I think as technologists we have a natural preference for clean technological solutions that don't involve politics or messy interpersonal relationship issues. Failing to confront that whole human sphere of life leads to negative consequences everywhere, and the talk does a great job of pointing out some of the macro social effects.
On a similar note but a more localized scale I'm reminded of this talk on one of the downsides of moving to microservices being that people can avoid difficult conversations https://youtu.be/kb-m2fasdDY?t=8m2s
By the way, everything on his site is absolute gold, I can't remember a time I was so awed by someone's blog. Even his resume is brilliant and hilarious!
I'm not a SV resident, but I've visited several times, over decades, and this rings true...
"Hop on BART after the conference and take a look at Oakland, or take a stroll through downtown San Francisco and try to persuade yourself you're in the heart of a boom that has lasted for forty years. You'll see a residential theme park for tech workers, surrounded by areas of poverty and misery that have seen no benefit and ample harm from our presence. We pretend that by maximizing our convenience and productivity, we're hastening the day when we finally make life better for all those other people."
And historically there's a lot of evidence this can work. But not if gains are concentrated in the hands of the already rich. Which had been the case in America for decades.
That seems a bit of a truism, though - if wealth gains go to the poor, naturally they will become wealthier. Trickle-down economics imagines exactly that if the wealthy get the pie, they'll dole it out.
We could debate the point of trickle-down, but the rich are not creating the poor. Something else is.
SF has spent 330 million dollars to tend to their 15k homeless people. 22k a year. I dont know about you but I dont see them living 22k a year lives. And I see syringes, severely sick and mentally ill people walking the streets everyday.
I mean, I feel like tech people have this sense of moral exceptionalism and don't understand the value of moral consensus in a democracy, as opposed to moral obligation by noblesse oblige.
Something isn't only right or wrong in a democracy. You have to also build consensus through boots on the ground, not by expecting everyone to have moral revelations by magic. So, is there consensus in America that inequity and poverty are issues that rise to the level of national discussion?
I would say no.
So how do you go about having an interesting and durable agenda of attacking American inequity and poverty in an extra-legal way when you don't have consensus in a democracy? By flexing tech money muscle? Do tech people, excluding elites, have that much money, that they can buy their way to morality without first doing consensus work?
Tech people in SV and the Bay live in a world where they experience consensus as a series of badly broken systems. The deep, durable pockets of poverty and misery? Built and maintained by democratic consensus. The needs-much-improvement public schools of SF and Oakland? Built and maintained by consensus. The state of BART? Built and maintained by consensus.
People on average don't like poverty, misery, gentrification, unreliable public transit, and spiking rents. But on average, people around here do seem to like all the things that contribute to those. Prop 13, "neighborhood character" concerns halting housing construction, and bus stops every block are all popular.
I can't claim to speak for everyone here. But personally? I see a lot of broken systems that I cannot fix. Where my attempts to help fix things for everyone are not only ineffectual, but actively unwelcome.
In light of that, I do what I can. I make myself comfortable. If I cannot address the suffering of others, I can at least limit my own suffering. No lives are improved if I join in the suffering of others to indulge some bizarre impulse to self-flagellate. My own suffering is not a moral imperative.
Well said. Rent control, prop 13, NIMBY all things that deepen the terrible consequences that are seen everyday.
And its crazy to me to visit Seattle right now and also find my way walking next to so many mentally ill homeless people.
I'm not big on most socialized solutions, but taking care of mentally ill people roaming the streets is one of those that are very easy to agree to. Yet the topic everyday is Trump this or that,while you had to walk past 10-20 people to get to the office.
There is some massive cognitive dissonance going on, that I think I can easily think and talk about because im a foreigner.
It's hard to take care of the mentally ill who live on the streets, especially in California. They cannot be legally confined for treatment until they are deemed a threat to themselves or others, and that's a pretty high bar. Until then, it's up to them to seek treatment. Many don't want it or can't handle the organization / discipline to stay treated.
