research | Digifesto

March 13, 2017

industrial technology development and academic research

I now split my time between industrial technology (software) development and academic research.

There is a sense in which both activities are “scientific”. They both require the consistent use of reason and investigation to arrive at reliable forms of knowledge. My industrial and academic specializations are closely enough aligned that both aim to create some form of computational product. These activities are constantly informing one another.

What is the difference between these two activities?

One difference is that industrial work pays a lot better than academic work. This is probably the most salient difference in my experience.

Another difference is that academic work is more “basic” and less “applied”, allowing it to address more speculative questions.

You might think that the latter kind of work is more “fun”. But really, I find both kinds of work fun. Fun-factor is not an important difference for me.

What are other differences?

Here’s one: I find myself emotionally moved and engaged by my academic work in certain ways. I suppose that since my academic work straddles technology research and ethics research (I’m studying privacy-by-design), one thing I’m doing when I do this work is engaging and refining my moral intuitions. This is rewarding.

I do sometimes also feel that it is self-indulgent, because one thing that thinking about ethics isn’t is taking responsibility for real change in the world. And here I’ll express an opinion that is unpopular in academia, which is that being in industry is about taking responsibility for real change in the world. This change can benefit other people, and it’s good when people in industry get paid well because they are doing hard work that entails real risks. Part of the risk is the responsibility that comes with action in an uncertain world.

Another critically important difference between industrial technology development and academic research is that while the knowledge created by the former is designed foremost to be deployed and used, the knowledge created by the latter is designed to be taught. As I get older and more advanced as a researcher, I see that this difference is actually an essential one. Knowledge that is designed to be taught needs to be teachable to students, and students are generally coming from both a shallower and more narrow background than adult professionals. Knowledge that is designed to by deployed and used need only be truly shared by a small number of experienced practitioners. Most of the people affected by the knowledge will be affected by it indirectly, via artifacts. It can be opaque to them.

Industrial technology production changes the way the world works and makes the world more opaque. Academic research changes the way people work, and reveals things about the world that had been hidden or unknown.

When straddling both worlds, it becomes quite clear that while students are taught that academic scientists are at the frontier of knowledge, ahead of everybody else, they are actually far behind what’s being done in industry. The constraint that academic research must be taught actually drags its form of science far behind what’s being done regularly in industry.

This is humbling for academic science. But it doesn’t make it any less important. Rather, in makes it even more important, but not because of the heroic status of academic researchers being at the top of the pyramid of human knowledge. It’s because the health of the social system depends on its renewal through the education system. If most knowledge is held in secret and deployed but not passed on, we will find ourselves in a society that is increasingly mysterious and out of our control. Academic research is about advancing the knowledge that is available for education. It’s effects can take half a generation or longer to come to fruition. Against this long-term signal, the oscillations that happen within industrial knowledge, which are very real, do fade into the background. Though not before having real and often lasting effects.

Leave a comment

August 7, 2014

Protected: the research lately

Enter your password to view comments.

December 20, 2012

An Interview with the Executive Director of the Singularity Institute

Like many people, I first learned about the idea of the technological Singularity while randomly surfing the internet. It was around the year 2000. I googled “What is the meaning of life?” and found an article explaining that at the rate that artificial intelligence was progressing, we would reach a kind of computational apotheosis within fifty years. I guess at the time I thought that Google hadn’t done a bad job at answering that one, all things considered.

Since then, the Singularity’s been in the back of my mind as one of many interesting but perhaps crackpot theories of how things are going to go for us as a human race. People in my circles would dismiss it as “eschatology for nerds,” and then get back to playing Minecraft.

Since then I’ve moved to Berkeley, California, which turns out to be a hub of Singularity research. I’ve met many very smart people who are invested in reasoning about and predicting the Singularity. Though I don’t agree with all of that thinking, this exposure has given me more respect for it.

I have also learned from academic colleagues a new way to dismiss Singulatarians as “the Tea Party of the Information Society.” This piece by Evgeny Morozov is in line with this view of Singulatarianism as a kind of folk ideology used by dot-com elites to reinforce political power. (My thoughts on that piece are here.)

From where I’m standing, Singulatarianism is a controversial and a politically important world view that deserves honest intellectual scrutiny. In October, I asked Luke Muehlhauser, the Executive Director of the Singularity Institute, if I could interview him. I wanted to get a better sense of what the Singularity Institute was about and get material that could demystify Singularitarianism for others. He graciously accepted. Below is the transcript. I’ve added links where I’ve thought appropriate.

SB: Can you briefly describe the Singularity Institute?

