On Die Puny Humans, a selection of folks have created statements for 2004. I was pleasantly surprised to read Cory Doctorow’s call to the toolmakers of 2004:
Stop making tools that magnify and multilply awkward social situations (“A total stranger asserts that he is your friend: click here to tell a reassuring lie; click here to break his heart!”) (“Someone you don’t know very well has invited you to a party: click here to advertise whether or not you’ll be there!”) (“A ‘friend’ has exposed your location, down to the meter, on a map of people in his social network, using this keen new location-description protocol — on the same day that you announced that you were leaving town for a week!”). I don’t need more “tools” like that, thank you very much.
Now, i don’t know much about science fiction, but i read it once in a while to understand the models that technologists are trying to mimic. When i asked Cory about the relationship between scifi and technology, he told me that scifi is not supposed to be prescriptive. Scifi is modeled after what exists today and is not a representation of the future. Quite often, very little in the way of technology is fully fleshed out. In this regard, he’s quite accurate. Even his own Whuffie (which i hear about in way too many meetings on reputation) is barely detailed in “Down and Out in the Magic Kingdom.” Still, while scifi shouldn’t be prescriptive, many technologists interpolate the ideas presented and flesh it out to be beyond problematic. Often, they have the nerve to refer to the fiction books as their model for why it is a good idea.
Given his role as a science fiction writer, i’m quite pleased to see him call out to technologists. All too often, the omniscient technologies that appear in the science fiction novels are not representations of good things, but embedded in a discussion of the pros/cons of changing social interaction through technology. Take Cory’s Whuffie and his examples of people scorning others because they are not worthy enough of interaction. C’mon now. All of us geeks have experienced a form of that, being chastised for not being cool enough, good looking enough, whatever enough. Why on earth would we want to develop a technology that encourages that? Oh, right, because if _we_ build it, we can be the ones in power, right? Hrmfpt. Seriously now, such a creation creates a whole new level of social awkwardness, new hierarchies that constrain us. Just because it’s an idea for a novel does not make it an idea for life.
So, in fleshing out Cory’s call to technologists, i’d ask all technologists to consider not only what problems a technology solves, but what new ones could emerge. Start thinking like a writer or an abuser of technology. Imagine how people could misuse a technology to hurt others. Consider who gains and loses power from such technology. It’s a fascinating exercise and far more fulfilling than just thinking about who benefits from something. And besides, then you won’t always be thinking “but the users shouldn’t do THAT with this technology.”
Hi danah… happiest of holidays…
Don’t you think that, in most design scenarios, it is hard for techies to *not* think of second and third order consequences of their developments? That is, I’d bet that they have already thought of a good deal of nefarious uses for a given tech. but that they try their best to keep them under wraps so as to not affect marketability.
If all the time spent writing dumb code like viruses was funnelled into interoperability issues or the like, the world would be a better place. Punks…
This actually goes to the second part of Cory’s comment. The trick isn’t really to try to envision how to stop nefarious uses, but to figure out how to design the technology so that those uses will be handled in a way that won’t destroy the technology for everyone.
In society, there is a live and let live approach to things. Not everyone uses “tools” as intended… if you want to stick a banana up your ass, by all means, go ahead… we’re not going to stop the sale of bananas because some folks stick them up their ass and that’s not what’s expected by the grocers.
Furthermore, just because you see the grocery mart as your sex toy store doesn’t mean that your approach affects others. Of course, imagine now that everyone had to tell everyone in the store why they were purchasing that item and what they intended to do with it. Can you imagine “Aisle 1, guy in green purchasing banana to stick up his ass.” This would be socially awkward for many, and probably more so for the little old lady than the guy in green.
Sometimes, not knowing is better for everyone. And the trick is to have structues available to let people repurpose the technology without impinging on others’ uses. This is what i mean by considering secondary effects… Consider how to allow for repurposing that won’t tear the whole system down rather than just trying to constrain the behavior.
I agree with your point danah. On the other hand, a lot of the consequences of technology are not predictable and emerge as the technology develops and is adopted widely. I think that in addition to trying to have a vision about the negative effects of technology (which I agree is important) and trying to design around the issues, I think that identifying tensions as they arise and providing feedback to the toolbuilders is important. One of the problem of commercial enterprise is that technologists are often forced to sweep these tensions or problems under the carpet for the better good of profits or commercial interests. Also the cost of changing a design or an architecture often makes such change difficult. I think designing systems to assume they will need to be changed is important. This does get difficult as technologies mature. This is why I think the social software / blog space is interesting. We can still change a lot of the basic architecture of this space. So although I agree it is important to call our to technologists to think, I think that the dialog between technologists and people like you and Cory is more important.
Feh! Who needs discourse and collaboraton when Al-Qaueda has released this flu virus – which is rampaging through the world – as we speak.
Or….
