Metadata Generation Research Project:
"The metadata generation research project is developing a model that will facilitate the most efficient and effective means of metadata production by integrating human and automatic processes."

Short film wows voters in seconds: "A surreal 15-second black movie comedy about an escapologist has won a short film contest."

<a title="Language Log: like is , like, not really like if you will" href="http://itre.cis.upenn.edu/~myl/languagelog/archives/000141.html">Language Log: like is , like, not really like if you will: "like is definitely a more powerful (and useful) expression than if you will. Perhaps that's why some people use it, like, too much?"

Orange Cone: A photogeoblog sketch: I like these stories that envision how things might work.

Sam points out on Afongen that O'Reilly have a Blog This button, for example on this page.

Clicking it shows a popup window that lists some code and suggested text to paste in your blog entry. I don't like it.

Bloug: "So a modest proposal: what if everyone involved in content management--the publications, the web sites, the meetings and conferences--banned CMS vendors for, say, one quarter? No vendor exhibitions at meetings, no product mentions on discussion lists, no CMS purchases, no nothing. Just discussion about all there is to content management besides the technologies."

Lou is right: the CMS discussion is dominated by the CMS vendors - that needs to change.

Many-to-Many: Otlet: Some ideas die because they are wrong: "The failure of universal subject classification working in concert with the mutable forces of scholarship didn't happen because that idea fell out of fashion - it was fashionable as recently as 1998, with people being paid fabulous sums of money to pursue it. It failed because it does not work."

Simon Willison implemented daily links at the top of his blog. I really like the CSS treatment of the visited links: in a dense list like this it makes sense. The strikethrough links are the ones I just visited:

A newborn: InformationScienceTheoryWiki

On the drawingcenter.org, I saw these directions:

If you are traveling by subway, you can take the trains to the Canal Street station.

In the code, the trains are a bunch of image tags. This can be easily made acessible by adding ALT tags, but I wanted to try something more semantic (not sure if it's useful, just for fun, after my BLOCKQUOTE experiments).

My best take was to use SPAN tags and the letters of the trains, but I couldn't get the CSS to replace the letter with the image... I'd be thrilled if someone could crack this!

James pointed out that the Drawing Center have an exhibition about Mark Lombardi (runs until December 18). Check it out if you are in NYC and interested in social network mapping.

(via Danny Ayers) SchemaWeb - RDF Schemas Directory: "SchemaWeb is a place for developers and designers working with RDF. It provides a comprehensive directory of RDF schemas to be browsed and searched by human agents and also an extensive set of web services to be used by RDF agents and reasoning software applications that wish to obtain real-time schema information whilst processing RDF data."

I got Tom's picture as well now!

I found a picture of Dare Obasanjo, still looking for Joe Gregorio, JayT, and Tom Hoffman.

Center For the Ethnography of Every Day Life: "Before the abstractions of social science, there are people's stories, the emotional worlds of disappointment and uncertainty, and the brave coping of everyday life. Established in 1998 with a grant from the Alfred P. Sloan Foundation, the Center for the Ethnography of Everyday Life fosters research and training to document the challenges of American working families. Working people, everyday lives explored in the tradition where ethnography and documentary come together."

Hey, so it's ugly, but that's all my fault. Really. Thanks to the excellent Bloglines service that I've been using lately (it just works for me), I've got a blogroll!

Jon is experimenting with automated categorization of blog posts. XML.com: Working with Bayesian Categorizers: "There's been some discussion in the blog world about using a Bayesian categorizer to enable a person to discriminate along various interest/non-interest axes."

Seb's Open Research: "I'm sure Marc will love the way it's crafted." I'm not so sure, but I'd love some feedback. I suck at CSS (just look at the source of this page!).

If you want to order prints of your digital pictures online, I can really recommend Shutterfly. You can order 15 prints for free when you sign up, just to try them out. Although I found myself ordering a lot more on my first order because it was so easy.

Victor coins the taxonomy dance.

Basic level categories

I've been waiting for someone to write about basic level categories as they relate to information architecture. No luck so far (apart from a 1999 Peterme post in which he says: 'the trick would seem to be to get people to the basic-level as quickly as possible.'). So I'm picking this up again. There's gold in them mountains folks!