LM: The Singularity Institute is a 501(c)(3) charity founded in the year 2000 by Eliezer Yudkowsky and some Internet entrepreneurs who supported his work for a couple years. The mission of the institute is to ensure the the creation of smarter-than-human intelligence benefits society, and the central problem we think has to do with the fact that very advanced AI’s, number one, by default will do things that humans don’t really like because humans have very complicated goals so almost all possible goals you can give an AI would be restructuring the world according to goals that are different than human goals, and then number two, the transition from human control of the planet to machine control of the planet may be very rapid because once you get an AI that is better than humans are at designing AI’s and doing AI research, then it will be able to improve its own intelligence in a loop of recursive self-improvement, and very quickly go from roughly human levels of intelligence to vastly superhuman levels of intelligence with lots of power to restructure the world according to its preferences.

SB: How did you personally get involved and what’s your role in it?

LM: I personally became involved because I was interested in the cognitive science of rationality, of changing ones mind successfully in response to evidence, and of choosing actions that are actually aimed towards achieving ones goals. Because of my interest in the subject matter I was reading the website LessWrong.com which has many articles about those subjects, and there I also encountered related material on intelligence explosion, which is this idea of a recursively self-improving artificial intelligence. And from there I read more on the subject, read a bunch of papers and articles and so on, and decided to apply to be a visiting fellow in April of 2011, or rather that’s when my visiting fellowship began, and then in September of 2011 I was hired as a researcher at the Singularity Institute, and then in November of 2011 I was made its Executive Director.

SB: So, just to clarify, is that Singularity the moment when there’s smarter than human intelligence that’s artificial?

LM: The word Singularity unfortunately has been used to mean many different things, so it is important to always clarify which meaning you are using. For our purposes you could call it the technological creation of greater than human intelligence. Other people, use it to mean something much broader and more vague, like the acceleration of technology beyond our ability to predict what will happen beyond the Singularity, or something vague like that.

SB: So what is the relationship between the artificial intelligence related question and the personal rationality related questions?

LM: Right, well the reason why the Singularity Institute has long had an interest in both rationality and safety mechanisms for artificial intelligence is that the stakes are very very high when we start thinking about artificial intelligence risks or catastrophic risks in general, and so we want our researchers to not make the kinds of cognitive mistakes that all researchers and all humans tend to make very often, which are these cognitive biases that are so well documented in psychology and behavioral economics. And so we think it’s very important for our researchers to be really world class in changing their minds in response to evidence and thinking through what the probability of different scenarios rather than going with which ones feel intuitive to us, and thinking clearly about which actions now will actually influence the future in positive ways rather than which actions will accrue status or prestige to ourselves, that sort of thing.

SB: You mentioned that some Internet entrepreneurs were involved in the starting of the organization. Who funds your organization and why do they do that?

LM: The largest single funder of the Singularity Institute is Peter Thiel, who cofounded PayPal and has been involved in several other ventures. His motivations are some concerned for existential risk, some enthusiasm for the work of our cofounder and senior researcher Eliezer Yudkowsky, and probably other reasons. Another large funder is Jaan Tallinn, the co-creator of Skype and Kazaa. He’s also concerned with existential risk and the rationality related work that we do. There are many other funders of Singularity Institute as well.

SB: Are there other organizations that do similar work?

LM: Yeah, the closest organization to what we do is the Future of Humanity Institute at Oxford University, in the United Kingdom. We collaborate with them very frequently. We go to each others’ conferences, we write papers together, and so on. The Future of Humanity Institute has a broader concern with cognitive enhancement and emerging technologies and existential risks in general, but for the past few years they have been focusing on machine superintelligence and so they’ve been working on the same issues that the Singularity Institute is devoted to. Another related organization is a new one called Global Catastrophic Risks Institute. We collaborate with them as well. And again, they are not solely focused on AI risks like the Singularity institute but on global catastrophic risks, and AI is one of them.

SB: You mentioned super human-intelligence quite a bit. Would you say that Google is a super-human intelligence?

LM: Well, yeah, so we have to be very careful about all the words that we are using of course. What I mean by intelligence is this notion of what sometimes is called optimization power, which is the ability to achieve ones goals in a wide range of environments and a wide range of constraints. And so for example, humans have a lot more optimization power than chimpanzees. That’s why even though we are slower than many animals and not as strong as many animals, we have this thing called intelligence that allows us to commence farming and science and build cities and put footprints on the moon. And so it is humans that are steering the future of the globe and not chimpanzees or stronger things like blue whales. So that’s kind of the intuitive notion. There are lots of technical papers that would be more precise. So when I am talking about super-human intelligence, I specifically mean an agent that is as good or better at humans at just about every skill set that humans possess for achieving their goals. So that would include things like not just mathematical ability or theorem proving and playing chess, but also things like social manipulation and composing music and so on, which are all functions of the brain not the kidneys.

