Silicon Valley Techno-Utopianism

First off I don't know that Paul Ryan is such an expert on government. But second, I think that it makes a lot of sense that a) human behavior is simple and predictable enough that formulae can be developed to chart it, and that b) someone who made his bones designing a human relationship machine that worked so well particularly by treating social interaction like a calculation would be interested in expanding that extraordinary early experiment into something more universal. Like what a shock that Zuckerberg thinks that human interactions can be simplified into a mathematical formula - he's only observed that treating them as though they can has been tremendously lucrative and made his project ubiquitous.

I think that it makes a lot of sense that a) human behavior is simple and predictable enough that formulae can be developed to chart it

This is all well and good, if all you are envisioning is that one may predict certain statistical probabilities across masses of people. That's been going on for quite some time. I think what Zuckerberg suggested might be true was far more radical than that.

I think designing a system that attracts a billion users a day to essentially interact with each other suggests an insight into human behavior and relationships that selling a delicious cookie does not.

― Mordy, Monday, February 27, 2017 12:04 PM (one hour ago) Bookmark Flag Post Permalink

Through evolutionary design, though! While the core news feed style of interface is intact, there are a myriad of design decisions that have been made as tweaks, new feature rollouts, and proposed directions that have either done well and remained integrated or have dropped off the map. And again, while Zuckerberg has some insight as someone who sits on top of this, he's not the one who's had input other than approving or suggesting initiatives for most of the last decade.

And I don't mean to downplay the significance of facebook -- the amount of time and energy I've expended just scrolling through, making small comments, or debating in posts is not a small thing, although it's hardly the only platform where I've done that.

But facebook is completely quantitative as a "human relationship machine." There is little to no qualitative innovation going on -- there are clues via likes and emotions, the amount of traffic a single user gives to a particular post or page, but if it is determined:
- A post is popular
- Significant number of likes, but a much greater number of comments
- The posts receiving the most likes use mostly negative language
I might, as a programmer, think "hmm, this indicates it's a contentious subject and a number of people are against the original topic, maybe?" but that is one of many interpretations of the pattern we see. And we're back to the idea of interpretation -- what is my goal, here?

If I want qualitatively good interaction, I might have someone moderate the conversation. If I want to track issues that might be problems, I might archive the original post, especially if it's to a news article, so we can see if a trend develops. I might prioritize or deprioritize flagging (moderation requests) to this thread due to the high traffic it's getting.

But all of those are at the bottom of the list because the first question is: how many ad impressions are being served up to this conversation, are they relevant, and what's the clickthrough rate? If this is a topic that spurs discussion, how can I encourage this topic to come up regularly? Should I put a link to this topic in the sidebar?

Because while you might claim insight on human behavior, your actual research is mostly centered not on content but on how traffic can drive more traffic. And content that brings traffic. All without caring whether you're devising a technical system to funnel the most attention into the worst spaces, because you claim to be content agnostic, when in fact, the worst content drives the most attention so you are, in fact, biased in favor of horrible content.

this isn't to say facebook is some infinite development of the most effective content->traffic->monetization funnel, but that is the core of the business

(to the extent that actual content is ephemeral -- the traffic is the main product of facebook, and traffic is always positive and never negative)

Sure, but this is like if someone developed a new management technique that got 300% more productivity out of its workers. You might fairly ask whether it's a good thing for the world and the workers to have them working harder/longer hours or whatever but anything that maximizes X, where X is something associated with human behavior, is going to indicate some insight into the behavior of human beings. Like if your contention is that people are unhappier using facebook than they were before facebook, we don't disagree. But humans do all kinds of things that make them unhappy - we're kinda unhappiness inflicting machines, on ourselves, on others. No surprise that things that harness our power to make ourselves unhappy might have a stronger impact on our behavior.

