Q&A: Andrew Sempere

By: Joshua Glenn
February 18, 2010

Andrew Sempere is a Cambridge, Mass.-based design researcher with IBM’s Collaborative User Experience Group/Center for Social Software. He’s also an artist whose interactive works have been exhibited (most recently) at the 2009 Boston Cyberarts Festival and the 7th Champ Libre Manifestation Internationale Video et Art Electronique in Montreal. As background for a panel on “indirect collaboration” he’s moderating for this year’s SXSW Interactive conference, HiLobrow.com’s Joshua Glenn interviewed Sempere — about online collaboration, for business and aesthetic purposes alike. This Q&A was cross-posted from the SXSW blog Indirect Collaboration, where a slightly different version was published earlier today.

GLENN: You work with social software researchers at IBM, helping them make smart decisions about encouraging or discouraging content sharing and reuse. How is doing so important to successful, inclusive business collaboration?

SEMPERE: My role is mostly to help the researchers think through user experience to develop an interface that helps support the work they are trying to do. I personally have a particular interest in the content sharing idea, but it’s not always that, although it does usually come down to figuring out what design principles to apply in order to help the users and the researchers accomplish their goals. Since we mostly make prototype systems (and not customer facing products) I also consider a big part of my job to be keeping an eye on what is happening in the social software space outside of the corporation, to try and understand early and see if there is anything that can be adapted for use inside the enterprise.

GLENN: Some skeptics might say that sharing and reuse are antithetical to business culture…

SEMPERE: It depends on your definition of business, and of sharing! Usually when people say this, they mean “socialization” as in, “Isn’t allowing my employees to socialize giving them allowance to goof off.” Or the more formal “What’s the ROI on social software.” I think these questions are the wrong ones to ask. Companies have long recognized that social interaction is important to business — certainly at the executive level. Business is fundamentally about interpersonal relationships.

But if you are looking for a quantifiable cost savings as a result of providing your employees access to social software, you’re probably not going to find it. It doesn’t work that way.

This is a long way of saying: yes! If your definition of business culture is essentially Taylorist, then social software is completely antithetical, but so is Internet Culture. This collision leads to what I call Compliance Theatre (more on that later). Taylor’s model of standardized work practices is about getting rid of the outliers. Imagine a bell curve… the purpose of standardization is to focus single mindedly on the middle of the curve, where things are average. You don’t want the low end of the curve, and as a result you also get rid of the high end of the curve. In many organizations this turns out to be exactly what people want. You probably don’t, for example, want MBTA operators texting each other, no matter how boring it is to drive exactly the same route in exactly the same way. You need their focus, and you need standard operation for safety.

In other organizations, eliminating the outliers is disastrous. Research, design, any field where you want creativity or are particularly interested in high quality output: you need to allow your employees to take risks.

There are other issues more specifically related to sharing and distribution which are diametrically opposed to certain business models. The music industry, for example, and the publishing industry to a certain extent. These industries were predicated on the notion of gatekeeping by controlling distribution channels, but in an era of nearly free distribution via the internet, they are mostly useless. This is not to say that music or books are useless, just the idea that you can profit off of distribution by selling little containers of culture — it’s a broken idea.

GLENN: Speaking of gatekeeping, according to Eric Raymond’s 1997 formulation, there are two methods of developing software (and by extension, anything): the “cathedral” approach, in which a group of experts do the work, and the “bazaar” approach — a cooperative activity initiated by members of the public. It seems to me that crowdsourcing software — which permits experts/gatekeepers to outsource part of their project to members of the public via a call for contributions — has been received so enthusiastically in business circles because it promises a third way, a synthesis of the cathedral and the bazaar. But is this enthusiasm misguided?

SEMPERE: I’m not entirely sure it’s a synthesis, or just a case of the clergy of the cathedral occasionally buying things from the bazaar. TopCoder is the best example that I know of. The main takeaway for me from the cathedral and the bazaar argument is that it’s a symbiotic (maybe codependent) relationship. Both must exist. Open source software works because professionally employed software developers have spare cycles. Second Life works because most of the residents have other sources of income. I think business is becoming more comfortable with the idea that this means they can take advantage of these spare cycles to get high quality work out of motivated individuals. It makes sense and it does work, at the same time it’s not a sustainable business model, it’s just a good way to accomplish some tasks. There are also serious issues from a business perspective related to licensing and such, but in general I think it’s great that businesses are thinking flexibly. I do sometimes worry that it potentially devalues professional work, but I even if that is true, there’s no stopping it.

GLENN: You’re also interested in artistic collaboration. Last year, you published a paper analyzing the fraught psychological and cultural implications of making art within Second Life, whose object-creation tool assumes individual ownership as a prerequisite to the creative process. From the SL user’s perspective, why shouldn’t an artwork, like any other object, be governed by a commodity-style trading system?