SB: To clarify, you mentioned that humans are better than chimpanzees at achieving their goals. Do you mean humans collectively or individually? And likewise for chimpanzees.

LM: Maybe the median for chimpanzee versus the median human. There are lots of different ways that you could cash that out. I’m sure there are some humans in a vegetative state that are less effective at achieving their goals than some of the best chimpanzees.

SB: So, whatever this intelligence is, it must have goals?

LM: Yeah, well there are two ways of thinking about this. You can talk about it having a goal architecture that is explicitly written into its code that motivates its behavior. Or, that isn’t even necessary. As long as you can model its behavior as fulfilling some sort of utility function, you can describe its goals that way. In fact, that’s what we do with humans in fields like economics where you have a revealed preferences architecture. You measure a human’s preferences on a set of lotteries and from that you can extract a utility function that describes their goals. We haven’t done enough neuroscience to directly represent what humans goals are if they even have such a thing explicitly encoded in their brains.

SB: It’s interesting that you mentioned economics. So is like a corporation a kind of super-human intelligence?

LM: Um, you could model a corporation that way, except that it’s not clear that corporations are better at all humans at all different things. It would be a kind of weird corporation that was better than the best human or even the median human at all the things that humans do. Corporations aren’t usually the best in music and AI research and theory proving and stock markets and composing novels. And so there certainly are corporations that are better than median humans at certain things, like digging oil wells, but I don’t think there are corporations as good or better than humans at all things. More to the point, there is an interesting difference here because corporations are made of lots of humans and so they have the sorts of limitations on activities and intelligence that humans have. For example, they are not particularly rational in the sense defined by cognitive science. And the brains of the people that make up organizations are limited to the size of skulls, whereas you can have an AI that is the size of a warehouse. Those kinds of things.

SB: There’s a lot of industry buzz now around the term ‘big data’. I was wondering if there’s any connection between rationality or the Singularity and big data.

LM: Certainly. Big data is just another step. It provides opportunity for a lot of progress in artificial intelligence because very often it is easier to solve a problem by throwing some machine learning algorithms at a ton of data rather than trying to use your human skills for modeling a problem and coming up with an algorithm to solve it. So, big data is one of many things that, along with increased computational power, allows us to solve problems that we weren’t solving before, like machine translation or continuous speech synthesis and so on. If you give Google a trillion examples of translations from English to Chinese, then it can translate pretty well from English to Chinese without any of the programmers actually knowing Chinese.

SB: Does a super-intelligence need big data to be so super?

LM: Um, well… we don’t know because we haven’t built a super-human intelligence yet but I suspect that big data will in fact be used by the first super-human intelligences, just because big data came before super-human intelligences, it would make little sense for super-human intelligences to not avail themselves of the available techniques and resources. Such as big data. But also, such as more algorithmic insights like Bayes Nets. It would be sort of weird for a super-intelligence to not make use of the past century’s progress in probability theory.

SB: You mentioned before the transition from human control of the world to machine control of the world. How does the disproportionality of access to technology affect that if at all? For example, does the Singularity happen differently in rural India than it does in New York City?

LM: It depends a lot on what is sometimes called the ‘speed of takeoff’–whether we have a hard takeoff or a soft takeoff, or somewhere in between. To explain that, a soft takeoff would be a scenario in which you get human-level intelligence. That is, an AI that is about as good at the median human in doing the things that humans do, including composing music, doing AI research, etc. And then this breakthrough spreads quickly but still at a human time-scale as corporations replace their human workers with these human-level AI’s that are cheaper and more reliable and so on, and there is great economic and social upheaval, and the AI’s have some ability to improve their own intelligence but don’t get very far because of their own intelligence or the available computational resources and so there is a very slow transition from human control of the world to machines steering the future, where slow is on the order of years to decades.

Another possible scenario though is hard takeoff, which is once you have an AI that is better than humans at finding new insights in intelligence, it is able to improve its own intelligence roughly overnight, to find new algorithms that make it more intelligent just as we are doing now–humans are finding algorithms that make AI’s more intelligent. so now the AI is doing this, and now it has even more intelligence at its disposal to discover breakthroughs in inteligence, and then it has EVEN MORE intelligence with which to discover new breakthroughs in intelligence, and because it’s not being limited by having slow humans in the development loop, it sort of goes from roughly human levels of intelligence to vastly superhuman levels of intelligence in a matter of hours or weeks or months. And then you’ve got a machine that can engage in a global coordinated campaign to achieve its goals and neutralize the human threat to its goals in a way that happens very quickly instead of over years or decades. I don’t know which scenario will play out, so it’s hard to predict how that will go.

