (Remotes can actually get quite large...) Anyway, a Remote takes a large number of phone lines and relies on the fact that not all of them are going to be on the phone, to concentrate down to a smaller number of lines that go to the Central Office (CO). Typically, these conversations are sampled and sent digitally to the CO.
So, stepping back, DSL works by having your twisted pair connected up to a DSLAM (AM = Access Module, I believe), which strips out those higher frequencies and forwards your audio onto the CO switch.
Any remote will have an issue with DSL, because the DSLAM is typically located at the CO. Digital remotes don't even transfer the higher-frequency signals to the CO, and analog remotes (much less common) are set up so your twisted pair isn't always connected to the CO.
Now, phone companies will occasionally run a few twisted pairs straight from the CO to the site of the remote, often for future expansion. (A T1 trunk is physically just 2 twisted pairs.) It's possible for the CO to use those pairs to handle individual lines instead of configuring them as trunks.
More modern remotes use fiber to talk to the CO -- fiber carries many more conversations and doesn't have to be regenerated as often. So, if there's only fiber going to the CO, then it isn't possible to use DSLAMs at the CO.
Now, recently, it's become possible to install DSLAMs at the remote, extending the range of DSL. Not too common yet, but will be the way that many phone companies do it.
For fiber-to-the-curb scenarios (this is what I have), my local phone company (Sprint) is offering me what they call DSL, but which is actually ethernet straight from the ONU. (Optical Network Unit -- it's a smaller box that sits on our curb and translates from fiber to copper. We only have copper for the last 500 ft or so.) The ONU has a a router card in it that forwards my IP traffic into the rest of their network via fiber.
So, in short...
1. It is possible to get DSL if you're on a remote. Either use a pair straight to the CO, or use a DSLAM at the Remote.
2. With fiber, it's harder, but IP can be carried over fiber if the phone company wants to make the investment.
So, about 12 years ago, I was the head grader for a CS course at CMU, and I devised a similar approach to catching cheaters on computer programs. The class was around 100 students, so comparing each pair of programs was ruled out. Instead, I just ran the programs through a script that counted things like semi-colons, braces, parentheses, number of lines and so on. Then, it was just a matter of looking at the ones that were "close" based on those numbers.
It all works because cheating is, almost by definition, lazy -- if you're going to go through the effort to change the original sufficiently to make it hard to tell that you cheated, then why not just do the work yourself?
As it happened, among the people we caught were 2 people who not only happened to be roommates, but also happened to be alphabetically next to each other in the class, and who can been caught cheating once before.
Persons A,B,C...
Person A looks and if B is Red and C is Blue,
then Person A flips a coin and says red or blue.
Next, person B looks. If C is Blue, person B guesses Blue.
Now, person C guesses red.
If it gets by person A, then there are only 3 possibilities for B&C. By getting by B, there's only one left.
There's a 1/4 chance that A will have to guess, so a 1/8 chance that he will guess and guess wrong. If B or C guess, they'll be right.
So, you can do it with only a 1/8 chance of doing it wrong.
I know this doesn't follow the "everybody guess at the same time rule," but I didn't see that in the NYT article...
The "CD Protectors" readily admit that they can't be 100% successful, just because of the wide variety of CDROMs, rippers, &c out there. So, a few things are going to happen:
1. People will discover the technology that allows ripping, and it will actually be a feature of that technology. Knowledge of what technology is necessary will spread around and more people will use it.
2. Those who don't possess the technology will sometimes buy a "copy-protected CD," and try to convert it to mp3s for use in their portable player. Unable to do so, they'll go looking out on the 'net for mp3 versions of these songs, and will find them because some people will possess technology that allows these CDs to be ripped.
3. Once person X discovers how easy it is to get copies of mp3s, he may not see any reason to buy the original CDs to begin with -- after all, if all he wants to do is play it on his mp3 player, the original CD doesn't do him any good at all.
So, I expect that this will actually result in more piracy, rather than less.
Two other predictions:
1. Out of spite, a lot of people will buy the Charlie Pride CD and then return it to the store because "it didn't work in my car" or whatever.