As a result, it's not nearly as tractable a problem as it seems.
Yet cities like buenos aires with rampant poverty don't have this issue and it doesn't nearly as much to solve that particular problem.
It is a problem of policy and one that affects both the homeless and the non homeless people tremendously. I guess if the couple that had kept the presidio street and were to put up the homeless tents there , the problem would find a solution much faster.
As many others, its a problem of political will. What is unique, is that people everyday have to deal with homeless people in SF and yet its not a top agenda. Smoking pot is tho.
I didn't mean to place moral judgment on you; but I did mean to question why there is special moral onus being placed on the tech community.
Of course as an individual, one is naturally free to assist in the suffering that is proximal to them, given ability and desire.
But there's also a place for policy, which means organizing and coalition-building, as SF surely has no shortage of churches and non-profit organizations for social justice.
I also think that people aren't necessarily so "bearish" with regard to their attitude about democratic power. Surely President Trump is an example.
"As computer programmers, our formative intellectual experience is working with deterministic systems that have been designed by other human beings. These can be very complex, but the complexity is not the kind we find in the natural world. It is ultimately always tractable. Find the right abstractions, and the puzzle box opens before you.
The feeling of competence, control and delight in discovering a clever twist that solves a difficult problem is what makes being a computer programmer sometimes enjoyable."
I recently asked a friend whether he thought people were deterministic. He didn't think so (I don't either.) What do you think?
>I recently asked a friend whether he thought people were deterministic. He didn't think so (I don't either.) What do you think?
It's the wrong question. People are deterministic, but that doesn't mean they're always predictable.
In some sense, it's trivially true: we're physical systems, so of course* we're deterministic. The underlying physics that govern how the subatomic particles that make up the atoms that make up the molecules that make up the brain, is pretty well established at this point -- so if someone wants to assert some kind of special exemption for human brains: "everything else in the observable universe obeys physical law, but this is different!" the burden of proof is going to be pretty damn high. And nobody has ever put forth anything like real evidence for such an exemption -- the fact that "we don't understand how the brain works" doesn't mean it doesn't work in some fundamentally knowable way.
However, that doesn't mean that brains are (at least with current or reasonably-foreseeable-future technology) "ultimately tractable" even in principle, much less that we have any way of getting inside brains to see what people are going to do in real time.
And the reason people care about whether people are "deterministic" or not is because we want to know how to influence them, and for what we should assign praise/blame etc., and there's a common hidden assumption that determinism removes our ability to do that in some way. But I think that assumption's false (you don't see determinists just throwing up their hands and declaring ethics don't matter). There's a massive amount of physical evidence for determinism and no evidence at all against -- I'm not sure how anyone with a science background could believe otherwise at this point.
* "Predictable" here means something like "predictable by other humans, including via reasonably foreseeable technology". If we're in a simulation, the simulator can, by construct, "predict" the whole thing.
I'm not sure why you think the underlying physics is deterministic. That idea was the state of the art understanding in say the 19th century (billiard balls and Newtonian physics), but quantum mechanics, at the beginning of the 20th century, fundamentally changed the understanding of reality at the lowest level.
For a good layman's introduction I would suggest, "QED
The Strange Theory of Light and Matter", (1986) by Feynman. Hard ideas but clear prose. I like to read it every few years to remind myself of the fundamentals.
> I'm not sure why you think the underlying physics is deterministic.
If conclusive evidence were found that human consciousness is fundamentally dependent on QM processes to function, that would be a Nobel Prize-worthy discovery. Aside from some minor papers exploring the topic, no such evidence has been found AFAIK.
The functioning of chemistry in individual cells definitely are influenced by quantum effects. Over a short period of time the functioning of the neurons in the human brain will be influenced by QM enough that the billiard ball, clock work, deterministic view of the future unfolding is not correct.
It isn't enough that the processes are affected. The processes must be affected in such a way that the idea of determinism at other scales is compromised.
