Posts by Richard Bennett

America may end up with paid-for 5G fast lanes under net neutrality anyway

BVS still doesn't understand network performance

The big tell that NN bigots don't understand networking is their obsession with "speed" (bandwidth or throughput as I understand it) to the exclusion of latency, consistency, and reliability. It's legal for my ISP to offer me a 250 Mbps plan and a 1000 Mbps plan, at my discretion. For some apps I'd rather have 250 with low latency and high latency consistency (AKA "low jitter") and for others I want the high throughput with wildly varying latency.

Why can't my ISP and I agree on more than one service level per account?

My router is happy to mark QoS for me, all I need is for the ISP to honor my markings.

Gigabit? More like, you can gigabet the US will fall behind on super-fast broadband access

This story is nonsense and should taken down. The US already has gigabit coverage to a majority of the nation's homes, but few want it because web sites and video streaming services can't keep up with speeds higher than 20 Mbps.

Net neutrality freaks furious over lack of fury at FCC hearing

Karl Bode is a waste of time

If you've read one Karl Bode article you've read them all. Try it and see for yourself.

Google lobbies hard to derail new US privacy laws – using dodgy stats

Facebook is full of frauds and scams

Wow, Kieren, do you honestly believe frauds and scams are absent from Facebook? That is so, so wrong. Here are a few of the most notorious:

1. Joe Mercola, the man who paid the largest fine in history to the FTC for making false health claims, has 1.7 million likes for his FB page.

2. David Avocado Wolf, the notorious alternative health scammer, has 12 million likes.

3. Food Babe, food scammer, has 1.2 million.

4. Vaccine Re-Education Discussion Group - an anti-vax fraud - has 102K members.

5. Stop Mandatory Vaccination has 123K members.

And there are tons of groups claiming climate change is a hoax, alternative medicine works, and similar garbage.

FCC drops idiotic plans to downgrade entire nation's internet speeds

Re: Yes and no

You might want to look at Ofcom's definitions.

The UK's FCC, Ofcom, defines Super-Fast Broadband as 30 Mbps down, while the UK government defines it as 24Mbps. Ofcom defines "decent broadband" at 10/1/. (CONNECTED NATIONS 2017 report, 15 December 2017).

According to Kieren McCarthy, Ofcom is idiotic.

Funnily enough, small-town broadband cheaper than big cable packages, say Harvard eggheads

The data is more telling than the analysis. On average, the 27 community networks were $10/month cheaper but 7 Mbps slower than their commercial rivals. And community networks do offer teaser rates; Longmont, CO, for example, offers a lifetime low rate to early adopters.

See http://hightechforum.org for a look at the data.

It's official: Ejit – sorry – Ajit Pai is new FCC boss (he's the one who hates network neutrality)

Oh my, we seem to be a bit overexcited by the expected pick. Pai has been in the communications game for 15 years and at the FCC as a commissioner for four. He does have a reputation for working across the aisle at the Commission, although that is something that Mr. Bombast Tom Wheeler didn't appreciate. In one example, Pai and O'Riely made a deal with Commissioner Clyburn to expand universal service for broadband and raise the cap on contributions. Mr. Psycho Wheeler held up the meeting where the vote was to be taken by four hours so that members of Congress, including her dad, could twist her arm so as not to make Captain Google Wheeler look bad.

You can look it up.

Similarly, blaming Pai for Chmn. Puppet Wheeler's partisanship is like blaming hooker victims for making Jack the Ripper want to gut them.

Trump hires very best, greatest net neut haters to head FCC transition

Factual errors

Neither Mark Jamison nor Jeff Eisenach has ever been a lobbyist, and Jeff does not do "a lot of corporate work for Verizon". A NYT reporter declared Jeff a Verizon lobbyist because he wrote a paper for GSMA, a group that counts Verizon among its membership, but that's stretching the truth.

Kindly correct your copy.

PS: I'm a former co-worker of Mark and Jeff and a former Register writer.

Net neutrality in the US: Look out! It's Neut-gate! Or is that Net-ghazi?

The correct term is "ObamaNet".

BitTorrent net meltdown delayed

@Orclev

"The fact that some rudimentary congestion control was tacked on after the fact due to a flaw in the design of the internet does not in any way imply that traffic shaping is essential to the internet"

Oh?

So the congestion control code that was hacked into TCP by Van Jacobson back in 1987 was just for the fun of it and has no operational value to the Internet at all, just as the original method of signalling congestion by IP-level error message was just for kicks?

These Internet protocols are so full of jokes, you never know when you've found one.

@AC Persistence of Memory

"Except last week, when you ranted and raved that bypassing TCP CC would cause the immediate death of the 'net, film at 11. HTH."