2. Within a day of being released, mp3s will be available in 100 places on the net.
If the music companies want to stem piracy, they need to meet their customers needs: Distribute their own digital music on-line, Sell individual songs instead of entire CDs with 2 good songs and 8 crappy ones, and make the price reasonable enough and the process easy enough that buying from them just makes the most sense.
In short, their problem is that their customers are sick of spending $15 for a CD with 2 good songs on it.
So, how does one force an international loosly-organized volunteer organization to do anything?
Exactly what punishment would a US court enact? A fine? Who do you send the bill to? What if they just issue an order that the IETF do something, and then at the next IETF meeting (in London), they just ignore the order? Will the judge hold the IETF in contempt and order it arrested? Maybe its assets should be seized. (What assets does the IETF have? The only one I can think of, ironically, is the 'IETF' name.)
What law?
I understand that the US Congress authorized ICANN, but I don't recall that they ever passed anything that said everybody needed to pay attention to them.
The internet has always worked under rough consensus, where people generally agree on how things are going to work. The ICANN changes that model, really without consent.
So, two problems:
1. A 'billion million years' is effectively permanent, and would, I suggest be unconstitutional.
2. The dissenting judge has a point -- the Constitution grants the power to (A) "Promote the progress of science and useful arts", and (ONLY as a method of promoting science and useful arts) (B) to secure "for limited times to authors and inventors the exclusive right to their respective writings and discoveries." So, if Congress does (B) above in a manner inconsistant with (A) above, then it has acted unconstitutionally.
It's like I said: "I'm going to let you live in my house for the next year and I'll let you take responsibility for upkeep, by giving you access to my bank account." I come back a year later, and my house is in decent shape, but you've emptied my bank account and bought yourself a land rover. Under the majority's iterpretation, you'd say "Well, yeah, you gave me access to your bank account," but I gave you that access to take care of the house, not to buy yourself a car.
The Constitution allows the Congress to set laws instituting Patents and Copyrights "To promote the Progress of Science and useful Arts" "for limited times." How long do you believe copyright should reasonably exist on a work and what "fair-use" of copyrighted materials do you believe should be allowed? In addition, are there any types of work that are currently copyrightable which you either do not believe are worthy of copyright or believe should be treated differently in copyright law?
Obviously not, otherwise I could send you a copy of Microsoft office without any prior agreement between us and you would then be able to copy and distribute it at will.
The key point there is "obligation whatsoever *to the sender*." If Microsoft were to send me a copy of Office, then this regulation would allow me to do whatever I wanted to. However, there are other federal laws and regulations (ie copyright law) which prevent you from doing just that. I suspect that any reasonable court would find that copyright is still protected even in light of this postal regulation.
So, it seems to me that if the government wants to change how the US is defined, they need to submit a draft in the appropriate format to the appropriate IETF working group for consideration at the next meeting. The US government can't just bypass the procedures that have been set up for managing the internet. (Don't get me started about ICANN.)
.us doesn't need some changes, just that there's an established procedure for change and the commerce department can't just run rough-shod over that procedure.
Not saying that
As for the last question ("How did the fans find out [about the music]"), by-and-large, the answer is that record companies pay radio stations and MTV to play their music when it first comes out. In effect, they plant a seed of interest in the minds of consumers. Then, after a while, the music companies stop paying for the music to be played, instead demanding payment when it is. But, at that point, the consumer interest is already there. Then, the music companies use their distribution systems to get the same music out to the stores.
This method of advertising and distribution has been next to impossible for unsigned acts to accomplish. However, the internet & MP3s change all that -- suddenly, you don't need a massive distribution network, because the king of networks (the internet) does it for you. And, in a environment where people are well-connected, word-of-mouth becomes a much more effective method of getting known.
So, you can see why the major labels are worried: their entire raison d'etre is being challenged by the internet: If you don't need a record company to do your advertising, and you don't need a record company to get your music to consumers, what's left for them to do?
So, it seems to me that there are two ways of dealing with this:
1. Prevent your browser from accessing the offending site (doubleclick in this case...)
and...
