Slashdot Mirror

That's a pretty simplified way of putting it, but basically correct. Major equipment vendors have been slow to adopt more advanced queuing strategies (Stochastic Fair Queuing integrated with some of the more advanced flavors of early discard.)

Right. The problem is not big buffers, per se. It's big dumb FIFO queues. There's nothing wrong with one big flow, like a file transfer, having a long latency, provided that other flows with less data in flight aren't stuck behind it. That's what "fair queuing" is all about. Each flow has its own queue, and the queues are serviced in a round-robin fashion. (With stochastic fair queuing, some hashing is done to eliminate some of the bookkeeping on flows, but the effect is roughly the same.)

I figured this out in the early 1980s (see RFC 970) and by the late 1990s, it was an established technology. We shouldn't be having this problem at this late date.

I wonder how much of the trouble comes from devices that are doing TCP-level processing in the middle of the network. Stateful firewalls and ISP ad-insertion engines can introduce substantial latency.

If you want to test for bad behavior, try running two flows, one that never has more than one packet outstanding, and one that just does a big file-transfer like operation like a download. If the latency of the low-traffic flow goes up to the same as that of the bulk flow, there's a big dumb buffer in the middle. If the packet loss rate of the low-traffic flow goes up, there's a small dumb buffer in the middle.

Circa 1999, this was the advice.

Geez. learn to Google.

http://www.isp-planet.com/technology/2001/ipv6_nowagain.html
"While some talk of the slow adoption of IPv6, at least one expert, who is at ground zero of the impending IP implosion, says the melt down of the Internet is already underway"
[August 7, 2001]

9+ years ago. Close enough. Sheesh.

How much of that cost is the cable?

http://www.isp-planet.com/business/fiber_price_bol.html

On land rural jobs cost about $15K/mile. On land super-urban jobs cost about $500K/mile. The difference is permits, corruption, kickbacks, etc. Also scaling is important, "one job in Montana" may be hundreds of miles, and "one job in Manhatten" may be measured in feet, but the fixed costs are... fixed... so the cost per mile seems higher on the short jobs.

If you assume underwater fiber costs around as much as the total cost of cheap rural route, the 6200 mile route times 16K/mile equals about $100M. That makes sense, since the whole job is only supposed to cost about $300M.

Repairing fiber is somewhat more difficult than laying fiber because it's time sensitive. But then again they probably charge by the hour anyway. Since a "several day" repair job approaches $10M, if you assume that is 4 days at $10M total, that would be about $2.5M per day. The little row boat they're using is going to take about 40 days to paddle across the pond, 40 days * $2.5M a day conveniently works out to about $100M. That makes sense, since the whole job is only supposed to cost about $300M.

Add in the usual admin overhead, several multimillion dollar executive bonuses, engineering work, station gear at each endpoint, marketing and sales upfront expenses including slashvertisements, booze, coke, etc, I think they could blow somewhat less than $100M on that.

My labor estimate is probably about right for overtime repair work and a bit high for contracted construction work. My estimate for overhead may be a bit high. That means the cost of the cable itself probably is about $125M to $150M.

a) Raise their prices considerably on all their "unlimited" plans--sucks for the light users, who are basically subsidizing the heavy users who want to stream HD video and movies

b) Covertly start throttling back heavy users--sucks for everyone, since no one even knows how much they're being throttled and there is no option of paying a premium to escape it

c) Set download caps--sucks compared to the "free ride" heavy users are getting now, but at least it's out in the open with no throttling bullshit (and light users don't get penalized).

d) Everything stays priced the same as now, without throttling or download caps

Want to see what the future will be like with the proposed capping system? Step right up folks, and take a look at Australia's largest ISP. You get to pick from unbeatable offers such as US$28.85 for 200MB, and US$93.86/month for 60GB! Want more than 60GB? No problem. For the low cost of just US$110.94 per additional gigabyte, you can download to your heart's content! Oh, what was that? You want to watch online video? Don't worry. As part of this attractive offer, you will also have exclusive unmetered access to our partner network of music, movies, sports, games, and more! Getting excited yet? Seriously though fellas, those were not typos and this is not a joke.

Out of every Slashdot article I have seen in the past year, no single controversy has posed anywhere near this of a threat to rights online or free and open source software; and we've got an almost inconceivable "+5, Insightful" first post that effectively sympathizes with the offenders. At least take a moment to research before rushing to Time Warner's defense. Believe you me, if they are given an inch on this one, they (and all U.S. ISPs) will take a mile.