Whether a new scheme improves on the current, diverse implementations of TCP CC or degrades it depends completely on how it's done. Field reports from uTP indicate that it bypasses traffic shaping, which is great for the individual pirate and not so great for the VoIP or gaming user. While BT, Inc. say this is an accident, that's really beside the point.

The fact that BT, Inc. has abandoned TCP because it doesn't effectively manage congestion is a very telling admission, because it makes the argument that traffic shaping is an essential part of the Internet and not just an annoying means of monopoly rent-seeking.

@Jason King

Compound TCP is *one example* of a TCP implementation that has a latency-driven congestion avoidance state, and Vegas and Cubic are others. The point is that the alleged innovation of uTP is nothing new.

The modern, manageable local area network is Ethernet over twisted pair and Wi-Fi. And yes, I've had dinner at Bob Metcalfe's house, not that it's particularly relevant to the current discussion.

Before spouting off about petty details that fall outside the scope of your comprehension try to develop some analytical skills.

@Laird Popkin

Laird says: "...while some have speculated that this could lead to "net meltdown" the intention of the LEDBAT effort is actually the opposite - to move bulk data over a distinct protocol that allows applications to detect congestion and "back off" so that more time sensitive data can take higher priority, and so that congested routers can make more intelligent decisions than they can with TCP. This should ultimately be great for all involved."

Effects don't flow directly from intentions, they come from actions. I support the LEDBAT and ALTO efforts, as well as related efforts down through the years such as IntServ, RSVP, DiffServ, ECN, and Re-ECN. But we need a whole lot more data and wider experiments that compare differing approaches before we declare any of these schemes successful.

I believe the principal flaw in uTP is going to turn out to be its exclusive focus on end-point measurement. There's not enough information there to make the decisions it wants to make, and a better approach would allow routers and shapers to communicate with the path selection process, as ALTO does. Not that ALTO is perfect either, mind you; it's a bit too piracy-friendly for my taste.

But as I've said in The Register for more than a year, TCP doesn't do the job for P2P and we need an alternative. Whether that's a combination of existing standards or whole new thing remains to be seen.

Google - the world's first firewalled monopoly

Excellent article

You deserve a Pulitzer for this one, Cade, very nicely done.

Why the US faces broadband price hikes

Robb Topolski's correction

Robb wants me to mention that he's stopped using Comcast for his High-Speed Internet accessing needs. He was happy with the service, but he didn't like this comment in one of their FCC filings:

"The vast majority of the assertions Free Press makes are based entirely on the unsupported allegations of a single person: their consultant, Robert Topolski. Mr. Topolski, to whom Free Press refers as an “expert,” is not an engineer, has no apparent experience in designing or managing a network (his reported certifications and experience are in quality assurance and software testing), and has no first-hand knowledge of Comcast’s network. See Robert Topolski, HIRE ME! (My Resume), at http://www.funchords.com/Robert_Topolski-resume08-SQA_Testing.pdf (last visited July 21, 2008)."

I can't say that Comcast's characterization is wrong, but it hurt Robb's feelings.

Meanwhile, Uber-Neut Susan Crawford declares that she too is a Comcast customer: http://scrawford.net/blog/planning-for-the-future/1242/

@Andrew Norton

I was an invited witness at the first FCC hearing on broadband management, and went there on my own dime. I wrote about it for El Reg so it's hardly a secret. Does the fact that the FCC asked for my opinion disqualify me from holding opinions?

I actually work for a company that supplies home routers to Comcast's competitors, but I'd be happy to take Comcast's money if it were on offer. Or AT&T's, or BitTorrent's, or anyone else's. As Jess Unruh said of lobbyists and elected officials, "if you can't take their money, drink their whiskey, screw their women and still vote against them, you're in the wrong business."

Google unfurls less laughable Wikipedia

Google knows

@George Treasury Secretary:

Google knows if you're using a fake name, for it sees all and knols all.

No entries for Google or Google's hobby, "net neutrality". Can't wait to see the inevitable bias.

FCC boss reshapes hardline stance on Comcast

It's laying groundwork

The FCC has to establish that it has the authority to issue fines in the future if some company steps out of line and refuses to change its ways. Fines would be non-productive in this case because Comcast has already committed to change out their traffic management system to one that works with a bit more precision. So they're already paying a fine in effect, but to the manufacturer of the new system rather than to the FCC. And that's a good thing, actually.

FCC chief wants to throttle Comcast

Dark day for pirates

The Comcast complaint originated among the video pirates and the resolution isn't going to help them any. They're still going to find their file-sharing activities throttled.

It's an interesting tack for the FCC to make - rattle the saber and assert authority, but not levy any actual fines or punishments. I suspect it's intended to allow both sides to declare victory.

NebuAd makes meal of opt-out cookie

The Senate hearing today was interesting

It turns out NebuAd has been anonymizing its data all along. I don't know if it makes a difference legally, but it does change the aesthetics of the system. They make a one-way hash out of IP addresses, and pass anonymous information to advertisers tied to the hash.

