[Air-L] a question about privacy protection and copyright in Internet research

Tue May 10 07:50:59 PDT 2011

Well what actually happens when you put something on facebook?  it isn't like you are sending a private letter to a group of friends.  it is much closer to going to a shopping mall where all the video cameras are tracking your every move.   It isn't your property you are acting within... it is facebooks, in doing that acting you are agreeing with the terms that they put upon you.  It isn't about your or my ethics Michael at that point, it isn't about what we should do or our ethical systems, it is then about the agreement that we consent to.   

There are no subjects to ask what you want to ask if you are doing document based research.  You can ask the webpage, but your browser probably is not going to respond.  The assumption you are putting forth is one that... there are subjects and it matters what their opinion is, but I want to say, when I read and interpret a book, a newspaper, a public record or any other published document, I need not consider that.  You want to say perhaps that 'they don't know they are publishing it' and even if they publish it 'they might not want it researched'  but that is not something we afford authors of published documents.  We could, but I think it is going the wrong direction.  That debate, I'd argue ended, when they put on the web in a manner similar to a published document in which it is assigned a copyright and is freely available to anyone without login.  I've already stated that login may change that status.  

I think you can do research on a database with personally identifiable positions without running afowl of u.s. standards for research ethics and I mostly agree with that personally.  If the data was freely given to a corporation and then the corporation freely gives it to you, then I don't think anyone is going to have any issue with you using that data as it was a. previously gathered  and b. you are not interacting with the people.  It might not be something people like, but I think that is generally the way it works.   We have many colleagues doing this research right now actually, they all seem to have had their material passed by IRB's, most I suspect with waivers, but some perhaps with deeper reviews.   I am fairly certain that if Facebook gave me data that they collected with personally identifiable information in it, and it was given under contract, and I ran that by the irb they would treat it much like they treat personally identifiable census data or use of standard social research data sets with individually identifiable data, that is... unless it contains private information, which as I said is a problematic construct, it is probably going to get a waiver, and the debate about whether these are human subjects or not, won't come up

gotta run to an appt. but look forward to rejoinders.  

On May 10, 2011, at 10:33 AM, Michael Zimmer wrote:

