The Community Engine Blog: folksonomy Archives

Veg-o-matic: An alpha web service using xFolk

“It slices, it dices ...” More specifically, this post presents an alpha release of a web service based on greasemonkey and reblog that identifies, validates, cleans, and republishes xFolk microformatted content. All comments are welcome.

I'm releasing a very alpha hack of an xFolk web service for this labor day weekend. Recall that xFolk is a microformat for representing the sort of social bookmark data you would find on a site like del.icio.us. A lot of people are republishing their links from such sites in their blogs. The advantage of using a microformat like xFolk for this sort of thing is that it makes it easier to write software to do something useful with the data.

The hack I am presenting here is a combination of a greasemonkey script and an alpha version of the reBlog refeed tool by eyebeam research and stamen design. The main action of interest is in the greasemonkey script that:

Finds all instances of xFolk microformatted content in a page.
Determines which of those instances are valid (i.e., contain all the required elements as indicated in the xFolk spec).
Creats a clean clone of the xFolk entry that only contains elements specified in the spec.
Uses an xml object to serialize the cleaned clone into a form element for republishing.
Adds a script to the valid xFolk instances that makes it possible to republish their “cleaned” version via Mike Migurski's alpha rewrite of the reblog republishing tool released by eyebeam and stamen

Continue reading "Veg-o-matic: An alpha web service using xFolk"

Bud posted this on September 2, 2005 | Permalink | Comments (0) | TrackBack (0)

Imitation is the surest sign of success

I disagree with Jason Calacanis. IceRocket's adoption of tagging using a format invented by technorati indicates that technorati is achieving wide marketplace acceptance. By my estimates over a million bloggers are using the technorati format.

Today, Jason Calacanis wrote about how icerocket, a blog search engine, had added tagging support. Jason's remarks were, typically for him lately, riding technorati a bit.

But it's how IceRocket supports tags that's the real kicker. They're using the reltag microformat invented by Tantek Çelik, Kevin Marks, and Derek Powazek of technorati as explained on this IceRocket help page.

Continue reading "Imitation is the surest sign of success"

Bud posted this on July 20, 2005 | Permalink | Comments (2) | TrackBack (0)

Distributed Tagging Hell

When many pieces loosely joined break, all hell can break loose without too much effort. Apologies to Tim O'Reilly who I have tracked back to 4 times at least for just one of his posts and all of you who are getting duplicate posts in your feeds.

At O'Reilly's Where 2.0, Stephen Randall put it best in terms of what he wants for usability:

One hand (not complex for those who get it)
Two billion people (anyone can get it)
Three steps (no time commitment)

Neither Movable Type nor distributed tagging (via technorati) are like this, but they need to be.

Continue reading "Distributed Tagging Hell"

Bud posted this on June 30, 2005 | Permalink | Comments (0) | TrackBack (0)

Microformats and the remixable web

To answer Jon Udell and others, microformats enable the remixable web.

Yesterday, I spent the day at technorati and the evening with Tantek Çelik, Ryan King, Jeff Barr, Chris Messina, and a host of others at the second ever (?) microformats group dinner. A lot happened, and I'm not here to give a blow-by-blow account. Rather, I want to convey the one major insight I walked away with (there were many minor ones). To answer Jon Udell, Microformats enable the remixable web.

Continue reading "Microformats and the remixable web"

Bud posted this on June 30, 2005 | Permalink | Comments (0) | TrackBack (0)

Microformats and do-it-yourself vertical search aggregation

Vertical search aggregation allows sites to become known for particular topics and attain search engine visibility. To this end, it uses microformats as a glue to integrate blogging and folksonomy tagging.

Microformats are pre-agreed, human friendly ways of formatting web posts so that machines can process them. Microformats exist for calendar entries, reviews, tagging web posts, and tagging links in link blogs. In brief, microformats are like templates in Microsoft Word but aimed at the open format world-wide web.

It turns out that some arcane features of microformats can be quite determinant in creating search engine visibility. In particular, the reltag format (for tagging blog posts) requires that tags (labels you apply to posts so that people will know how you categorize them) point at a URL that acts as a repository or definition page for the tag. This requirement means that tag repositories are likely to receive a lot of links giving them high search visibility and effectively making tag pages aggregators for vertical search on specific topics.