SEMPERE: Second Life itself is not crucial to the fate of the universe, but it is very unique, and as such represents an excellent place to see where things might go (or might go wrong) as we continue down a path of virtualizing our culture. We humans, at least in the developed world, are moving anything and everything we can online, and developing new content, ideas and tropes that never existed in the analog world. This is not a bad thing at all, but I am deeply concerned that at the same time we are relinquishing ownership of our output to private organizations whose end goals and values are often accidentally preventing that culture from breathing properly, and in some cases smothering it.

A very small but significant portion of the content on Second Life is art. It is culturally significant, and it deserves a chance to play a part in art history. Even the “lowbrow” content is often part of someone’s personal history and deserves some kind of respect.

The way that SL is designed, content is created by individuals, but owned by a private for-profit entity whose goal is (and perhaps should be) to make money. This isn’t a bad thing, but the money making goal and the user created content goal interact in some strange ways:

I’m not a lawyer, but the ToS for Linden Lab (which apparently Lawrence Lessig consulted on), reads like a huge problem to me. Linden gives you full rights over your intellectual property, but they do NOT give you rights over the instantiation of that property. That is to say, if I build a house in SL, I own the idea of the house, but not the record of the house as it exists on Lindens servers. The idea is a good one — apparently it was Lessig’s influence that ensured it wasn’t the standard “LL owns everything.” But, Linden has backed itself in a corner here. They promise ownership of IP, but they have also styled themselves as the content police, and are in fact being sued for failing to properly protect the “rights” of certain content creators. These individuals represent an extreme minority of the SL user base, but they, by virtue of invoking lawyers and DRM, are making world-changing decisions within the walls of SL.

Here’s the doomsday scenario: The company goes under, the servers are sold. Since the ToS relinquish IP rights, whatever is left of the LL company cannot use the content, nor can any of the creditors. In fact, the content of the servers is worse than useless, it’s a legal liability, since the ToS (might) open them to lawsuits. The safest course of action from a business perspective would therefore be to delete all the content. Literally, they will burn down a library, and ten years of collective work will be gone forever.

I mentioned I’m not a lawyer, and I actually hope that I’m wrong about my reading of the situation, but even if I am I think the example stands: extremely complicated negotiations are occurring right now in spaces where we are entrusting our cultural artifacts. These discussions are of crucial importance and worth keeping an eye on, especially since we are seeing a dismantling of public libraries and other shared institutions in favor of virtual archives controlled by private organizations.

Incidentally all of this virtual talk might be hard to fathom for people who haven’t spent much time online, but try and imagine if Kodak asserted ownership over that box of super 8 movies in your parents’ attic just because they made the film. It’s already true that it’s hard to view these technologically, but imagine that one day you pulled down the box to discover that every frame of your recorded memory had been deliberately erased, not because of time, or rot, or bugs, or neglect, but because of a licensing dispute at the Kodak company.

GLENN: Are we really moving towards this possibility?

SEMPERE: Yes — for example, the recent #amazonfail where Amazon retracted versions of Orwell’s book that they didn’t have the rights to, AFTER they were sold. iPhones and presumably iPads all have a built-in remote “kill switch” which allows them to delete apps and data from your device. The company has promised not to use this, but they reserve the right to, not to mention the fact that these days when I buy DRM music from itunes it’s nothing like when my parents bought records. I won’t be able to leave them for my kids to listen to without making sure they have my login, password and the right hardware, and I have a hard time imagining finding iTunes version 7 at a garage sale (especially since software is now distributed online, rather than via “dead media”) the way you might find an old 8-track or phonograph.

We’re moving towards a model where were we can casually consume content from the cloud and most of our structures of control are embedded in software. In many ways this is wonderful, but we need to be very very careful about what we’re coding into the system if we don’t want to risk accidentally creating a dark age.

There is also a secondary shorter point to be made here: by forcefully insisting (as Second Life does) that creative practice be monetized you exclude the outliers and push towards lowest common denominator content. Rather than make something interesting and daring, you will trend towards saleable, which means copying the tried and true. It’s Taylorism again: terrific for selling widgets, terrible for culture. Linden has taken several steps to encourage more revenue (because they need it). I have friends who work there, and I don’t want them losing their jobs either, but aggressively chasing every revenue stream, especially in a micro economy, completely destroys innovation.

GLENN: From the online arts community Learning to Love You More, which takes Fluxus-style assignments (e.g., “Recreate an object from someone’s past”) out of the gallery and onto your desktop, to Takashi Kawashima and Aaron Koblin’s “Ten Thousand Cents,” which used Amazon’s Mechanical Turk service to outsource the digital painting of a $100 bill to thousands of turkers (who could each only see their own section of the picture), artists have started using “Web 2.0” (for lack of a better term) tools originally created for crowdsourcing or social networking as a spur to innovation and creativity. Are projects like these the antidote to Second Life’s approach? Or do they also fail, albeit in a different way?

