David Weinberger on why he hates Friendster. The most interesting bit, for me, is this one:

"Look, I want to say to the Friendsters of the world, we already invented a social network for friends and strangers. It's called the Internet. Why are you privatizing it? Why do we need a proprietary sub-network to do what the Internet has already done in an open way?"

Excellent questions, although one answer might be that the proprietary sub-networks have a gentler learning curve—these proprietary networks are appealing (I think) to people who aren't particularly web-savvy. Just follow the link in your invitation and fill in the blanks and within three minutes you can be enjoying some of the pleasures of connectivity. It's a closed, limited network, yeah, one that walls out most of the bewildering wilds of the Web, but if the success of AOL has taught us anything, it's that what some people want out of the Web is a managed (or manageable) experience.

This does beg the question of whether the more-managed (closed) parts of the Web might be hurting the less-managed (open) parts. The Web is not a finite resource like geography: each managed "area" does not take the "place" of an unmanaged "area." But attention is a finite resource, maybe the key one in talking about the Web, and when a service like AOL steers its users around and around in its own little content ghetto it siphons that resource away from the rest of the Web. Some of the better social networking websites (Flickr, Orkut, Tribe) are less guilty of this, since they allow you to include a link to your webpage in your profile, appends a little arrow that points "out"...

Related: Caterina's continued enthusiasm for social networks.

Labels: internet, networks, rants

Sunday, April 04, 2004
9:36 AM
0 comments

A while ago, in one of my social software critiques, I wrote:

"What would actually be useful is to be able to rate my friends in terms of the intensity of the bond between us. There are eight people currently listed as my friends in Orkut. I am, indeed, fond of them all. But two of them I've never met. Two more I've only met on one occasion. One is a casual acquaintance who I haven't had contact with since he moved away from Chicago six months (or so) ago. Two are people who I enjoy some degree of intimacy with, people who I could call on the phone just to say "how are you?," people who I have had dinner with (once or twice) within the past year. And one is a guy who's been my friend for fifteen years. This information has obvious relevance to anyone who's trying to make a meaningful use of my network, but to Orkut (or Friendster or Tribe) all of the bonds are of equal intensity, creating a picture of me and my network of friends which is weirdly distorted to the point where it is practically a fiction."

What's needed in order to solve this problem is a workable taxonomy of types of relationships, and I had given some preliminary thought of some basic categories that such a taxonomy would contain. But now I've learned that Ian Davis and Eric Vitiello have cooked one up already: Relationship (which appears to have been designed as an attempt to formalize some sort of metadata protocol?).

Clay Shirky critiques Relationship (he actually describes it as "self-critiquing"), and he does so mainly by pointing out some of the predictable difficulties of establishing any sort of taxonomy (grey areas, etc.). Still: to my mind even the flawed Relationship taxonomy is better than the current model used by most social software, where all human bonds are equivalent.

Shirky's final point in rejecting the taxonomy is as follows: "the madness of the age is to assume that people can spell out, in explicit detail, the messiest aspects of their lives, and that they will eagerly do so, in order to provide better inputs to cool new software."

Thoughtful, but I think he might be wrong. I think the assumption that people would (if they could) is actually an increasingly safe assumption. The madness may be not the assumption that people will do it, but the fact that people will do it.

Labels: internet, networks, rants

Wednesday, March 17, 2004
11:12 AM
0 comments

My old pal Brian S. sends along one link to this social software weblog and one to an article on Club Nexus, an online social network at Stanford. (The article is co-authored by Orkut Buyukokkten, the Orkut that gives Orkut its name, and Club Nexus appears to have inspired at least a few of the Big O's quirkier features, including the sexy/trusty/cool thing that I hated on last week.)

The bulk of the article is dedicated to a variety of statistical observations made from sorting and sifting the data-sets of user information in various ways. For instance, correlations can be observed between the way users describe their personalities (via personality-descriptors selected from a menu) and what those same users describe as their interests. (Appendix A, which begins on page 14, entertainingly records these correlations.)

This is an interesting use of the technology, although it's worth noting that this view of the "big picture" is only available to the system administrators, those who have the access and the tools necessary to organize the big picture into meaningful data.