As I mentioned last week, technorati seems to be one of the first organizations to actively build a business model around this phenomenon. A question raised in subsequent emails is how can an individual site or family of sites use tagging and microformats to get themselves on the map like technorati. In this post, I'd like to propose a light weight mechanism for doing just that.

Continue reading " Microformats and do-it-yourself vertical search aggregation"

Bud posted this on June 19, 2005 | Permalink | Comments (1) | TrackBack (1)

West Coast Mini-Tour, June 26 – 30

If you'd like to meet me during the mini-tour, please drop me a line. Email link is in the post body.

June 26 – 30, I'm going to be in the Bay Area (San Francisco, specifically) catching up with Tantek Çelik and “The” Ryan King on microformats, likely attending the Vertical Leap conference signaled to me by Dave McClure of simplyhired, and attending at least part of Where 2.0. It's going to be an exciting trip. As I signaled earlier this week, microformats have some real implications for distributed business models. Specialty search engines like technorati, simplyhired, and gataga would seem well positioned to take advantage of microformats in their business models.

If you would like to see me on this tour, drop me an email. I'm very interested in meeting readers. Please note that my geographic knowledge of the Bay Area is limited. For instance, I thought the SDForum where Vertical Leap will be held was in San Diego (SD, get it), so keep that in mind when suggesting meet-up locations. Fortunately, a quick phone call with Dave McClure cleared up the SDForum confusion before I booked a flight to San Diego.

Continue reading "West Coast Mini-Tour, June 26 – 30"

Bud posted this on June 18, 2005 | Permalink | Comments (0) | TrackBack (0)

Folksonomy makes tag aggregators king of search rankings

Technorati tag aggregation pages are achieving top ten search results for significant, niche terms like podcasting, folksonomy, and blogosphere. Technorati's introduction of the seemingly obscure reltag microformat is at the root of its rise in the search rankings. Competitors seeking to duplicate technorati's results need only adapt the reltag or xFolk microformats to their own ends.

Current wisdom has it that, in web publishing, content is king. As the web has grown to upward of 11.5 billion pages, search has become the default way of navigating the web. Search engines determine a site's ranking based on indexable content, so content as seen by search engines had better be your king if you want to be found.

What current wisdom has not allowed for is the idea that folksonomy tag aggregators like technorati might form an intervening layer between content providers and search engines. There is evidence that this has started to occur in topics related to social software during the six short months since technorati began to aggregate weblog posts by tag.

Continue reading "Folksonomy makes tag aggregators king of search rankings"

Bud posted this on June 13, 2005 | Permalink | Comments (2) | TrackBack (0)

xFolk — Blogmarks really steps up to the plate

Blogmarks has provided excellent support for xFolk in their Blog Sync functionality.

I meant to post about this earlier but have been very busy with client work. Blogmarks has really stepped up to the plate with xFolk support in their newly renovated blog sync functionality. Blog sync allows you to republish in your blog the links you bookmarked that day.

Continue reading "xFolk — Blogmarks really steps up to the plate"

Bud posted this on June 12, 2005 | Permalink | Comments (0) | TrackBack (0)

de.lirio.us implements xFolk

Steve Mallett of de.lirio.us has formatted their folksonomy listing pages using xFolk. When folksonomy publishers implement xFolk as their output format, their users still have the same hassle-free bookmarking interface and gain more highly semantic markup for no cost. The more semantic the markup, the better Internet visibility.

Steve Mallett, creator of de.lirio.us, has formatted their folksonomy listing pages in xFolk:

I set up de.lirio.us to do this [publish using xFolk markup] a few minutes ago. I must say I do like this a better than the technorati tag format thing. I like technorati, but it's a bit too technorati-centric.
Conversation on tags & liberating tagged data - I did it anyway | Fooworks - 'blog of Steve Mallett

Steve's major issue was how using xFolk would affect the end user, a point raised just today by Tantek Çellik in regard to gaining acceptance for microformats.