In any event, positing some sort of "QM consciousness" does nothing to save the concept of free will. Where is the "free will" in a fundamentally random phenomenon? Whether one accepts determinism or not, the notion of free will is incoherent.
> The functioning of chemistry in individual cells definitely are influenced by quantum effects.
So? Transistors used in modern computers directly depend on quantum effects (semiconductor physics was an outgrowth of QM) as well, yet we do not need to take into account QM when writing software or designing microprocessors.
Where will my body be in five years? Quantum effects will definitely be manifest enough that the answer to that is not deterministic. Maybe because I am not a professional philosopher I'm missing the special meaning of the work deterministic as used here. I think the point of the OP was that computers are a specially designed system that is very complicated, but at the bottom created by humans who have spent a lot of time and effort to make it deterministic unlike the the "real" world.
At my end, I wonder where the author ever saw deterministic computer systems. In my experience, you have to work really, really hard to get even a system computing outputs from inputs without processing any external events in-between to produce deterministic results, with things like floating point hardware and software and parallelism APIs, not to mention your own bugs, working overtime against you - to say nothing of faulty DRAMs and other such delightful phenomena.
I'm pretty sure that having to account for segfaults, keyboard interrupts, power outages, and other kinds of 'real world' exceptions that have to be caught means that the code might be deterministic but the computer is exposed to unexpected stimuli.
People probably act in a mostly deterministic when there are rules constraining them: sports, board games, social environments, lines. Maybe we are deterministic on a neurotransmitter level. But there's little control over external stimuli.
People are not 100% deterministic, but evidences also show that expert con man and similarly persuasive people can trigger emotions and thoughts in others reliably.
No contradiction between the two. Human beings have natures, habits, character, etc. That fact that, say, depriving people of food will tend to produce hunger does not contradict the idea that we have a measure of freedom in our choices. Of course, a simplistic, mechanistic view of the word will leave no space for anything but blind stuff bumping into other blind stuff, but that picture is wrong.
A lot of the criticisms against SV these days echo Jonathan Swift's portrayal of the scientific communities of his day: https://en.wikipedia.org/wiki/Laputa
> Techies will complain that trivial problems of life in the Bay Area are hard because they involve politics. But they should involve politics.
... SNIP ...
> In a world where everyone uses computers and software, we need to exercise democratic control over that software.
Tighter control over targeted advertising by Facebook won't suddenly help with crappy zoning or infrastructure projects designed to maximize the extraction of wealth from taxpayers while minimizing the results.
The author correctly points out that the Bay Area's problems are political in origin, and that tech companies and technologists in general have been reticent to get dirty with politics.
However, their solution, putting tech companies under more democratic control, does not solve the problems specified.
Raytheon and Lockheed Martin's work is under democratic control. As are the zoning policies which keep out poor and middle class citizens from the few parts of the country which have benefited from our economic growth. The companies that charge our local state and federal governments 10x what Europeans to build roads, bridges, mass transit and build other projects are very tightly controlled by our democratic systems.
Facebook and Google are not preventing people from building housing. They are not building tools for the government to round up "undesirables", nor are they building drones that blow up civilians. They are building technology that allows advertisers to more accurately sell people stuff.
It should be obvious to anyone at this point that this could be used for nefarious purposes. But so can any number of other technologies developed over the past two centuries. The 20th century was a testament to how much murder humans can commit when given the opportunity, and we did not need machine learning to make it happen.
I'm not sure I'm following your points. You're moving back and forth between enormous historical and current problems and sweeping the platitudinal meat under the rug.
It's easy to point to those problems and say,"Yeah, but this is nowhere near what it used to be!". If a global, location and interests tracking system looks like a duck, then it's probably a duck.
Disclaimer: I work for an advertisement company. Opinions are my own.
> I'm not sure I'm following your points. You're moving back and forth between enormous historical and current problems and sweeping the platitudinal meat under the rug.