Question: is there a social networking website that offers total information openness, by which I mean a site that allows all of its users to access, navigate, and sort all of its accumulated information? Because it's worth thinking about not only the value that these sorts of systems provide for their users, but also the value that they provide for their administrators. I would bet that the information that these sorts of systems may yield will turn out to be worth actual money, and as a result it's worth trying to develop a sense of exactly how much of a gap exists between what the administrators can learn from the system and what the users can learn from the system. Reading this article gave me the feeling that we may all be merrily participating in the world's largest market research scheme, and for the first time it made me think that there might be a political point to feeding noise into these systems.

Labels: internet, networks, rants

Tuesday, February 10, 2004
11:13 AM
0 comments

Follow-up to this post: Brian Eno has contacted Matt Jones. I admit that I'm impressed, although they haven't actually had the sit-down chat yet, and until they do I'm not convinced that the project has been successful. (My interest in the project is strictly to see how well a data-exchanging network can provide an experiential pleasure: one that is not exclusively data-oriented. Having Brian Eno post a comment on your blog is undoubtedly pleasurable, but it's a data pleasure, only one order of magnitude above the sorts of pleasures that the Web already excels at delivering. (If one is prone to thinking of all experience as being data, processed by the mind, then one could possibly argue that all experiential pleasures are also data pleasures, but I'll argue for the distinction, given that the data involved in an experience like an actual sit-down face-to-face conversation is massively richer than the data involved in a blog exchange.))

Other tidbits: Crystal sends along a Memepool link to an interesting essaylet about the social networking technology used by French dove breeders of the 17th century, complete with karma mechanisms and fake profiles.

The "state of the state of social software" posts on this blog and elsewhere have mostly been occasioned by the launch of Orkut. I've been fooling around with Orkut for the course of the past week, and I've made some observations:

On the plus side, Orkut does one thing well: it provides me with the meaningful-blurb-under-the-photo that Friendster doesn't give me (This is the third of the three "things I want to do that Friendster won't let me do." Orkut doesn't allow me to do the first two, but at least they're managing one out of three.)

On the drawback side, Orkut has what strikes me as a truly useless friend-ranking mechanism, where you can rate friends as "sexy," "very sexy," or "super sexy." (Also cool / very cool / super cool and trustworthy / very trustworthy / super trustworthy.)

First off, this mechanism bizarrely makes ranking someone as merely "sexy" into kind of an insult... my guess is that this will lead to the kind of inflation known to characterize EBay's feedback reports, where people are unusually free with the perfect scores, because anything less seems critical. (Orkut attempts to minimize this problem by making the rankings anonymous, but, still, I find it difficult to visualize myself giving my friends anything less than the highest rankings.) If everyone in the system ends up being 90% cool (even typing out such a stupid phrase makes me gnash my teeth) then the listing that indicates a person's "coolness level" is meaningless, noise in the signal.

There's the germ of a good idea here, though: in order to provide meaningful data, social networking websites need to allow you to rank the people you're connected to. But rating them in terms of coolness or sexiness is absolutely the wrong way to go about this (not least because of the vagueness: what's the difference between "very trustworthy" and "super trustworthy" anyway?).

What would actually be useful is to be able to rate my friends in terms of the intensity of the bond between us. There are eight people currently listed as my friends in Orkut. I am, indeed, fond of them all. But two of them I've never met. Two more I've only met on one occasion. One is a casual acquaintance who I haven't had contact with since he moved away from Chicago six months (or so) ago. Two are people who I enjoy some degree of intimacy with, people who I could call on the phone just to say "how are you?," people who I have had dinner with (once or twice) within the past year. And one is a guy who's been my friend for fifteen years. This information has obvious relevance to anyone who's trying to make a meaningful use of my network, but to Orkut (or Friendster or Tribe) all of the bonds are of equal intensity, creating a picture of me and my network of friends which is weirdly distorted to the point where it is practically a fiction.