Continue reading "de.lirio.us implements xFolk"

Bud posted this on June 5, 2005 | Permalink | Comments (0) | TrackBack (1)

An xFolk 0.4 Implementation for Blogmarks

xhtml microformats like xFolk make sharing information easier. Their greatest value is realized when they are used by large numbers of people. To that end, I am pushing for xFolk's implementation as a publishing format for several social bookmarking tools. Blogmarks is the first to respond.

As put so well by Peter Merholz among others, xFolk is:

... a markup extension to allow tags to be created anywhere, not just through systems like del.icio.us.
You’re It! » xFolk - technology for decentralized tagging

Just to be clear, the “tagging” he is talking about is when people bookmark things they find on the web under informal categories (tags). These tags can make that bookmark easier to find later and to quickly find related (similarly tagged) resources. Systems like del.icio.us enable the tagging (labeling) itself, and they let users share the tagged bookmarks among themselves. Tagging has proven to be quite popular, at least among a certain group. You might think of it as appealing to the same urge that causes people to become coin, art, or music collectors.

The point of xFolk is to make it possible to easily share your bookmarks outside of systems like del.icio.us. One important step in getting this vision to reality is to get services that enable tagging to use the xFolk format. BlogMarks.net (a social bookmarking tool originally signaled to me by Jonas Luster) has been kind enough to share template code that you can use with their blog sync feature to make xFolk formatted bookmarks from their service appear in your blog.

Continue reading "An xFolk 0.4 Implementation for Blogmarks"

Bud posted this on June 3, 2005 | Permalink | Comments (0) | TrackBack (1)

Archives are at the heart of decentralized communities

In decentralized, emergent communities, the community archive defines the community over time. Therefore, designers of such communities need to pay attention to the processes by which these archives emerge. The ongoing debate over folksonomy provides us with a public record of decentralized archiving strategies that do and don't work.

In describing one of the fundamental building blocks of decentralized blog communities, James Farmer remarks:

If you’re implementing blogs in your school, college or university or even organisation one of the first things you’re going to want to think about is aggregation.
BLOGSAVVY » Savvy public aggreation tools, the key to binding your blogs together

Nothing could be more true. The “River of News” style public aggregators that James then goes on to review all do a good job of displaying a decentralized community's zeitgeist. But, by itself, zeitgeist as captured by near real-time blog aggregation is insufficient to sustain decentralized communities for the long term. For that, these communities need archives.

Archives crystallize what such communities found important over time and ultimately define the community.

Continue reading "Archives are at the heart of decentralized communities"

Bud posted this on June 2, 2005 | Permalink | Comments (0) | TrackBack (0)

Web 2.0 can and will be commoditized

Many information business models rely on network effects derived from their exclusive access to social networks. Standardized xhtml microformats for sharing social data level the playing field for smaller players by making it possible for them to pool their data. These pooled data can provide the same positive network effects as the exclusive networks possessed by larger players.

Peter Merholz makes an interesting observation regarding what many are starting to call Web 2.0:

On a sales call with a potential client, I tried to impress upon her the need to fundamentally reconsider how her company approaches what they do, and I used the analogy of Snapfish/Ofoto/Shutterfly and Flickr. The former were stuck in pre-Web, pre-networked-world ways of thinking about people, things, and relationships. The latter is built, ground-up, *of* the Web, and recognizes that the “value-add” (as business types like to call) lies not in the production of things (which inevitably get commoditized and provide negligible margins), but in the provision of services that provide an experience you simply can't get anywhere else.
peterme.com: Death Throes of a Business Model

I've been giving this topic a lot of thought lately as I have moved forward with developing and trying to sell business owners on the xFolk xhtml microformat, a format designed specifically to commoditize large aspects of the types of services Peter is describing.