The problems America faces are largely controlled by democratic systems.
Google, Facebook, et al are neither the cause nor the solution to the problems described by the author of the post.
If we decide to turn ad networks into surveillance systems used to commit crimes against humanity it will be because our democratically elected representatives were told to do something by voters and they did so.
If you don't mind me asking, what sort of political, civics or activism work have you done? What you're saying rings true to a point, but it's not useful to the conversation and comes off as armchair naivete.
One thing I dislike about articles like this is that they get you to agree on some basic point, such as, "Hey did you notice that there is a lot of wealth AND poverty in San Francisco?" Yes, I agree with that. "Well, actually, it's libertarians' and techies' fault!" Oh okay, you were correct about the first part so you're correct about the second too.
"Also, by the way, they're just like Hitler / Stalin / thinly veiled shout-out to something really bad!"
> Earlier attempts to create a rationalist Utopia failed for interesting reasons, and since we bought those lessons at a great price, it would be a shame not to learn them.
I do agree that San Francisco has this strange juxtaposition of impoverished hellscape versus obscene wealth. It's the main reason I plan to leave here within the next couple of years. But I'd do myself a disservice if I latched onto a thought that this is the fault of some political ideological minority in the area, or even the majority (bleeding hearts). This "poverty and misery" is the techies' fault in the same way that alcoholism is the fault of bartenders.
All that said, I agree with the conclusion about data. There should be a moral obligation to limit or stop this storage. It creeps me the hell out, and it's why I don't use Facebook or Google any more, knowing that this doesn't fully solve the problem. I'll add that the metaphor he depicts is really great:
> As a technologist, this state of affairs gives me the feeling of living in a forest that is filling up with dry, dead wood. The very personal, very potent information we're gathering about people never goes away, only accumulates. I don't want to see the fire come, but at the same time, I can't figure out a way to persuade other people of the great danger.
I just wish the writer were mature enough to make that argument without the scapegoating of ideology, or the wrong ideology. This isn't about libertarianism or "trickle down" (from HN comments) or capitalism or a dislike for politics, it's about technological progress. Ideologies that are opposed to the writer's scapegoats will also act in service of technological progress when it is convenient for them.
It's not hard to read his point. For startups, the bay area is nearly almost the first market to launch in. Uber, Lyft, Airbnb, numerous food delivery services, or anything aimed at the SMB market, all of this goes in production in bay area first and this place is closer to the ultimate digital utopia than anywhere else in the country. His point is that it doesn't make "the world a better place", presumably because all value is captured by the capital owners [his comparison to industrial revolution]
47 comments
[ 2.7 ms ] story [ 107 ms ] threadOn a similar note but a more localized scale I'm reminded of this talk on one of the downsides of moving to microservices being that people can avoid difficult conversations https://youtu.be/kb-m2fasdDY?t=8m2s
https://news.ycombinator.com/user?id=idlewords
I highly recommend browsing, it's great writing.
Apparently, "Paul Graham is a weenis," though. http://www.idlewords.com/2005/04/dabblers_and_blowhards.htm#...
N.B. if you click on the parenthesized .com you get a link of everything from that site
"Hop on BART after the conference and take a look at Oakland, or take a stroll through downtown San Francisco and try to persuade yourself you're in the heart of a boom that has lasted for forty years. You'll see a residential theme park for tech workers, surrounded by areas of poverty and misery that have seen no benefit and ample harm from our presence. We pretend that by maximizing our convenience and productivity, we're hastening the day when we finally make life better for all those other people."
"the best way to end poverty is for me personally to be rich"
SF has spent 330 million dollars to tend to their 15k homeless people. 22k a year. I dont know about you but I dont see them living 22k a year lives. And I see syringes, severely sick and mentally ill people walking the streets everyday.
Something isn't only right or wrong in a democracy. You have to also build consensus through boots on the ground, not by expecting everyone to have moral revelations by magic. So, is there consensus in America that inequity and poverty are issues that rise to the level of national discussion?