"Why does this really matter? ISPs in other countries are doing it, and businesses should be allowed to maximize their profit," you might say. Well, for starters, internet access has become a vital lifeline that is second-to-none. It has superseded all other forms of communication and media. Restrictive bandwidth policies do nothing more than perpetuate the digital divide by putting financial strain on the people who are already on the brink. This means that when Johnny's parents have home (telephone, or) cable service with a major U.S. company that offers package deals, they will likely opt to conservatively use one of the most inexpensive service plans. At this point, experimenting with things as simple as Ubuntu and Folding@Home become impractical or impossible for Johnny, unless he really wants to go out on a limb by asking for permission.

As of 2008, 5 ISPs control 56% of the U.S. market share. This means that half of the country will be coerced into using the unmetered media networks offered by their provider. What happened to the vision of net neutrality?

Here's the bottom line: if Japan and South Korea can figure out a way to provide blazing speeds at a low cost, then so can the United States.

P.S. For those opposed to the proposals, please contact your elected officials, or request that it be done on your behalf.

For clarification, the FCC's ruling was not that comcast could not throttle BitTorrent traffic, but the method they used (packet injection -- causing connections to drop/terminate) was unlawful.

Thanks for the clarification. I didn't read the FCC's ruling, just the CNet article, which like a lot of /.'s articles, is mis-titled.

While YOU may want throttling to be illegal, it isn't.

No, I don't want throttling illegal. Or traffic shaping. What I want is the ISPs to hold up the way they billed their services. They sold unlimited access but now that they oversold the service they want to renege on it. I also want them to do what they were given taxpayer dollars to do, build out broadband. Telcos were given $200 billion to buildout broadband but all they did was use the money to pad their bottom line. I also want they to stop trying to block competition.

Falcon

I was as confused as you were by all this, since the article didn't say what the actual attack is. I eventually found something that explains it.
http://www.doxpara.com/DMK_BO2K8.ppt linked from http://www.isp-planet.com/equipment/2008/nominum+vantio.html

  See slide 17 in the presentation. But the trick is, your forged reply to a query for 83.foo.com is:

83.foo.com IN NS www.foo.com (83.foo.com is a subdomain, whose name server is www.foo.com)
www.foo.com IN A 6.6.6.6 (glue)

  So I guess I still agree with you, that BIND must be trusting more than it needs to. Caches could distrust any glue except when it has to. (I think the glue is only necessary when the name server is part of the domain it's serving. e.g. The glue would be needed if the name server was ns1.83.foo.com. Otherwise why not ask the .foo.com server for www.foo.com's A record, or use a cached one, instead of trusting the glue?)

  IIRC, djbdns is skeptical of glue. I remember reading a big rant about glue on DJB's web site years ago. I'm not sure if that's why it's not vulnerable, or if it's because it already did source-port randomization.

  Anyway, that presentation seems to cover a lot of what I wanted to know. In the worst case, you have a cache that trusts glue, and you can poison it by guessing a 16bit ID. In the even worse case, multiple requests for the same name leave the cache willing to accept more than 1 ID for a single response, leading to a birthday attack.

    The trick is to generate lots of DNS queries for names you choose when the server isn't run by idiots (accepting recursive queries from the whole Internet). Web log analyzers are one possible vector. Otherwise if you can feed HTML to something that will resolve names for IMG tags, or presumably javascript could go nuts...

The only thing ISPs care about by default is that their users aren't abusing access to the Internet through means like taking up too much bandwidth or doing illegal things.

What's optimal for [ISP] is not necessarily optimal or beneficial for an open and neutral internet.

When you read something like *"X works better than !X" the question you should immediately ask is this:
Works better for whom?

4 of the top 5 ISPs have what could only charitably called a conflict of interest between their tv/pay-per-view offerings and their internet offerings.

*X = a neutral internet

Why does this story sound familiar...right, because I've heard it twice before. In 2003 it was AOL who cut them off, then in 2005 Level 3 did the same thing.

While it seems Sprint is to blame here, when I see Cogent on the bad end of this so many times I can't help but wonder how many of these problems are brought on by their own management. It's not too often you get to see a pair of N/A results on the health report, but as you can read that's exactly what happened in 2005 as well.

Hell the fibre you want is already there in most places,
they just need to light it.

Over 90% of the fibre already in the ground is Dark Fibre.

http://en.wikipedia.org/wiki/Dark_fibre#Dark_fiber_overcapacity

It is all about the greed.

We paid 200 billion dollars in Tax money for the Bandwidth upgrade
already and we got boned.

http://www.isp-planet.com/politics/2006/broadband_scandal_book.html

The ppl running these companies are little more than leeches
on the ass of society, thieves in nice clothes.

