Slashdot Mirror
Telcos Propose 2-Tier Internet
cshirky writes "Boston.com is reporting that 'AT&T Inc. and BellSouth Corp. are lobbying Capitol Hill for the right to create a two-tiered Internet, where the telecom carriers' own Internet services would be transmitted faster and more efficiently than those of their competitors.' The telcos basic fear, of course, is that the end to end design of the net (PDF version) will erode the telcos ability to use service charges to generate revenue for delivering video and voice; the proposed solution is to break end-to-end in order to protect pricing leverage over the users." We reported on this at the beginning of the month, when it was just speculation. Not any more.
For one thing, it would require a radical change in how the internet currently works. TCP/IP was designed around the whole idea of having no central routing (note, I didn't say naming) authority. This is one of the features which make it resilient to damage, since the network can adapt to nodes which suddenly might go dark.
This, after all, was the whole purpose of it, since ARPANET was intended to be resilient to enemy attack if parts of it were taken out.
Gregory Casamento
## Chief Maintainer for GNUstep
Because both Bellsouth and AT&T operate under a monopoly status granted by the federal government to provide local telephone service. That is why they have to ask Congress if they are going to change the terms of the service.
This is an old issue - Lessig has been writing about it since 2001.
The idea pops up every few months, but in the end, it is economic suicide for a market that already has an open, neutral standard to splinter into a set of closed, preferential standards.
In short, the competition between providers will reduce their profit below the current 'tacit agreement' point it is currently at, thanks to the neutral standard. This is especially true as long as they are not offering any additional value with their service, and only destroying the value of the current network effects.
The economically feasible solution is to price discriminate (as much as existing customers hate it, it does reduce deadweight loss and increase revenue). Simply, charge by bandwidth provided, and charge less for 'preferred' types of bandwidth, such as traffic internal to their network.
[Recommended Reading: The Innovator's Solution (which addresses closed vs. open standards) and anything about Nash-Bridges Equilibrium (which addresses tacit agreement among competing parties).]
If companies want to try to create supernets for their customers to better access each other, I say allow them to.
It isn't about them trying to create a supernet, it's about them breaking the current 'Net and inserting them selves between the end points. then they can prioritize traffic based on who coughs up the most money to them. No $$ = no access.
This isn't a suppliment to what has become, in essence, a Utility.
Unfortunately, with the current Administration's track record on pro-corporation, pro-Internet regulation, this proposal should look mighty juicy indeed. This will give them two things they always wanted. An easy wat to regulate/control the Internet, and more $$ for their friends.
See my post from last time this was discussed on slashdot:
t hreshold=1&commentsort=0&tid=95&mode=thread&pid=14 162317%2314165101rel=url2html-32438http://slashdot .org/comments.pl?sid=169910&threshold=1&commentsor t=0&tid=95&mode=thread&pid=14162317#14165101>
ahref=http://slashdot.org/comments.pl?sid=169910&
As long as they offer the same deals to everyone, without individualized contracts, they'd probably meet the nondiscrimination requirement of common carrier status.
Fedex does this, for example. Their volume discount is determined by formula, and can't be negotiated off of the formula. No matter who you are, you'll get the same deal based on your volume -- so Fedex can keep its CC status.
"Trolls they were, but filled with the evil will of their master: a fell race..." -- J.R.R. Tolkien on Olog-hai
Common Carrier status is granted when the provider doesn't filter content. However misguided this attempt is, it does not violate common carrier status, because they are not passing judgement or denying certain content. They're still allowing it all, albeit at different levels of service.
It already happened on November 21st of this year.
http://www.schwabpt.com/downloads/support/T_SBC_21 Nov05v3.pdf
"Important Information about the new AT&T Inc. The AT&T Corp. ("AT&T") and SBC Communications Inc. ("SBC") merger completed effective November 21, 2005. The newly formed company is known as AT&T Inc. Initially AT&T shares will be exchanged for SBC shares under the 'SBC' ticker symbol. On December 1st, 2005, the newly formed company will take back the symbol 'T'"
And making it law is what they are trying to do.
Good judgement comes from experience, and experience comes from bad judgement.
- W. Wriston, former Citibank CEO
The DOJ's budget was $23.4 billion dollars last year, as opposed to $21 billion in 2000. By the DOJ's AntiTrust division's own reports, its budget has gone up, even with respect to inflation: DOJ Budget Trend Data, Antitrust Division.