I would say no.
So how do you go about having an interesting and durable agenda of attacking American inequity and poverty in an extra-legal way when you don't have consensus in a democracy? By flexing tech money muscle? Do tech people, excluding elites, have that much money, that they can buy their way to morality without first doing consensus work?
Which is somewhat ironic considering the many parallels between that and how colonialism was justified.
People on average don't like poverty, misery, gentrification, unreliable public transit, and spiking rents. But on average, people around here do seem to like all the things that contribute to those. Prop 13, "neighborhood character" concerns halting housing construction, and bus stops every block are all popular.
I can't claim to speak for everyone here. But personally? I see a lot of broken systems that I cannot fix. Where my attempts to help fix things for everyone are not only ineffectual, but actively unwelcome.
In light of that, I do what I can. I make myself comfortable. If I cannot address the suffering of others, I can at least limit my own suffering. No lives are improved if I join in the suffering of others to indulge some bizarre impulse to self-flagellate. My own suffering is not a moral imperative.
And its crazy to me to visit Seattle right now and also find my way walking next to so many mentally ill homeless people.
I'm not big on most socialized solutions, but taking care of mentally ill people roaming the streets is one of those that are very easy to agree to. Yet the topic everyday is Trump this or that,while you had to walk past 10-20 people to get to the office.
There is some massive cognitive dissonance going on, that I think I can easily think and talk about because im a foreigner.
As a result, it's not nearly as tractable a problem as it seems.
It is a problem of policy and one that affects both the homeless and the non homeless people tremendously. I guess if the couple that had kept the presidio street and were to put up the homeless tents there , the problem would find a solution much faster.
As many others, its a problem of political will. What is unique, is that people everyday have to deal with homeless people in SF and yet its not a top agenda. Smoking pot is tho.
Of course as an individual, one is naturally free to assist in the suffering that is proximal to them, given ability and desire.
But there's also a place for policy, which means organizing and coalition-building, as SF surely has no shortage of churches and non-profit organizations for social justice.
I also think that people aren't necessarily so "bearish" with regard to their attitude about democratic power. Surely President Trump is an example.
"As computer programmers, our formative intellectual experience is working with deterministic systems that have been designed by other human beings. These can be very complex, but the complexity is not the kind we find in the natural world. It is ultimately always tractable. Find the right abstractions, and the puzzle box opens before you.
The feeling of competence, control and delight in discovering a clever twist that solves a difficult problem is what makes being a computer programmer sometimes enjoyable."
I recently asked a friend whether he thought people were deterministic. He didn't think so (I don't either.) What do you think?
It's the wrong question. People are deterministic, but that doesn't mean they're always predictable .
In some sense, it's trivially true: we're physical systems, so of course* we're deterministic. The underlying physics that govern how the subatomic particles that make up the atoms that make up the molecules that make up the brain, is pretty well established at this point -- so if someone wants to assert some kind of special exemption for human brains: "everything else in the observable universe obeys physical law, but this is different!" the burden of proof is going to be pretty damn high. And nobody has ever put forth anything like real evidence for such an exemption -- the fact that "we don't understand how the brain works" doesn't mean it doesn't work in some fundamentally knowable way.
However, that doesn't mean that brains are (at least with current or reasonably-foreseeable-future technology) "ultimately tractable" even in principle, much less that we have any way of getting inside brains to see what people are going to do in real time.
And the reason people care about whether people are "deterministic" or not is because we want to know how to influence them, and for what we should assign praise/blame etc., and there's a common hidden assumption that determinism removes our ability to do that in some way. But I think that assumption's false (you don't see determinists just throwing up their hands and declaring ethics don't matter). There's a massive amount of physical evidence for determinism and no evidence at all against -- I'm not sure how anyone with a science background could believe otherwise at this point.
* "Predictable" here means something like "predictable by other humans, including via reasonably foreseeable technology". If we're in a simulation, the simulator can, by construct, "predict" the whole thing.