2. Sending so many bogus requests to the site that any real data they collect would be totally obscured by the gazillions of bogus html requests they get in. (Hmm... an IP address associated with Senator so-n-so's office has visited the "Curing Impotence" site 5 million times in the past hour.)
How hard would it be to write a process that spoofed IP addresses on HTTP requests? You could even make it part of a virus -- imagine that, a virus that actually did something good.
--
C
I work for a large telecom company, and we typically sell networks consisting of many products that were developed independently, possibly including 3rd party equipment and/or equipment from companies where we just bought the company out-right.
So, we have the same problem on a smaller scale -- we may have 10 different implementations of the same protocol -- some in firmware, some in (ack) hardware, some in one language, some in another, some off-the-shelf, etc.... (We try to avoid having multiple people doing the same thing, but sometimes it just can't be helped.)
To enforce adherence to the standards: we write that requirement into contracts with third-party vendors; we have testers who pound on the different protocol implementations and generate a lot of negative press when something isn't up to standards; and we do a lot of interoperability testing -- if an implementation isn't doing what it's supposed to do, then the people who did it get a *lot* of heat put upon them to fix it.
A few thoughts come to mind....
1. I'm a software guy making more than twice the starting salary our company offers to kids right out of school.
2. A big chunk of the value of experience depends on the *industry* you're in. For example, I'm in the telecom industry, and it takes a good couple of years just to learn all the terminology and special requirements in a telecom environment. Companies like Cisco, Lucent & Nortel are continually raiding each other for experienced people.
3. Who really wants to be programming in 20 years? I'm 32 years old, and if I haven't advanced beyond programming, I'm going to be very frustrated.
4. Your value goes down once you graduate, not because of your salary, but because the industry is moving so quick -- if you aren't continually updating your skills, next years college graduate is going to know more than you do. They say that the half-life of an education is 5 years.
So, in the telephony industry, this is a big requirement... The problem is that you don't want the outage of a single network link to take your machine out of service, or to interrupt any existing transactions.
So, for example, if you have a TCP stream going to a specific NIC and the link between the NIC and the switch gets cut, or the NIC fails or something, then you need to be able to continue the same TCP stream on a second interface.
You end up with several issues: On a lot of NICs, it's not that easy to figure out when it's having problems. Secondly, the second NIC is typically at a different hardware address, so you need to update the ARP cache of any machine sending to you. And, you have to figure out how to tell when the first NIC is working again.
So... It seems to me that it's only illegal to distribute copies of a work to people who are not licensed for that work: I can take my favorite book to the photocopier, but I can't give the photocopy of the book to somebody who doesn't already own a copy of the book. (unless I also give that person the original and every other copy I've made.)
My understanding is that the my.mp3.com service verifies that the clients actually own the CD, so I don't see where the problem is. It sounds like the recording industry has successfully convinced a judge that fair use should be banned.
So, it seems to me that there's a actually a good argument in the text of the constitution to say that the Congress' retroactive extending of copyright is unconstitutional. Last year, in the "Sonny Bono Copyright Act," Congress extended copyright to 95(!!) years, primarily in response to lobbying from Disney and other media companies.
The problem is that extending copyright retroactively doesn't encourage the arts -- the work has already been produced, quite often by somebody who is already dead! And, since the constitution allows Congress to institute copyright to encourage the arts, retroactive copyright extension doesn't do it.
Important for things like project Gutenburg.
I wonder what their response is going to be when doing a web-search for "Leonardo" comes up with a bunch of "Leonardo Finance is a bunch of fascist bastards" websites.
So, a few things to ponder:
Article 1, Section 9, Clause 5 of the US Constitution:
"No Tax or Duty shall be laid on Articles exported from any State."