Your description made me envision facebook as a gigantic maze, where the main objectives are to draw people in, then agitate them to run about as much as possible, while constantly redirecting them deeper and deeper into the maze in the hopes they will never leave again. As with every discussion of facebook I have ever read on ilx, this discussion does not inspire me to regret for one second my not joining fb.

lol otm

yeah i don't disagree with any part of that description xp

lol I was just typing an analogy about mazes

you learn about human beings *will do in a maze* and you learn about the crafting of mazes

only if you particularly endeavor to study, say, happiness or fulfillment do you get insight into the human condition. but you have to actually run the stats to look for that! if you sit there optimizing for "engagement" and nothing else, then you're only going to learn about how to engage people. and not to action, necessarily, but to click on an article again or comment

thanks for the interaction, though, guys

bleeding off a lot of emotional energy today, and I'm glad it's going to ilx instead of a site with ad impressions

huh, kinda looks like there's another theranos, this one in nuclear energy (!??!?!???!)

https://www.technologyreview.com/s/603731/nuclear-energy-startup-transatomic-backtracks-on-key-promises/

pretty much all of facebook and human behavior research is boiled down to that one map a guy made where he showed friend linkages across the globe, and it was all the cities connected to all the other cities by a shitload of little lines, except china and iran etc. are all dark - SCIENCE BITCH

you could gain instant credibility by saying "hey guys, this isn't a scam -- we're not promising cold fusion!" xp

to clarify I mean all the human behavior research at facebook - not all of facebook AND all of human behavior research. duh.

I used to follow all kinds of sociologists and human behavior ppl studying online interaction, social network structure, ad hoc hierarchies in the google reader days and I never got the impression anyone doing interesting work was actually at any of the large corps, and if they were, they ended up leaving when their research was curtailed

I think your explanation above plus the maze runner analogy does a great job explaining why anything interesting in those fields is NOT of interest to these firms - they might pay lip service, and maybe even real money for a while, but then it probably becomes quickly apparent that novel social science findings based on proprietary and un-publishable data aren't really useful to anyone - the scientist can't go to conferences except to lurk and the company can't do anything with it because uh sure so that's how we know people like or don't like something? lol we already KNEW THAT

of course I'm talking completely out the butt here so I would love for an actual social scientist / behavioral person at some big network to come correct my horrible assumptions but they're probably not allowed to

I'd imagine a few projects have taken off where someone's proposed, hey, let's set up this new feature and let it ride for a given period and we'll only look at the depth of user interaction and whether they contribute to the community over time rather than how much bulk traffic the page gets. But those are pretty far out outliers!

maybe you aren't familiar with the great Facebook ethics experiment of 2013, where news feed content was manipulated to show specific test individuals only a subset of emotional expression from their friends, to see if that would make their emotion echo what they perceived as the community tone

http://www.pnas.org/content/111/24/8788.full

Can anybody find the results of this?

https://www.cnet.com/news/sorry-facebook-friends-our-brains-cant-keep-up/

lol wtf I had forgotten about that PNAS thing but I think I remember reading about it at the time. fucking asswipes

https://motherboard.vice.com/en_us/article/how-our-likes-helped-trump-win

relevant to yesterday's conversation: third-party company has a facebook app that people happily opt-in to get their personality type, company then gets a bunch of user data (love when you use a site and they want everything, might as well click ok) and then they're able to draw all kinds of inferences based on datapoints

reminds me of a few weeks ago when Uber stepped in it over some issue and they had an apology-style ad on facebook. click on it to block, and it turns out they had paid to show it to people who had liked the ACLU page

yeah that's the Cambridge Analytica psychosocial stuff that's getting sent around. Pretty sure our fucked up Electoral College is still how Trump won and not any bog standard social media data mining tools but every little bit helps I guess, plus Cambridge Analytica is apparently backed by another psycho billionaire and Bannon's on the board so that helps write headlines too (despite the fact that Ted Cruz also spent shitloads of money on CA stuff so how come he isn't presidetn?)

there is no evidence that CA helped or are good at their job, much less that they won the election
https://www.bloomberg.com/view/articles/2017-02-01/trump-s-secret-sauce-is-just-more-ketchup

big data without any understanding of context and behavior doesn't magically produce real world results wtf I don't understand.

bring me more data. and some equations. maybe just one equation actually.

that is fair

I was more into the fact they were able to predict different small aspects of individuals based on affinities, which isn't surprising or complex, but just more jerk stuff you can do because privacy controls are garbage and people don't have a handle on anonymity.