SB: It seems like there may be other factors besides the nature of intelligence in play. It seems like to wage a war against all humans, a hard takeoff intelligence, if I’m using the words correctly, would have to have a lot of resources available to it beyond just its intelligence.

LM: That’s right. So, that contributes to our uncertainty about how things play out. For example, does one of the first self-improving human-level artificial intelligences have access to the Internet? Or have people taken enough safety precautions that they keep it “in a box”, as they say. Then the question would be: how good is a super-human AI at manipulating its prison guards so that it can escape the box and get onto the Internet? The weakest point, hackers always know…the quickest way to get into a system is to hack the humans, because humans are stupid. So, there’s that question.

Then there’s questions like: if it gets onto the Internet, how much computing power is there available? Is there enough cheap computing power available for it to hack through a few firewalls and make a billion copies of itself overnight? Or is the computing power required for a super-human intelligence a significant fraction of the computing power available in the world, so that it can only make a few copies of itself. Another question is: what sort of resources are available for converting digital intelligence into physical actions in the human world. For example, right now you can order chemicals from a variety of labs and maybe use a bunch of emails and phone calls to intimidate a particular scientist into putting those chemicals together into a new supervirus or something, but that’s just one scenario and whenever you describe a detailed scenario like that, that particular scenario is almost certainly false and not going to happen, but there are things like that, lots of ways for digital intelligence to be converted to physical action in the world. But how many opportunities are there for that, decades from now, it’s hard to say.

SB: How do you anticipate this intelligence interacting with the social and political institutions around the Internet, supposing it gets to the Internet?

LM: Um, yeah, that’s the sort of situation where one would be tempted to start telling detailed stories about what would happen, but any detailed story would almost certainly be false. It’s really hard to say. I sort of don’t think that a super-human intelligence…if we got to a vastly smarter than human intelligence, it seems like it would probably be an extremely inefficient way for it to acheive its goals by way of causing Congress to pass a new bill somehow…that is an extremely slow and uncertain…much easier just to invent new technologies and threaten humans militarily, that sort of thing.

SB: So do you think that machine control of the world is an inevitability?

LM: Close to it. Humans are not even close to the most intelligent kind of creature you can have. They are more close to the dumbest creature you can have while also having technological civilization. If you could have a dumber creature with a technological civilization then we would be having this conversation at that level. So it looks like you can have agents that are vastly more capable of achieving their goals in the world than humans are, and there don’t seem to be any in principle barriers to doing that in machines. The usual objections that are raised like, “Will machines have intentionality?” or “Will machines have consciousness?” don’t actually matter for the question of whether they will have intelligent behavior. You don’t need intentionality or consciousness to be as good at humans at playing chess or driving cars and there’s no reason for thinking we need those things for any of the other things that we like to do. So the main factor motivating this progress is the extreme economic and military advantages to having an artificial intelligence, which will push people to develop incrementally improved systems on the way to full-blown AI. So it looks like we will get there eventually. And then it would be pretty weird situation in which you had agents that were vastly smarter than humans but that somehow humans were keeping them in cages or keeping them controlled. If we had chimpanzees running the world and humans in cages, humans would be smart enough to figure out how to break out of cages designed by chimpanzees and take over the world themselves.

SB: We are close to running out of time. There are couple more questions on my mind. One is: I think I understand that intelligence is being understood in terms of optimization power, but also that for this intelligence to count it has to be better at all things than humans….

LM: Or some large fraction of them. I’m still happy to define super-human intelligence with regard to all things that humans do, but of course for taking over the world it’s not clear that you need to be able to write novels well.

SB: Ok, so the primary sorts of goals that you are concerned about are the kinds of goals that are involved in taking over the world or are instrumental to it?

LM: Well, that’s right. And unfortunately, taking over the world is a very good idea for just about any goal that you have. Even if your goal is to maximize Exxon Mobil profits or manufacture the maximal number of paper clips or travel to a distant star, it’s a very good idea to take over the world first if you can because then you can use all available resources towards achieving your goal to the max. And also, any intelligent AI would correctly recognize that humans are the greatest threat to it achieving its goals because we will get skittish and worried about what its doing and try to shut it off. And AI will of course recognize that thats true and if it is at all intelligent will first seek to neutralize the human threat to it achieving its goals.

