Godwin’s Bot is a good essay from Misha Lepetic on 3QuarksDaily on artificial intelligence (AI). The essay reflects on the recent Microsoft debacle with @TayandYou, an AI chat bot that was “targeted at 18 to 24 year old in the US.” (About Tay & Privacy) For a New Yorker story on how Microsoft shut it down after Twitter trolls trained it to be offensive see I’ve Seen the Greatest A.I. Minds of My Generation Destroyed By Twitter. Lepetic calls her Godwin’s Bot after Godwin’s Law that asserts that in any online conversation there will eventually be a comparison to Hitler.
What is interesting about the essay is that it then moves to an interview wtih Stephen Wolfram on AI & The Future of Civilization where Wolfram distinguishes between inventing a goal, which is difficult to automate, and (once one can articulate a goal clearly) executing it, which can be automated.
How do we figure out goals for ourselves? How are goals defined? They tend to be defined for a given human by their own personal history, their cultural environment, the history of our civilization. Goals are something that are uniquely human.
Lepetic then asks if Tay had a goal or who had goals for Tay. Microsoft had a goal, and that had to do with “learning” from and about a demographic that uses social media. Lepetic sees it as a “vacuum cleaner for data.” In many ways the trolls did us a favor by misleading it.
Or … TayandYou was troll-bait to train a troll filter.
My question is whether anyone has done a good analysis of how the Tay campaign actually worked?
On the ethos of digital presence: I participated today in a panel launching the Italian version of Paolo Sordi’s book I Am: Remix Your Web Identity. (The Italian title is Bloggo Con WordPress Dunque Sono.) The panel included people like Domenico Fiormonte, Luisa Capelli, Daniela Guardamangna, Raul Mordenti, and, of course, Paolo Sordi.
Continue reading Paolo Sordi: I blog therefore I am
I’ve been meaning to blog this 2014 use of Voyant Tools for some time. Which Words Are Used To Describe White And Black NFL Prospects?. Deadspin did a neat project where they gathered pre-drafting scout reports on black and white football players and then analyzed them with Voyant showing how some words are used more for white or black players.
Continue reading Which Words Are Used To Describe White And Black NFL Prospects?
There are a number of stories about The Malware Museum on the Internet Archive. This archive gathers a number of 1980s and 1990s viruses (just the animated parts) with emulators so you can run them and see their visual effects. The Toronto Star story has a quote from Hyponnen on the art of the early viruses,
“You could call it an art form,” he said in an interview. “These early virus-writers were expressing themselves with animations and sounds.”
It wasn’t until later that viruses started encrypting things and blackmailing you to decrypt them or doing other things to make money.
There is an extended talk (50 minutes) by Mikko Hypponen, the security specialist who gave this collection, on The History and the Evolution of Computer Viruses. The talk starts with the first PC virus BRAIN that he traced back to two brothers in Packistan to Stuxnet. (For a good book on Stuxnet see Kim Zetter’s Countdown to Zero Day.)
The NPR show Planet Money aired a show in 2014 on When Women Stopped Coding that looks at why the participation of women in computer science changed in 1984 after rising for a decade. Unlike other professional programs like medical school and law school, the percent participation of women when from about 37% in 1984 down to under 20% today. The NPR story suggests that the problem is the promotion of the personal computer at the moment when it became affordable. In the 1980s they were heavily marketed to boys which meant that far more men came to computer science in college with significant experience with computing, something that wasn’t true in the 70s when there weren’t that many computers in the home and math is what mattered. The story builds on research by Jane Margolis and in particular her book Unlocking the Clubhouse.
This fits with my memories of the time. I remember being jealous of the one or two kids who had Apple IIs in college (in the late 70s) and bought an Apple II clone (a Lemon?) as soon has I had a job just to start playing with programming. At college I ended up getting 24/7 access to the computing lab in order to be able to use the word processing available (a Pascal editor and Diablo daisy wheel printer for final copy.) I hated typing and retyping my papers and fell in love with the backspace key and editing of word processing. I also remember the sense of comradery among those who spent all night in the lab typing papers in the face of our teacher’s mistrust of processed text. Was it coincidence that the two of us who shared the best senior thesis prize in philosophy in 1892 wrote our theses in the lab on computers? What the story doesn’t deal with, that Margolis does, is the homosocial club-like atmosphere around computing. This still persists. I’m embarrassed to think of how much I’ve felt a sense of belonging to these informal clubs without asking who was excluded.
Feminist Frequency has posted an excellent Speak Up & Stay Safe(r): A Guide to Protecting Yourself From Online Harassment. This is clearly written and thorough discussion of how to protect yourself better from the sorts of harassment Anita Sarkeesian has documented in blog entries like Harassment Through Impersonation: The Creation of a Cyber Mob.
As the title suggests the guide doesn’t guarantee complete protection – all you can do is get better at it. The guide is also clear that it is not for protection against government surveillance. For those worried about government harassment they provide links to other resources like the Workbook on Security.