Continue reading "Web 2.0 can and will be commoditized"

Bud posted this on June 1, 2005 | Permalink | Comments (8) | TrackBack (1)

xFolk Entry 0.4 — Microformat for decentralized tagging

xFolk Entry 0.4 is a new iteration of the xFolk microformat that is extremely easy to implement. It enables the publication of tagged bookmarks so that they can be harvested on the web and aggregated into folksonomies. As such, xFolk eliminates the need to rely on centralized data repositories to create folksonomies.

xFolk is an xhtml microformat that enables users to tag and share bookmarks on the Internet without using a centralized system such as del.icio.us or flickr. To give a concrete idea of how xFolk facilitates decentralized bookmark tagging, consider a writer who wishes to publish a list of related links at the end of a web article. At the end of the article, the writer simply formats the links and tags in xFolk. Then a web crawler that understands xFolk can digest the page and extract the link information, placing it in one of possibly many directories. In this example, xFolk's underlying use of well understood standards already widely used by publishers and crawlers eliminates the need for the centralized services that currently exist. A similar scenario exists for link blogs.

xFolk 0.4 vastly simplifies the xFolk 0.3 microformat proposed six weeks ago so that it can be implemented in under five minutes on a variety of web publishing platforms. The simplification was achieved in part by splitting xFolk 0.3 in two: (1) xFolk Entry, described here, for indicating folksonomy tags and annotations for links in web articles and (2) xFolk Definition, described in a later post, for providing folksonomy tag semantics. Additionally, xFolk Entry 0.4 now makes use of the widely adopted reltag format, further limiting the scope that xFolk itself must cover. Finally, xFolk Entry 0.4 reduces the number of class attributes to three without a reduction in expressive power.

Continue reading "xFolk Entry 0.4 — Microformat for decentralized tagging"

Bud posted this on May 31, 2005 | Permalink | Comments (10) | TrackBack (7)

Review of Google Personalized Home Page

I like the Google Personalized Home Page, but it just provides an artificial sample of what I want to watch to anyone interested. Give me the ability to add my own RSS feeds, and you'll get a real picture of the top things I pay attention to. Oh, and by the way, throw in tagging too.

John Battelle reports on Google's new personalized home page which appears to be called just that, “Personalized Home Page”.

Google is launching the kind of personalized integration tool that many thought they'd never do. At first it was thought to be called iGoogle, but the name is uncertain at this point.
John Battelle's Searchblog: MyGoogle Is Coming Today

Well, I've tried it, and for a proof of concept, I think it is great. Basically, it lets you add a limited set of sources to your main Google search page including Google News, Slashdot, BBC, Wired News, Weather, GMail, and Driving Directions. These fit my “quick hit” information needs.

There are no ads!

But, I would like just one more thing.

Continue reading "Review of Google Personalized Home Page"

Bud posted this on May 19, 2005 | Permalink | Comments (0) | TrackBack (0)

Blogging and Folksonomy: Experts, Fear, and Why You Can't Go Back

Fear is a powerful tool experts use to get you to pay them for their opinion. Is a lot of the fear, uncertainty, and doubt we hear about blogging, tagging, and other social media just an attempt by threatened experts to re-establish their market value in the face of revolutionary change?

Blaise Cronin, Dean of the School of Library and Information Science at Indiana University notes:

The present generation of bloggers seems to imagine that such crassly egotistical behavior is socially acceptable and that time-honored editorial and filtering functions have no place in cyberspace. Undoubtedly, these are the same individuals who believe that the free-for-all, communitarian approach of Wikipedia is the way forward. Librarians, of course, know better.
via Robert Scoble in The Red Couch: Blog backlash, academian style

Although many don't realize it, Information Science has played a central role in the evolution of the Internet and is one of the fields that finds itself the most altered by it. Lately we've seen a lot of respected professionals like Dean Cronin weigh in on the value of blogging, tagging, and other social media. Invariably, the refrain is a call to return to a higher, earlier standard, or at least a plaint about the current degradation of standards and lack of true intellectual content.

I have a few reactions:

Continue reading "Blogging and Folksonomy: Experts, Fear, and Why You Can't Go Back"

Bud posted this on May 8, 2005 | Permalink | Comments (1) | TrackBack (0)

xFolk 0.3 — xhtml microformat for emergence

In this post, I have described an iteration of xFolk that is much more similar to previous microformat efforts in how it specifies and uses attribute values. This version should be easy to implement in templates and tools. Ten examples from current web sites illustrate how xFolk might be implemented in functioning web pages.

xFolk is an xhtml microformat that allows people to apply tags to links in web articles and even whole articles themselves. These tags can be thought of as subject headings that help readers understand what the linked reference is all about and perhaps lead to whole repositories of references on that heading. By providing this service to readers, xFolk also enhances the ability of search engines to process and index the pages in which the links occur as well as the pages the links point at. In this way, xFolk also provides the benefits of structured blogging without introducing any markup beyond standard xhtml.