> I find this logic troubling, Jeremy. 
> 
> In our earlier exchange about Facebook suddenly making something public that was originally shared to a limited group, you seem to be placing strong preference on what a platform might afford at a given time (does it make something publicly available) rather than what the subject expected/intended/currently desires. That's not how I wish to have research ethics determined.
> 
> And below you again now that "Texts are not people.  Representations of people are not people", and appear to state that unless you are directly engaging with a human, you're not engaging in human subject research. Elizabeth has replied in part on this, but just to add: your logic seems to indicate that I can do research using a database of personally-identifiable and sensitive information, publish my results, etc without ever having to worry about ethics related to "human subjects" since I'm merely using a database, a "representation of people".  Is this your position?
> 
> -mz
> 
> 
> -- 
> Michael Zimmer, PhD
> Assistant Professor, School of Information Studies
> Co-Director, Center for Information Policy Research
> University of Wisconsin-Milwaukee
> e: zimmerm at uwm.edu
> w: www.michaelzimmer.org
> 
> 
> On May 10, 2011, at 8:33 AM, jeremy hunsinger wrote:
> 
>> I think i already mentioned the private information exception.   But past that, as I said, instead of complicating it, i argue that we we need to simplify it.  Texts are not people.  Representations of people are not people.  Whatever claims people have on those objects are not claims against their nature, they are claims to ownership or copyright.  That is the line I think we have to take here.  Things like documents are not research subjects.  They are research objects, research quasi-objects, but they don't rise to the level of being a research subject no matter what exterior subjective claims people put on them.  If they have private information embedded in them, that is still an object, the issue with private information is not that the object becomes a research subject, but that the claim of a subject overrides the interests of the researcher in that case.  So my argument is that the private information claim is not that it transforms the document, but it creates a clai
>> m by one person upon another based on the claim of privacy, however... very few things once published rise to that claim of private, especially once they are published, for instance a person's name on a document published is not private, it isn't private in the newspaper, it isn't private on public documents, nor are other identifying characteristics, height, demographic features, even salary might be public.  So what rises to the claim of private in a public document?  Medical records are one... though even then medical records can be made public by any number of sources and once public, might not be made private again.   so... this line... seems really to be far, quite far away.    Treat a published source as a published source, it is not a research subject.
>> 
>> 
>> 
>> On May 10, 2011, at 8:58 AM, Elizabeth Buchanan wrote:
>> 
>>> Hey, all,
>>> 
>>> I'm late in adding a couple of points about this discussion. From a US-human subjects perspectives (talking more regulatory), I disagree with Jeremy's hard line "if you are dealing with texts, you are not dealing with research subjects." I disagree with it from  disciplinary and philosophical perspectives, but, that would take us down the road of another discussion. 
>>> 
>>> But--from a US regulatory perspective,(f) Human subject means a living individual about whom an investigator (whether professional or student) conducting research obtains
>>> 
>>> (1) Data through intervention or interaction with the individual, or
>>> (2) Identifiable private information.
>>> 
>>> So, I would argue that to patently reject "texts" as subjects is a risky choice and not a line that holds a lot of ethical weight; disciplinary specificity is very important for this discussion--these decisions are not, should no,t be based on only one set of criteria; with online research, as we know, there are community (or venue) norms, ethics, practices; disciplinary norms, ethics, practices; extant research ethics guidelines at the federal/country/commission level; and, increasingly, third party (eg, Facebook, etc etc) norms, "ethics", practices. All of those must be considered in making these decisions about research practices. 
>>> 
>>> Regarding the issue of public v private (which I agree with Michael is a fluid and moving dichotomy), again, from a regulatory perspective, there are boundaries:
>>> 
>>> From  101 (b)(2) and 101(b)(4) in the CFR 45: 
>>> 
>>> Regarding online observation and exemptions for existing data, the issue of researcher "entry requirements" and "threshold of access" is significant. The OHRP standard is "what is readily ascertainable." If there are costs, restrictions, or other measures that prevent data from being "readily ascertainable, "the exemption does not apply and the data should be considered private. (This comes from a report in draft form, written with a colleague from the Office for Human Research Protections here in the US.If you are looking for more regulatory guidance, I can share this report).
>>> 
>>> Recently, I gave a few talks with the lawyer who advises OHRP, and while they have issued no formal guidance on Internet research in particular, some interpretations have been presented. I can share our powerpoints on Internetresearchethics.org for those interested.
>>> 
>>> 
>>> Best regards,
>>> 
>>> e.
>>> 
>>> 
>>> Elizabeth A. Buchanan, Ph.D.
>>> Endowed Chair in Ethics
>>> Director, Center for Applied Ethics
>>> University of Wisconsin-Stout
>>> PO Box 790
>>> Menomonie, Wisconsin, 54751-0790
>>> 
>>> buchanane at uwstout.edu
>>> 
>>> 
>>> Principal Investigator, 
>>> Internet Research Ethics Commons
>>> internetresearchethics.org
>>> (NSF Project # 0924604)
>>> 
>>> elizabeth at internetresearchethics.org
>>> 
>>> _______________________________________________
>>> The Air-L at listserv.aoir.org mailing list
>>> is provided by the Association of Internet Researchers http://aoir.org
>>> Subscribe, change options or unsubscribe at: http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
>>> 
>>> Join the Association of Internet Researchers:
>>> http://www.aoir.org/
>> 
>> Jeremy Hunsinger
>> Center for Digital Discourse and Culture
>> Virginia Tech
>> 
>> 
>> 
>> Computers are useless. They can only give you answers.
>> --Pablo Picasso
>> 
>> 
>> 
>> 
>> _______________________________________________
>> The Air-L at listserv.aoir.org mailing list
>> is provided by the Association of Internet Researchers http://aoir.org
>> Subscribe, change options or unsubscribe at: http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
>> 
>> Join the Association of Internet Researchers:
>> http://www.aoir.org/
> 

Jeremy Hunsinger
Center for Digital Discourse and Culture
Virginia Tech

Words are things; and a small drop of ink, falling like dew upon a thought, produces that which makes thousands, perhaps millions, think. --Byron