Which isn't predictive data! It's teasing out descriptions based on known correlations. But for writers this is somehow still buzzworthy, and can be confused with predictive analytics

journalists are doing the cambridge analytica pr dept's job for them by loudly assuming they have solved social science with maths.

your prior should be: CA is the Big Data money pit the winning party happened to be throwing money into at the time they won the election

(just workshopping a couple of tweets, let me know if you have any feedback before i tweet them)

good points

there are no particular incidents or locations evaluated, so.. if the losing party had been using CA, would we be still be assuming their methods work in articles?

bring me more data. and some equations. maybe just one equation actually.

http://vignette3.wikia.nocookie.net/evil/images/a/ac/Anti-Life_Equation_Schematic.jpg/revision/latest?cb=20150406013607

https://pbs.twimg.com/media/C58KIXWU4AEBV5U.jpg:small

strong lede

http://www.reuters.com/article/us-snap-ipo-breakingviews-idUSKBN1692E8

time to fire up the old economy into the shitbin thread

That thread is one of your crowning moments Tombot.

Also mookie I think that whole Reuters article was pretty solid.

true

https://www.theatlantic.com/magazine/archive/2017/04/why-is-silicon-valley-so-awful-to-women/517788/

tech utopian bullshit siren sounded at this section:

At long last, the industry that has transformed how we learn, think, buy, travel, cook, socialize, live, love, and work seemed ready to turn its disruptive instincts to its own gender inequities—and in the process develop tools and best practices that other, less forward-looking industries could copy, thus improving the lives of working women everywhere.

I er think you might need to read further (spoiler: it hasn't).

oh I realise there's more said later on, it's the idea that Silicon Valley could ever have been a leader on this front given how ingrained sexism was and is. Also the idea that the various technologies it has proudced are in some sense "disruptive" is also BS, IMO

Also at the Ringer

https://theringer.com/women-sexism-discrimination-harassment-investigations-tech-silicon-valley-uber-452d03ec530b#.mxy3jigk1

A recent New York magazine piece detailed how, often, it is easier for established white women working within tech to speak up. “Fowler’s story, like [former GitHub employee Julie Ann] Horvath’s, comes with implicit privilege. Both women were, fortunately, able to speak out without completely derailing their careers,” Madison Malone Kircher wrote.

before websites you would see that there were no women in the room, or that they were silent or shouted down. on the internet, we screen out the women for you, or you can't tell if they've left

https://www.theguardian.com/technology/2017/mar/14/tech-women-code-workshops-developer-jobs

Serious Q: were there (proportionally) more women posting on ILX during the "zing crew" epoch than now?

do you mean during, or _before_

Techno-utopianism seems like the natural outgrowth of an industry that strongly self-identifies with sci-fi, but has far less intellectual discipline than, say, astrophysics or the space program, and has an incestuous relationship with its own PR hype.

That's a pretty nice formulation, Aimless. I think the causality is more chicken-and-egg than that, though - historically, the techno-utopianism goes back much further than the PR hype. Your 70s hacker types were deep into people like Stewart Brand, Bucky Fuller and maybe Reyner Banham - so the blind spots around technocratic and military-industrial allegiances (and cognitive dissonance with a related hippie libertarianism) are actually baked in from an early point.

despite some elements of science fiction, Ayn Rand isn't usually shelved there

https://www.theatlantic.com/magazine/archive/2017/04/why-is-silicon-valley-so-awful-to-women/517788/

snapchat was literally started in a frat house, so thats a thing

Real question: Is the Valley really worse than other industries, or are they just getting singled out because they claim to value inclusiveness, diversity etc.? I found this paragraph interesting:

"Such bias may be particularly rife in Silicon Valley because of another of its foundational beliefs: that success in tech depends almost entirely on innate genius. Nobody thinks that of lawyers or accountants or even brain surgeons; while some people clearly have more aptitude than others, it’s accepted that law school is where you learn law and that preparing for and passing the CPA exam is how you become a certified accountant. Surgeons are trained, not born. In contrast, a 2015 study published in Science confirmed that computer science and certain other fields, including physics, math, and philosophy, fetishize “brilliance,” cultivating the idea that potential is inborn. The report concluded that these fields tend to be problematic for women, owing to a stubborn assumption that genius is a male trait."

Are we supposed to believe that women don't face the same level of discrimination in law, or medicine? Is that true?

there's discrimination and there's massive under-representation. that sounds like a (plausible) theory for the latter.