Intervention: Intervenor 258

Document Name: 2015-134.223968.2394514.Intervention(1fbm@01!).html

Raisons pour comparaitre / Reasons for appearanceI would be willing to appear in order to help interpret the submitted data. I would be able to provide expertise in Internet measurement data in Canada and key performance metrics.

Intervention: Intervenor 258

Document Name: 2015-134.223968.2394513.Intervention(1fbm901!).pdf
14 July 2015
John Traversy
Secretary General

Canadian Radio-television and Telecommunications CommissionOttawa, ON *** ***

**** Mr. Traversy,
Re:
Telecom Notice of Consultation CRTC 2015-134
Review of basic telecommunications services

The intervention submits geo-located Internet performance data to assist in the evaluation of wireline Internet provision in Canada. It submits public-domain data from Internet measurements collected across Canada during 2014. These measurements of actual Internet performance result from the Network Diagnostic Tool run through the Measurement Lab Consortium. The intervention joins these measurements to pre-existing geographic information systems in Canada. The intervention at this point does not interpret data rather it hopes to contribute to the overall discussion at the hearings by submitting open data in a manner accessible to stakeholders and connected to other demographic data in Canada.

The intervention has two objectives:

1. Submit existing Measurement Lab data to inform interventions in CRTC Review of basic telecommunications services (File number: 8663-C12-201503186)2. Interpret the Measurement Lab Network Diagnostics Tool as a tool to measure actual speed and performance metrics during the hearings Submitted data help evaluate the sectors’ obligation to Section 7.b of the Telecommunications Act “to render reliable and affordable telecommunications services of high quality accessible to Canadians in both urban and rural areas in all regions of Canada”.

Data includes indicators of network performance such as latency and throughput that reflect the quality of Internet access.

The intervention has four components:

1. Data reference document that explains the Internet measurement methodology, the sampling procedure and the Internet performance metrics such as upload throughput, download throughput and round-trip-time.

2. Itemized data of each NDT test run in Canada for 2014 submitted as two CSV files ofupload performance for 2014 and download performance for 2014.

3. Aggregated data of average upload and download performance metrics. Data aggregates itemized data using the Industry Canada hexagon and the Statistics Canada Dissemination Area geographic information systems. Aggregated data is submitted as four CSV files of 2014 upload performance per hexagon, 2014 download performance per hexagon, 2014 upload performance per dissemination area, 2014 download performance per dissemination area.

4. Map files linking itemized data to Industry Canada hexagon and the Statistics CanadaDissemination Areas geographic information systems. Map files are accessible in QGIS, a cross-platform, free-software geographic information system application.

Due to the large size of the data files, the latter items have been uploaded to the Concordia University Spectrum system. The data may be accessed at:

https :// spectrum . library . concordia . ca /980168/.

Submitted data have limitations that should be considered in its interpretation. These limitations are as follows:

1. Data do not determine responsibility for Internet performance. Many factors might degrade Internet performance from the home user to the core routing. Aggregated data, as a result, emphasize regional performance rather than carrier-specific performance.

2. All tests measured Internet performance against servers located in core Internet aggregation points in the United States. The greater geographic distance from home tocore might degrade performance however Internet routing has a complex relationship https://spectrum.library.concordia.ca/980168/

https://spectrum.library.concordia.ca/980168/
https://spectrum.library.concordia.ca/980168/
https://spectrum.library.concordia.ca/980168/
https://spectrum.library.concordia.ca/980168/
https://spectrum.library.concordia.ca/980168/
https://spectrum.library.concordia.ca/980168/
https://spectrum.library.concordia.ca/980168/
https://spectrum.library.concordia.ca/980168/
https://spectrum.library.concordia.ca/980168/

to geography. Much of Canadian Internet traffic routes through the United States1, so data reflect a common, real world experience of Internet use.23. Users self-select to run the tests. The Measurement Lab relies on crowdsourcing3 to generate data. Some regions have no or low number of tests as a result. The number of tests should be considered when interpreting data.

While these limitations must be considered in interpreting the data, the intervention remains important because it contributes data about regional performance using the well-established, international Measurement Lab tools matched to pre-existing geographic information systems used by Statistics Canada and Industry Canada. These issues are not unique to this data sample. Internet measurement remains an emerging field especially in regulatory affairs.

The intervention hopes to contribute to the use Internet measurement data in Canada in the CRTC hearing.

By submitting data to the hearings, this intervention also hopes to explain the Network Diagnostics Tool, enhance understanding of its results and familiarizes the policy communitywith an initiative by the Canadian Internet Registration Authority (CIRA) known as the Internet Performance **** (IPT). The initiative has built and installed testing nodes within key Internet exchange points in Canada and developed a web-based test Internet Performance****. The initiative relies on the same Measurement Lab Consortium standards and the Network Diagnostics Tool that generated the data in this intervention. The intervention hopesthat other intervenors may in the future elect to conduct their own tests with the IPT drawing on the data described in this intervention.

All Measurement Lab data is in the public domain. Dissemination blocks have been adapted from Statistics Canada, Census Boundary Files (Dissemination Blocks and Dissemination Areas), 2011. This does not constitute an endorsement by Statistics Canada of this product.

Hexagons by Industry Canada were released as part of the Connecting Canadians program.

1 ****, **** A. and ****, ****, Internet Surveillance and Boomerang Routing: A Call for Canadian Network Sovereignty (July 1, 2013). TEM 2013: Proceedings of the Technology & Emerging Media Track - Annual Conference of the Canadian Communication Association (Victoria, **** 5-7, 2012). Available at SSRN:

http :// ssrn . com / abstract =2311792

2 No M-Lab testing points existed in Canada in 2014. The Canadian Internet Registration Authority has built three Measurement Lab in Canada in May 2015. Future testing could use the same methodology with the CIRA-supported infrastructure to test performance in Canada.

3 Brabham, **** C. Crowdsourcing. The MIT **** Essential Knowledge Series. Cambridge: The MIT ****, 2013.

http://ssrn.com/abstract=2311792
http://ssrn.com/abstract=2311792
http://ssrn.com/abstract=2311792
http://ssrn.com/abstract=2311792
http://ssrn.com/abstract=2311792
http://ssrn.com/abstract=2311792
http://ssrn.com/abstract=2311792
http://ssrn.com/abstract=2311792

The intervention wishes to thank **** Ritzo at the Open Technology Institute at the New America Foundation and **** Smith at Concordia University for their support.

Sincerely,
**** McKelvey
Assistant Professor, Communication Studies
Concordia University
fenwick . mckelvey @ concordia . ca
*-***-***-**** ext. 8673
mailto:******@***.com
mailto:******@***.com
mailto:******@***.com
mailto:******@***.com
mailto:******@***.com
mailto:******@***.com
mailto:******@***.com