Episode One Hundred and Seventy Eight: Who’s Afraid Of Infrastructure; White Hats; Utopias; Chinese Room Mobs
by danhon
0.0 Sitrep
1.0 Who’s Afraid of Infrastructure?
So it wasn’t that there was a US citizen in Texas who had contracted and died of ebola, it was that there was a nurse who had now contracted it who had been involved in treating him. It was that the existing systems of notification had failed: the hospital initially said that the information about their patients travel had been captured but not communicated to the doctor, and then that lapse was blamed, in a way, on the electronic healthcare record software used by the hospital. But we know that software is built to reflect need and working practices.
Instead, it feels like another one of those vulnerabilities in how the human mind works and processes risk: I can read articles about how checklists work and how their adoption in hospitals is doing wonders for patient safety. About how sharpies being used to mark *this leg*, not that leg, are helping to reduce the (rare!) occurrences of wrong-limb-surgery. And, you know, I can read about how MRSA was totally a thing that required changing working practices and a reminder that it might be a good idea to keep hospitals clean-clean, as opposed to cheaply-clean.
2.0 White Hats
[1] Your next project needs a white-hat jerk – Ross Floate
3.0 Utopias
I’d say that my intent was not necessarily to shift the *blame* on to marketing and advertising, just that I got super excited about what Raven had said, and was looking for other factors as well.
Raven’s key point here, I think, is this:
Facetime solves the problem of being able to say goodnight to your kids when travelling on business, but it doesn’t solve the problem of a business culture that expects you to spend shit-loads of time away from your young family — which is a systemic problem with many other side-effects besides keeping you away from home, and one that technology tends to exacerbate at the molar scale, even (if not especially) when it seems to ameliorate it at the molecular scale. Or, to put it crudely: in order for an iPhone to make your life easier, a number of Chinese workers must make their lives rather more difficult. The benefits of technology are not at all evenly distributed, and neither are the downsides.
I wonder if the proposition that Raven is putting forward is, or has to be true: that for an iPhone to make my life easier, a number of Chinese workers must may their lives more difficult. Someone like Marc Andreessen might say that the same Chinese workers are more or less lining up to have their lives made rather more difficult, but that those difficult lives are easier (or maybe just differently difficult) than the rural lives that they might have had before and progress, of a sort, is being made. Or that the example of the manufacture of the Neo Lucida[6] from Golan Levin and Pablo Garcia in their XOXO talk when they went to China to meet the people making it – by hand – who were very proud of the work that they were doing. And that’s without even considering the fact that Chinese workers might not have had to make the iPhone in the first place, but that’s the price we’re willing to pay. As Levin and Garcia remind us, pretty much *everything* we buy now, is hand-made and touched by human hands. The idea that robots assemble things for us is more or less a fallacy.
Raven’s call is for a flattening and a more vigorous, honest definition of the problem rather than a vague waving at a sort-of-everything. So what if it’s difficult for us to keep the whole thing in our heads? Make new things to make it easier to keep the thing in our heads. Try to understand that which we need understanding. Build those giant data-viz rooms as Bret Victor appears to be calling for so that we can use our pattern-matching to tease apart relationships in ways that our brain architecture might be better suited to understanding.
[1] (DIS)ASSEMBLING #STACKTIVISM; POKING HOLES IN UTOPIA – Paul Raven
[2] Episode One Hundred and Seventy Seven: We Will Sell it To You Wholesale (TinyLetter archive link, because I am so far behind on the archives at newsletter.danhon.com)
[3] Make technological utopia easier with this one weird trick – Paul Raven
4.0 Chinese Room Mobs
From Elementary Penguin on Metafilter[1]:
So GamerGaters prowl Twitter for any negative mention of GamerGate, find appropriate copypasta from their secret stash that used to be on Girhub, and spam replies, right? I just realized that this is exactly Searle’s Chinese Room. So does any part of the system actually understand what it’s doing?
So, here’s the gist: a mob evolves or discovers the behaviour that to communicate, propagate and distribute its message whilst its members perceive that it’s under attack, it uses an elementary form of pattern matching. When it sees the enemy saying a certain thing, match against that pattern, look up a response that’s been pre-written and simply post that response.
