Shared publicly  - 
Here's my TEDxVancouver talk on thinking about data in a human context. I'm really pleased with the way this came out.

If you like the talk, please share it - and as always I'd love to hear feedback/questions/ridicule.
Ben Bogart's profile photoRajiv Vaid Basaiawmoit's profile photoCharles Barouch's profile photoAndreas Haardt's profile photo
Thanks for posting, I'm sad I was not able to make it in person.
Hi Jer, I really enjoyed your talk. I always wonder how people can speak so naturally in these contexts without a laptop or notes to cue (teleprompter??)

I can't really relate to the Apple history part, and I never used hypercard, as I was one of those super niche Amiga people. I only retired my Amiga 3000 in 2001 (running NetBSD) when I bought my first PC. Anyone reading this use AmigaVision? It was a interactive multimedia authoring environment where the UI centred on a flowchart on which you drag and drop icons representing events, media etc.. (Amiga Vision (1990))

I specifically wanted to ask about your approach regarding the intersection of art, science and design. There is much hype and effort around visual analytics these days, and how visualization can be used to solve all kinds of problems, and that visualizations should be interactive because there is no one optimum view of data.

I took a class in this ilk, and it could easily have been reduced to a perceptual science class. We spent all of our time considering the science of seeing, and ways of optimizing visual representations for the visual (and cognitive to some degree) system. Why pure blue on black is a bad idea, etc.. Most of this knowledge was already known and implicit in art, it was just not 'objectively' verified.

How do you resolve this low level sensory consideration of a visualization with the much more complex and nuanced notion of constructing meaning through visualization?

I consider science and technology as cultural productions no different than art, just with different methods and purposes. What could be the purpose of science beyond the construction of meaning? Which is also centrally the purpose (or at least question) of art.

How do you resolve science and art in your practise?

I suppose considering science as unitary as I have here is problematic, so perhaps the question is less meaningful. Clearly you use mathematical tools, even models, in your practise, but is that science, or math (which some have argued is more like art than science)? How a scientist or engineer thinks about programming is very different than how an artist does. (As I am constantly reminded by my hardcore AI PhD supervisor)

Somehow we seem to have this baggage that working with technology makes you an engineer (as the applicator of science), bypassing the notion that any technology can be used freely and creatively outside of a scientific methodology. For example the programs of +Pall Thayer as philosophical enquiry and poetry.

I'd love to hear your thoughts.
"For example the programs of +Pall Thayer as philosophical enquiry and poetry." which happen to be on exhibition at the Pace Digital Gallery in New York right now.
Nice synchronicity Pall! You in NYC now?
I'm in the NYC area. I work at Purchase College in Westchester County and therefore live in the 'burbs.
Great talk Jer. My favourite part is pointing out how often we provide data about ourselves to others (2nd and 3rd parties) and how rarely we get access to that same data. I'm surprised there aren't more "social data" services like your project that provide this kind of give-and-get-back data model.
Thanks for the comments - I'm in the midst of an install but will have some time to give some answers to questions sometime tomorrow...!
I finally took the time to watch the talk. Obviously, due to your current employment, it has a very capitalistic bent. It sounds like "we have all of this data, let's turn it into money", which is understandable. I'm sure that's why your employers hired you. But the existence of all of this data reaches far deeper. We have more information about individual beings today than we've ever had before. Our current culture paints a more exact and recordable picture of contemporary culture than anything that has ever been known before. It's not "just" a commodity to be sold to popular media. We can take a snapshot of contemporary culture from so many angles, simultaneously. However, the representation of that data will always be colored by the one who paints it. Let's not forget that.

In many ways, I would say that hypercard was the beginning of the end. Before hypercard, you actually had to gain an understanding of how computers worked. Hypercard removed several levels of that. It was a move towards the machines dictating what you should do with them as opposed to the user telling the machine what to do.
Congrats on the splendid TEDx talk! Very glad that I had the chance to hear you speak in Belgium a couple of years ago. Keep up the good work - Information is beautiful :)

It's pretty hard for me to understand how you could extract a 'capitalistic' message from my talk. Of the handful of projects I showed, only one of those (Cascade, for the NYT - which was conceived and developed solely as a research project) has any kind of profitable motive. Everything else (the Kepler visualization, Colour Economy, Just Landed, the 9/11 Memorial work, OpenPaths) is decidedly non-commercial.