This post provides a new iteration of xFolk markup based on extensive feedback from Tantek Çelik, Kevin Marks, and Thomas Vander Wal as well as various postings from around the web. The purpose of the new iteration is to provide a version that: (1) fits within the standard approach for defining such formats as provided by examples like XFN, Votelinks, XOXO, hCard, and hCalendar; and (2) can be easily implemented in weblog and other softare. The new iteration also retains the original goal of allowing authors to express explicit semantics (meanings) for their tags if they choose.

The rest of this post proceeds as follows. I'll simply lay out the new version of the microformat. Then I'll very briefly lay out the conceptual use cases and finally go through ten applications pulled from working sites on the web.

Continue reading "xFolk 0.3 — xhtml microformat for emergence"

Bud posted this on April 19, 2005 | Permalink | Comments (2) | TrackBack (2)

To work, folksonomy tags need to be distributed

Distributed folksonomy tagging services like technorati should be more spam resistant because they rely on people's natural linking behavior to determine the quality of the information they are presenting. Centralized services require a dedicated group of users to perform this same function.

Folksonomy tagging is a growing practice on the web where people apply one or more one word tags to things like bookmarks and photos to help them recall them later. Services such as flickr and del.icio.us provide centralized community web sites where users store these tagged items. As a value-added feature, both services then provide aggregations for tags so that users can see what they and everybody else has given a tag of say “Boston”. In today's Search Engine Blog, Danny Sullivan raises the flag regarding potential abuses of tagging:

Wide-open tagging, where anyone can get their pages to the top of a list just by labeling it so, is going to be a giant spam magnet.
Another Poke At Tags As Search Savior

To which Steve Rubel responds:

Continue reading "To work, folksonomy tags need to be distributed"

Bud posted this on April 14, 2005 | Permalink | Comments (2) | TrackBack (4)

xFolk — schema for emergence

Folksonomy provides value through emergent structure and direct personal value to the user. In reformulating xFolk, this post focuses on the data schema needed to support emergent structure. The next post will focus on the xhtml microformat.

Yesterday, Dave Sifry announced Technorati's related tag feature. Tags are simple, one word descriptors that people apply to their own and others' web articles for later recall. Services like Technorati, del.icio.us, de.lirio.us, and scuttle aggregate these tags into folksonomies that exhibit useful emergent structure. The simplest example of one such emergent structure arises from assuming that people mean the same thing when they use the same tag and then using the aggregated tags as a sort of navigation hierarchy. All of the services mentioned do this. With technorati's announcement, they all now also offer a related tags feature. The simplest version of related tags is to say that tags are related when enough people use them to label the same set of artifacts. For instance, if one set of users are labeling a set of sites with “baseball” and another group are labeling the same sites with “MLb”, it is reasonable to assume the two tags are related. These relations can be added to enrich the navigation structure.

There are many more examples of such emergence through aggregation, currently less explored, that have potentially greater commercial value. xFolk is an xhtml microformat that makes it easier for people to share their vision of the world by tagging URLs in their web articles. People will derive value from xFolk in two ways: (1) from information that emerges by aggregating individual tags across many people or even just one person; (2) from the ease of being able to tag things for their own personal recall. Note that these two value propositions require xFolk to address two very different concerns. The first is that of aggregators who must collect together all of these tagged artifacts. The second is that of users who must wrestle with the microformat to tag URLs in their web articles.

This article focuses on enabling the first part of the value proposition to users; specifically, what is the underlying data schema that allows for the highest value aggregations? This schema provides an indication of the desired data elements to be included in any xhtml microformat that end-users will employ. The next article in this series, “xhtml microformat for Emergence” will consider how this schema can be translated into a format that end-users and, more likely, end-user tools can employ in a wide variety of use cases.