Our best bet is Coop ISPs.

The rest of the game is rigged, and it always has been.

That is why Google decided to light their own network because
the Telecoms started whining about how they needed more of
Google's money.

If anyone can muster the willpower of all the techs nationwide
into an effort to bypass the bullshit, we might see Internet
Access like the have in Europe and Japan.

In the mean time, status quo will rule the day.

Hopes of Google's nationwide Wifi rollout have all but died.

I'd rather pay Google $50 a month for a low latency Wifi
connection that is 3 - 4 hops of the Backbone than these
Cable companies that have me going close to 30 hops
just to connect to a friend's server across town on a
different ISP.

Wow. I'm surprised AOL still has that many customers.

I'm not, though they were at 9.3 million at the end of 2007.

http://www.isp-planet.com/research/rankings/usa.html

They bought out Compuserve IIRC, which i'm sure is included in those numbers. In fact, the solution to Vista was to switch to AOL.

AOL has been around a long time. It's been well, a decade since I looked into it, but for number of access numbers they rivaled Compuserve, and compuserve was huge in the 1980s. If you were a world traveler, and needed to access your e-mail, AOL was a legit option.

Another AOL perk is keeping your old e-mail address. Ok, most ISPs will be happy to forward e-mail, or maintain a mail box for a nominal fee, but this certainly adds to their numbers.

Yet another factor is the fact that you can get an AOL & Cable bundle. Earthlink is rather the same, though I've only heard of this offering over Dish.

But likely the biggest factor was their flood of floppy disks and CD-roms to anyone and everyone.

according to CALEA, having a wifi router MAKES you an ISP. If you provide the 'last link' you're the ISP. As such you must provide the feds with real time wiretapping capabilities or suffer a 10K a day fine when they ask. You also can't disable the service if you realize the error of your ways when you get a tap request.

That would be my defense - excuse me Mr. Wifi provider, are you registered under CALEA? OOoh so I'm not guilty of theft, I'm actually guilty of helping the feds find someone who isn't a registered wifi provider

I'm AC on this because that law downright scares me.

I ran The Sync.com, an Internet video company that among other things helped to launch and hosted the Slashdot "Geeks in Space" audio webcasts. We had some angel money from folks involved in early ISPs (who did make lots of money). We started getting serious ad revenue from banner ad sites in 1998, but by the end of 1999 the banner market collapsed. In 2000, we were in talks for a few months to be purchased by a company in San Francisco. Tens of thousands of dollars of lawyer time into the contract process, they pulled out, we went under, and shortly afterwards they went under as well.

Towards the end of 2000, I ended up working at SkyCache/Cidera a satellite provider of USENET feeds and streaming media distribution. Unfortunately, after raising $75 million, they also had challenges, two layoffs with 50% staff cuts each time (one was originally scheduled for Sept. 11, 2001, but had to be postponed), and eventually went under.

So I left the Internet, and made the transition to broadcast television engineering (where it is all going IP anyway)...

That's really not fair. The cost of maintaining a network infrastructure is tiny compared with the cost of maintaining all of the nation's roads. The cost per mile of fiber is estimated at $16,000 per mile in rural areas. (Source: isp-planet.com) The cost of a two-lane road in Florida is $2.7 million dollars per mile. (Source: Florida DOT) Then, there's the maintenance. With fiber, you get to fix occasional cuts. With roads, you have to repave them every few years at a cost of almost another half million dollars for that same two-lane road. Then, there's the cost of buying the land for the road in the first place, whereas with fiber pulls, you don't have that initial cost since it is pulled along existing rights of way.

So the initial cost of roads is two orders of magnitude greater, and the maintenance costs are probably on average five or six orders of magnitude greater. Also, you only need one fiber to reach each house. This means that you don't need to be able to get from any house to any other house. You just need the main arteries and a single way to get to every house. That means that you only need to cover a road if you have at least one house that cannot be served by another road that is already served. This means that many short roads don't need poles or fiber pulls at all, but they still need asphalt.

Further, the cost of pulling the bundle doesn't depend that much on how many fibers you pull, so you pull extra infrastructure for decades of build-out. With roads, the cost is almost exactly proportional to the square footage covered, so you tend to build extra capacity on an as-needed basis.

I can't imagine the government doing a worse job than the telcos do now frankly, but the best solution would be to have the government buy it out and spin off a nonprofit to manage it similar to the way TVA works now. That's a generally well managed power generation and distribution infrastructure that provides cheap power to several states, all from a nonprofit that was created by the government and spun off.

http://www.isp-planet.com/business/2007/bandwidth_price_2008_bol.html

Bulk bandwidth can be bought at prices between $5 and $10 per mbit. 1mbit is approximatly 324GB (and that would be bidirectional, so double that when discussing residential bandwidth limits).