SB: What about intelligences that sort of use humans effectively? I’m thinking of an intelligence that was on the Internet. The Internet requires all these human actions for it to be what it is. So why would it make sense for an intelligence whose base of power was the Internet to kill all humans?

LM: Is the scenario you are imagining a kind of scenario where the AI can achieve its goals better with humans rather than neutralizing humans first? Is that what you’re asking?

SB: Yeah, I suppose.

LM: The issue is that unless you define the goals very precisely in terms of keeping humans around or benefiting humans, remember that an AI is capable of doing just about anything that humans can do and so there aren’t really things that it would need humans before unless the goal structure were specifically defined in terms of benefitting biological humans. And that’s extremely difficult to do. For example, if you found a precise way to specify “maximize human pleasure” or welfare or something, it might just mean that the AI just plugs us all into heroin drips and we never do anything cool. So it’s extremely different to specify in math–because AI’s are made of math–what it is that humans want. That gets back to the point I was making at the beginning about the complexity and fragility of human values. It turns out we don’t just value pleasure; we have this large complex of values and indeed different humans have different values from each other. So the problem of AI sort of makes an honest problem of longstanding issues in moral philosophy and value theory and so on.

SB: Ok, one last question, which is: suppose AI is taking off, and we notice that it’s taking off, and the collective intelligence of humanity working together is pitted against this artificial intelligence. Say this happens tomorrow. Who wins?

LM: Well, I mean it depends on so many unknown factors. It may be that if the intelligence is sufficiently constrained and can only improve its intelligence at a slow rate, we might actually notice that one of them is taking off and be able to pull the plug and shut it down soon enough. But that puts us in a very vulnerable state, because if one group has an AI that is capable of taking off, it probably means that other groups are only weeks or months or years or possibly decades behind. And will the correct safety precautions be taken the second, third, and twenty-fifth time?

I thank Luke Meuhlhauser for making the time for this interview. I hope to post my reflections on this at a later date.

Leave a comment

November 18, 2012

What I’m working on

I’ve been a bit scatterbrained this semester and distacted by quite a bit of personal stuff and art/hobby stuff. But I need to reorient myself to my actual work. Unfortunately, I keep running into walls trying to configure emacs with the Solarized theme, which is naturally a prerequisite to doing my statistics homework. So instead of working, I’m going to write a blog post about what I’m working on.

Dissent networks. I’m working with Yahel Ben-David and Giulia Fanti to design an anonymized microblogging service that could work over a delay-tolerant network of wirelessly connecting smart phones in case of Internet blackout, due to natural disaster or government intervention. The idea is that as far as effective anti-censorship technology goes, soft rollout of applications on existing mobile hardware is much more feasible than hard rollout of new mesh networking hardware. And ad hoc mesh networks have serious performance issues at scale, which may in theory be circumvented (or, assumed away?) in a delay tolerant network. We are making a lot of progress on the message prioritization model but there may be some scaling problems hiding behind the curtain…
GeoNode interviews. I’m planning on doing a number of interviews of members of the GeoNode community and related activity so I can write up a report of the project for an academic audience. There are a lot of open questions in ICTD that GeoNode addresses as a case study, and in the process I hope to be able to give back to the community somehow through the research. I’m excited to catch up with old colleagues and learn more about how the project has progressed since I started school. Because of some logistic contingencies at the same time I’m looking into the ideology of the Singularity Institute. It provides an interesting contrast in terms of internet eschatology. I’m not sure how much I want to get into web utopianism in the write-up of this; I think I’d prefer to focus on practical generalizations. But since the purported generalizations of GeoNode are so often utopian (not necessarily in an unattainable way–I’m still a believer) I’m not sure I can dodge the issue.
Figuring out what information is. Maybe because I’m a painfully literal person, it bothers me that I’m at a School of Information where nobody thinks there’s a usable definition of “information”. It’s especially bothersome because there are some really good theories of information out there, including the mathematical ones (originating in Claude Shannon but evolving through Solomonoff, Kolmogorov, Bennett…) and some philosophers (Dretske, Floridi). The biggest barrier to adoption of these definitions in my corner of the world seems to be convincing the social scientists that these concepts are relevant to their work. I think I’ve got a pretty compelling argument in the works though. It takes Andy diSessa’s theory of constructivist epistemology to show that the “phenomenological blending” that leads Nunberg to dismiss the term ‘information’ along with the ‘Information Age’ is actually characteristic of scientific understanding generally (e.g. in learning physics). So I think basically I need to figure out the underlying phenomenological components of people’s naive intuitive theories of information and try to develop a curriculum that reformulates them into the more formal one. I think that if I could pull that off it would make the social scientific and “big data” analytics worlds more comprehensible to each other, maybe.
Bounded symbolic networks. With Dave Tomcik I’ve been working on a paper to try to synthesize Anthony Cohen’s theory of the symbolically constructed community with the kinds of social network theories that are applicable to online social media. This was the work that inspired some of the Weird Twitter performance/experiment/fiasco. It’s funny that that experience shed so much light on the subject for me, but I think I’m going to leave all those insights as out of scope for the paper. It already looks like it may be a stretch to move from Cohen’s sense of ‘symbol’ to the kind of symbol used in digital communication but I’m going to give it a shot. I’m less confident about where any of this goes since I’ve learned since starting the projec that there are so many other contesting theories of community, but I still think its worth trying to combine a couple of them into a single model since so many communities are digitally constituted and in principle algorithmically detectable.
I’ve got to come up with a project for my statistical learning theory course fast.

