BattlefortheNet study

[Steve Bauer]

I saw a variety of news articles referencing a BattlefortheNet study
based upon M-lab data but I haven't been able to find the study
itself.

http://www.theguardian.com/technology/2015/jun/22/major-internet-providers-slowing-traffic-speeds

Is it available somewhere?

Thanks,
Steven Bauer
MIT

[Livingood, Jason]

I was told via Twitter that the report is what is attached, originally shared via Google Docs with the reporter (I exported it into Word), now substantially edited down to two URLs at https://docs.google.com/document/d/10DvT4OYpJ_Lc3Zz-8AURgB07MvMqRBbllwrnwqJb_3M/edit by Tim Karr at Free Press this afternoon.

Colin at M-Lab posted a detailed update today (6/24/15) at http://www.measurementlab.net/blog/interconnection_and_measurement_update and FreePress (which seems to have originated or edited the Google Docs document) posted at http://www.freepress.net/blog/2015/06/24/internet-health-test-reveals-network-slowdowns. By the way, one day following (6/23/15) the publishing of the Guardian article, according to the attached ex parte filing, New America Foundation / OTI had a conference call with the FCC to discuss the report that was released today on the M-Lab blog. It looks like the M-Lab blog post is reasonably aligned with the Word doc from Google Docs.

Hope this helps!

Jason

--

You received this message because you are subscribed to the Google Groups "discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to discuss+u...@measurementlab.net.

To post to this group, send email to dis...@measurementlab.net.

Visit this group at http://groups.google.com/a/measurementlab.net/group/discuss/.

[Juju]

That's great work. A pity there is nothing for other countries to compare like Canada or Europeans'

Anyone knows if there is plan to extend the observatory?

Thanks

Julien

[jpartr...@gmail.com]

Dan Rayburn posted a piece last night correcting the erroneous conclusions that had been speculated about the M-Lab "study" - seems that nobody bothered to talk to GTT until Dan Rayburn did

“Study” of ISP Slowdowns Conveniently Ignored Real Cause, ISPs Not At Fault

http://blog.streamingmedia.com/2015/06/isps-not-causing-network-slowdowns.html

[Livingood, Jason]

In related news, AT&T and GTT today announced a new interconnect agreement.

http://www.marketwatch.com/story/gtt-and-att-enter-into-an-interconnection-agreement-2015-06-30?mod=mw_share_twitter

--

[Don Waterloo]

CIRA is underway 

http://cira.ca/content/internet-performance-test-advanced-results

--

[Zubair Shafiq]

Many news articles covering M-Lab's study are not aware of its limitations, and are misinterpreting the findings. 

The methodology used by M-Lab cannot locate congested links or determine the root-cause of performance degradation experienced by users. Nick Feamster, in a recent blog article (https://freedom-to-tinker.com/blog/feamster/where-is-internet-congestion-occurring), also points out these issues with the M-Lab study. 

M-Lab blog (http://www.measurementlab.net/blog) and the detailed technical report (http://www.measurementlab.net/static/observatory/M-Lab_Interconnection_Study_US.pdf) acknowledge that they cannot determine the root-causes of congestion. For example, here is an excerpt from the M-Lab blog article:

"It is important to note that while we are able to observe and record these episodes of performance degradation, nothing in the data allows us to draw conclusions about who is responsible for the performance degradation. We leave determining the underlying cause of the degradation to others, and focus solely on the data, which tells us about consumer conditions irrespective of cause."

Unfortunately, some news articles (like the following) are misinforming the public with sweeping (but false) conclusions. I believe that M-Lab should be clear and upfront about the limitations and nuances of its study to avoid widespread misunderstanding. 

- The Gaurdian: Major internet providers slowing traffic speeds for thousands across US (http://www.theguardian.com/technology/2015/jun/22/major-internet-providers-slowing-traffic-speeds)

- NetworkWorld: ISPs accused of violating net neutrality rules (http://www.networkworld.com/article/2940115/lan-wan/isps-att-comcast-verizon-accused-violating-net-neutrality-rules.html)

- FreePress: The Internet Health Test Reveals Network Slowdowns (http://www.freepress.net/blog/2015/06/24/internet-health-test-reveals-network-slowdowns)

- CBS: Largest ISP Companies Slowing Data For 75 Percent Of US (http://washington.cbslocal.com/2015/06/23/study-largest-isp-companies-slowing-data-for-75-percent-of-us/)

Huffington Post: Internet Test Reveals Americans Are Not Getting the Speeds They Paid For (http://www.huffingtonpost.com/timothy-karr/internet-test-reveals-ame_b_7660542.html)

-- 

M. Zubair Shafiq 

----------------------------------------------- 

Assistant Professor 

Department of Computer Science 

The University of Iowa 

Iowa City, IA 52242-1419 

Office: 201J MacLean Hall 

Tel: 319-335-0742 

Fax: 319-335-3624 

Email: zubair...@uiowa.edu 

Web: http://www.cs.uiowa.edu/~mshafiq 

-----------------------------------------------

[jpartr...@gmail.com]

This piece by Richard Bennett also does a good job of correcting the erroneous speculations surrounding GTT interconnection issues.

GTT’s Growing Pains Behind Over-Hyped Congestion Claims

http://hightechforum.org/gtts-growing-pains-behind-over-hyped-congestion-claims/

On Wednesday, June 24, 2015 at 6:16:29 PM UTC-4, Steven wrote:

[Zubair Shafiq]

Right, this is a well-done piece by Richard Bennett. Thanks for sharing. 

[Adam Rothschild]

I don't think it's necessary to speculate on the speculation (which is
precisely what Messrs Bennett and Rayburn are doing in their editorial
pieces, not dropping hard science), when GTT's former CTO laid it out
in the comments section:

http://www.huffingtonpost.com/dan-rayburn/study-of-isp-slowdowns-co_b_7698056.html

Financial motives on the part of the "journalists" should come into
question as well.

$0.02,
-a

[Richard A Steenbergen]

That would be me. Obviously I'm limited on what I can say with regards to interconnection agreements and capacity between GTT/nLayer and AT&T, but I can definitely confirm that Dan Rayburn's theory is incorrect and baseless. I think the situation is probably best explained by the joint press release announcing additional capacity. How often do you see Tier 1 networks putting out a joint press releases about peering upgrades, and why might that have occurred in this circumstance? To a logical observer, the answer should be obvious.