Sure, that is for bulk bandwidth at good locations. It is obviously more expensive to get it out to residential areas. But don't come with your bullshit that bandwidth is expensive.

Here's where your analogy fails: this commons isn't finite!

The money to buy bandwidth isn't either. Most people have no idea that Comcast has to buy bandwidth and think the answer is simply to upgrade equipment to handle more bandwidth. Buying the bits is the problem. There is a point where the bits used exceed many users contribution to the purchased bits.

http://en.wikipedia.org/wiki/Bandwidth_cap

Here is an idea of what ISP's pay for broadband.
http://isp-lists.isp-planet.com/isp-bandwidth/0608/msg00015.html
Buying a lot more bandwidth without gaining a lot more subscribers is not a money maker. Limiting the high bandwidth users enable your investment to serve more users.
And... $8/meg isn't really a super great price... I deal with an ISP
currently that has 700 meg committ to two different tier 1 carriers and
pays slightly less... This person is asking for 10GE with an unknown
commit but for 10GE it's usually at least 1 Gig for sure...

If you were an ISP and one user was using the same bandwidth as 200 other users combined, I'm sure you would be looking at options. Especially if the applications were spreading to the rest of the users so your number of high usage users grew like clockwork. It will be the time to decide to up the subscription cost 200X or limit high bandwidth use. Maybe just put the penalty on the non-profitable users so they will simply go elsewhere. Sucking a little water from the drinking fountain is fine. Connecting a fire pump and using the fountain for flood irrigation because the water is "Free and Unlimited" is not OK regardless of the free and unlimited claim. Home unlimited accounts were never sold for 24X7 saturated connections. When your ISP has to plumb for bigger pipes and buy the bits, they are getting sticker shock. They noticed the system is leaking large amounts of money to a few high demand users and the numbers of those users is multiplying. Taking no action is not an option, unless you consider simply canceling broadband service altogether as it becomes unprofitable.

First try and find out if there is a Wireless-ISP providing service in your area. You can check with people on the ISP-Wireless list:
http://isp-lists.isp-planet.com/isp-wireless/

You may have to be willing to pay for your own equipment to get the signal that you need through the trees to your house. The lower the frequency the better it will go through trees. You will also want to have directional antennas on both sides of the system (grid or Yagi for 900Mhz). There are a lot of options for equipment to do this with.

Canopy, Trango, and Tranzeo are just a few, Here is a link to one from Tranzeo:
http://www.tranzeo.com/products/radios/TR-902-Series

This one is from Trango, probably a good choice:
http://www.trangobroadband.com/wireless_products/m900s.shtml

You will want to get good antennas, here is a 15dBi 900Mhz grid:
http://wisp-router.com/itemdesc.asp?ic=GD9-15P-NF&eq=&Tp=
Wisp-Router can also sell you a coax jumper that you will need to connect the radio to the antenna.

You will want as much elevation as you can reasonably get to install the radio and antenna. Put the radios outside next to the antennas and run quality CAT5 to the inside for your connection.

Now you need to find someone to connect up to. Either an ISP or another person who you can get a broadband connection setup at their location and link to with wireless. Maybe you can get a second Internet connection installed at their location or increase their service level so you both can share it.

If you use the equipment above I think it is quite possible you will be able to get access. This depends on how far you have to go to find someone who has access and is willing to work with you of course.

Good luck!

Years ago, there was talk of structural separation. I wonder if this might re-open those discussions? If this had happened when the linked article was written (2002), I wonder if we'd even be talking on and on about net neutrality? I'd say not.

United States: 80 people per square mile.

South Korea: 1,274 people per square mile.

This is an old, tired argument. Sure, North Korea is more densely populated. The Netherlands too. But the density does not really say it all, it's just an average. My dad got top marks in his course in statistics, and he went on all his life pontificating about the "chicken average": I eat two chickens, you eat none, on average we ate one each". There are surely immense areas of the US without broadband (like Yellowstone park, say), but what about areas as dense as NYC?

The question is better put as: how many Americans live in high-density areas? Quite a few. The overall density is low because there is a damn half of the country that is uninhabited, and that's before counting in Alaska.