I'm sure they're doing something totally illegal, but there's a certain evil cleverness to the system that I admire (no, this is not an endorsement.)

I don't know why the movie pirates are so torqued about NebuAd, as it doesn't seem to affect them.

P2P site cries traffic shaping foul at Canuck ISP

Total Hysteria

Traffic shaping has been an operational imperative on IP networks from the beginning. See RFC 896 (http://www.faqs.org/rfcs/rfc896.html) from 1984. Everybody does it, and always has.

Next thing we'll be hearing that IP networks packetize data to promote capitalism.

And BTW, Robb Toploski is not an "independent researcher," he's an unemployed sofware tester and fan of piracy tools; few people, if any, use Gnutella and EDonkey for legal downloads.

We're still waiting for him to make his packet captures public. It's one thing to make allegations, and another to back them up with hard data.

Everyone's a winner in the Comcast - BitTorrent detente

@AC

Comcast will have to install some new equipment to apply per-user priority quotas enforced by the CMTS. This scheme wasn't actually suggested by EFF as you claim; they asked for across-the-board floating bandwidth caps in the cable modems and metered pricing.

The big change is new equipment that enables control of the CMTS scheduler, something Comcast couldn't do before. And in fact I proposed a system of this sort in the talk I gave at the ITIF on March 12th, so you can make of that what you will.

Interesting comments

@Robb Topolski: I suggest you read the Nagle RFCs I mentioned in the article for an understanding of how fair queuing and weighted fair queuing work and their status in Internet operation for these past 20+ years. Most networking guys have this stuff cold, but if you ever return to SQA with Intel in Hillsboro you might ask some of the Digital Home guys how their Wi-Fi WMM system prioritizes gaming and video streams over web surfing. I can't buy the logic that "company x is evil, therefore all they do is evil." If it looks like a good deal, it is a good deal, and if it looks like a rook it is a rook. I'm easy to please.

@AC: How was I wrong exactly? I said that I couldn't figure out a better way than RSTs to handle the traffic load given the equipment limitations Comcast had pending the DOCSIS 3.0 upgrade, but they gave themselves a better option by installing some gear that handles traffic in-line with the CMTS. This stuff is expensive, but they're willing to shoulder the expense on your behalf. That means that Comcast is better at running their network than I am, and I more or less expect that.

@b shubin: Corporations are rewarded for rational behavior by making sales, and that's what makes the world go round. Comcast certainly isn't in the same game as Jobs and crew, they're just a simple little TV company. I appreciate them because they're willing to sell me a broadband connection I can abuse to my heart's content, while AT&T doesn't even want to offer me a DSL option. Nothing that they're doing involves switches vs. hubs, BTW, because they have a single-cable network. You've apparently missed the signature difference between DSL and cable.

--

How long until we get the usual caped crusader saying "they sell me X bandwidth and I should be able to use all of it 24x7?" I always appreciate that one.

Comcast admits it can do the impossible

More confused than ever

Let's try to get a few things straight, one last time. In the first place, Comcast and BitTorrent, Inc. have reached an agreement, not an "agreement" with scare quotes. You may not like it, Cade, but it is what it is.

In the second place, not a single person on the planet has ever argued at any time that TCP RSTs were the only way to manage a cable network overloaded with traffic. It may be the best way, or the quickest way, or the most reasonable way, or the cheapest way, but no one has ever claimed it's the only way. So your title is ridiculous.

And finally, nobody in their right mind wants to see a hard cap on upstream traffic rates in a cable modem network that's suitable for the overload case in which everybody is seeding BitTorrent, and that isn't what Comcast is implementing. What they're doing is using the power of the CMTS scheduler to control the rates at which level 2 frames enter the cable channel. Briscoe wasn't aware that this was an option.

Robb Topolski proposed this method in a comment to one of your previous National Enquirer-inspired scare pieces on the wickedness of Comcast's capitalistic, American ways, so it must be whole grain and organic, if not utterly vegan. It's interesting that he's now flip-flopped and is calling it the end of the world. Pretty much like you are, Cade.

Funny how that works.

Cool Rules for the FCC: In the Lions's Den

@Charles Swiger

The point is that *on average* other apps use much fewer connections than BT, largely because they all operate in bursts of single operations, while BT runs many connections in parallel persistently. Because BT's connections are persistent, they're not in slow start and have a higher bandwidth quota *on aggregate* before and after any single packet drop against any one of them. That's how BT beats the system.

So the only way to deal with this from a system fairness point of view is to police BT streams specifically.