Continue reading "xFolk — schema for emergence"

Bud posted this on April 10, 2005 | Permalink | Comments (0) | TrackBack (1)

The business value of folksonomy and xFolk

The value of folksonomy is that it increases the information value of content for all concerned. The value of xhtml microformats like xFolk is that they lower the costs and barriers to transmit the content better than competing approaches.

In my spare time for the past several weeks, I've been working on xFolk, an xhtml microformat to allow people to tag the links they put in their blog posts and share those tags, often referred to as folksonomy. The tags consist of one or more one word descriptors that tell what the link is about. The altruistic value of tagging links is that the tag can help people better determine what you intend for them to draw from the link, and the tag can tell them where to look for more information. Fine, but what's the business proposition?

Continue reading "The business value of folksonomy and xFolk"

Bud posted this on April 5, 2005 | Permalink | Comments (0) | TrackBack (0)

Why xFolk? Folksonomy enhances search

Folksonomy is not a threat to search. Rather, it enhances search by making the meaning of links more explicit. Yahoo seems to have realized this in their purchase of flickr.

Steve Rubel makes an excellent point regarding the power of folksonomy in response to a Business Week article on tagging

The BW article also underscores the rising importance of tags for marketers. Often I am asked why I think tags are a big deal. The reason is simple - it makes consumer-generated content a lot more discoverable.
Micro Persuasion: When It Comes to Tagging, You're It

Continue reading "Why xFolk? Folksonomy enhances search"

Bud posted this on April 3, 2005 | Permalink | Comments (1) | TrackBack (1)

Newsgator: Taxonomy and Folksonomy

Newsgator is experimenting with folksonomy in their feed archive. If done right, consumers will get to see the archive on their own terms, and Newsgator will get to see what those terms are.

I spoke with Mike Lombardo of Newsgator today. He wanted some hints on how Newsgator might be able to fit into the educational market. The interesting angle in all of this is the content archive Newsgator has amassed since they created Newsgator online.

Continue reading "Newsgator: Taxonomy and Folksonomy"

Bud posted this on March 30, 2005 | Permalink | Comments (8) | TrackBack (2)

Folksonomy — Practical Application and xFolk

Folksonomy is one means of self-expression in a group, a sort of: “let me share with you my vision of the world by tagging parts of it”. xFolk makes this type of expression easier in web pages by adapting and generalizing semantic markup for a number of use cases.

In folksonomy, people tag digital artifacts (web articles, media files, etc.) and share the tagged artifacts with others. Want to see the latest web articles on folksonomy, look under the technorati or del.icio.us aggregations for the tag folksonomy. Want to find the latest pictures of Boston, look under the flickr aggregation for the tag Boston.

Either of the experiments I just suggested will show both the strengths and weaknesses of organizing archives based on common tag. You will discover things you consider personally relevant mixed in with a lot of dreck. The debate on the quality of archives organized by tag has reached the point where David Weinberger finds it possible to debate himself at length on the topic without reaching a solid conclusion.

Rather than just focusing on folksonomy for navigation, it may be more useful to adopt the perspective that folksonomy is one means of self-expression in a group, a sort of: “let me share with you my vision of the world by tagging parts of it”. Further, considering folksonomy as individual expression opens the mind to many useful aggregation and analysis possibilities. For instance, what does the aggregate of tags applied to given artifacts tell us about how people perceive them? How can individual tagging patterns across artifacts help us figure out a translation between personal tag vocabularies to better communicate with one another?

When I first proposed xFolk a week ago, I mainly saw its utility in helping people share this type of individual expression. To do so, xFolk has to meet a number of technical and social desiderata. Feedback over the past few days has helped me narrow down issues with the first proposal, the state of other efforts to achieve goals in line with xFolk, and various ways of better expressing the microformat for semantic “correctness” and generalization.

In the rest of this post, I want to lay out a revised view of the issues xFolk should address and a strategy for doing so. In a later post to follow shortly, I will return to the nitty gritty of xFolk's markup.