Coginitive science has been making many discoveries about how humans categorize, like: that categories have fuzzy boundaries, that members of a category may be related to one another without all members having any property in common (this is called Family resemblance), that some members of a category may be �better examples� than others (this is called centralicity), and most interestingly, that categories are organized into a hierarchy from the most general to the most specific, but the level that is most cognitively basic is �in the middle� of the hierarchy. These categories in the middle are called basic level categories.

For example, "cat" is a basic level category, "feline" or "Siamese cat" are not.

Basic level categories have some characteristics that make them interesting for information architects:

- Things are remembered more readily at basic level.
- People name things more readily at basic level.
- The basic level name for things is learned earliest in childhood.
- Languages have simpler names at basic level.

In short, people naturally, at a deep cognitive level, deal easier with basic level categories.

It is important to understand that basic level categories are not just easier on a superficial level, because they are shorter or something. Cognitive scientists say that basic level categories are cognitively real. They seem to be ingrained in the human mind somehow, in a way that makes it easier for us to deal with basic level categories.

Does this mean that information architects should be aware of the basic levelness of the categories they use? I think so, but I'm not sure how exactly. Remember that basic level cateogories are processed more easily, faster. That has got to mean something to us!

The only research I found about basic level categories in information retrieval is Using 'basic level categories' to retrieve multimedia from the World-Wide-Web Hoenkamp, E.C.M. (1999). Proceedings of the 21st Annual Conference of the Cognitive Science Society, 1999, 796.

Other interesting things I came across while doing research for this:
- one user�s classes are another user�s attributes.
- To test whether a category member is more or less central to the category, you can ask a series of questions, compare how long it takes people to answer.

Learn more:
- It's all Eleanor Rosch's fault, well explained by George Lakoff's in Women, Fire and Dangerous things.
- More goodies.

Boundary objects are everywhere. Back in October, Denham Grey wrote about Boundary objects and KM. Judith Meskill adds a long and yummie list of research links.

Interaction design is discovering boundary objects: Shared boundaries. (Interaction Design Hub): "It is quite obvious that ours is a “community of interest” rather than a “community of practice”, and that “boundary objects” abound." Sweet.

A keeper: How to shop for a house.

It just occurred to me we should have a tag to indicate where in a quote we are doing this: "[the program] crashed twice" (it's not paraphrasing - what is this called?)

Search Engine Decoder: who provides what information to who?

Good Experience - The ROSE framework: "Business results are metrics that the CEO can understand." Must be a typo!

Kottke is redesigning his site, and trying out giving different types of content (book reviews, links, ...) a different look on his homepage.

Jason identifies 5 content types on his site:

- Movie Review
- Book Review
- Remaindered Links ("hey look at this thing")
- A Comment (a comment posted on another site)
- A Regular Post (Jason says this means "here's what I think about this thing". It's really an 'everything else' category - classify here if a post fits in none of the other categories.)

Any given post can be easily classified in 1 and only 1 of these 5 categories - there are no obvious overlaps, especially if we treat A Regular Post as an 'everything else' category. No post will belong to more than 1 content type.

Next, apart from fields that are generic to all content (author, creation date, ...), different content types each have different fields, "microcontent-specific entry fields". For example, a movie review has a title, a link, a rating, a photo, and some text.

The discussion talks about how this can be supported by weblog tools, while staying generic enough. I've been doing this semantic stuff for the past year at my job, and it's hard, but doable and useful.

BBC news: "The designers say the 70-mm-tall device could be used as a "flying camera" to enter earthquake-shattered buildings. " Plain cool.

The X-Bar: Child Language Acquisition: "In my experience, not only are children not explicitly taught language, but correction is a fruitless task."

Why is it that on many all-CSS sites, selecting text becomes almost impossible? (See for example the new Sprint site.) I compulsively select some text when browsing the web (don't ask), and CSS websites often don't let me. Apart from my obsession, breaking text selection seems like breaking a fundamental UI interaction pattern to me.

Murray Altheim asks the Topicmap community what they mean when they talk about facets in [topicmapmail] Two Models of Facets. Facets in topicmaps are not what you'd think they would be.

I'm OK with Outlook, and adding contacts to my contact list is easy, but the interface to browse/use the contacts is useless. So: am I missing something, or is there some plugin to Outlook I can use to browse my contact info? Export it and use another app? All tips welcome.