The Internet Tax Freedom Act, sect 1101(a):
"No State or poilitical subdivison thereof shall impose any of the following taxes during the period beginning on October 1, 1998 and ending 3 years after the date of the anactment of this Act --
[..] (2) multiple or discriminatory taxes on electronic commerce"
Internet Tax Freedom Act, section 1104 ("Definitions"), section 2:
Discriminatory tax -- The term "Discriminatory tax" means --
(A) Any tax imposed by a State or political subdivision thereof on electronic commerce that --
(i) is not generally imposed and legally collectible by [...] on transactions [...] accomplished through other means;
[...]
(iii) imposed an obligation to collect or pay the tax on a different person or tentity than in the case of transactions [...] accomplished through other means."
There are other problems: Purchasing items sent as gifts to people in other states, purchasing items that are not subject to state sales tax.
So, toward the beginning of the letter, the authors make the statement that having an "end-to-end" architecture (ie intelligence on the edge), with generic protocols in the network and applications running on clients is one of the strengths of the internet. Then, it goes on to say that ISPs can provide intelligence in the network, through things like web-page cacheing, and that you should be able to choose an ISP that does that.
This entire argument seems hokey to me. At most, it seems to me that AT&T should allow users to say "I don't want to use AT&T's web servers or usenet servers or mail servers. Please take $15 off my monthly bill." The user could then configure their software to point to somebody else's servers.
This repeatedly goes back to my point (made in this space before) that ISPs are an anachronism caused by trying to route a packet-based network over a switched network: You have to have gateways that convert from one to the other. This, currently, is the main function of ISPs. In a world where you don't need the gateway -- ie where the packet-based network extends all the way to the machines that want to see it (as is the case with packet-over-cable), then the main purpose of the ISP goes away.
Now, I would argue that if Mindspring & AOL want to stay relevant, they need to transform themselves into "ASPs" -- Application Service Providers. At the low end, this means running the same servers that the ISPs ran before -- mail, usenet, web, &c. But, there's a lot of opportunity beyond that -- PC backup services, video servers, network file storage and so on. These are the logical place for the ISPs to move into. But, they're so entrenched in what they've been doing that they can't see to move beyond it.
I keep trying to make this point:
The ISP provides the following services:
1. (*main purpose*) Provides the other end for your modem to connect to, establishing IP connectivity to your house, effectively extending the internet to the home.
2. Mail Server
3. Web Server space
4. Usenet setver
etc...
I would argue that cable modems, by themselves, do the job of #1, above -- they effectively extend the internet into the home. With cable, there is no set of corresponding cable modems on the other end that you're switched to, depending on what phone number you dialed. By attempting to force cable systems into the ISP model, you enormously drive up the cost. because the cable system would need to know "Ok. This house is a mindspring customer. Route his packets to the mindspring router." Just plain silly.
As for the other 3 things I listed above, I see no reason why they should be grouped together -- why can't I use Hotmail as my mail server, some other usenet server and buy server space from a third place.
So, I argue that if the government is going to mess around with the broadband market, it should prevent forced bundling of services -- The company proving access via cable modem should not be able to force people to also pay to use its mail servers, for example. (This is similar to how Local & Long Distance services are set up -- Sprint is my local carrier, but they don't force me to use Sprint as my LD carrier.)
Let's not confuse the "Red Hat Linux Distribution v6.0" (ie an ordered collection of 1's and 0's) with "Red Hat Linux v6.0," the product. Red Hat Linux, the product includes a CD, manual, box, assorted packing material and the right to obtain support under certain conditions. People who take a CD and burn the 1's and 0's onto it are not creating "Red Hat Linux" -- they're creating a CD containing the "Red Hat Linux Distribution." The two are *not* the same thing.
So, I actually went and *looked* at the mediametrix web site, and saw their list of the top-50 web sites. Note the following observations:
1. Sites #2, #3, & #4 are AOL, MSN and Geocities -- sites where individuals can put up their own web pages.
2. Out of the top 10 sites, 5 are primarily search engines (yahoo, go, lycos, excite, angelfire).
3. Another 2 of the top sites, netscape.com and microsoft.com are (I believe) the default start page for two very popular browsers.