More than that, the entire message of the talk is that we need to think about data in a human context. I talk about data being tethered to the real world, and carrying a 'weight' that makes it more than just a collection of numbers. At the end I argue that we need to have artists, poets, writers, etc. in the dialogue around data if we want to do this as effectively as possible. I've re-watched the video and I really can't find a single point that speaks to "turning data into money." Maybe you can help me here?

As for HyperCard being the 'beginning of the end' - I have a really difficult time seeing your point. As I say in the talk, I think HyperCard was the last piece of software that shipped with a mass market computer that encouraged users to actually program their machines. The intent of HyperCard was for the public to build and program their own applications. Yes, it had a GUI interface, but it also had a powerful, innovative language (HyperTalk) behind it which made programmers out of a lot of people who might not of otherwise ever wrote a line of code.

Lastly, I'll take issue with your suggestion that my employment or my employers somehow forced or encouraged this 'capitalistic bent' that you managed to see in my talk. I have a half-time research position at the NYTimes. I'm there two days a week, and I have an open mandate to pursue projects and research topics at my own discretion. Outside of that, I'm an adjunct professor at ITP, and I have a software art practice that I've been cultivating for the better part of a decade. I've never said anything in public as the mouthpiece for an organization, and I never will.


Jer, there are a few researchers in my lab (at UBC) trying to mathematically model the spread of information in a social network. As you said in your talk, good data is rare and valuable, so I think they've had trouble assessing the validity of these models.
Pall, I would say that perception of Jer being capitalistic in any way is quite bent. Having watched Jer's video, he's pretty clear on bringing the human element into data and his work demonstrates the very thing that you are saying needs to happen. There are only a hand full of people out there today taking snapshots of the different perspectives of our lives and culture and Jer is one of the avant-garde artists of today working in this very realm.
Michael - yes, it is tricky to find good data. We're really interested in (and actively pursuing) data sets to apply Cascade to apart from the one (Twitter) that we are currently using. Ethan Zuckerman has made the point that Twitter is very much like the fruit fly - while it's a great experimental animal (mostly due to ease of access to data), it's dangerous to extrapolate our findings there into the broader 'social space'.
+Ben Bogart - haven't forgotten about you. Just waiting until I have a few more free minutes to get my thoughts collected around your questions.
Great talk, +Jer Thorp!
But please do not use that "dinosaur analogy". Archaeologist don't "dust off dinosaur bones". That would be paleontologists...
Andreas - I know, I know! Literally as that came out of my mouth I knew it was wrong. :) My apologies to paleontologists (and achaeologists!) everywhere.
Regrading Pall's comments... Controversial or not, these comments seem to have blossomed into a discussion! (always good.) While I can't specifically say that I agree with Pall that your talk is about "turning data into money", I think he does have a point.

As Jer discusses, data does have meaning, it is a consequence of a lived experience. The question of capitalism is not so much in the turning data into money because data is already money (an abstraction of value in the world that can be manipulated for specific aims). Analysis of all kinds of data are the only way for capitalism to continue to grow and be effective (and arguably any other enterprise that depends on money).

Where I think Pall's point falls is the question of where this data comes from. The organizations with the resources, and the most to gain from collecting this kind of data, are for-profit enterprises. Of course governments are also included here, but there is a move toward greater and greater privatization in the US and Canada. In fact the current government in Canada has scraped the "long form" census, which greatly decreases the amount of data about the people of Canada available to government to inform policy. Perhaps the grand notion is that even the data governments need will be increasingly offloaded to private collection and analysis. Corporate entities increasingly broker our communication and social relations (google+ case in point), and lets not kid ourselves, this is all about the data. How much data really comes from non-corporate non-profit enterprise? Even scientific data and knowledge is increasingly private property.

Of course Jer did mention OpenPaths as a alternative to corporate data-collection. Which is really a nice idea, and I'm happy that it emphasizes the notion of donating data (as apposed to unwittingly giving it).

