Confirming your sources

I remember speaking with a reporter at one of my conferences who told me she always makes sure she gets the story right, even if it means that she doesn’t necessarily get the story first. That often means needing to get multiple sources, which is a challenge when trying to get a scoop on a major breaking story.

The need to apply that kind of rigour has stuck with me.

I generally try to go back to primary sources to confirm statements, rather than rely on reports that cite those sources. As a result, I’m often the family fact-checker when someone wants to confirm something they saw “on the internet”.

As many of us know, a lot of times those things turn out to be false. As it turns out that you can’t even trust official government websites to always get it right.

For example, last May, in an edition of Statistics Canada’s Daily looking at “Access to the Internet in Canada, 2020”, the agency wrote:

Over two-thirds of Canadians have Internet download speeds of 50 megabits per second or more
For those respondents who knew their advertised Internet connection speed, 72% reported Internet download speeds of 50 megabits per second (Mbps) or more. An Internet download speed of 50 Mbps or more allows users to do online activities such as supporting multiple users at a time, streaming ultra-high-definition video with more than one connected device, or quickly downloading a high-definition movie. This indicator can be used by the Government of Canada to track its objective to make broadband connectivity with a download speed of 50 Mbps and an upload speed of 10 Mbps available to 95% of Canadians by 2026.

The last sentence, the government’s objective (mirroring the CRTC’s objective) is to make broadband connectivity with a download speed of 50 Mbps and an upload speed of 10 Mbps available to 95% of Canadians by 2026. However, the first part of the paragraph indicates Statistics Canada is tracking the speed of the service to which people subscribe, not the speeds to which they have access. I may have access to extremely fast speeds, but choose to subscribe to a lower speed for many different reasons. The indicator discussed in that paragraph is clearly looking at subscribed speeds, not available speeds.

We can quibble about what speeds people need for various online activities, but tracking subscribed speeds simply does not and cannot help the Government track its objective of available speeds.

This isn’t the first time that people have confused adoption with access in the government (and CRTC’s) broadband objective. I wonder if the word “access” is the cause of the misunderstanding. We want everyone to have the ability to choose a 50 Mbps connection, which some people write as “we want everyone to be able to access a 50 Mbps internet access”. Having the ability to access is the objective; having a 50 Mbps access isn’t. Statistics Canada needs to correct its website.

In another instance, also from last May, a University of Calgary student relied on political hyperbole on a government website to claim “The federal government took the lead in providing service to low-income Canadians”, when referring to low income broadband programs that were in fact created by the private sector (Rogers and TELUS). It took years of cajoling to get the government to participate, despite virtually no government funding being involved. The paper was later amended to read “Federal government participation was key in expanding service provision to low-income Canadians.” The student had relied on hyperbole from a government website that had said “Connecting Families is an initiative of the Government of Canada”. The website now refers to the “Connecting Families initiative, in partnership with Internet service providers”. The difference between government partnering, versus government leadership, was significant because of the paper’s thesis promoting increased government intervention in telecom markets.

Even the CRTC website has been known to contain the odd error. Its Communications Market Reports website has a “changelog” that tracks changes and corrections. A few weeks ago, we noticed that the reported “Average Internet Data Downloaded/Uploaded per month, per subscription” was reported as the average of the uploaded data and the downloaded data, instead of the sum of those. The Commission now reports data uploaded and downloaded separately.

I am also reminded of some popular commentators who frequently display innumeracy, perhaps cheered on by those who share their perspective, if not their propensity for propagandizing. Sometimes, their basic data may be correct but accompanied by incomplete analysis, such as we saw last week in the false equivalency posted by TekSavvy, comparing 398 lobbyist registration reports by “Big Telecom” (defined as TELUS, Rogers, Shaw, Bell, Cogeco, and Videotron), with the 32 meetings by TekSavvy and CNOC. There are many problems with this post that implies the smaller ISPs aren’t getting a fair share of face time in Ottawa.

On the other hand, one might ask why Teksavvy and CNOC, with just than 10% of the internet services market had just about as many meetings (32) as Quebecor (34), considering the latter is a diversified company, a facilities-based carrier with more than 1.8M internet connections, more than 1.5M mobile subscribers, and more than 1.4M TV subscribers, on top of a range of other media operations.

A quick look at the scope of subjects being reviewed by the lobbyists demonstrates why it was completely reasonable for 6 major conglomerates with revenues of more than $60B and more than 150,000 employees to have had more meetings than the smaller wholesale-based internet service sector. Let’s take a look at the range of subjects that are covered in the government filings. Teksavvy reported that it talks to government about Broadcasting, Consumer Issues, and Telecommunications. CNOC reported the same 3 areas of interest. Rogers’ Lobbyist filing shows that discussions cover: Arts and Culture, Broadcasting, Budget, Consumer Issues, Economic Development, Employment and Training, Government Procurement, Health, Immigration, Industry, Infrastructure, Intellectual Property, International Relations, International Trade, Justice and Law Enforcement, Labour, National Security, Privacy and Access to Information, Sports, Taxation and Finance, and oh yes, Telecommunications. Bell reports 20 subject areas; TELUS reports 19; Quebecor shows 14; Shaw has 12.

Maybe, just maybe, this helps to show the fallacy of Teksavvy’s false equivalency.

All of this is to say that information provided on websites – even government websites – should be viewed with a grain of salt.

It’s important to question and even challenge sources.

Other than the information on this website, of course.