the (un)certain future of competitions
Ok, so this is my entry to the Next Stop Design “competition” for a bus stop on the University of Utah campus. I’d just spent a fair amount of time in Utah so the setting piqued my interest. The boulders are recycled from national parks around Utah and brought into Salt Lake to form the shelter of the bus stop. Whatever, right, pretty straightforward and not bad for a lazy afternoon’s work. What’s actually much more pertinent for discussion is how NextStopDesign understands the application of the buzzword “crowdsourcing” in relation to the future of architectural competitions.
Wikipedia defines crowdsourcing as “the act of taking tasks traditionally performed by an employee or contractor, and outsourcing it to an undefined, generally large group of people or community in the form of an open call.” Wikipedia is an example of crowdsourcing. Here’s another: Netflix uses an algorithm to recommend other movies you’d like based on your past viewing habits. This algorithm could certainly be improved upon so it becomes more accurate and stops trying to get me to watch The Benjamin Button movie. So to design a better recommendation algorithm, Netflix didn’t hire some movie-algorithm-predicter company, but rather put out an open competition – with a prize of 1 million dollars – with the idea that anyone out there can come up with a better algorithm that will more truthfully predict what kind of movies I’ll like. And anyone and everyone has tried: academics, laypersons, programmers, etc.
Now here’s how NextStopDesign competition organizer and researcher Daren Brabham defines crowdsourcing: “a company posts a problem online, a vast number of individuals offer solutions to the problem, the winning ideas are awarded some form of a bounty, and the company mass produces the idea for its own gain.” Sweet, so I can be compensated a nominal amount for my work, and then some other company reaps massive profits. Score! Where do I sign up?!?! But maybe the problem is that this dude’s narrow and cynical reading of crowdsourcing isn’t actually inaccurate, but actually pretty well describes the exploitive nature of crowdsourcing.
This is how NextStopDesign attempts to apply the principle of crowdsourcing into the design of a bus stop in Utah: Anonymous users post one to three images of their proposed design on the site, and other registered anonymous users rank said designs on a scale of one to five stars. When the voting ends, the highest rated design “wins.” What you win is completely undefined, but hidden deep in the bowels of the site is the statement that NextStopDesign will present the portions of the highest rated designs as possible qualities the Utah Planning Division could consider implementing in the future. In this scenario NextStopDesign acts an unnecessary parasitic gatekeeper. Now, since being highly rated is predicated on how others rank you, it is in each users own best interest to vote everyone else as low as possible as they jockey for a higher position. This leads to a cutthroat environment where everyone leaves absurdly irrelevant and overly harsh criticisms on other designs, and depresses the entire vote score. Out of around 200 designs the median score is a paltry 1.6 out of 5. Therefore, the highest ranked design is the one has garnered the most goodwill amongst a loose network of vindictive users that are each looking out for their own vested self-interest. Since all comments and ratings are anonymous you can’t trust anybody. This ultimately leads to the major misunderstanding NextStopDesign must confront regarding crowdsourcing and urban planning which is this: When Netflix crowns a winner they will be able to quantifiably judge that someone in the crowd has designed a more efficient algorithm. It can be tested, verified, and agreed upon by all. The design of a bus stop is different, and must address a whole slew of realities such as siting, fabrication, cost, etc. that are unable to be processed from even the most beautifully rendered image. On the other hand, what could make this competition interesting is if NSD were attempting an experiment to quantify the intangible qualities of architecture via a participatory network, or using the performative values of a proposed design (using program, energy, structural)as a means to rank and determine what’s “better.” But based on their repeated and simplisitic definition of crowdsourcing (step1_competition, step2…., step3_profit!) as simply a buzzworthy potential means to realize a new profit model it falls flat, and it is narrow and cynical. I’m torn here, because I think a lot of the submitted designs are really clever and inventive (I’m looking at you Bus-Shroom!). It’s the means to which they’re being used that bothers me.
There’s absolutely nothing groundbreaking about letting the general population participate in the results of an architectural competition. I’ve worked on at least two competitions where the voting results of the public became a factor in the jury’s deliberations. Also, archinect just ran a completely awesome competition for a Michael Jackson memorial where online contributors could rate and comment on the submissions. But here, the participatory aspect is but one component used in addition to a jury of experts in fields of design, architecture and engineering. And in lieu of this weeks amazing presentation by Usman Haque, in which he presented a survey of his projects that utilize a true participatory network in inspiring ways, NextStopDesign can’t help but come off as cloying and depressing.