I don't think that power and data can be separated. Even having data proves a certain degree power over that being measured, and the guise of objectivity (for me objectivity is about considering what is measured as equivalent to what is real and True, deemphasizing the filtering and intentional effort to choose something to be measured, and how). Jer is self-identified as a "data" artist, and as such is choosing to be deeply steeped these issues of control.

While I'm really happy that Jer is moving in this direction of meaning and the human aspect of data, I don't think that the power issues can be ignored. I don't think that you can be an artist who visualizes data X without dealing with the politics of how X was constructed. There seem to be only two options in the case of big data: (A) A powerful entity (government) collects data in order to make the world better for the people data is collected from. (B) A powerful entity (for-profit corporation) collects data in order to extract wealth (exploit) those from whom data is collected.

Jer, how do you (if you do) deal with these issues of power and objectivity as a data-artist?
Ben: I partly disagree with your 3rd paragraph. While I can see your point about a movement towards greater privatization of personal data, the fact is that much of this data wouldn't exist without the enterprises recording it. Responsible enterprises are transparent about what data is collected, and allow the end-user to easily download a copy.

Where I disagree is the comment about governments and science. For public data, there's a huge movement towards open data, with governments and government organizations at all levels publishing data and sponsoring developer competitions ( and
Michael, I'm not sure where you are going with "the fact is that much of this data wouldn't exist without the enterprises recording it." It seems like your saying that its better the data is there than if its not, no matter who collects it? If this is your argument, then you've missed the point about power structures. No data is objective, it is always laden with ideology that manifests in am implicit choice as to what is important and what is not. Without knowing those assumptions the data is misleading.

The ability for the end-user to download their own data is irrelevant, as it makes no difference to how the data is collected and why. It does (have the potential) to tell the user what is being collected (what the collector deems important), and therefore open the door for the user making critical choices, but if they can't opt-out of the system where are those choices manifest?

You are right about a trend for open data in government. I think this is really great. There is a big gap between meaningful use of the data and providing links to huge CSV files. This is complex and required a bridge made up of tools, education and general data literacy. My fear is that private interest will fill this gap with apps that help you find the Urban Outfitters in the neighbourhood with the least crime.
I was trying to make the distinction between a) ownership of existing personal data is transferred to a private enterprise, and b) personal data is created/collected by a private enterprise.

My understanding is that in your previous post you were talking about a), where as I believe b) is more common. I don't think I made a moral judgement on either of these.

I also agree entirely with your highlighting that choice is an important component alone with transparency in data collection.

On open data: I think you can rest easy that there are plenty of socially-motivated programmers willing to fill that gap, and they're off to a great start:,,,
Michael, Thanks for the open-data links. I have not been keeping up to date.

I don't really make a distinction between A and B, because often it seems like data is passively collected about us (by our phones for example) without an explicit choice (regardless if its transferred to apple or third parties, or not) I'm not sure about how copyright (ownership) fits in with this. I think its a fair analogy to photographing a person. The photographer is collecting identify information about a person. In many jurisdictions if that photograph is taken in public space, there is no privacy issue, because the information (identity) was already placed in the public, so photographing it is acceptable. The act of photography does mean that the information (one representation of your face, but not all representations if it) is the copyright of the photographer. Photographing private space is another issue and strictly controlled by whoever owns the space.

Apparently in France the law has changed and your image (even in public space) is yours. So no photographer has the right to even capture it, let alone own a representation of it. I think this discussion is important in relation to personal data. Presumably the space of your phone is private, because it is not accessible to anyone, but you don't really own it entirely yourself either, because it does things (like collect data) that you did not ask it to. Who "created" the data?

In a photograph, your face was not created by anyone, its just yours because its attached to your body/identity, but the photographer makes a representation of it, which he owns.

In the phone, your data is created by software not written by you, but is attached to your body and identity (your location in space for example). The visualization of that data could be like photographs, owned by someone else easily.

What about grouping data from multiple people together, is that a representation that can be owned as independently of the pieces? Because the software on your phone technically created a representation of the movement, is that data owned by the company that created it?

In the case of a social network it appears to be somewhat public (many people can see your data) but its not really public, because the space is owned and controlled. We retain the copyright to what we post here, but google+ can do pretty much anything they want with it until the end of time.

How is posting words on google+ like moving around in space with a phone that tracks your location? Do you own your movements in space the way we own our specific arrangements of words?