SEMPERE: They fail spectacularly but are beautiful for it. Fluxus is close, but yeah, I think the whole point of crowdsourcing an art piece is to play with the idea of an artwork as singular vision. The people who are contributing to the work (especially in the Mechanical Turk example) have unknown motivation, or are motivated by micropayments. Second Life artwork is still incredibly traditional in its approach (i.e., a single individual or close team working towards a single idea). None of this is problematic exactly, there’s room for both, but what Second Life is missing is exactly what the other examples have — the ability to share and collaborate in a “Web 2.0” manner. As I said earlier, I don’t think SL is the end-all of art in 2010, but as an example of what is possible it frustrates me for exactly that reason. There is a collision of world views, namely the Linden business model you are obliged to internalize (effectively the old commodity model, which requires single owner/creator permission systems and DRM to exist ) and the possibilities of a universe in which everyone can have infinite virtual resources. I think this categorizes the debate going forward on how we attempt to monetize creative work in an age where we have rendered distribution control effectively free. Devices like the iPhone and iPad and the Kindle seem to be aimed at this space. In particular I find it interesting that Apple created their own processor for the iPad. The only reason I can see for this is to control DRM down to the hardware level. We’ll see, I guess.

GLENN: You recently gave a talk at the 2010 ACM conference on CSCW (Computer Supported Cooperative Work) on the topic, “What Changes Since Computation: A Design Manifesto for 2010.” You took issue with the received wisdom that the work style of “digital natives” (those who’ve grown up using computers, the Internet, mobile phones, MP3s, etc.) is different in important ways — specifically having to do with social networking and online collaboration — from that of their elders. How is this by now widely accepted bit of common sense incorrect?

SEMPERE: I discussed the Digital Natives argument because I always find it frustrating. First and foremost, kids are no more born digital natives then they are born literate. Literacy (reading or computational) is something that is learned, taught, part of a culture. Also, pretty much anyone who uses email and IM (standard business tools these days) is already living in a virtual world. The first MMPORGs date to the early 1970s! So I believe the digital natives argument is baloney, but I also believe it is invoked as a kind of shorthand to describe a complex anxiety, and that’s where it gets interesting.

People, especially business folks of a certain age, invoke this argument when they have faith in the structured, top down model of business and they begin to discover that work, actually, is occurring in a completely different way then they thought. Employees routinely socialize using websites and exchange information outside of the sanctioned tools. Employees are more comfortable crossing hierarchies, or with casual interpersonal communication that would be unheard of in their experience. It is all very confusing to someone who has internalized the myth that you work 9-5 and pass everything up and down the command chain.

As a result, a huge amount of effort is spent on compliance theatre: generating the illusion that business is proceeding along standard lines, while successful employees produce their work by circumventing the culture of control.

I borrow the term from Security Theatre, and here’s an important subtlety: in both security and compliance theatre, it isn’t that the theatre is pointless. Theatre is never pointless: It serves an extremely important role in re-enforcing organizational culture and defining mental boundaries. It also, ironically, enables circumvention by making a production of what is “important.”

GLENN: I have experienced that problem — circumventing the culture of control — almost everywhere I’ve worked, even before Web 2.0 tools made doing so commonplace. Which is anecdotal evidence supporting your point that anti-gatekeeping trends making older people nervous in the business world aren’t new — but they’re newly exacerbated by new tools and software.

SEMPERE: I couldn’t agree more — it’s not new at all, just working at computational speed, and so more visible, and sometimes more effective (although counter measures can also move at this speed).

GLENN: In the art world, too, indirect collaboration is as old as the parlor game Consequences, whose descendants include the Surrealists’ “Exquisite Corpse” technique, not to mention Mad Libs. Now that new tools and software have (unintentionally) made it simple to invite the general public to participate in various sorts of art projects, will we see (what you call) Compliance Theatre in art?

SEMPERE: Compliance Theatre in art? Why not! I suspect it’s already there, and it probably depends on art versus Art. Certainly the art world has all kinds of theatricalities… I’ll have to think on that one more!


This Q&A was first published at the blog Indirect Collaboration, on Feb. 18, 2010. The blog is published by the moderator and panelists of “Indirect Collaboration,” a panel at this year’s SXSW Interactive. Click here to read a Q&A that Joshua Glenn did with Margaret Wertheim for Indirect Collaboration. Click here to read a Q&A that HILOBROW’s Joe Alterio did with Charles Burns and Gary Panter for the same blog.