Sci-Fi, schmi-Fi – geeks and jocks alike are puppets to teh perceptions that marketing mizards plant in our brain. Witness the Cult of the Mac or the videogame business. At best technologists can ‘guess’ what is needed. We’ll never know – for sure.
Anticipating problems, solving perceived problems, or creating new problems – unfortunately it all comes down to economics. If there’s someone who will pay for it, it’ll exist. Building castles of sand on the backs of free labor – can, at best only keep dreams alive.
It takes cold, hard cash – and that’s why it’s the Mark Pincus’ or Reid Hoffman’s of teh world who are dictating terms, conditions and features.
Unitl there is economic support for ‘right’, ‘truth’ and ‘discourse’ = we’ll all be victim to crass commercialism. Nothing happens without money.
Part of the problem with the social networking services is that they are so darn centralized.
Meetup, Friendster etc are programmed by a few people whose assumptions define the system. You can game the system, but you can’t change it.
By contrast, there are lots of open source and open code blog and wiki tools — people who want to write their own rules can.
I’ve been hoping in a LazyWeb, not-enough-free-time sort of way, that someone would put together peer to peer or otherwise open social networking tools.
Marc – i disagree with you. There is more to life than the entrepreneur’s vision of it. And, frankly, visions can exist beyond that. Just as music has been made without the RIAA, technology has been built without the entrepreneurs. It’s a question of the creators goal: creation or money. I’m sorry that you feel like a victim to crass commercialism.
Regarding whuffie, and the various social tools based on trust networks: the problems with these are typical of scale free networks, which have a few highly connected nodes and many more which are less connected.
Scale free networks can occur as a result of new nodes preferrentially connecting to well connected nodes. They are fairly robust to random failures, have low hop counts between nodes, but are vulnerable to attacks on the highly connected nodes.
There is another kind of network with low hop count, called (i think) a “Small World” network. In this, nodes all tend to have about the same number of connections. Most of the connections are between nearby (in some arbitrary sense) nodes, but there are a few long distance connections. These networks are not as vulnerable to attack, and are, i think, to be preferred.
How to do this?
* Limit the number of links from and, especially, to a person.
* Encourage deletion of old links as new and better ones are found, so that power doesn’t become entrenched. For example, require that links be explicitly renewed regularly.
* Make links two-way, ie they have to be accepted by both parties, and are accorded the same status/strength by both.
For example, Google could look at only the top 5 incoming links to a site to calculate its pagerank, rather than adding up *all* incoming links.
http://en.wikipedia.org/wiki/Scale-free_network
http://www.ams.org/new-in-math/small-world.html
Building norms together with the technologies
danah has a good rant in response to Cory’s thoughts on technologists that create technologies which cause awkward social situations.danah…
“tools” are mostly created by engineers in the course of doing what they’ve been taught to do: solving problems.
Contextualizing the problem or the solution in a broad way is anathema to problem-solving.
In product engineering, the problem is often defined by the marketplace. I daresay the “technologists” are one of the last cogs in the wheel and not the best targets at which to direct this pretentious rant.
Cory obviously got up on the wrong side of his social network that morning… however this discussion might be a good segue into the panel we’re doing in March at SXSW on The Aesthetics of Social Networks. I think it’s interesting, and might be revealing, to approach SNs as objets d’art, and not as “social technology.”
Joi has it right… we need perspective on the various technologies we’re building, and we need to think about what they represent at a higher level, apart from the applications themselves.
I should add that I never even thought to take the dim view of social network apps that Cory expressed, partly because the SN sites I’m associated with (Ryze, Friendster, LinkedIn, and Tribe) all attracted me because some part of my own virtual neighborhood was already there. I almost never form a connection to someone I didn’t know before. The real value of those networks for me has been in reinforcing relationships with people I already know, especially those that I don’t see often. I’ve found people I haven’t seen or connected with in years, or who I thought had fallen off the virtual map.
I can accept that social networks apps are ugly, as long as we agree that they’re beautiful.
Aesthetics of Social Networks
Artistamp by Byron Grush, from the EMMA Gallery It’s probably premature for me to talk about an aesthetics of social networks, which is a concept that Honoria and I came up with while riffing on ideas for courses or…
I would somewhat disagree with danah on this one.
I am wondering what’s better, a geek who has zero interest in social issues or a geek who fancies her/himself an expert in those matters. Some of the socially-awkward echnologies mentioned above are being developed precisely by the latter group. Those are also the people who go on developing things like the golden rice and BT corn, hoping to solve all of world’s problems with technology. So, sometimes I wonder if perhaps it would be better for everyone if those of us interested in social consequences of technology and willing to spend time to understand the issues, would just leave the engineers alone, and let them build stuff that they find interesting. We can then try to figure out what would be good and bad uses of such technologies and work on influencing the public policy. This might come out better than if the engineers feel pressured to make their work socially relevant and proceed to work on technical solutions to “real” problem guided by an inadequate understanding of social issues.