Personally I don't believe in the ownership of the immaterial at all:

I think the bottom line about data is two-fold: (1) Data is real, its meaningful and human, like Jer says. (2) The spaces in which we construct our identities are increasingly private. If the collector of data owns it (as apposed to the person who manifests it) then data is increasingly "owned" by large institutions who by definition have the power to control those spaces. This reflects the continuing erosion of public space in all aspects of life.
You've shown me that the definition of who creates the personal data is much more nuanced than I would have initially thought (the photography analogy is useful). In the case of the phone, I think there could be a strong argument that you, as owner of the phone and licenser of the software, created the data. It just so happens that the license terms usually specify that data is shared with the company.
Oops. "Capitalistic" was a very bad choice of words. It sounds far more offensive than I intended. "Commercial bent" is what I meant.
I don't think it's much of a secret that TED, under Chris Anderson, is a profit-making enterprise. I don't think it's a perfect organization by any means - their recent 'Ads Worth Spreading' program in particular raises my hackles.

2 things, then:

This talk was from an independently organized TEDx event, which was 100% run by volunteers and from which (as far as I know) no one made any money. Of course it's wrapped in a brand that gets some benefit from the event, but I know at the heart of the event the organizers had very little (or no) commercial intent.

I do a lot of public speaking in a year - lectures at NYU and other schools, workshops, talks at various conferences, artsist talks, etc. Part of this is because I think well on stage - my talks aren't rehearsed and I find I get some insight from talking through my ideas.

Mostly, though, I want to balance the 99% commercial discourse we're having around data with some discussion about ethics, humanity, critical thinking. So, my intent in saying yes to these things (which rarely pay any money - and when they do it's usually in the low hundreds of dollars) is to try to get these conversations happening more broadly. I think this works - the video that we're talking about in this thread has been viewed 3,000 times in three days, which is represents generously more people than I'd be able to get in front of in person during any given year.

Which I guess leads us to a hypothetical question for you, Ben - would you speak at TED if they asked you to come and talk about your work? Knowing what you do about their corporate practices? And also knowing that 500k - 1M people would hear what you wanted to say?
I like your talk there. We need a less technical approach to data. Thanks.
Hi Jer, I would like to hear more about the art-science relation in your work, and how you relate to the power structures in play in the collection of data.

As for your question. I've thought about it over the last few days, without doing any further reading/research. I would certainly be inclined for the opportunity. I would also make a distinction between TED and TEDx, since there seems to be little, or no, monetary connection between them.

The final answer is that I would certainly give serious though to presenting and not reject it outright. That being said I would be more than tempted to be critical of any TED policies I find ethically troublesome (if there indeed are such policies).
I think the discussion about power systems and data is a subset of the main one that (I hope) comes out of my talk - that data are not stranded numbers but instead are tied to the real world.

I always strongly encourage my students to think about the provenance of data - where did the numbers come from? How were the measurements made? What are the politics driving their production? These questions are essential to a successful negotiation with any source of data.

The Openpaths project is certainly an attempt to reverse the typical power relationship that exists with personal data these days - in which a corporate entity has ownership and control over the data. We see the project in many ways as a proof of concept for a system in which this ownership order can be reversed; and we hope that this might give people a little taste for what it's like to own and control their own data.
Hi Jer, Indeed your talk did refer to it, but I don't recall you taking a strong stand on how much data is unwittingly collected about us. Indeed its a large and complex discussion. It would be interesting to see a side note on the documentation of these data artworks which includes an analysis of power relations in play with the data used.

The art-science question was my main one...
Thanks for the link, +Ben Bogart it certainly brings up some alarming cases of the misuse of curation; one of the reasons I prefer an open model (e.g. Android) vs. walled garden (e.g. Apple). As we discussed earlier, it takes hacktivist/whistleblowers to actually make use of this openness to discover and report misuses.

