Anybody know what this page has decided to stop wrapping? My other comment pages seem to be fine…

5 Comments »

TEAM = Tame meta-meat

1 Comment »

One last report on etech. I led a “birds of a feather” discussion on Tuesday night, a circle-the-chairs sort of thing for thirty people. The topic was “From Trees to Tags,” and I began with about 12 minutes on the topic. This is from the speaking notes I took into the session, so what follows is highly approximate. It’s also highly redundant with other blog posts of mine:

I’m writing a book called “Everything Is Miscellaneous,” so I convened this BoF selflishly, in order to use you and what you know.

One way of getting at what the book is about is to ask: Why is knowledge organized the way the real world is? It’s not at all obvious that we should organize our ideas the way we organize our socks.

We’ve organized knowledge into trees, from Aristotle to Linnaeus to Dewey. You get a tree by doing the basic thing of lumping and splitting, and then splitting the lumps until you get to a lump that is too unitary or miscellaneous to bear any more splitting. But lumping and splitting has been constrained by physical limitations. For example:

1. A thing has to go in one pile or another. For Aristotle, this was expressed as the Law of Identity (A is A and A is not not-A), a pretty basic rule.

2. The way we lump and split is the same for everyone: If you own a clothing store and separate it into men’s and women’s departments, it’s separated that way for everyone who enters.

3. The lumping and splitting is done by experts.

4. The person who owns the stuff also owns the organization of the stuff. You can’t come into the clothing store and rearrange it the way that suits you.

5. Lumping and splitting results in a neat and clean order. It’s clean-edged.

But now we’re digitizing information, resulting in a third order of order in which we break the rules of real-world order:

1. Things can go in more than one pile – You put your e-store’s hiking boots under shoes, men’s and women’s apparel, outdoor wear, popular items, items on sale, etc.

2. The arrangement can be different for each person.

3. You or your social group are the experts.

4. Users get to control the organization of the stuff.

5. Messiness is a virtue on the Web.

You can see much of this in the rise of tagging: Users create the metadata and anyone can figure out how to sort through it and organize it. It’s out of the hands of the owners of the stuff being classified.

So, what I’m saying is that we’re moving from thinking that the right way to arrange — and understand — things is to figure out the taxonomic tree ahead of time. Instead, make a big pile of leaves, each with lots of metadata, and allow users to add more metadata and to sort and categorize it as they need.

But there are problems with this, especially with regard to tags:

– One word can have many meanings, and one meaning can have many words. As tagging gets more popular, that’ll be a bigger issue.

– If we form social groups based around how we use words, we run the risk of fragmenting ourselves further, this time around semantics.

– Folksonomies can reinforce homogeneity.

But, I’m hopeful. Ask why tagging is happening now. After all, we’ve been able to tag Word documents forever, but we don’t. Why now? I think in part it’s because we are tagging not just for ourselves. We’re doing it socially, aware that we’re making the Web better for others. Together we’re building a new infrastructure of meaning, created by and for one another. We will figure out amazing things to do with this new social semantic construct.

(Unfortunately, I didn’t take notes during the discussion. Drat!) [Technorati tags: tags etech taxonomy]

13 Comments »

RageBoy has discovered that Amazon seems to be rolling out a feature that shows you for any particular book which phrases in it are “statistically improbable.” For example, Chris’ own Gonzo Marketing uses the phrase “public journalism” and “market advocacy.” Obviously those are not phrases unique to Chris’ book, so Amazon is doing some sort of statistical analysis to find phrases that have some prominence within a book and across books. Fascinating.

Unfortunately, apparently you need to be using the Safari browser to see this on Amazon. Or perhaps you need to be taking the same drugs as RB. Either way, get a Mac…before it gets you.

3 Comments »

There’s no “I” in “team” but there are two “me”s in meme.

4 Comments »

RageBoy, everyone’s Chief Blogging Officer, has a good example (involving something I wrote) of how errors creep into the media: The humor was missed, the main point was ignored in favor of the inflammatory one, and the nesting of the quote was flattened.

The point isn’t that the media sometimes make mistakes. We all know that. For me, the point is that it was too small an error for the medium to acknowledge. I suppose I could have written a letter and they would have run it in their corrections box. But that would have been so long after the event that, with a mistake this size, who would have remembered the context or cared?

But, had it been a blog that got it wrong, the blogger would have fixed it immediately, probably with a little embarrassed aside.

So, while bloggers may get more facts wrong, our corrections are finer-grained. And, of course funnier.

4 Comments »

Another BBC talk. Gavin Bell, Matt Biddulph, and Tom Coates. How can you make media objects — i.e., all the programs ‘n’ stuff the BBC has — addressable? And then what could you do with them?

To make a program addressable, they say, you give it an identifier. (Their first ever identifier was kr7rm.) All BBC and radio programming will be addressable. 40,000 hours of national tv broadcast across 8 tv stations every year. 76,000 hours of national radio across ten networks.

