Evidence of meeting #43 for Access to Information, Privacy and Ethics in the 41st Parliament, 1st Session. (The original version is on Parliament’s site, as are the minutes.) The winning word was data.

A recording is available from Parliament.

On the agenda

MPs speaking

Also speaking

  • Warren Everson  Senior Vice-President, Policy, Canadian Chamber of Commerce
  • Brendan Wycks  Executive Director, Marketing Research and Intelligence Association
  • Annie Pettit  Vice-President, Marketing Research and Intelligence Association

11:25 a.m.

NDP

Charmaine Borg Terrebonne—Blainville, QC

I have another question about anonymity. How do you ensure that the information that you obtain and use remains anonymous?

11:25 a.m.

Vice-President, Marketing Research and Intelligence Association

Annie Pettit

That's an interesting question. A lot of people assume that when they write something in social media, their whole life is present for anyone to see.

Pieces are present. For example, you might sign up with a user name, you might provide an e-mail address, you might provide geo-location—Canada, Alberta, or something like that. In most cases it's a very tiny bit of information that is released. As social media researchers, we pick up that data, where it is available, for the purpose of aggregated data, so that we can say this percentage of people from Alberta share this opinion or this percentage of people from Ontario share this opinion.

If the information goes beyond that, in terms of a written report to a client, we take great efforts to make sure that whatever personal information shows up in it has been completely masked. There are no user names in reports, no photos, no e-mail addresses. Even in what someone has written as a tweet or a status update, the wording has been tweaked so that you can't identify what the phrase was originally. If someone were to have said “I really love Nike shoes”, the phrase might be transferred into “I really like Nike shoes”. It's just a few minor words, so that the general phrase is still there but there's really no way to match it back to the person who originally gave that information.

11:25 a.m.

NDP

The Chair Pierre-Luc Dusseault

Unfortunately, your time is up.

We now go to Mr. Del Mastro, for seven minutes.

11:25 a.m.

Conservative

Dean Del Mastro Peterborough, ON

Thank you very much, Mr. Chairman. Thank you very much to the witnesses.

First, Mr. Everson, I agree with your assessment. These are very important economic developments. I think the benefits are far- and wide-reaching. We see them in every region of the country and in every city and community in the country. I think it's important for business and I think it's important as far as a growth tool for business is concerned.

However, I'm interested in whether the Chamber of Commerce has looked at the issue of informed consent. I'm going to raise the same question with the marketing research association as well.

In my opinion, because so much legalese goes into social media sites and an awful lot of laymen use them, including me, who may not understand the impact of the boxes they're checking and what the intent of the statements is that they're signing off on, an awful lot of people scroll through disclaimers to get down to the box that says “Yes, I agree” and move on.

I'm just wondering whether the Chamber of Commerce looked at the issue of informed consent with respect to privacy.

11:30 a.m.

Senior Vice-President, Policy, Canadian Chamber of Commerce

Warren Everson

We didn't opine on it in a resolution, but of course we have a lot of debate about this in the relevant committee and within the chamber. I noticed that when the commissioner was here she was using the term “meaningful consent” to deal with a constituency of people who may not be competent to give their consent or understand the implications.

I would make a couple of points on what you said. On the legal gobbledygook issue, I think we all understand why it exists: somewhere a lawyer is trying to make his or her client bullet-proof against any possible action against them. It would be nice if the committee were to opine that it was time to clean up that language and make it simpler, putting that challenge before the attorneys.

As for the issue of meaningful consent, I don't know a more thorny issue in the world of cyber-commerce or one that I will be watching with more excitement to see what you come up with. The nice thing about being a member of Parliament is that if you're not an expert in any one field, you're an expert in the application of common sense, and this is where you're going to be with this issue.

If somebody is too young, is it the vendor or the carrier's responsibility to ascertain that? And in that case, how can they do so without unwarranted intrusion into their privacy? It's an extremely difficult challenge. I keep saying that someone bought them a computer, because if they're 13, they likely didn't buy the computer themselves, so there has to be some societal construct around them that might be employed.

It is an extremely demanding issue and one for which, as I say, the most obvious solutions, for the middlemen involved, are quite significant intrusions into our privacy, and we probably wouldn't be very happy about that.

Have I answered your question?

11:30 a.m.

Conservative

Dean Del Mastro Peterborough, ON

I think you did. I can say, having taken a number of university-level business law courses myself, that one of the first things you learn is that terms in law do not necessarily mean what they mean in everyday life. This is why, when you actually sit down and read one of the privacy statements or one of the statements you're signing off on, if you actually read them, you may have a lot more questions if you've taken a couple of law courses than you do if you haven't.

I can tell you, having read them, that I'm not entirely clear on what some of the subsections are getting at. I will ask that of companies when they come in, because I don't think it's your place to answer that. But I do think there is a role for clear language in working with the public, especially when we're talking about children or young people who might be using social media.

To the Marketing Research and Intelligence Association, I agree with you that research is very important, but again it comes back to an issue of informed consent. I respect that you're saying you don't sell anything and don't advertise anything. But you're giving all your research to people who do sell things and advertise things—that's who the customers are—and they're looking to use your research.

It's very important research, because one of the things that retailers are looking for today is how to reach a customer. And it's not just retailers, but others. How do you get a message to somebody in an era when we're not sure they're watching television, not sure they're listening to local radio, or reading local newspapers. They might be doing all of those things; they might be doing it online.... So the research you're doing is very important.