As a computer scientist though, I can't help but point out the inaccuracy in the descriptions of Turing's halting problem and Godel's incompleteness. It actually is possible to prove that a program is correct, although impractical for all but the most trivial programs!
+Michael Lawrence Cory is no computer scientist, but certainly a very competent media theorist (amazing how few of them I agree with at all). How many media theorists even know of Godel's theorm(s)? You should comment on locusmag to set him straight. ;) Seems to me these statements were setting up a context of complexity for the article, and not core arguments.
+Ben Bogart Done. I agree that I was being nit-picky. The inaccuracy doesn't in any way detract from the strength of the article.
I am an omnivore in the computer world. I spent time on Hypercard (Mac), CanDo (Amiga), Delphi (Windows), and on and on. I've been in the business database world for most of my life. I have to call a bunch of you out one point. Please note: THERE IS NOTHING EVIL ABOUT GETTING PAID. Please stop apologizing for making money by keeping important topics and conversations alive.

My kids like to eat nearly every day. To feed them, I use the F[r]ee model, where I use what works, instead of waging war on free software of fee software. I donate to freeware projects because those folks need to eat, too. There is noting noble about condemning capitalism while using Google+. You did not pay to be here, but someone is certainly paying for the servers, bandwidth, et al.

Just to be clear, I am a career technologist. I have a pony tail. I have extensive knowledge of PHP & MySQL. I am OLDER thatn PHP & MySQL. I still believe that good people can give some time away AND make money without having to beg forgiveness for either. When we abandon ourselves in the pursuit of money... that is evil. Working for a living is quite honorable.
+Charles Barouch There is nothing evil about getting paid. There is something evil about getting paid via the exploitation of others. Money just abstracts production from the buyer, so you don't have to look who you are exploiting in the eye. Your brand of pragmatism is the very same that leads to exploitation. Without idealism there is no ethics.
+Ben Bogart You assume all income is "equally evil" and that misses the point. When I worked in garment, we imported from third world countries and we consistently raised the standard of living in each place we went. At the same time, we also manufactured domestically, creating jobs in our own country.
I'm sure you can decide that what we did overseas was exploitation and that what we did domestically was pollution, but you've probably worn some of our shirts. See, I'm betting you eat food, wear clothes, and use a computer. That makes you culpable for all the evil you have funded. Surprise. It also makes you culpable for the good that money has done.
Some of it fed families that would have gone hungry, some rebuilt communities. You can sit there and buy from 'evil' companies all day long while trying to hold the rest of us to some theoretic standard, but the facts is: not all income is evil, not all companies are unethical, and you probably don't think of your own role as you buy socks at Walmart.
Love the Bogart Vs. Barouch discussion - and in many ways it could be just a question of perspective. Nevertheless, I agree with both on certain points and being a scientist, a social worker and a recent MBA graduate all at the same time I can share that business has always been there - and without business life, technology and all developments that humanity benefits from today are of course not possible. However, the issue of money being evil only comes when income generation becomes your sole focus and greed takes over. However, while there are examples of greed and profit-above-all-else industries, there are an equal number of examples of industries taking care of social impacts, non-profits, volunteers - and I believe it is this balance (or counter force) which makes the world an enjoyable place to live in :)
Anyway, without detracting much from the topic of this post - which is about Jeremy's talk on humanising data - I would actually like to ask Jer Thorp one question. Is this only about making data more visual and empathic or is it also about making it simple? The reason I ask is - as a scientist - I can very well see and appreciate the artistic impact, the humanising of the data and of getting it all down into a relatively more manageable scale. However, wouldn't the data still remain too complex still in a layman's perspective. What are your thoughts or experiences on this? Have you encountered comments from your general readers for example, that your visualisation has actually made more sense for them or made it easier for them to grasp the data or did they still remain complex for them?
+Rajiv Vaid Basaiawmoit - I didn't have the heart to tell Ben that I work for an institution that uses charitable contributions to protect 254 acres of plant life, maintain one of the largest nature libraries in the world, and defend seventeen acres of old growth forest. In my other job, I usually end up working with companies who are in financial trouble and keep them running so they can keep paying their employees. Oh, and I speak at events without compensation beyond the occasional 'we will pay for your room.'
F[r]ee is a model which works. Fee & Free both have a place in arts, sciences, technologies, everywhere.
That was a breathtaking TED-Talk +Jer Thorp I have seen it now three times. Thank you for that. Everything is clear now.
I'm a database programmer in a company, that has a lot of flight-data available. I'm dreaming of a day, to put this data in such beautiful animations.
You have done that very well!
Add a comment...