The gist of Searle’s Chinese Room argument, one that Searle used against those who believed that human consciousness could be replicated as a series of simple-to-understand processes, was that you get a bunch of people, lock them in a room and give them a set of easy to understand instructions like, if you see this symbol, show this other symbol. No one person in the room *understands* Chinese, but if you get the rules right (and enough people) you can feed the room Chinese and out comes English and it kind of works. Searle says: this is a system, and it’s not conscious.
The argument about AI and consciousness isn’t really the point of this particular observation, more that once you have people who are treating a medium as broadcast and aren’t involved in understanding or conversation, they appear to be acting like a bot, a thought that Fred Scharmen is pretty sure that he’s had since *at least* 2009[5] in the form of social media encouraging bots (also: brands) to act like humans and humans to act like bots.
Saitta’s point from her tweets – that not all communications mediums are about communications, that there are channels that people use for broadcasting and explicitly not listening, and that we’ve seen this before in speech divided by class where there are private channels for listening.
But, I still think there’s a thing here, which is this: when individual components of a mob start *acting* like a Chinese Room in a certain medium (ie: they aren’t just Chinese Room components everywhere, just in a certain space) conversation disappears from that medium. In a way, this presents itself as a the “these people are impossible to have a conversation with” problem because the conversation isn’t one: it’s a conscious choice of the other party deciding to not understand what’s being said by the responding party, and to instead resort to pattern-matching and talking points and then…
And then when we talk about talking-points and staying on message, isn’t that something that we learned from PR management of politicians? Find a keyword in your interviewer’s spiel, match it to a talking-point and then spew out that talking point. There’s no wonder people like Jeremy Paxman get annoyed and ask the same question and Ian Katz, the former editor of BBC’s Newsnight has written about this death of the political interview when one side has been reduced to performing like automatons[6]
Because: how do you argue with something that isn’t conscious, or, in other words, is *choosing* to not be conscious?
Saitta’s suggestion – that you go to where you can be assured that you’ll be listened to and *understood* – in respect of Gamergate is what you get when Jesse Singal goes back to Reddit to counter accusations of his article on the same being poorly researched[7]. Singal eviscerates and engages with members of the movement on their own turf, which, for what it’s worth, doesn’t *appear* to degenerate into name-calling.
So part of the question is this: why does this happen on Twitter? Is this something that only happens on Twitter, or on all broadcast networks? Is this behaviour – copy-paste-Chinese-Room-ish behaviour an emergent behaviour of limited-content networks that have a high susceptibility to virality and pass-along and where certain types of content like hashtags are rewarded when they get repeated with high frequency?
[1] Insomnia Thoughts
[2] https://twitter.com/hondanhon/status/523922780631617537
[3] https://twitter.com/sarahjeong/status/523923453275930624
[4] https://twitter.com/dymaxion/status/523992887025864704
[5] https://twitter.com/sevensixfive/status/524054829849919488
[6] The death of the political interview – Ian Katz, The Financial Times
[7] Another poorly-researched hit-piece, from the Boston Globe – singal commenting on Reddit
—
3:21pm, Central Time, somewhere on the way to Salt Lake City after having given the opening keynote at the HOW Interactive Design Conference in Chicago yesterday. Opening keynotes are pretty scary – you don’t have anything to refer back to and it’s kind of up to you to kickstart things and make them happen. There’s nothing to react to yet. And then the stereotypical hiding in the hotel room afterwards for the post-talk comedown.
I hadn’t written for a while and it was becoming a Thing in my head, and I was worried that I just wouldn’t anymore. Or that people would accuse me of not writing anymore because the writing had “done its job” and I’d, well, gotten a job, neither of which are strictly 100% true – well, I *do* have a job, but it’s not like the writing has lost all value to me. And it’s true that the writing did help me get the job, but I’d like to think a whole bunch of other things helped me get the job, too. But, honestly, more a matter of my brain being overloaded with essentially being jacked-into-the-matrix of the new job, learning a whole bunch of new relationships and concepts and teasing them all back and forth until I get some sort of clarity.
So. An endeavour for more writing.
Dan