When Twitter cut off commercial relations with our NSF-funded, academic-driven, Twitter service, thereby undermining a product that lost money for 6 years in the midst of the first year it was about to make a small profit they pointed to specific studies on this list (below) as proof of why academic disregard for the corporate legal terms that govern their property (ie., all Tweets) justified killing a very unique service that academics ask to use every week. https://discovertext.com/publications/ It is goes against the views of many, but when you use the massive and ecologically-destructive, energy-intensive platform like Twitter, even just to Tweet a picture of a great pizza, someone has to pay for every part of the experience, including the indefinite storage, the fact it was super easy, the massive number of pixels in a modern phone image to be stored, and on it goes. I would be surprised if most academics understood the cost of simple search in Twitter. It is very expensive to operate Twitter. When users "agree" to the ToS, they get many advantages, incredibly rich information, a sweet interface, some other specific technological affordances, but no ownership of the data they create beyond the right the delete or personally export the content. Among the TOPICS banned for research by Twitter: IDENTITY. I don't make the point to say we should all bow down before the corporate overlord and stop using Twitter data for research. Quite the opposite, however, because Twitter is an amazing resource. I do think that many academics have a libertarian (at times indefensibly self righteous) viewpoint about the storage, display, and use of Twitter data (especially deleted Tweets) that contributes to the conflict with social data platforms. For sure it was and remains a learning curve for everyone, especially with shifting goalposts moving closer together over time. I want to encourage continued debate as we are only in the birth stages of the ecosystem; the wild west of Internet in history. I have spent years encouraging Facebook & Twitter to let academic teams help them solve critical problems in politics, policy, and beyond. Similarly every week I explain to academics on web meetings why these rules exist and how we try to engineer compliance methods that IRBs can approve. For the most part, the response from Twitter and other platforms is, "we have this covered," and "if an academic was really smart they would be HIRED to work inside the platform." Again, I am not endorsing this viewpoint, far from it, I am just reporting it, but it has been expressed often enough directly to me that it must be a key part of platform management culture. To fundamentally change the playing field, it may come down to resistance versus cooperation with a partner who does not want to play. Just remember, this works both ways. Twitter views academia as stubbornly non-compliant. Academia views Twitter as an evil embodiment of capitalism that just happens to spit out interesting and well formatted data in units that are highly amenable to (too often) fully automated analysis. As Mark Z would say, it is complicated! It reminds me of this animated debate from 2010: https://www.youtube.com/watch?v=PEQQ9r0aC34 Clearly now I have too much time on my hands since Twitter ended our relationship. Not bitter, just sobered. We were studying Digital Citizenship in 2002 with NSF funding when Twitter and Facebook were yet to be born. We have watched the arc of the API-enabled data flow rise and fall. We helped build a 501 c(6) "The Big Boulder Initiative" dedicated to the long term preservation of the social data ecosystem. There was a sustained effort to bring academia into the corporate mix. It eventually failed after Twitter acquired Gnip. Long story. The take away: Academic fingerprints are all over the good, bad, and ugly parts of this history. To just blame Twitter is an ahistorical or ideological act. The truth is much more complicated. If you have a spreadsheet with any potentially deleted Tweets on your hard drive or in the cloud, this is your part of the bigger dilemma and your obligation to fix if you agree with (or comply with) the right to be forgotten. PS: The good new is RSS feeds are free and abundant ;-) and full of valuable data as well. Stu Shulman <https://twitter.com/StuartWShulman>US Soccer C-Licensed Coach On Thu, Feb 28, 2019 at 6:00 AM Alex Gekker <gekker.alex@gmail.com> wrote:
Dear Chris,
I never try to be clever, it's no longer fundable :) Seriously, maybe it came out a bit more flippant than intended - probably partially due to medium-specificity: I typed the original response on a smartphone, thus brevity. I was really just trying to give a contextual example of social media public-ness and subsequent gradient of use in various forms, using this thread as a handy example. Your follow up examples corroborate my initial intent. That being said, I *would* like to screenshot this thread for our MA students and discuss this with them. I would invite the thread to think me with about the ethics of such act, in relation to the original questions. I could email you all a consent form or do an "opt-out" comment (which might be truer to the spirit of the mailing-list-as-platform) or I can crop out the names of the respondents. These are choices that researchers have to engage with (especially in places, like ours, that lack ethics boards and clear procedures).
Alex.
On Thu, Feb 28, 2019 at 9:05 AM Chris Leslie <chrisleslienyc@hotmail.com> wrote:
Dear Alex Gekker,
I think you are making an effort to be clever, humorous, or provocative. However, I don’t think the ethical question posed by the original query goes away because it relates to something we do every day (or that the human subject are doing among themselves). I also think the difference you suggest about private messages and public posts is worth thinking about more.
For instance, what if I published an academic article about the attitudes of Internet researchers and quote your email as an example that reflects x number of posts on the AOIR listserv? Depending on my thesis, you could rightly be offended and potentially harmed by my article, even if forwarding the email chain to your grad students or summarizing the discussion for their benefit might have seemed ok. In the journal article, I am using your statement as a research finding - a generalization of a human trend I have observed that is backed up by quantitative data. Yet in sending your email to the group, you didn’t consent to being a human subject and you were not given the opportunity to give permission afterward. Even if I changed your name and rearranged the words, you could still feel like you were identifiable to members of AOIR who witnessed the conversation.
I interact publicly with humans in everyday life - asking people their opinions about politics or their jobs, for example - in ways that are ethical in interpersonal or educational settings. Those activities don’t involve the IRB or raise ethical concerns because they are not the foundation for research findings. Everything changes when I publish an article with my name and academic affiliation and the implied endorsement of the academic journal. So, to me, the dilemma you pose doesn’t seem to reflect the need for IRB clearance in the first place.
Chris
The Air-L@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/