First, horrifying bullshit!
The freakin' Supreme Court, which is usually pretty on-point... well, there's no non-contentious way to put this. They blamed a woman for her own rape because she was unable to say "No." This woman, L.K., has some pretty severe cerebral palsy and mental retardation; nevertheless, the court ruled that since she couldn't say "No," she may as well have said "Yes." The reason this is bad is because it supports the (false) view that people live in a constant state of "yes". This is, in a really shitty nutshell, what rape culture is all about: it is the responsibility of victims to avoid rape, it is not the responsibility of rapists not to rape. (That's rape culture.) This stems from a backwards idea of "consent" that pervades our society, the idea that "yes" is the default and you have to "opt out" of anything that comes your way. This is backwards because actual consent is opting in. Which means you have to clear something before you do it. Jeez. What is so hard to understand about this?
There is some excellent supplementary reading here: The Pervocracy, on Consent Culture and on Rescripting Sex. Read up, yo.
Second, a crazy shitstorm! (Thanks to Z for linking me Scalzi's follow-up.)
A legitimately awful (but anonymous) Internet Person was recently outed, spoiling his internet anonymity and jeopardizing the livelihood of his wife and kids. Reddit, predictably, blew up that one of their own would be so rocked with a reality check (this is teh intarwebs, after all, where you ought to be able to do anything you want with impunity!). John Scalzi gut-checks them right in their reality, with a multi-point knock-down argument as to why you can't be a shitbag - even on teh intarwebs - and expect to get away with it. Boo-fuckin'-yah.
In case you're not clicking on all the links, which is fine, here's the money quote:
"It's amazing how much you can sweat in a 60 degree office," he said with a nervous laugh.
Judging from his internet footprint, Brutsch, 49, has a lot to sweat over. If you are capable of being offended, Brutsch has almost certainly done something that would offend you, then did his best to rub your face in it. His speciality is distributing images of scantily-clad underage girls, but as Violentacrez he also issued an unending fountain of racism, porn, gore, misogyny, incest, and exotic abominations yet unnamed, all on the sprawling online community Reddit. At the time I called Brutsch, his latest project was moderating a new section of Reddit where users posted covert photos they had taken of women in public, usually close-ups of their asses or breasts, for a voyeuristic sexual thrill. It was called "Creepshots." Now Brutsch was the one feeling exposed and it didn't suit him very well.
You should really click all the links. :)
Third, and finally, Watson is becoming House.
Watson recently beat the pants off of some foolish humans on Jeopardy! Now Watson's natural language processing architecture and immense computational power is being used to crunch through our absurdly large medical literature to calculate confidence intervals for treatment options. All for the benefit of medical patients.
The Jeopardy! competition has a pretty interesting foible - you can't answer less than a quarter-second (I think) after the reader finishes reading the clue. I might be wrong on the exact time interval, but the point is that there's a timing game at work. So Watson, being a computer, is able to "do the timing thing" better than any other human, and in a fundamentally unfair way. The thing is, both Watson and the humans are able to "see"/process the question immediately, and so while a human may "grok" the answer "immediately", Watson has the entire clue's reading time to crunch numbers. For long clues, this gives Watson a distinct advantage. For short clues... I think the playing field is level. It's hard to say, especially when we try to hash out what exactly "short" means. Point is, there's no possible way to universally and unambiguously gauge "understanding the question" among respondents and thereby introduce "genuine and fair" competition in response time. It's the nature of the beast. Meh.
Watson still rocked socks. And in hospitals, this isn't really the kind of problem we're concerned with. And best of all, there are legit humans checking Watson's answers in our real-life hospitals, so we don't have to worry about the odd nonsense answer cropping up because Watson didn't have a better one - human beings will double-check any answer Watson proposes, and if it's something that's "probably" wrong but still might be right, we can look into it. This is the sort of future I look forward to: not rule by humans or rule by machines, but a genuine dialogue between the two. Woo! We live in exciting times!