After the 1984 breakup, there was AT&T for long distance. And AT&T's competitors, mainly MCI and Sprint. And the seven "Baby Bells": Ameritech, Bell Atlantic, BellSouth, NYNEX, Pacific Telesys, Southwestern Bell, and U S WEST. Southwestern Bell changed their name to SBC, and along the way bought up Pacific Telesys and Ameritech. Bell Atlantic bought NYNEX, and then merged with GTE (a non-Bell-System provider of local phone service, serving roughly as many customers as any of the baby bells, but geographically dispersed) to become Verizon. Qwest bought U S WEST. And SBC's offer to purchase the remains of AT&T were recently approved.
AT&T was broken up because you only had one choice for long distance: AT&T. After the breakup, long distance was a competitive market, and now you can get Long Distance for a tenth of the cost it was before... and that's without inflation.
Absolutely untrue. The original telegraph companies had government-backed eminent domain powers. Further, they often relied on railroad landed (acquired through eminent domain). There were constant battles between the two; see, for instance, Western Union Tel Co v. Pennsylvania R Co, 195 U.S. 594 (1904), available at: http://caselaw.lp.findlaw.com/scripts/getcase.pl?
The Pennsylvania statute (mentioned in that ruling) granting eminent domain to the telegraph company was absolutely typical, and telegraph companies in the US relied on such mandates. Normally such power was granted to a single company, giving it a monopoly in the state or region.
rage, rage against the dying of the light
So the last ones standing are AT&T and SBC
Amoung the many things wrong with your post: AT&T was not one of the "last ones standing". AT&T was an empty shell, and was bought by SBC just for the name. TFA talks about "AT&T and BellSouth". BellSouth is not SBC. AT&T is SBC's new name, but isn't the old AT&T. The old AT&T is history.
The breakup of Ma Bell did nothing to offer any consumer more than one choice for local service. It was only about long distance, and the plan worked as far as that went. It also allowed the cell phone industry to emerge unhindered, and consumers have many choices there.
AT&T for long distance only, and Bell Labs became Lucent Technologies
More than a decade after the breakup of Ma Bell into the Baby Bells, AT&T decided on its own to spin off both Lucent (which got Bell Labs, later spinning off part of Bell Labs as BellCore, and spinning off Avaya which would hire away monay of the remaining Bell Labs people) and NCR (which they bought in 1991). Many years after that (2001) they voluntarily spun off AT&T Wireless. AT&T Broadband was spun off in 2002. Not quite the monopolist cospiracy theory you seem to be peddling. In fact, I can't even keep all of the spin-offs and their spin-offs straight without a diagram.
Socialism: a lie told by totalitarians and believed by fools.
This means that my customer is not getting the full 1.54mbps bandwidth their SLA guarantees, and by effect neither would I. This is {potentially} interference with interstate commerce and is also discriminatory in deciding whose traffic goes where, not to mention breach of contract (violating the SLA).
Wrong.
You'll get the full T1 from your termination point to theirs. That is ALL the SLA covers. You are not guaranteed any type on link to other networks at all. Never ever. Telcos don't guarantee service on their competitors networks.
What most people dont' realize is that the common carriers DO THIS ALREADY. The connection equipment of choice is ATM, and that supports QoS. Leased circuits were configured with QoS depending on what was paid for by the customer. As a field engineer at Lucent, it was explicitly explained to us "see that level there, marked 'no guarantee, best effort'? That is all the Internet traffic -- lowest priority there is."
However, all this is done at the network level and not the transport level. Major carriers routinely ran their own circuits high priority. Anyone else who paid for one, also got high priority circuits. Everone else got 'best effort' links. Links where they didn't control both endpoints, like to a competitor thru a peering agreement, were 'best effort'.
The fuss is not that the carriers are doing this, it is that they want to do this further up the stack. They want to become more than carriers and get into the realm of "content providers". Thus, not just provide the wires, but the stuff on the wires as well. This is where they run afoul of the existing laws.
In essence, they want to do QoS at the TCP level. Personally, I think that is fine by me as long as it is TARRIFED like services are now. If SBC wants to do it for SBC produced content, they have to charge that division the same as if it was a Google, Yahoo or NBC service. The "premium" costs the same no matter WHO you are.
I'd love to have end-to-end QoS available, even if at a premium.
-Charles
Learning HOW to think is more important than learning WHAT to think.
The article mostly talks about the telcos trying to offer Video services at a "premium" to subscribers, not much else. This way they can take a huge chunk of the next-gen Video-over-IP market, by having better quality videos (and most likely live video feeds) over their network. How they would stop video.google.com if Google decides to up the ante and offer H.264 quality videos is beyong me... that would probably be illegal.
So basically they are asking Congress to let them bundle a Video & Internet in one package sans the legal troubles. In my opinion, Verizon (a TELCO) is already doing something like that with their FiOS service (Broadband + Video), although in limited areas.
Mozilla stole tabs from NetCaptor. So what? Right?