4. The first e-commerce site is amazon.com, at #11 with barnesandnoble.com at #35. The second most-popular commerce site is ebay.com
5. Free-email sites, such as hotmail.com (and yahoo.com) are prominent in the top 50.
6. The first 'news' site is msnbc, at #24, followed by zdnet at #26, pathfinder at #28, wather.com at #30, cnn.com at #31.
7. There are a few unexpected entries -- passport.com (????), ivillage.com ? Lends to questioning how broad of a sample the study used.
So, I would argue that the following are more adequate conclusions:
** Many of the most popular web sites are used only to find other websites -- either as search engines or start pages.
** The sites that provide access for user webpages, such as aol, msn and geocities, are very popular -- meaning that lots of people are doing their own web pages.
So, I would argue that rather than a contraction in web sites, we're actually seeing an expansion -- sure the number of companies hosting these sites is pretty compact, but there's a lot of interest in what amateurs are putting on the net! AOL was able to beat the pants off Amazon, BarnesandNoble, ebay and expedia, COMBINED, just by allowing their users to publish their own web pages!
The fact that the search engines are so popular indicates that many people are visiting sites that aren't the big popular ones (those are typically already bookmarked, or obvious: cnn.com).
So, rather than saying that 'the web is contracting', with the connotation of usage on the web shifting toward a few sites, I'd argue that it's expanding because individual people are creating sites and other people are visiting those sites *more* than they are most of the 'mainline' sites.
My first encounter with the internet was in 1986, when I first went to college. And, the university's ISP was... Well, I suppose it was the university itself, but in reality, we had a "direct connection" to the internet. ISPs didn't exist because at the time 2400 baud modems were the fastest available, and 2400 baud just isn't fast enough for IP.
Fast forward a few years.... Modems get faster, and people start using them to make IP connections across phone lines. Still not incredibly fast, but still somewhat convenient. At this point, most IP addresses were still assigned to specific machines.
With invention of the World Wide Web, interest in the internet started to grow, and a cottage industry erupted around being able to provide IP access to people at home. About the same time, protocols became established to provide temporary IP addresses to a dialed-in machine. Voila! The ISP was born. And, as ISPs grew, they began to offer more services to their customers: e-mail, usenet news, web server space, and so on.
So, ISPs began as a solution to a problem: Nobody had the money for a direct internet connection to their home, but a lot of people wanted internet access. As a result, the primary purpose of an ISP is to provide a place where people can dial their modems and connect to the internet. The other big thing that ISPs do is provide a method for client machines to obtain IP addresses. All the other services ISPs provide: Mail server, news server, web space and so on are not intrinsic to the ISP itself -- it's perfectly possible and reasonable for a person to use one company to dial into the 'net, another company as their mail server, and a third company as a usenet server.
So, let's suppose for a second that it was possible and economically feasible for a person to have a "direct connection" to the internet from their home -- what does the ISP provide?? The only thing that I see that they still do is hand out IP addresses. And, for a short time, that would be useful. But, in the next few years, IPv6 is expected to take off, providing enough IP addresses for every man, woman and child to have several dozen. (I'm not sure of the exact ratio, but it's big.) In that case, there's no reason to borrow IP addresses at all, and the only remaining reason for ISPs to exist goes away.
I would argue that 'always on' internet access, via cable modem and DSL is, in fact, a direct connection to the internet. And, with that, nobody needs an "Internet Service Provider," because they already have internet service.
My opinions. Not my employer's.
The problem isn't really when telecomm companies buy each other -- in fact, I'd argue that because of economies of scale, that's a good thing: they're able to reduce costs while putting together a large network. The problem comes in when these companies don't have much competition. We still have AT&T, MCI/Worldcom, Level 3, Bell Atlantic/Nynex, Bell South, Ameritec, Sprint, GTE (well... didn't they just merge with somebody?) and a few dozen CLECs.
My guess is that you'll continue to see consolidation along the way, and we'll end up with a handful (3-5) of companies which will compete fiercely with each other.
The thing I think is the most bizarre about this is the sheer difference in size. According to the WSJ, Global Crossing has 148 employees -- they're a *START-UP* company that's trying to buy US West (~54,500 employees) and Frontier ( ~8100 employees).