</grouchy>

Labels: internet, networks, rants

Tuesday, February 03, 2004
10:36 AM
0 comments

Over at Black Belt Jones, Matt is attempting to use social networking technology to attain a specific goal: he wants to sit down and talk to Brian Eno within the next two months.

This experiment reminds me of the chapter in Hakim Bey's Temporary Autonomous Zone where he critiques the Web because it primarily provides its users with information, and fails to fulfill desires for anything other than more information. For Bey, information is useful only inasmuch as it enables the fulfillment of real-world desires (especially forbidden / illegal ones), and he views most of the information available through the Web as irrelevant to attaining these desires. Dating sites like Nerve, or social networking sites like Friendster, which hadn't emerged at the time of Bey's writing, present themselves as tools for the fulfillment of specific real-world desires (friendship, sex, lunch with Brian Eno), and, as such, they seem to be offering the right carrot, but the real question is how effective they are at delivering on this promise. So I'll be intrigued to see whether Matt Jones gets his wish.

Labels: internet, networks, rants

Wednesday, January 28, 2004
3:54 PM
0 comments

Darren B writes in to ask "What is it people expect of networking websites that isn't being delivered?"

Let's take an example of a real-world situation that I'd like to be able to use this sort of site for.

Let's say a friend invites me to a party at her place. I don't know many of her friends, but I know she has a Friendster profile, so I think "maybe if I browse her friends via Friendster I'll be able to see who might be coming to this party, so that I can find someone who might share my interests, someone who I might be interested in talking to." So I go check out her profile. There are 50 friends listed there (commonly more). I click on the first one, oops, that's someone from Maine. I click on the second one, that person's from Somerville, MA. I click on the third one, that's a fakester profile for "Beer." Suddenly clicking through the other 47 is seeming less appealing.

At this juncture there are three things I'd like to do that Friendster won't let me do:

1. Arrange the list. In this case I'd like to put the people who live closest to me at the top. Sorting it so that people who shared multiple connections to me might be interesting; I can also see wanting to sort the list by gender, or by who's single. But I can't do any of these things.

2. Search within the list. Within this group of people, who lives in Chicago? Who shares my interests? Who's single? Is there a way to find out without having to click 50 times?

3. Barring a way to arrange the list or perform a sub-search, I'd like to be able to at least quickly browse all the friends linked to my friend's page for some useful basic information. Which of them live in Chicago? A little blurb under the photo with just their location would at least enable me to know more-or-less at a glance which of these people I have a realistic opportunity to get to know. But all I can see are their names and, in some cases, their faces: information that holds no meaning given what I am trying to do.

I think Friendster has lost sight of the fact that the information that's most useful to us when looking for friends or people to date is local information. Friendster should be continually sifting by geographic proximity as well as degrees-of-separation proximity. Not "here's a random person in your network" but "here's someone in your network who lives next door."

Internet networking makes it easy to connect to people all over the globe, but unless I'm looking for correspondents, knowing that I'm two degrees of separation from someone who shares my interests but lives in Wyoming does me very little good. What I want Friendster to tell me is that I'm two degrees of separation from someone who shares my interests and lives two subway stops away. And the really frustrating thing is that Friendster has that information, but its interface isn't smart enough to let me ask for it.

To be continued (when Friendster repairs its broken search feature).

Labels: internet, networks, rants

Tuesday, January 27, 2004
2:39 PM
0 comments

I just joined Orkut, as I joined Tribe before that, and Friendster before that.

But I have to admit, I don't quite get the appeal of networking websites.

And it looks like Warren Ellis feels the same way: "[W]hat can you actually do aside from invite all your friends and piss about on a couple of small message boards? ... What happens after that? After you've gotten all your friends in—whom you send email to or IM regularly in any case, presumably. That's it. All done. Until, I guess, yet another social network system opens and you start all over again."

In fairness, I can see how they'd be interesting to study (see Danah Boyd's Apophenia, or, even more relevantly, the archives of her defunct Connected Selves blog). But, even with my interest in networking technology, my experience is similar to Ellis': I go to the sites, sign up, wander around for a bit browsing the profiles of strangers, and then wander away.

I might feel differently if I were actively looking to date someone new (I'm not) or if I were trying to do business networking (?).