They say: “This is what we might be doing after we get bored of broadcast.”

The content they deal with is complex. What’s a program? There are versions and edits, different broadcasts of the same program. There are series. There are specials. Documentaries. Films. There are “time slots.” Bulletins.

The data is scattered throughout the organization. There are maybe 15 systems in the BBC to track this data.

How about the logistical problems? Schedule changes, breaking news, legal issues, the sheer scale of the BBC.

The core of the representation is the episode. e.g., episode 2 of Absolutely Fabulous in Series 1. They create a web page for every episode of every program the BBC creates. They want to make the episodes as rich with information as possible, but the Web page has to be simple. Every individual episode will be uniquely identifiable and addressable forever via a URL. They also have persistent schedules so you can see any program that aired on any day.

They move everything into a big database using a complex data format, Standard Media Exchange Framework (SMEF) that includes more firleds than they care about, including awards. Anyone in BBC can connect to it via an API.

Things you could build on top of this: Statistics. Audio/visual on demand. RSS feed to tell you when an episode is. Social software: What’s the most watched program in your circle, buddy lists, etc. [Technorati tags: etech bbc]

1 Comment »

Clay talks about how taxonomies always have values built in. Even the periodic table’s “noble gases” division reflects an assumption about the “essential” state of elements. He points to the Dewey Decimal System’s skewed religion category. [Yikes! I’ve been doing that, too! I probably heard it from Clay first. I will attribute it from now on. Ack!] Even the Library of Congress puts the Balkan Peninsula and African on equal footing because it’s measuring the number of books on the shelves. The categorization reflects not the ideas but the physical storage.

He points out, that even though Yahoo has cross links, it has a concept of which is the real categorization; it only shows you the number of links in a category if you are seeing the category in its “real” environment.

Hierarchical systems, he says, inevitably adds hyperlinks to cut across the taxonomy. (Yahoo only allows three, he says, because they didn’t want to be spammed.) If you have enough links, you don’t need the hierarchy. “There is no shelf.”

When does ontological organization work? When you don’t have a lot of stuff, it’s stable, things have clear edges, an authoritative source and trained users. I.e., the opposite of the Web.

People have assumed that tags that mean the same thing are actually the same, but (Clay says) “movie” people don’t want to hang out with “cinema” people, and “queer” people certainly don’t want to hang out with “homosexual” people. There is information in the differences that thesauruses and categorization schemes miss.

He shows graphs that provide evidence that tagging forms power laws — who tags, and the tags that individuals use.

Organic categorization uses market logic, merged from URLs not categories. The categories overlap, not synch. The mergers are probabilistic, not binary. User and time are core attributes; you can do grouping, inclusion/exclusion, and decay. The semantics are in the users, not in the system; this is not a way to get computers to understand one another.

[Great talk. As always.] [I spoke with Clay afterwards and he talked about Dewey two years ago during an etech talk I must have been at. I apologized. He was, of course, gracious. I will be adding the phrase, “As Clay says” to yet more of what I say. Happily.] [Technorati tags: etech shirky taxonomy]

8 Comments »

Tom describes student projects. [I missed many of these]

– A purse displays when wifi is present

– A protest button initiates a DOS attack on a nearby malevolent corporation

– “Needies” — stuffed animals with mp3 players. If two get together, they talk shit about others behind their backs.

– CareNet displays grandma’s life signs around the edge of an electronic photo of her

– Junkie’s Little Helper: If levels of meds in a med cabinet drops low, it goes on line and alerts IRC chats that the person is high

– Ku: It communicates sadness over the Net. You sit in a chair waiting for the ceiling to cry on you because someone is sad. (A water pump is involved.)

– White stone: A stone gets warm when someone sends you a hug.

[Technorati tags: etech igoe]

1 Comment »

Clay begins a segment on tech and education.

He says he thinks of his group at NYU as “The Department of the Recently Possible.” A few years ago they noticed that students were increasingly integrating phones into their apps. So they started looking into it. One experiment: PacManhattan that mates the urban grid and the game grid. The runners are controlled by people in a control room via mobile phones.

DodgeBall was an experiment in mobile social networking. “Mobile phones are the first things since keys that everyone carries,” Clay says, citing Marko Ahtisaari. DodgeBall alerted him that there was someone one degree removed from his social circle in a bar he was in. The system initially suffered from “the ex-girlfriend bug”: She would still be in your circle friends of friends so it would keep suggesting you meet up with her, so you need a way to “denominate” links. [Apparently this was a problem with ex-boyfriends as well.]

Phones are not becoming PCs because: Only the minimum platform is widespread. There’s no “ocean of practice.” Manufacturers don’t understand the benefits of allowing hacking. The US carriers work against ubiquity. So, phones will continue to augment PCs. E.g., upload from phone and view on PC (e.g., flickr).

Mesh is coming, but not soon, Clay says. [Technorati tags: etech shirky]

Be the first to comment »