(I used to ask this stuff on mailing lists, I don't know what happened ...)

This has happened to me before and seems to be a recurring problem at Apple.com: this page links (the Take a Closer Look link) to Apple - Page Not Found.

Tanya confirms my understanding of boundary objects. I'm happy, because I wasn't entirely sure I was getting it right. She also points to a study that looks at the wireframe as a boundary object.

For me, it's been useful to think of many of our deliverables (taxonomies, wireframes and such) as boundary objects that build bridges between different communities of practice. It alleviates much frustration.

I am collecting a list of Companies That Do Ethnograohy.

The Guide To Ethnography wiki is back.

Does anyone have real-life experience with working with a programmer in countries like India through a service like Technical Outsourcing or Coders4rent.com? Tell me your stories. Was it easy to convey requirements? Were you happy with the communication? With the quality of the work?

(via Catalagoblog) Library Juice 6:24: "I've made a stab at a typology of the "amusing search," with
lessons and questions arising from each."

(via languagelog) Wiktionary: Colours. A wiki dictionary with translations of thousands of words in hundreds of languages.

LANGUAGE = DISEASE? (in the comments): "The fact that for thirty years or more it's been possible to get certified as a linguist without knowing any languages beyond your native one seems to me a perversion of the order of things that likely presages Armageddon."

So now you can make a documentary or a movie all by yourself and sell it online.

Welcome to the TEI Website: "Initially launched in 1987, the TEI is an international and interdisciplinary standard that helps libraries, museums, publishers, and individual scholars represent all kinds of literary and linguistic texts for online research and teaching, using an encoding scheme that is maximally expressive and minimally obsolescent."

Numbers in Over 4000 Languages

I have a bunch of books I'd give away of someone would want them (if they come and pick them up). Is there something online where I can do that?

I wrote an article called Themes and metaphors in the semantic web discussion. I tried out a new approach to semantics in blockquotes, but don't know much about CSS, so let me know how it can be improved...

I'm looking for a political cartoonist who is interested in doing a cartoon or 2 a month for my redesigned website about Colombia. Subject matter will be much about Colombian politics, and how it relates to US politics (so characters include Bush and the Colombian president Uribe). I can work with you to develop the cartoons (ie, you don't have to know too much about Colombian politics, we can work together).

Anyone know where to look?

(via Simon) PHP.net have implemented a javascript autocomplete for their function search. Note that this is not the browser autocomplete that lists things you typed in before. Instead, this dropdown lists all valid functions starting with "my". Very, very useful.

Simon notes that the javascript is not being shared (it's scrambled), and in the comments Sergi points to this autocomplete script (which isn't half as nice). A quick Google reveals:
- How to make your own.
- You can ask IE to show the dropdown on your form elements.
- How to make your own (another one).

I am very dissapointed that the W3C isn't putting these controls in the HTML specs (correct me if I'm wrong here). Things like autocomplete, list or table sorting, ordering lists or validation should be taken care of by the browser by specifying a simple html tag or attribute.

All this javascript is nice but it isn't really going anywhere - if a truly useful and compatible javascript library for this stuff that didn't require lots of customization was going to happen, it would be here already. (With my respect to the authors of the various libraries out there - they're doing a great job - but it's not good enough. Not easy enough to implement. Not standardized enough.)

CMS Feature Directory: a unusual inerface to a faceted classification system where you can order "facet layers". I'm not too crazy about this interface myself.

Paolo Valdemarin Weblog:
"As an example, in Dave's taxonomy you can find:
- Politics
-- Presidential Election of 2004
--- Dean Campaign
--- Clark Campaign
--- ...
If Howard Dean would have his own taxonomy, it might contain something like:
- Howard
-- Presidential Election of 2004
--- Politics
--- Budget
--- Ideas
--- ...
Same topics, different nesting according to the different points of view."

Wrong. Same terms, different topics (although you could come up with some level of cross mapping that might work for you). Dean's topic "Presidential Election of 2004" is not the same as Winer's topic "Presidential Election of 2004", although they use the same term. Dean's "Presidential Election of 2004" is probably somewhat similar to Winer's "Dean Campaign" topic.