Salaries and marketing notwithstanding, the rest of those costs are all pretty much up front costs that the infrastructure provider pays once. You act as though they have to replace this stuff on an ongoing basis just for fun. The only time that stuff has to be replaced is when it is necessary to roll out a new technology (e.g. DSL), and even then, it can mostly be rolled out on an as-needed basis, upgrading a single rack of new hardware that will handle a few thousand customers, then waiting until it is mostly full of customers on the new service before upgrading another rack.

For the ongoing maintenance, don't forget that when you own as much gear as the telcos own, if something breaks and is discontinued, you just do a working pull from a unused card slot in a nearby cage and go on with your business. At such point as the hardware fails to the point that there aren't enough working boards, then you replace the equipment with new gear as it dies, but that's likely many, many years after the hardware is discontinued. Oh, and every time they do end up upgrading a rack of gear, all that old gear is now surplus parts that they can and will substitute in as parts go bad, so the effective cost of maintaining old gear actually drops as it gets decommissioned.

The reality is that the bulk of the wire infrastructure in most areas was put in decades ago and has more than paid for itself many times over. Thus, with the exception of having to wire new neighborhoods as cities expand, the cost of operating a telco is almost entirely profit, personnel and marketing costs notwithstanding. Adding DSL to an existing voice line also adds very little incremental cost. A DSLAM port costs on the order of $70-110 per port in the quantities purchased by telcos (source: isp-planet.com). At an average price of $30 per month, for a DSL connection, the hardware costs are covered within the first couple of months, and the cost of sending someone out to hook it up is covered shortly thereafter. Your customer equipment is paid for with the few months after that, and by the time you're at the one year mark, you've likely paid for all of the infrastructure and personnel costs associated with your account, customer service and marketing costs notwithstanding. From there on, beyond pennies per month for power to the DSLAM, the entire cost of your network connection is profit for the telcos---maybe not your local telco, since they have to pay money to an upstream provider for their bandwidth---but profit for telcos, nonetheless.

Distracted by HillBilly's reply, I misread your question and replied about a more general consumer advocacy group than you were probably asking about. But it appears that CR has covered ISPs.

Well according to this link we will start seeing 100 mbps downloads much sooner than that.
http://www.isp-planet.com/cplanet/tech/2006/prime_ letter_060703.html

I am writing to strongly discourage you from engaging in regulation or taxation of the internet. The internet has thrived and become a vital force in our economy thanks to its free, unregulated nature. Don't make the internet subject to the ultimate failure of other economic regulation as shown in so many other things that are regulated. The Soviet Union showed us that ordering the economy to work a certain way doesn't work. Please learn that lesson and don't regulate the internet with the so-called "net neutrality" bill. This regulation will only open the door for myriad other regulations on the method, usage, and structure of the internet.

Well here is the full list as of Q4 2005:

http://www.isp-planet.com/research/rankings/usa.ht ml

~S

Snort is open-source.... SourceFire makes money off the other things they've created to work with/around Snort...

Quoted from here

"Roesch sees Snort and Sourcefire as two different solutions aimed at distinctive markets. "The idea of Snort was to give people the best free, open source intrusion detection system we could, and we were pretty successful at that," he said. "The idea of Sourcefire is to say, 'Okay, we've got good intrusion detection technology: let's add everything else people need to use these systems effectively in large organizations.'"

And that's not to say that large organizations can't use Snort without the backing of Sourcefire. Roesch says some of the biggest companies in the world use Snort. Sourcefire just adds the manageability along with ease of use and deployment that many enterprise customers are looking for in an intrusion detection system.

Sourcefire's OpenSnort Sensors cost $9,995 each, and the OpenSnort Management Console costs $19,995. Various service contracts are available, ranging from a platinum level with around-the-clock support to a standard contract with per-incident support and e-mail discussion list access. Training on Sourcefire's products is also available. Training on IDS and forensic analysis in general is planned for the near future"

Also, the Federal Information Security Management Act might have a lot to do with this decision as well:

"The Federal Information Security Management Act (FISMA), Title III of the E-Government Act of 2002, outlines requirements to secure Federal information. Each Federal Agency, including contractors or other organizations who work with the agency, must develop, document, and implement an agency-wide information security program. Detailed guidance and recommendations are provided by the National Institute for Standards and Technology (NIST) encompassing all aspects of information security."

From 2001:

http://www.isp-planet.com/news/2001/messagelabs_01 1126.html

"SkyScan AP uses Image Composition Software (ICA), which decomposes an image," White explained. "It runs 22,000 algorithms and in addition to skin tone textures, it can decipher porn through other features such as facial expressions.""

In practice these tools are simply filtering by URL, then by colour gamut analysis.