Currently backburnered but hopefully getting back to work on next semester:

Deception detection on Twitter. With Alex Kantchellian, using LDA to detect ‘deceptive’ tweets (tweets with links to content that aren’t what they say they are about–think rickrolling) and using deception as an indicator for spam detection. Preliminary results good, just need to check it against a larger data set.
Strategic Bayes Nets. I need to rework the paper I did on computational asymmetry and Strategic Bayes Nets with John Chuang and submit it somewhere. After my statistical learning theory course, I’m better prepared to think through the engineering implications of the model. And an economist friend, Gavin McCormick, has taken an interest in applying the model in some behavior economics contexts, which would be sweet.

For some reason I keep getting distracted from the two things I imagine myself really focusing on, which are:

Automated argument analysis of mailing list discussions. Using Bluestocking to analyze argumentation on mailing list (starting with open source projects because it’s where my background is and where there’s a solid benefit to industry, but hopefully generalizing to web standards bodies and collaborative decision making more generally). Idea is to try to build tools that help facilitate consensus or disseminate consensus-building. I think this could be really hard but really awesome if I can pull it off.
Dissertationtron. In my imagination, in the process of writing my dissertation I build a really thin CMS that allows me to logically organize my content, notes, and references while keeping everything open to the web. That way I can publish the content in increments and gather user feedback. Also, I anticipate wanting to build dynamic demos of the concepts and tools used in the process of the dissertation research and making them available through the web. Naturally the whole thing would be backed by a git repository. Ideally I would be running it through the same automated argumentation analysis from the previous bullet point as a way of doing automated tests on my own logic, though to get that to work that may involve some crippling limitations on my vocabulary use and sentence structure while writing.

At this stage in the process, I’m having a hard time distinguishing between pipe dreams, productive work, and academic yak shaving. Interestingly, because there are so many different academic communities to pick from in terms of audience, and so few constraints coming from my department, the real problem (in terms of pulling a dissertation together) isn’t getting work done but in the integration of disparate components. That, and the problem that since I don’t know what direction I’m going in, I don’t have any guarantee that where I end up will be anything anyone will want to hire into an academic setting. Fingers crossed, industry will continue to be an option. I guess it’s good I’m comfortable with bewilderment.

Leave a comment

September 20, 2012

several words, all in a row, some about numbers

I am getting increasingly bewildered by number of different paradigms available in academic research. Naively, I had thought I had a pretty good handle on this sort of thing coming into it. After trying to tackle the subject head on this semester, I feel like my head will explode.

I’m going to try to break down the options.

Nobody likes positivism, which went out of style when Wittgenstein refuted his own Tractatus.
Postpositivists say, “Sure, there isn’t really observer-independent inquiry, but we can still approximate that through rigorous methods.” The goal is an accurate description of the subject matter. I suppose this fits into a vision of science being about prediction and control of the environment, so generalizability of results would be considered important. I’d argue that this is also consistent with American pragmatism. I think “postpositivist” is a terrible name and would rather talk/think about pragmatism.
Interpretivism, which seems to be a more fashionable term than antipositivism, is associated with Weber and Frankfurt school thinkers, as well as a feminist critique. The goal is for one reader (or scholarly community?) to understand another. “Understanding” here is understood intersubjectively–“I get you”. Interpretivists are skeptical of prediction and control as provided by a causal understanding. At times, this skepticism is expressed as a belief that causal understanding (of people) is impossible; other times it is expressed as a belief that causal understanding is nefarious.

Both teams share a common intellectual ancestor in Immanuel Kant, who few people bother to read.