Have any of you had rewarding experiences with these sites? E-mail me at jeremy [at] invisible-city.com.

Related: Village Voice article which draws rather lazy analogies between Mark Lombardi's drawings, Friendster, and a species of invasive blackberry (Rubus armeniacus).

Labels: internet, networks, rants

Monday, January 26, 2004
7:02 PM
0 comments

The Tucson issue of poetry webmagazine Can We Have Our Ball Back.

Disclaimer: I know ten of the people in this issue, and at least one of them I dated. Put that in your social network analysis visualization schema and smoke it.

Labels: internet, networks

Saturday, January 24, 2004
12:49 PM
0 comments

Friendship Choices Among Fourth Graders

An article documenting examples of pictorial images in social network analysis.

Related: Connecting the Dots, a use of social network analysis to identify terrorist networks (documented in the Business 2.0 article "Six Degrees of Mohamed Atta," now available to subscribers only.)

And no discussion of attempts to map "covert networks" is complete without mention of Mark Lombardi's hand-drawn documentations of capitalist power and influence. Iran-Contra makes the friendship choices among fourth graders seem like, well, child's play in comparison.

Labels: internet, networks

Friday, January 23, 2004
4:07 PM
0 comments

I can tell that it's getting warm enough that my brain has begun to turn off. One symptom is that I have actually begun to look forward to seeing the following movies:
The Hulk
Charlie's Angels Full Throttle
Terminator 3

Another symptom is that I've gotten sucked into browsing through Friendster profiles. If you're a Raccoon reader and a Friendster user, feel free to add me to your network. My e-mail is jeremy@invisible-city.com.

When I was in New York recently I saw Friendster-related graffiti on the Bowery. I wish I'd taken a picture of it: it was a tag that said "Friendster Demons!" or something similar. Rabid fans using Friendster as tribal identification? Guerrila marketing? Anti-Internet religious fundamentalist? Your guess is as good as mine.

Village Voice article on Friendster's impact on real-world socializing, found through this interesting-looking "social software" blog.

Related: in this post (which points to the "relational aesthetics" stuff I blogged on Saturday), Test writes: "As our understanding of social uses of the internet matures from 'is anybody out there?' to 'so what are we going to *do* here?', so networked art is maturing from its initial investigation of the form-factors and politics of the network to an interest in how people are using networked media to connect, and participate in social exchange."

In terms of how network media can inspire exchange, you could do worse than to examine Art of the Mix, a site "dedicated to making mixed tapes and cds" which has the nice side effect of encouraging a gift economy among music geeks... Thanks to Dirk for the heads-up on that one.

Labels: internet, networks

Tuesday, June 24, 2003
11:38 AM
0 comments

For a long time now I've been eyeing A Year With Swollen Appendices, the published version of Brian Eno's 1995 diary. I'd always hedged on buying it before, but I saw a used copy available on Amazon for around three bucks and decided to go for it.

I'm enjoying it immensely. (I can't figure out why I delayed, exactly—I've always been very attracted to the way Eno looks at the world, and I love the diary form in general.) Every page contains some little handy epigraph or interesting observation or mental tool, and it is difficult for me to resist annotating the entire book here. But I just reached a passage written after news of the Aum Shinrikyo gas attacks broke, and I thought Eno's thoughts on terrorism then bear some relevance to our current situation:

"It struck me forcefully (again) that the more 'richly connected' we make our world the more vulnerable we make it. Empowerment cuts both ways: as the complexity of things increases, so does the ability of an increasingly minute people to destabilize it. This, it strikes me, is the real limit on development—that we will accept the threat of terrorism as a limit on how complex we make things. So the Utopian techie vision of a richly connected future will not happen—not because we can't (technically) do it, but because we will recognize its vulnerability and shy away from it.

So I expect a limit to be reached, a sense of pulling back from what is possible. And this will be followed by waves of nostalgia for the future-that-could-have-been. Country songs that say 'we could have had it all,' etc., etc. A sense of disappointment with ourselves—perhaps like the sense that pervaded Europe on the failure of the League of Nations."