Local wireline telecom has never *NOT* been a monopoly. Note that 'the new AT&T' is entirely unrelated to 'Bell Labs', which was renamed Lucent a long time ago. (And I think since merged with Ascend)

Also, it takes months to get a DSL line activated specifically *becuase* the ILEC now you dont have any choice but to wait for them. If there were true competition, then they might just have some motiviation to get on the ball.

Its about CHOICE, and service providers recognizing that customers have it, and motivation them to provide better service to keep them. When customers have no choice, the monopoly provider have no motiviation to provide good service.

When they broke up AT&T, they only market that was de-monopolized was the long distance market (which brought great benefits to the market, LD is now damn cheap or often free).

Unfortunately, all they did in regards to the local wireline was create a number of smaller geographic monopolies. The 'ILEC' baby-bell companies still had (and still HAVE) a complete stranglehold over wired last-mile. That a hard-to-enforce law said they had to 'share' access to it, they did so as grudgingly as possible, stalling and stumbling anywhere they could. The claimed the new startups should have to 'build their own network', despite that doing so would cost billions and take decades to do (which as it happened, is how the ILEC's built the existing one, with not a little govt-guaranteed captive customer base, exclusve access to right of way, etc). Now they have brushed even that away in most markets, and I find it fitting that the new monopoly bears the 'mark of the beast' of the old one. I only hope that they meet with the same fate, only done correctly this time (see: http://www.isp-planet.com/politics/2002/structural _separation.html )

First off, owning an mp3 is not illegal as long as you have the rights to own it (i.e. you legally own the compact disc, cassette or record of it).

If you really want me to name people charged with illegally acquiring mp3s:
Diana Li
Daniel Peng
Joe Nievelt
... etc. ...

The list is a long one, I hope three names will suffice, if not, you know where Google is ...

I understand what you're saying, but it serves no purpose in this conversation.

Then again, we have no real data on how they intend to work this new business model. We just know they're greedy enough to want to do it.

BTW: thanks for the BGP and AS Path-length idea. I'm off to read now after some googling :)

ISP-Planet's most recent rankings of U.S. ISP market share show AOL still way out in front of everyone else, at 20.8 million U.S. subscribers. Say that Google decides to use AOL's backbone, ATDN, for at least some of the traffic to and from Google servers. Assume also that SBC, BellSouth, or whoever - let's call such a company "Jerkweed" - begins to charge traffic to and from Google at a higher rate than anywhere else. What are Jerkweed's options?

1) Have some sort of metering at Jerkweed<=>ATDN transit points. That's a shitload of traffic to meter. I mean, you have Google traffic, then you have traffic from all those AOL subs. Does Jerkweed count individual packets bound for Google, or does it just say fuckit and count them all, even though a lot of those packets will be for AOL subs, not Google? Either way, it costs money to do that sort of metering, and it'll probably slow down the transit. If people complain, ATDN can point the finger squarely at Jerkweed. Winner: ATDN.

2) Rewrite the peering agreement between Jerkweed and ATDN to charge higher rates. Well then, you have contract negotiations, during which ATDN will say to Jerkweed, "OK, we will charge you more to access our 20.8 million AOL subscribers in the U.S. Oh, and don't forget our even larger number of worldwide subscribers, either." Winner: ATDN.

So, with this move, Google is bringing the noise, both vs. MS and vs. Jerkweeds everywhere. The message is clear: "If you try to pin us down, you'll only be hurting yourself." Turnabout, after all, is fair play.

Yes, the ILEC's *do* still have monopolies in most areas of the US, since they still have exlucive control over the buried cables in the ground. Since that *is* a natural monopoly, as long as one company controls that which has incentive to not allow other companies access to it, they always will. CLEC's should not be required to rewire every block in the US when it has already been done, and at taxpayer and captive-customer expense.

The only solution that will ever work is:

http://isp-planet.com/politics/2002/structural_sep aration.html

So pissed I didn't check my links.

Server logs: 1, 2, 3

ISP logs: 1, 2, 3 related directly to P2P apps

System logs: A textbook on the subject you might like to read, Explanation of how to read a system log

This is a wet dream for individuals and you are griping about it. This is exactly what I'd love to see in the US - the Incumbent bells seperated from their monopoly control over the existing wired infrastructure that was paid for by captive customers under their guaranteed monopoly status, so that it is available on an equal basis to many service providers (of dialtone, internet, whatever), who can them compete on even footings. Unfortunately, it seems like the US FCC is going the other way here, which will kill off any chance competition had before it even got off the ground.