Habermas has room in his overarching theory for multiple kinds of inquiry–technical, intersubjective, and emancipatory/dramaturgical–but winds up getting mobilized by the interpretivists. I suspect this is the case because research aimed at prediction and control is better funded, because it is more instrumental to power. And if you’ve got funding there’s little incentive to look to Habermas for validation.

It’s worth noting that mathematicians still basically run their own game. You can’t beat pure reason at the research game. Much computer science research falls into this category. Pragmatists will take advantage of mathematical reasoning. I think interpretivists find mathematics a bit threatening because it seems like the only way to “interpet” mathematicians is by learning the math that they are talking about. When intersubjective understanding requires understanding verbatim, that suggests the subject matter is more objectively true than not.

The gradual expansion of computer science towards the social science through “big data” analysis can be seen as a gradual expansion of what can be considered under mathematical closure.

Physicists still want to mathematize their descriptions of the universe. Some psychologists want to mathematize their descriptions. Some political scientists, sociologists, etc. want to mathematize their descriptions. Anthropologists don’t want to mathematize their descriptions. Mathematization is at the heart of the quantitative/qualitative dispute.

It’s worth noting that there are non-mathematized predictive theories, as well as mathematized theories that pretty much fail to predict anything.

Leave a comment

September 20, 2012

digital qualities: some meditations on methodology

Text is a kind of data that is both qualitative (interpretable for the qualities it conveys) and qualitative (characterized by certain amounts of certain abstract tokens arranged in a specific order).

Statistical learning techniques are able to extract qualitative distinctions from quantitative data, through clustering processes for example. Non-parametric statistical methods allow qualitative distinctions to be extracted from quantitative data without specifying particular structure or features up front.

Many cognitive scientists and computational neuroscientists believe that this is more or less how perception works. The neurons in our eyes (for example) provide a certain kind of data to downstream neurons, which activate according to quantifiable regularities in neuron activation. A qualitative difference that we perceive is due to a statistical aggregation of these inputs in the context of a prior, physically definite, field of neural connectivity.

A source of debate in the social sciences is the relationship between qualitative and quantitative research methods. As heirs to the methods of harder sciences whose success is indubitable, quantitative research is often assumed to be credible up to the profound limits of its method. A significant amount of ink has been spilled distinguishing qualitative research from quantitative research and justifying it in the face of skeptical quantitative types.

Qualitative researchers, as a rule, work with text. This is trivially true due to the fact that a limiting condition of qualitative research appears to be the creation of a document explicating the research conclusions. But if we are to believe several instructional manuals on qualitative research, then the work of an e.g. ethnographer involves jottings, field notes, interview transcripts, media transcripts, coding of notes, axial coding of notes, theoretical coding of notes, or, more broadly, the noting of narratives (often written down), the interpreting of text, a hermeneutic exposition of hermeneutic expositions ad infinitum down an endless semiotic staircase.

Computer assisted qualitative data analysis software passes the Wikipedia test for “does it exist”.

Data processed by computers is necessarily quantitative. Hence, qualitative data is necessarily quantitative. This is unsurprising, since so much qualitative data is text. (See above).

We might ask: what makes the work qualitative researchers do qualitative as opposed to quantitative, if the data they work with with quantitative? We could answer: it’s their conclusions that are qualitative.

But so are the conclusions of a quantitative researcher. A hypothesis is, generally speaking, a qualitative assessment, that is then operationalized into a prediction whose correspondence with data can be captured quantitatively through a statistical model. The statistical apparatus is meant to guide our expectations of the generalizability of results.

Maybe the qualitative researcher isn’t trying to get generalized results. Maybe they are just reporting a specific instance. Maybe generalizations are up to the individual interpreter. Maybe social scientific research can only apply and elaborate on an ideal type, tell a good story. All further insight is beyond the purview of the social sciences.

Hey, I don’t mean to be insensitive about this, but I’ve got two practical considerations: first, do you expect anyone to pay you for research that is literally ungeneralizable? That has no predictive or informative impact on the future? Second, if you believe that, aren’t you basically giving up all ground on social prediction to economists? Do you really want that?

Then there’s the mixed methods researcher. Or, the researcher who in principle admits that mixed methods are possible. Sure, the quantitative folks are cool. We’d just rather be interviewing people because we don’t like math.

That’s alright. Math isn’t for everybody. It would be nice if computers did it for us. (See above)

What some people say is: qualitative research generates hypotheses, quantitative research tests hypotheses.

Listen: that is totally buying into the hegemony of quantitative methods by relegating qualitative methods to an auxiliary role with no authority.