Engineers do stop to think about how their technology might affect the world. It’s that it actually takes more than just a moment of reflection.
Yuri – i don’t know if we disagree. Perhaps my rash blog entry spiraled a bit too far out of control.
I agree that most engineers are not able to understand the social consequences. But i also believed that many of those best equipped to make sense out of the impact are not prepared to build technology. One fallacy about techno culture is that the technologists can do it all on their own, understand everything, etc.
That said, creators do have the power of making technologies flexible / extensible. Anyone who has gone through a software engineering program knows that extensibility is highly valued. No technology is perfect as-is. It must grow with its users.
Unfortunately, this is where i watch things fail. Creators are not prepared to alter the path of their technology as users change. More frequently, they try to adamently configure the user to play within their boundaries. This is where i get frustrated.
In order to prepare for that level of extensibility, creators must first ask what problems they are solving. Then they must look at how users are repurposing the technology, why and with what consequences. The why is the most important step. If creators don’t think about these issues at first, they are not prepared to extend them; they are only prepared to stick to their vision of what works. This is where i think problems stem from.
[Of course, larger problems stem from me trying to make arguments with a belly full of holiday chocolate…]
Paul – the problem is that you don’t want to create artificial boundaries. This makes the information even more inaccurate. This also makes you have to choose between people you know and people you have to be considerate to. Furthermore, no one wants to update any of this information. Even when given that option, users often balk and say that it’s rude to remove someone or otherwise indicate that they don’t know the person. Two-ways links are what most social network systems do as is… Also, no one wants to give away the status of any relationship. This is considered problematic socially…
Of course, i’m talking about social networks not Google networks.
Jon – the problem is that social networks are concepts with valuable meaning in other domains (sociology for example). What is happening is that people want to take the concept and then mess with it in the creation process and then expect all underlying theories to still play out. I’d hesitate to call this an object d’art because this is not what people are doing with these things. They are playing out a whole new level of social norms.
Should engineers weigh the social impacts of they design descisions: see Victor Papenek, Bonnie Nardi [http://www.firstmonday.dk/issues/issue4_5/nardi_contents.html], Peter Rice [http://tinyurl.com/3bbrs ] for cheerleading.
For cons of conscience, see Sloman’s excellent: ‘existential pleasures of engineering’: [http://tinyurl.com/yus8d]
For designing for [ab]use-cases, see Fiona Raby and Tony Dunne’s ‘design-noir’:
http://www.doorsofperception.com/Features/details/9/
Technological Determinism
Though I’ve been spending less time online lately, I couldn’t help but notice that Joi Ito and danah boyd (in…
ah matt — you read my mind and posted the dunne and raby for me. thanks.
i wanted to get back to something you mentioned at the start of this post, danah, and it’s something i chatted about with cory when i drove him to the airport really really early in the morning when he visited ivrea — about sci-fi writers and scenarios and new visions of the world.
it’s not prescriptive but it plays an important role in visualizing, in creating new scenarios for the ways the future could look. this is a tool that’s really important for us in design(ezio manzini is a good person to think of in this regard).
one of my favorite historical examples of sci-fi inspiring people is with paul scheerbart (who actually predates sci-fi and who died in 1916)… it was his work that got bruno taut and a host of modernist architects and artists to work in new materials and in new ways. in 1914, taut’s glass pavilion was born out of that inspiration and collaboration.
Dunno why Ross’s article didn’t trackback, but i liked the comparison so i’m noting it here.
http://ross.typepad.com/blog/2004/01/from_culture_to.html
zephoria – I agree about not creating barriers, let people enter what data they will. I’m more worried about emphasis. Sure you can link to 1000 people, or google can return a billion results, but only the top ten really matter to someone browsing.
The Stable Marriage problem has been tickling at my unconscious for some time…
http://mathworld.wolfram.com/StableMarriageProblem.html
…as applied to networks (except polygamous to some set level).
I’m slightly worried by Google at the moment. People seem to be finding exploits. For example, if there’s a page that lots of people link to, you can buy it and put your own links there and be confident that not everyone will bother updating their links. Google is fighting this with some ad-hoc stuff like filtering by certain words, but i don’t think thats going to work in the long term.
So what i’m thinking google might do instead is solve a stable n-way marriage type problem. This means pages will tend to be considered to link to pages of about the same page-rank stratum as themselves. A link from Joe-random-blogger to slashdot is ignored and vice versa.
This corresponds somewhat to how people actually browse (which is what google is meant to simulate with pagerank). You tend not to follow the link to slashdot from a random page, because you know what it’s about already. And you tend not to follow the link by the guy saying “first post!” too.
With this system, you can’t get high pagerank without endorsement by other pages of (nearly as) high pagerank. So you can’t feed off lots of stale links from pages with low pagerank, and pages with high pagerank will tend to be well maintained and so delete the link to you fairly quickly.