Labels: networks, war_on_terror

Sunday, September 29, 2002
12:24 PM
0 comments

This Kevin Kelly article, "Where Music Will Be Coming From," is interesting and mostly well-reasoned. (It's a New York Times article, so if you don't have an account with them, use member ID invisible-city, password reader.)

I think this is accurate:

"Digital copies are not only perfect and free, they are also fluid. Once music is digitized it becomes a liquid that can be morphed and migrated and flexed and linked. You can filter it, bend it, archive it, rearrange it, remix it, mess with it. And you can do this to music that you write, or music that you listen to, or music that you borrow.

At first glance it seems audiences were drawn to online music because of the power of the free, but in reality the rush to online music came from digitized sound's ever-expanding power of liquidity. Once music could swirl around one's life unencumbered, the millions of people who downloaded peer-to-peer file-sharing software suddenly and simultaneously imagined a thousand ways to conjure with music's liquidity. It wasn't only that it was free; it was all the things you could do with it.

Once music is digitized, new behaviors emerge. With liquid music you have the power to reorder the sequence of tunes on an album, or among albums. To surgically morph a sound until it is suitable for a new use. To precisely extract from someone else's music a sample of notes to use oneself. To X-ray the guts of music and outline its structure, and then alter it. To substitute new lyrics. To rearrange a piece so that its parts yield a different voice. To re-engineer a piece so that it sounds better on a car woofer. To meld and marry music together into hybrid breeds. ... With digitization, music went from being a noun, to a verb, once again."

And I think his conclusion is accurate:

"Music could experience a similarly exuberant, irrational flowering of the amateur spirit.

Part of the reason people have been inspired to create text, graphics and action in the digital realm has been the arrival of new tools. Fans of music are already ... making music in the way that a camera makes an image -- by starting with what is there and adding a unique view to it. Just as the introduction of the Brownie camera changed photography from an expert's art to a ubiquitous public expression, with the right tools in hand it is not a very long hop from now to a time when everyone makes music in a small, amateur way."

But I disagree with his thoughts about how people will sift through the ensuing mountian of music. These thoughts (which appear at least three separate times on this bullet-pointed list) don't go beyond "people will pay others to sift for them," an unsubtle hint to people looking to make money off the future. (Unsurprising, given Kelly's long-running tendency to operate as a corporate-friendly visionary.) But there has always been more music than people can possibly listen to; this is not new. And, yes, people have made money by providing accessible guides to that music, and, yes, this will continue to be true into the future. But I think that the real trend here, which Kelly misses, is the increase in noncommercial fan networks.

Mailing lists, suggested discography pages, weblogs, Amazon collaborative filtering systems and Listmania lists, submonitions pages: these things are proliferating wildly right now, and they all serve to amplify basic word-of-mouth. As a result, it is easier than ever to find music you will like, not harder than ever, and this makes it less likely that recommending music will be lucrative in the future.

Think of MP3 swap networks, which have all the benefits of the old underground cassette networks, only without the expense of postage. And, like the cassette swap networks, MP3 swap networks are participated in by both fans of obscure music and obscure musicians seeking kindred souls. And it's easier than ever to find kindred souls right now. I see the amateur musicians out there finding other musicians, who are pursuing similar experiments, and then swapping tracks with them, eventually perhaps bonding together as an identifiable unit. Anticon and Elephant 6 are perhaps recent successful examples of this phenomenon...

I am indebted to Dirk Hine for the original link to the Kelly article.

Labels: music_commentary, networks, technology

Thursday, March 21, 2002
11:23 AM
0 comments

I like this quote from Kobena Mercer:

"[A] profusion of rhizomatic connections ... implies another way of conceiving 'the role of the intellectual,' not as a heroic leader or patriarchal master, but as a connector located at the hyphenated intersection of disparate discourses and carrying out the task of translation."

I plan, sometime soon, on reading his Welcome to the Jungle : New Positions In Black Cultural Studies. Expect more notes in this space.

In other news: it is so windy today that my house is shaking back and forth.

Labels: networks

Saturday, March 09, 2002
1:32 PM
0 comments