Comments on: Cable guys to FCC: ISPs aren’t the bottleneck, Google is!

By: Antonio Dueñas Rodriguez

Antonio Dueñas Rodriguez — Sat, 21 Jul 2012 03:33:14 +0000

They think we are idiots.

By: Virtuous

Virtuous — Fri, 20 Jul 2012 16:27:48 +0000

The ISPs try to pretend they have competition. If the Verizon – cable deal goes through there will be even less competition. Verizon and Comcast are essentially monopolies.

By: Stacey Higginbotham

Stacey Higginbotham — Fri, 20 Jul 2012 13:41:15 +0000

DRL, I’m glad you brought that up. My intention wasn’t to conflate the two issues, but I didn’t address them in depth to avoid making the article an epic. PErhaps with another paragraph and some careful linking I could have gone deeper. However, I’m glad you brought it up in the comments. Although I could have done without the dig at my editing :)

By: Stacey Higginbotham

Stacey Higginbotham — Fri, 20 Jul 2012 13:37:39 +0000

YES! I’m glad you brought it up, because I really didn’t want to make this any longer. But 300 Mbps is a fatter pipe, not just faster, and when you have a lot of people in a home you need more capacity.

By: Chris

Chris — Fri, 20 Jul 2012 10:05:02 +0000

By their own admission, they “self-edit” the blog posts here, and nobody on the staff can validate them for accuracy because none of them have ever been educated or employed in any facet of networking. It’s just someone’s opinion, and yes, it’s full of false information and ill conceived conclusions.

If the writer is so smart on the subject and has the answers to the world’s network challenges, she should welcome testing end to end performance instead of spreading FUD and making ad hominem attacks against those suggesting it.

The NCTA PR actually committed a major gaffe by naming YouTube, for a bunch of complicated reasons that would be way out of her league to understand.

If you seek medical or financial advice from bloggers who have no practical experience in medicine or finance, then you’re at home here trying to understand complex subjects like packet networks.

By: i2fun

i2fun — Fri, 20 Jul 2012 06:14:19 +0000

Now that’s what I call in your face reporting. You got a 10 for 10 from me on this. So don’t listen to the “Oh noessss the SkyNet is falling” hater I see here. They are usually just that Orwellian Doublethink Trolls just doing what those AntiGOOG voices in their heads tells them to day!!!

By: ButcherChop

ButcherChop — Fri, 20 Jul 2012 06:02:39 +0000

Oh boy, Brett.

Oh wait, i almost fed the troll.

By: David Gest

David Gest — Fri, 20 Jul 2012 05:40:28 +0000

How could Comcast ever support 300mb traffic flows with their peering policy as broken as it is, and ports as hot as they are?

By: drl

drl — Fri, 20 Jul 2012 02:47:25 +0000

Thanks for introducing some sanity and insight into this discussion. The author has written a lot of good stuff that makes sense and is well thought out, but this piece was not in that category. How in the world can you argue that something is wrong with the logic behind the statement that you don’t need 300 Mbs to deliver 20Mbs streams? How are the fact that web services don’t run at 300 Mbs and the fact that web data centers (the implementation of the web services) don’t run at 300 Mbs two different issues? Even though most readers aren’t smart enough to juggle two different concepts at once, there really are two things going on here and Ms. Higgenbotham is conflating the two. The world can’t consume any single application at hundreds of megabits because there aren’t any single services producing that or compelling applications identified that need that. I’ve spent lots of time working with ISPs who would love to identify such an application to drive demand and so far they’ve struck out. Simultaneously many of the services looking for single digit megabits to the end-station can’t get it because access networks are being massively oversubscribed (under-provisioned) by cheap/greedy ILEC/ISPs. The major thesis, that local access providers are (yet again) disingenuously deflecting scrutiny of their anti-consumer practices, is important to document, but the supporting arguments are confusing and ill-considered. Also lacking is exactly the kind of use case you provided that actually captures a real world-scenario where the content consumer might need massive aggregate bandwidth to the home that is not driven by any single service. Throw in typos and errors that suggest this was never proof-read and the result is a product that is not worthy of the author’s talent and reputation.

By: fgoodwin

fgoodwin — Fri, 20 Jul 2012 01:34:00 +0000

“Between your computer and Google’s servers are a lot of steps. A lot happens between Google’s servers and your router when you request a YouTube video.”

As far as I’m concerned, those two statements are the crux of the entire issue, and why the FCC’s speed measurements are so pointless. When ISPs sell me access, it’s just that: ACCESS. By definition, they CANNOT ensure the speed or quality of my connection to an infinite number of end-points on the world wide web.

If the FCC wants to truly test the ISP’s speed claims, they need to measure the speed of MY access connection (or yours) and NOTHING MORE because that is all the ISP is providing me.

My ISP cannot be responsible for how quickly (or slowly) a Netflix server responds to my request, so what is the point in measuring that and holding ISPs accountable for that?

It’s ridiculous.