http://isp-planet.com/politics/2002/structural_sep aration.html

The breakup of AT&T was only effective for long distance - and look how cheap that is now, if you shop around. Unfortunately, that breakup left intact lots of smaller monopolies, who still have captive customers with no real choice. As far as territory wars - I have never seen a single shread of evidence that any ILEC ever competed with another ILEC in the same area - try getting service from Verizon while living in an SBC area, or vice versa (*wired* service, that is, I understand competition is thriving in the wireless market)

The only way real competition (where companies have to actually do their best to keep customers) for local wired phone service would ever flourish would be exactly what Australia us doing - seperate the monopoly telco from the wires in the ground and heavily regulate the organization that has the wires, forcing them to be fair to all comers, allowing the wires to be used by any provider of service without letting the monopoly company gouge them (if they let them use it at all)

http://isp-planet.com/politics/2002/structural_sep aration.html

If you truly are a consumer of phone service (and not a shill for an incumbent) then you truly dont understand whats holding back better service and prices.

As any fule kno, the most notorious spam blacklist is SPEWS.

ORBS, and its later reincarnation, ORBZ, also weren't exactly the nicest players on the field. I remember one incident where I couldn't send email to someone from a GMX account, because GMX - a webmail provider not unlike Hotmail etc., with several million users - had ended up on their blacklists (I'm not sure anymore whether it was ORBS or ORBZ at the point that happened, but it matters little, anyway).

This articleon the death of ORBZ has some more interesting points regarding the controversy surrounding these lists.

That might actually be a fantastic idea, it would be fantastic to use the MAC to seed the WEP keys.

The MAC address sits outside of part that's encrypted, so it doesn't really work for that. Fire up a copy of tcpdump, snort, or etherial and look at some packets coming across a wireless connection. They're trivial to find out and trivial to spoof.

Here's a visual analysis of a packet.

The Fed is actually working to try and help get broadband into rural areas as well. You can find some basic info on it here. Basically... the Department of Agriculture, the same department which helped fund getting electricity into Rural areas, has earmarked about $2bil for helping ISP's offer high-speed internet into those same rural areas.

From the two articles, i don't get everything.

First, why do you need IP for TV ? over ADSL, it's a lot better to send it in raw ATM. Of course, you can use IP to broadcast TV to the DSLAMs. And if it's IPTV to play TV on the computer, what exactly is the use ? Isn't it better to get it directly on TV ?

Then, why do you need Microsoft for that ? Are these Bells not using MPEG2 or MPEG4 for TV ?

And, 1mln users in Europe for TV over ADSL ? It's very very low ! There are about 700 000-800 000 only in France: France #2 ISP provides TV over ADSL as part of their triple play solution, and they have reached 600 000 people subscribed to the triple play offer.
Also, their triple play offer, and especially the freebox, is running Linux, like most of their whole architecture, so how exactly the Bells' choice is a coup for Microsoft: thet are entering a market very late. In France, all of it has already been taken, with the 3 major ISPs already offering TV over ADSL. And I can't see how Italy could top that, with their currently expensive ADSL.
Moreover, they're already working on providing HDTV through their triple play offering.

So i think either i missed the point, or both articles are (at least partially) wrong: some other people explain here that several ISPs are also offering triple play offers in the US or Canada. Can someone explain me ?

I guess once they got their high-speed net to all the city buildings and schools, their interest pretty much fizzled, leaving the city-zens still not quite on of the game... I still can't get DSL.

Using this example it should be easy to see that this reaction from telecom (i.e. Bell) is natural survival instinct. They _will_ kick and scream louder and louder until they get their way. POTS *may* be able to support higher then rolled out data speeds but with fiber being rolled out is it worth it? Or are they just trying to make it look like they don't really need fibre optics so they don't look desperate.

Examples of simple Application Service Providers (ASPs) even you may use today:
- Virtual webhosting
- E-mail providors
- Online Fax providers (delivered to e-mail type)

This has been talked about for years now (practically as long as I remember the internet being around, maybe before). From an artical written in 1999, "IBM rolled out a series of new hosted business applications that support critical accounting, human resource and sales automation services for small and midsize businesses."

Really, it's just an old idea made new . In 1999 they were asking if it would work. however it sorta went by the side when it came to the sort of all-in-one solution we are talking about now due to the lack of high speed connections for enough people. There are however lots of firewall providors, managed VPNs, managed intrusion detection services, managed anti-virus and content filtering, managed vulnerability assessment and emergency response, and some that to a bit of all of the managed security.

If you want to read more about HOW they are talking about doing it here is a link to a few white papers to puruse on ISPs providing these services.