For a good layman's introduction I would suggest, "QED The Strange Theory of Light and Matter", (1986) by Feynman. Hard ideas but clear prose. I like to read it every few years to remind myself of the fundamentals.
If conclusive evidence were found that human consciousness is fundamentally dependent on QM processes to function, that would be a Nobel Prize-worthy discovery. Aside from some minor papers exploring the topic, no such evidence has been found AFAIK.
So? Transistors used in modern computers directly depend on quantum effects (semiconductor physics was an outgrowth of QM) as well, yet we do not need to take into account QM when writing software or designing microprocessors.
People probably act in a mostly deterministic when there are rules constraining them: sports, board games, social environments, lines. Maybe we are deterministic on a neurotransmitter level. But there's little control over external stimuli.
... SNIP ...
> In a world where everyone uses computers and software, we need to exercise democratic control over that software.
Tighter control over targeted advertising by Facebook won't suddenly help with crappy zoning or infrastructure projects designed to maximize the extraction of wealth from taxpayers while minimizing the results.
However, their solution, putting tech companies under more democratic control, does not solve the problems specified.
Raytheon and Lockheed Martin's work is under democratic control. As are the zoning policies which keep out poor and middle class citizens from the few parts of the country which have benefited from our economic growth. The companies that charge our local state and federal governments 10x what Europeans to build roads, bridges, mass transit and build other projects are very tightly controlled by our democratic systems.
Facebook and Google are not preventing people from building housing. They are not building tools for the government to round up "undesirables", nor are they building drones that blow up civilians. They are building technology that allows advertisers to more accurately sell people stuff.
It should be obvious to anyone at this point that this could be used for nefarious purposes. But so can any number of other technologies developed over the past two centuries. The 20th century was a testament to how much murder humans can commit when given the opportunity, and we did not need machine learning to make it happen.
It's easy to point to those problems and say,"Yeah, but this is nowhere near what it used to be!". If a global, location and interests tracking system looks like a duck, then it's probably a duck.
Disclaimer: I work for an advertisement company. Opinions are my own.
The problems America faces are largely controlled by democratic systems.
Google, Facebook, et al are neither the cause nor the solution to the problems described by the author of the post.
If we decide to turn ad networks into surveillance systems used to commit crimes against humanity it will be because our democratically elected representatives were told to do something by voters and they did so.
The tools they're building and the data they're collecting could certainly be used for that purpose.
"Also, by the way, they're just like Hitler / Stalin / thinly veiled shout-out to something really bad!"
> Earlier attempts to create a rationalist Utopia failed for interesting reasons, and since we bought those lessons at a great price, it would be a shame not to learn them.
I do agree that San Francisco has this strange juxtaposition of impoverished hellscape versus obscene wealth. It's the main reason I plan to leave here within the next couple of years. But I'd do myself a disservice if I latched onto a thought that this is the fault of some political ideological minority in the area, or even the majority (bleeding hearts). This "poverty and misery" is the techies' fault in the same way that alcoholism is the fault of bartenders.
All that said, I agree with the conclusion about data. There should be a moral obligation to limit or stop this storage. It creeps me the hell out, and it's why I don't use Facebook or Google any more, knowing that this doesn't fully solve the problem. I'll add that the metaphor he depicts is really great:
> As a technologist, this state of affairs gives me the feeling of living in a forest that is filling up with dry, dead wood. The very personal, very potent information we're gathering about people never goes away, only accumulates. I don't want to see the fire come, but at the same time, I can't figure out a way to persuade other people of the great danger.
I just wish the writer were mature enough to make that argument without the scapegoating of ideology, or the wrong ideology. This isn't about libertarianism or "trickle down" (from HN comments) or capitalism or a dislike for politics, it's about technological progress. Ideologies that are opposed to the writer's scapegoats will also act in service of technological progress when it is convenient for them.