In her blog entry announcing the guide, Anita Sarkeesian explains the need for this guide thus and costs of harassment thus:
Speak Up & Stay Safe(r): A Guide to Protecting Yourself From Online Harassment was made necessary by the failure of social media services to adequately prevent and deal with the hateful targeting of their more marginalized users. As this guide details, forcing individual victims or potential targets to shoulder the costs of digital security amounts to a disproportionate tax of in time, money, and emotional labor. It is a tax that is levied disproportionately against women, people of color, queer and trans people and other oppressed groups for daring to express an opinion in public.
How did we get to this point? What happened to the dreams of internet democracy and open discourse? What does it say about our society that such harassment has become commonplace? What can we do about it?
Eric Raymond, widely admired for his The Cathedral and the Bazaar, is now peddling social justice paranoia. See Why Hackers Must Eject the SJWs. He starts with the following,
The hacker culture, and STEM in general, are under ideological attack. Recently I blogged a safety warning that according to a source I consider reliable, a “women in tech” pressure group has made multiple efforts to set Linus Torvalds up for a sexual assault accusation. I interpreted this as an attempt to beat the hacker culture into political pliability, and advised anyone in a leadership position to beware of similar attempts.
See his “safety warning” at From kafkatrap to honeytrap. His evidence for this ideological attack seems to be gossip from trusted sources – gossip that confirms his views about “women in tech” and pressure groups and so on. This sort of war rhetoric closes any opportunity for discussion around the issues of women in technology. For Raymond it is now a (culture) war between those on the side of hacker culture and STEM, against “Social Justice Warriors” and what is at stake is the “entire civilization that we serve.”
Why are these important issues being militarized instead of aired respectfully? When did the people we live with and love become the other? Just how confident are we that we objectively know what merit is in the hurly burly of life? What civilization is this really about?
Other reactions to this story include Linus Torvalds targeted by honeytraps, claims Eric S. Raymond in The Register and Is This the Perfect Insane Anti-Feminist Rumor? from New York Magazine.
Jeremie alerted me to a strange debate raging about Dead or Alive Xtreme 3, a sexist beach volleyball game that Koei Tecmo decided not to release in the West, apparently because of concerns about a feminist backlash according to an employee’s comments on Facebook:
Do you know many issues happening in video game industry with regard to how to treat female in video game industry? We do not want to talk those things here. But certainly we have gone through in last year or two to come to our decision. Thank you.
The Guardian has a nice article about the issue with background on adult genres common in Japan, Dead or Alive and otaku culture: why sensitivity is not the same as censorship. Ars Technical has an article too with an update, Dead or Alive publisher denies game is too sexist for Western audiences, that mentions how the publisher Koei Techmo has released an official statement that sort of backtracks on the comments that triggered the issue. Gamespot also has an updated story Dead or Alive Xtreme 3 Won’t Ship Worldwide Due to Sexism Backlash Fears.
Needless to say, this has animated the SJW (Social Justice Warrior) discussion around the representation of women in games and censorship of games. (I should note that it isn’t censorship if a publisher decides to not publish something.) Interestingly, this is not the first time we have had this debate about Japanese adult games across cultures. Brian Ashcraft has an article in Kotaku on Why Is CNN Talking About Rapelay? which documented how the Japanese publishers of adult games were adapting to attention from the West by changing titles and not localizing titles. What has changed is how the West is arguing with itself through Japan. The Japanese seem to be trying desperately not to be part of our culture war.
Today I deposited a Data Management Plan Recommendation for Social Science and Humanities Funding Agencies (http://hdl.handle.net/10402/era.42201) in our institutional repository ERA. This report/recommendation was written by Sonja Sapach with help from me and Catherine Middleton. We recommended that:
Agencies that fund social science and humanities (SSH) research should move towards requiring a Data Management Plan (DMP) as part of their application processes in cases where research data will be gathered, generated, or curated. In developing policies, funding agencies should consult the community on the values of stewardship and research that would be strengthened by requiring DMPs. Funding agencies should also gather examples and data about reuse of archived data in the social sciences and humanities and encourage due diligence among researchers to make themselves aware of reusable data.
On the surface the recommendation seems rather bland. SSHRC has required the deposit of research data they fund for decades. The problem, however, is that few of us pay attention because it is one more thing to do, and something that shares hard-won data with others that you may want to continue milking for research. What we lack is a culture of thinking of the deposit of research data as a scholarly contribution the way the translation and edition of important cultural texts is. We need a culture of stewardship as a TC3+ (tri-council) document put it. See Capitalizing on Big Data: Toward a Policy Framework for Advancing Digital Scholarship in Canada (PDF).
Given the potential resistance of colleagues it is important that we understand the arguments for requiring planning around data management and that is one of the things we do in this report. Another issue is how to effectively require at the funding proposal end something (like a Data Management Plan) that would show how the researchers are thinking through the issue. To that end we document the approaches of other funding bodies. The point is that this is not actually that new and some research communities are further ahead.
At the end of the day, what we really need is a recognition that depositing data so that it can be used by other researchers is a form of scholarship. Such scholarship can be assessed like any other scholarship. What is the data deposited and what is its quality? How is the data deposited? How is it documented? Can it have an impact?
You can find this document also at Catherine Middleton’s web site and Sonja Sapach’s web site.