Examples of simple Application Service Providers (ASPs) even you may use today:
- Virtual webhosting
- E-mail providors
- Online Fax providers (delivered to e-mail type)

This has been talked about for years now (practically as long as I remember the internet being around, maybe before). From an artical written in 1999, "IBM rolled out a series of new hosted business applications that support critical accounting, human resource and sales automation services for small and midsize businesses."

Really, it's just an old idea made new . In 1999 they were asking if it would work. however it sorta went by the side when it came to the sort of all-in-one solution we are talking about now due to the lack of high speed connections for enough people. There are however lots of firewall providors, managed VPNs, managed intrusion detection services, managed anti-virus and content filtering, managed vulnerability assessment and emergency response, and some that to a bit of all of the managed security.

If you want to read more about HOW they are talking about doing it here is a link to a few white papers to puruse on ISPs providing these services.

Examples of simple Application Service Providers (ASPs) even you may use today:
- Virtual webhosting
- E-mail providors
- Online Fax providers (delivered to e-mail type)

This has been talked about for years now (practically as long as I remember the internet being around, maybe before). From an artical written in 1999, "IBM rolled out a series of new hosted business applications that support critical accounting, human resource and sales automation services for small and midsize businesses."

Really, it's just an old idea made new . In 1999 they were asking if it would work. however it sorta went by the side when it came to the sort of all-in-one solution we are talking about now due to the lack of high speed connections for enough people. There are however lots of firewall providors, managed VPNs, managed intrusion detection services, managed anti-virus and content filtering, managed vulnerability assessment and emergency response, and some that to a bit of all of the managed security.

If you want to read more about HOW they are talking about doing it here is a link to a few white papers to puruse on ISPs providing these services.

Examples of simple Application Service Providers (ASPs) even you may use today:
- Virtual webhosting
- E-mail providors
- Online Fax providers (delivered to e-mail type)

This has been talked about for years now (practically as long as I remember the internet being around, maybe before). From an artical written in 1999, "IBM rolled out a series of new hosted business applications that support critical accounting, human resource and sales automation services for small and midsize businesses."

Really, it's just an old idea made new . In 1999 they were asking if it would work. however it sorta went by the side when it came to the sort of all-in-one solution we are talking about now due to the lack of high speed connections for enough people. There are however lots of firewall providors, managed VPNs, managed intrusion detection services, managed anti-virus and content filtering, managed vulnerability assessment and emergency response, and some that to a bit of all of the managed security.

If you want to read more about HOW they are talking about doing it here is a link to a few white papers to puruse on ISPs providing these services.

Examples of simple Application Service Providers (ASPs) even you may use today:
- Virtual webhosting
- E-mail providors
- Online Fax providers (delivered to e-mail type)

This has been talked about for years now (practically as long as I remember the internet being around, maybe before). From an artical written in 1999, "IBM rolled out a series of new hosted business applications that support critical accounting, human resource and sales automation services for small and midsize businesses."

Really, it's just an old idea made new . In 1999 they were asking if it would work. however it sorta went by the side when it came to the sort of all-in-one solution we are talking about now due to the lack of high speed connections for enough people. There are however lots of firewall providors, managed VPNs, managed intrusion detection services, managed anti-virus and content filtering, managed vulnerability assessment and emergency response, and some that to a bit of all of the managed security.

If you want to read more about HOW they are talking about doing it here is a link to a few white papers to puruse on ISPs providing these services.

Examples of simple Application Service Providers (ASPs) even you may use today:
- Virtual webhosting
- E-mail providors
- Online Fax providers (delivered to e-mail type)

This has been talked about for years now (practically as long as I remember the internet being around, maybe before). From an artical written in 1999, "IBM rolled out a series of new hosted business applications that support critical accounting, human resource and sales automation services for small and midsize businesses."

Really, it's just an old idea made new . In 1999 they were asking if it would work. however it sorta went by the side when it came to the sort of all-in-one solution we are talking about now due to the lack of high speed connections for enough people. There are however lots of firewall providors, managed VPNs, managed intrusion detection services, managed anti-virus and content filtering, managed vulnerability assessment and emergency response, and some that to a bit of all of the managed security.

If you want to read more about HOW they are talking about doing it here is a link to a few white papers to puruse on ISPs providing these services.

Good ideas are not adopted automatically. They must be driven into practice with courageous patience.
-- Admiral Hyman G. Rickover

Google it like I did if you don't believe me. Tabloids can hide behind free speech most of the time, you and I cannot.

Bully for SBC if they can see this push for fiber through to the end, but their history has shown a decided lack of resolve in this area in the past.