But it comes back to this question, and this is my concern. There's all kinds of medical research we could do that could be very important. We could in fact look at the medical records of every member of Parliament to determine whether running 16 hours a day and eating whatever is put in front of you is good for your liver. I suspect it's not. But you'd have to have their consent to look at those issues and to look at their medical records, to see whether it is okay.

If you just looked at the medical records and came back with helpful information, it might be good that you have the information, but the way you received that information in order to process it might be entirely wrong. This is the issue with respect to informed consent. If you're researching things from social media and are producing good data from it but are doing so in a fashion whereby a lot of folks....

If you had a focus group and you sat down with folks and said, “Did you understand that it meant this, this, and this?”, I expect most people in the room would say “Yes, I knew that”, because people aren't about to tell you that they had no idea what the legalese meant and they don't want to seem, for whatever reason, incapable of understanding it. They're probably going to answer affirmatively, if you just ask them if they understood that this is what they were signing on to. But I'm not confident that people always do understand the implications of what they're signing off on.

Would you support a move toward more common language and clarifications with respect to privacy and then abide by it? It sounds as though you're very keen on abiding by all of the privacy guidelines of social media. Do you see that there is an opportunity for people to be confused about people providing consent that they don't mean to be providing, and that there's a role for this committee in working to clarify it?

11:35 a.m.

Vice-President, Marketing Research and Intelligence Association

Annie Pettit

We absolutely support plain language. To some extent in the industry there is already a move towards it. There have been a few companies so far that have published their plain-language terms of service. Essentially the language is along the lines of saying “We're going to share your information with third-party companies. Are you okay with that?”—using simple statements, simple language that people can read and understand very quickly. The companies that are doing this are getting a lot of praise from those around them. There's a lot of push to actually do this.

Beyond that, within the market research space we pride ourselves with the “do no harm” phrase. We know that a lot of people don't read those, don't understand them, and we take it upon ourselves to be a sort of overseer. We respect that not everybody knows or understands. We will look out for you on your behalf. When we see that something should be done in a certain way, even if it's technically legal, if we don't think it meets our higher ethical standards, then we'll make sure to do what we know is the right thing, even though it goes beyond checking the box that nobody has read.

One final point is that observational research, which is essentially what social media research is all about—looking at what people are saying, looking at what they're doing—has a long-standing tradition as a legitimate research method. Sociologists, anthropologists, psychologists have been doing this for more than a hundred years. It's a standard practice. It's becoming more and more popular now with social media—it's easier to do it and you can observe a lot more people doing it—but we still take pride in ensuring that what we do is following the “do no harm” methodology.

11:35 a.m.

NDP

The Chair Pierre-Luc Dusseault

Thank you. I am going to have to stop you there, unfortunately.

We now go to Mr. Andrews, for seven minutes.

June 5th, 2012 / 11:35 a.m.

Liberal

Scott Andrews Avalon, NL

Thank you, Mr. Chair.

I'm trying to get my head around something here this morning. I'm going to make a statement, and I want you to correct me if I'm wrong or clarify it.

Listening to you this morning, I'm getting the impression that there are two types of data that we're looking at. We have Facebook, Twitter, and the companies that actually own the data that is put into it by people and then they either resell it or they market it themselves.

Then what I think I heard you say this morning, Annie, is you guys observe data online. You're not actually in possession of it. From your perspective, and from the place you guys come today, you're looking at that data that's online. You're not actually talking about the data that is owned by the companies.

Is that a fair statement? I'm trying to compartmentalize this into two different things here and I don't know if I've done that.

11:35 a.m.

Vice-President, Marketing Research and Intelligence Association

Annie Pettit

We indeed collect that data, so what we see online we bring it in-house, so that we can measure the opinions, aggregate the opinions, and come to the research conclusion. So we do collect that data.

11:35 a.m.

Liberal

Scott Andrews Avalon, NL

You collect that. So what you're advocating today is separate from what these companies do with their data. There are two distinctive types. Facebook could have its data on individuals and they can sell it, push it, and market it. You do it outside. Okay.

You mention that you don't go into a password, but most of these sites, the social media sites, have passwords. You have to access them. How do you get the data without accessing...? When you Google something, what comes up, comes up, and often you're linked to the site to go get it. So how do you square that circle?

11:40 a.m.

Vice-President, Marketing Research and Intelligence Association

Annie Pettit

There are two kinds of passwords. There is one kind where you must enter the password before you can actually enter the website and see anything at all. It is completely blind. You don't see anybody's names, photos, user IDs, comments, nothing, until you create a password and go inside. That is a large portion of Facebook and medical kinds of sites where patients talk to each other.

Then there are the other kinds of passwords that are simply there, so that I have my space and my friends see what I have written, and everyone can follow each other. We know that because there is a name associated with each comment. That's the second kind of password.

Those are searchable by whatever browser you want to use, the Googles, and that's the kind of data we collect—only the data that is physically viewable if you were to go online and not have your own password.

11:40 a.m.

Liberal

Scott Andrews Avalon, NL

How many different social media companies are we talking about? We know the big ones, but how many are out there that would have to comply with all the privacy issues?

11:40 a.m.

Vice-President, Marketing Research and Intelligence Association

Annie Pettit

Strictly social media research companies—there are not a lot. Certainly, under 100 are social media research.

Then there are untold hundreds of other companies in the social media monitoring space. They're not members of any kind of research organization, no industry organizations, but they're doing similar kinds of things. They're counting, reading, and listening. Some of them are providing pieces of measurements, but they don't classify themselves as a market research company as mine is.

11:40 a.m.

Liberal

Scott Andrews Avalon, NL

I'm going to come back to that, but here is the second part of my question. How many social media companies are there, like the Facebooks, the Twitters?