BitTorrent busts Comcast BitTorrent busting

@Robb Topolski

OK, Robb, let's be very clear about what we're saying. You seem to believe that it's impossible to seed on the Comcast network, and therefore BitTorrent downloaders are severely affected across the whole Internet. If this were true, we'd have reports from all over about how Vuze and BitTorrent, Inc. were having trouble delivering HDTV and movies to their customers, but we don't. To the contrary, Vuze said in their FCC petition that they haven't been affected by Comcast in any way.

And moreover, Linux distros are still moving quite nicely across the net, even inside of Comcast. My last Fedora 8 download ran at 4 Mb/s the whole way to completion. So your argument is obviously lacking substance.

In fact, there's no relationship between your MIDI seeding and the kinds of transactions that are typical of legal uses of BitTorrent. You've had trouble seeding because you don't have enough of a swarm going to reach the critical mass that BT needs to run really fast. And in fact, BT is not even the appropriate tool for the job you're trying to do. You'd have much better luck putting your music files on your free Comcast web space and letting people download them with HTTP.

So why don't you do that?

@Robb Topolski

One thing you should clarify, Robb, is that the only trackers that care about seeding/leeching ratios are private trackers used predominately by pirates. Legitimate downloads, like Linux distros and licensed content from BitTorrent, Inc., don't care about how much seeding you do. You shouldn't have misled poor Cade about that, he's a trusting soul who admires and respects you. So be honest with him.

I did another experiment on Comcast with BitTorrent last night, downloading a talking book version of the King James Bible I found in Mininova. After the download completed, I seeded quite successfully all night long. I started this around 8:00 PM and the pure seeding started at around Midnight and ran at 160 Kb/s.. I had no problems at all, and it's on my OGG player right now.

I've attacked you personally on this issue because I don't think you've been forthright about it, and I'm encouraging you to become a more responsible citizen.

Gullible

"But don't let Comcast fool you. The BitTorrent community should do everything it can to bust Comcast's BitTorrent busting. And so should the FCC."

Please. If you're going to make a bald assertion like this, at least pretend to back it up with a fact, and I don't consider poor Robb Topolski's addle-brained theories "facts." If he wants to maintain high standing with the private trackers that support piracy, he should do his seeding in the off hours when network management is not in effect on Comcast's network.

The new obfuscation scheme isn't going to work, it's simply another nuisance that all the ISPs will have to deal with on the way to metered pricing.

Comcast cops to BitTorrent busting

Odd remark.

"So there you have it. Comcast throttles your BitTorrent uploads only when you're most interested uploading them."

No, they only throttle your uploads after your download has finished, when you're least interested in them. It's not very complicated, Cade.

Hogging the Trough: The EFF Strikes Back

Let's repeat a few points

Some of the commentors seem not to have read the article (or the previous one) and are raising objections that have already been answered, so a few facts:

1. Comcast doesn't advertise an unlimited service, their Terms of Use say you can't run file servers from a standard residential account. BitTorrent running as a seeder is in the category of prohibited applications.

2. VoIP is not a bandwidth hog, it's an application with very modest bandwidth needs that is destroyed by unchecked BitTorrent.

3. Comcast's high upload speed relative to DSL is what makes seeders running on Comcast more attractive to other downloaders than DSL-based seeders. This brings connections into the Comcast network.

4. The bottleneck on the upstream side isn't bandwidth per se, it's the packet rate. So a number of connection requests use up the cable modem's contention slots before raw bandwidth is maxed out. It's not about bandwidth, it's about duty cycle.

5. TCP packet drop doesn't address the fairness problem at all.

I hope that helps.

Richard Bennett

Dismantling a Religion: The EFF's Faith-Based Internet

@Robb Topolski

I'm sure glad you don't engage in personal attacks, Robb.

Saying "We need not list these errors again, because Richard believes he is correct despite any facts to the contrary" doesn't cut the mustard; post a factual claim.

I never cease to be amazed

My goodness, I seem to have upset some people. Let's try and get to some facts, OK?

* Comcast has already announced they'll be upgrading their network to DOCSIS 3.0 starting in Q4 2008. This means the existing 40 Mb/s down/10 Mb/s up configuration will be replaced by a 130/100. That doesn't mean each user gets that much, it just means that the pool will get much larger, especially in the upload direction.

* Comcast doesn't advertise their service as "unlimited" or guarantee any rate. They publish the caps, and they're not shy about them.

* The Internet lacks a system to enforce per-user fairness, as its entire congestion avoidance system has been engineered on a per-flow basis.

* The lack of any per-user fairness system and the limitations of DOCSIS 1.1 explain why Comcast does what it does.

* Calling RST Spoofing "fraud" doesn't address the question of whether it's reasonable or not.

I talked to the EFF report's authors before they published, and I know for a fact that they understand the Internet's fairness problem better than they let on.

Don't let your rancor at the cable companies/phone companies cloud your better judgment. The Internet is not magic, and bandwidth isn't free.