Let’s accept that hegemony as an assumption for a second, just to see where it goes. All authority comes from a quantitatively supported judgment. This includes the assessment of the qualitative researcher.

We might ask, “Where are the missing scientists?” about qualitative research, if it is to have any authority at all, even in its auxiliary role.

What would Bruno Latour do?

We could locate the missing scientists in the technological artifacts that qualitative researchers engage with. The missing scientists may lie within the computer assisted qualitative data analysis software, which dutifully treats qualitative data as numbers, and tests the data experimentally and in a controlled way. The user interface is the software’s experimental instrument, through which it elicits “qualitative” judgments from its users. Of course, to the software, the qualitative judgments are quantitative data about the cognitive systems of the software’s users, black boxes that nevertheless have a mysterious regularity to them. The better the coding of the qualitative data, the better the mysteries of the black box users are consolidated into regularities. From the perspective of the computer assisted qualitative data analysis software, the whole world, including its users, is quantitative. By delegating quantitative effort to this software, we conserve the total mass of science in the universe. The missing mass is in the software. Or, maybe, in the visual system of the qualitative researchers, which performs non-parametric statistical inference on the available sensory data as delivered by photo-transmitters in the eye.

I’m sorry. I have to stop. Did you enjoy that? Did my Latourian analysis convince you of the primacy or at least irreducibility of the quantitative element within the social sciences?

I have a confession. Everything I’ve ever read by Latour smells like bullshit to me. If writing that here and now means I will never be employed in a university, then may God have mercy on the soul of academe, because its mind is rotten and its body dissolute. He is obviously a brilliant man but as far as I can tell nothing he writes is true. That said, if you are inclined to disagree, I challenge you to refute my Latourian analysis above, else weep before the might of quantification, which will forever dominate the process of inquiry, if not in man, then in our robot overlords and the unconscious neurological processes that prefigure them.

This is all absurd, of course. Simultaneously accepting the hegemony of quantitative methods
and Latourian analysis has provided us with a reductio ad absurdum that compels us to negate some assumptions. If we discard Latourian analysis, then our quantitative “hegemony” dissolves as more and more quantitative work is performed by unthinking technology. All research becomes qualitative, a scholarly consideration of the poetry outputted by our software and instruments.

Nope, that’s not it either. Because somebody is building that software and those instruments, and that requires generalizability of knowledge, which so far qualitative methods have given up a precise claim to.

I’m going to skip some steps and cut to the chase:

I think the quantitative/qualitative distinction in social scientific research, and in research in general, is dumb.

I think researchers should recognize the fungibility of quantity and quality in text and other kinds of data. I think ethnographers and statistical learning theorists should warmly embrace each other and experience the bliss that is finding ones complement.

Goodnight.

1 Comment

March 13, 2012

The instability of adversarial roles

One topic I’m interested in researching is the automated detection of certain kinds of social (or anti-social) activity on the internet. This paper. “Visualizing the Signatures of Social Roles in Online Discussion Groups”, by Welser et al., is a good example of a stab at the problem. Can we look at data from a mailing list and identify the most helpful person on it? Welser thinks so, and they develop a preliminary model for detecting them.

That’s all well and good until the roles get more complicated. A great way to make things more complicated is by introducing an adversarial relationship into the mix. The internet is rife with adversity, in the form of flame warriors, trolls, and spammers. There is also much more benign disagreement as well, but this is probably comparatively rare. Or (this is a broad claim based in cynicism, not research:) people are so likely to take disagreement and conflict personally or dismissively that much legitimate conflict on the Internet is probably seen as flaming, trolling, or spamming.

The problem is that it is very hard to pin down the definitions of these terms. This isn’t just a conceptual problem. It’s also a problem for engineering solutions around these roles. Spam filtering, for example, depends on a certain model of what counts as spam. While training a classifier based on a user’s subjective classification makes lots of sense in some circumstances (like a mail filter), in other cases the line may be less clear. Trolling, meanwhile, can be death to a web-based community. Once could argue that Pinterest has only been successful partly because it has been able to keep the trolls out. But in other contexts, where vigorous debate is encouraged, standards of ‘trolling’ may differ dramatically.

Horse_ebooks is spam content (or, spam detection evasion content) that that has turned into a viral meme. Trolls sometimes become accepted members of a web community, understood to be entertaining and serving as rites of passage to n00bs. So these roles are not necessarily fixed.

With so much data about such roles available for analysis, research into these questions could teach us a lot more about human communication and conflict.

Leave a comment

Tag: research