In the end, though, it might be worth re-reading the Wired article that NextStopDesign quotes from liberally, and there, one can find the murky origins of NextStopDesign in the form of the earliest use of crowdsourcing, here in the interest of cheap, mass-marketed television programming: America’s Funniest Home Videos. Yes Bob Saget and ABC did make a fortune, but just because America voted for the hysterically zany toothless kid, doesn’t make him as great as the Simpsons. Maybe you really do want to look at reality show programming as the new paradigm for urban planning and architecture, but I’m pretty sure nobody wants more Wiffle Bats to the proverbial crotch.
September 19th, 2009 at 4:16 pm
A friend of mine just emailed me one of your articles from a while back. I read that one a few more. Really enjoy your blog. Thanks
September 20th, 2009 at 9:00 am
I think your understanding of crowdsourcing is a bit too narrow, and that you aren’t reading the research. Had you read some of Brabham’s peer reviewed and published articles – you might have a different understanding of crowdsourcing. I’m not entirely sure he is praising the crowdsourcing model as “good” but is more researching it as a way of understanding how crowds participate online, and more importantly, if this can be a model used for public good – in a way that for profit companies use it to make money. My understanding is that crowdsourcing can be democratic BECAUSE the crowd chooses, not just offers suggestions, but actually sorts through the options and picks a winner. If you look at other crowdsourcing sites like threadless.com or innocentive you can see that the crowd doesn’t always operate in a simply self motivated way, but that they are genuinely offering critique and feedback for other designers. Furthermore, nextstopdesign is a prototype, and a means for opening up the design process for public good – therefore, if it is for the public – should there necessarily be a major incentive? What does it mean to have a panel of experts anymore, if they are disconnected from the people who are actually using the bus? Shouldn’t those who use the bus have a more accessible means for putting forth ideas and designs – and shouldn’t those ideas carry as much weight as the architects who don’t take the bus? Just some questions to think about….
I suppose I see nextstopdesign as more of a step toward transparency and public participation, than a way of getting the “best design”. I have read several of Brabham’s articles, and the research and work he is doing on crowdsourcing is some of the most innovative out there. I highly recommend you read some of it – instead of simply looking at wikipedia – to get a more holistic idea of how he views and hopes to use crowdsourcing. He has an article in Planning Theory and Convergence that both approach the idea of crowdsourcing with enthusiasm and caution…….http://darenbrabham.com/research.html
check it out
September 20th, 2009 at 12:47 pm
Dear Christina, thank you for your response, there’s a lot of great questions to think about in there.
Starting with Threadless.com, which is an amazingly successful business model. I actually did read Brabham’s thoughts on Threadless (I have researched some of Brabham’s work on crowdsourcing, I even linked to it in the third paragraph. And I totally agree with you that it is both fascinating and well thought-out.) but again, from the gamut of “public good” to massive profits, I get the impression that Brabham’s sympathies are more closely aligned with the profit potential. This is evidenced by his conclusion re: threadless that they pulled in gross of “18 million in 2006, all with fewer than 20 employees.” (http://www.darenbrabham.com/files/brabhamplanning.pdf , page 251, paragraph 2). And would closely align with his own definition (ibid, 251, paragraph 1) of crowdsourcing that I quoted: “….the winning ideas are awarded some form of a bounty, and the company mass produces the idea for its own gain.” But anyway, trying to discern Brabham’s intentions on such a vast topic is probably futile and not nearly as interesting as looking at how transparency and public participation can be achieved in architecture.
So, using the Threadless analogy, how does that apply to NextStopDesign? I get the first part, people vote on something, but where does the implementation happen? Will NextStopDesign eventually manufacture bus stops, only in Utah, or everywhere in the country? What if the design the crowd votes for is wildly impractical and requires materials and manufacturing assistance from a number of discontinuous building trades and the assistance of consultants in the fields of engineering, fabrication, lighting, etc. There’s a fundamental difference between Threadless producing and selling the same shirt with a different graphic and building a bus stop. This may be why some architectural competitions welcome public input, but also rely on the consultation of a “panel of experts” to advise on the thousands of variables that are necessary to actually build something.
I agree with you that NextStopDesign is a step toward transparency and public participation, and the democratization of design through the internet can be a great thing. Check out this competition by the AIA that is held through facebook: http://www.facebook.com/event.php?eid=171304305336, here they’re using an existing network – Facebook – so there’s already a built in respectability rating, no one is anonymous, and the winner is chosen by peer voting. But again, like NextStopDesign, actually implementing the winning scheme is outside the scope of the competition.
This is a really interesting question you asked: “What does it mean to have a panel of experts anymore, if they are disconnected from the people who are actually using the bus?” and I’ll answer it with another question. Do you think the internet, with its global reach, is an appropriate means to solve a localized problem: the design of a Utah bus stop? For instance, via the NextStopDesign twitter feed, they are receiving votes and submissions from Kazakhstan. And while I’m sure the Kazakhs are great people (with the possible exception of Borat), there is no way you could argue that they are the ones using the bus. Wouldn’t a more appropriate response simply be to place a suggestion box outside the existing bus stop on the University of Utah’s campus where the connection between user and designer would be the strongest? But maybe because of the nature of the crowd, the overseas votes are simply a small part of the overall mass?
Anyway, I’m looking forward to the results of the NextStopDesign competition and seeing the conclusions that Brabham and others parse out from them.
Best, John
September 20th, 2009 at 6:55 pm
First – I think Brabham is a bit more critical of the profit element. The quote about profit at Threadless was directed from Jeff Howe who originally coined the term crowdsourcing. I believe he has quite a bit more enthusiasm for the profitability element associated with crowdsourcing. Brabham is much more critical of the profit element, and he definitely sees the exploitive element associated with the model. His main goal is in figuring out how it can be worked into a usable model for public good – for nonprofits and government to use to increase public participation.
I don’t think a comment box is anywhwere near as engaging or thoughtful as simply asking people to come up with something on their own and sort through those submissions – suggestion boxes assume that we have to work within the structures that already exist, where nextstopdesign allows people to imagine, think outside the box, and throw some of the rules away. Isn’t that how innovation happens?
I also think that local problems require both global and local reach. We need to start thinking more systemically, especially in the way we plan urban environments. Although someone in another country might not use this particular bus – they may have a valid experience that can shape the way transit happens. It is the exchange of these ideas, the crowd sifting, and the imagination that I am intrigued with – and it is that crowd sifting – where the crowd produces and the crowd chooses that really makes this a viable model. Lakhani noted through his research with Innocentive that it was people from outside of the discipline that solved the problems placed before them. So, if we know we can communicate beyond borders, and that an interdisciplinary approach can aid solutions – it seems to me that global reach is entirely appropriate.
I too will be interested in the results of nextstopdesign, especially since there will be another project proposed in October. I wonder how/what will change reflecting what they have learned so far.
cheers
September 20th, 2009 at 9:24 pm
Ok, how’s this sound, and you can stop me if I’m being too simplistic, but it seems like we’ve identified two broad models of how crowdsourcing can operate. On the one hand we have something akin to threadless.com or Wikipedia. In this model, a crowd works to generate ideas or submissions, and a company profits. Simple I know, but bear with me. On the other hand – the second model – we have Netflix or Innocentive, where it is “people from outside of the discipline that solved the problems placed before them.”
Now for the sake of argument, if we accept those two models, where on that spectrum does NextStopDesign fall? If it’s the first model, I can’t help but find it slightly exploitive. This is where the company mass produces the design for its own gain. But if it’s the second model, you are hinting that NextStopDesign is making a value judgment, that a problem is being solved by people outside the discipline. And as imaginative as the designs are, I don’t think that NextStopDesign is proposing that these designs are better, simply because a crowd produced and voted on it. Or are they? Will NextStopDesign determine that the winning design solved the problem better than if the Utah City of Metropolitan Planning Board hired some local architects? And if so, what are your criteria for judgment. But again, I think it’s a question of implementation, and the murky nature of NextStopDesign’s intent. The website says it is an experiment in crowdsourcing “the best ideas and designs to build a better bus stop.” Which would certainly imply the second model.
Anyway, it’s a fascinating discussion and obviously something that, as an architect, I’d have a vested interest in following.