Continue reading "Folksonomy — Practical Application and xFolk"

Bud posted this on March 29, 2005 | Permalink | Comments (0) | TrackBack (0)

xFolk: Iteration in Response to Comments

Readers raised several useful issues with the first iteration of the microformat. This new iteration addresses those issues. Further comment is appreciated.

This post is somewhat of an insider's post. I am assuming you have read and are following with interest the xFolk topic I started here.

After some feedback from Thomas Vander Wal and Tantek Çelik, it appears that xFolk is going to require some iterations. There were eight issues with the initial proposal:

My creative use of the rel attribute might be hard to get accepted (several issues here)
How would uses of attribute values in xFolk mix in with other microformats. In particular, what would I do if my uses of the an attribute collided with other uses?
It would be nice to constrain attribute value terms before they are used. People expect to know possible attribute values before they are used. Listing them in the XMDP after they are used breaks standard web processing models.
In xhtml (and to some extent generally in xml), attribute values tend not to contain user data. Tags are best considered user data.
Regardless of whether one agrees with the prior point, people prefer that user data be contained in elements for the practical reason that it is more visible that way and therefore less likely to be a spam attempt.
It would be nice if a more direct tie could be made between this microformat and people's current practices with link blogs and exporting their del.icio.us entries.
How would non-hypertechnical users deal with this format?
What about privacy? You're publishing everything on the web.

Believe it or not, these all seem surmountable, but they will require redefining important aspects of the microformat.

Continue reading "xFolk: Iteration in Response to Comments"

Bud posted this on March 23, 2005 | Permalink | Comments (0) | TrackBack (0)

xFolk: An xhtml microformat for folksonomy

xFolk is an open xhtml microformat that allows users to publish their own folksonomy classifications for aggregation by the services they choose. As such, it starts to give back users control of their own data. A side benefit of xFolk is that users may designate explicit semantics for their folksonomy tags, providing a link to the more formal practices of information architects.

Folksonomy is an emerging practice on the Internet where people tag digital artifacts (e.g., pictures, bookmarks) with their own labels and then share them. Of course, people in the real world have been tagging and sharing collections for years; consider record and coin collections with tags like classical, rock, jazz, Merovingian, etc. One factor that makes tagging on the Internet different is that items and their tags can be shared with the world at large very easily, allowing for an emergent understanding of how common, tagged items are viewed by a large number of people. Further, folksonomy tagging with web-based tools (e.g., flickr and del.icio.us) is also very easy.

As a result of ease of tagging and sharing, folksonomy is spreading on the web like wild fire. Fickr, the folksonomy-based photo sharing service, has over 300,000 users, and Robert Scoble has signaled folksonomy tagging to Microsoft's senior management as an area requiring strategic investment.

Currently, all folksonomy data formats are proprietary and idiosyncratic to the service providing them, likely as a result of folksonomy's ad hoc development to date. There is no easy way to transfer one's data from one service to another. Further, there is currently no format for sharing or even expressing one's explicit understanding of the meaning of his/her own folksonomy tags. Tag and object tagged mutually and implicitly define each other without an explicit anchor point.

In this post I will motivate and sketch xFolk, a microformat for specifying and publishing folksonomies using components already part of xhtml. Since xhtml is already a well-established web standard, a compelling motivation for creating the xFolk microformat is that it will enable users to independently maintain their own folksonomy data while still being able to easily share it. A byproduct is that xFolk will allow those who wish to define explicit semantics for their folksonomy tags to do so, and these semantics will also be easily shared.

xFolk is inspired by Eric Meyer's and Tantek Çelik's panels at SXSW 2005 on emergent semantics and extending xhtml. My thoughts are at a very early stage and might even be termed a pre-proposal, but I am publishing them here because I think they are sufficiently developed to profit from community interaction.

Continue reading "xFolk: An xhtml microformat for folksonomy"

Bud posted this on March 22, 2005 | Permalink | Comments (3) | TrackBack (2)

Emergent Semantics and Folksonomy

The idea of emergent semantics is to provide a light weight mechanism for defining a standard usage in markup. The mechanism is available to anyone with a web page. It might be a nice incremental mechanism for folksonomists to use as they themselves attempt to impose some order on their own emergent tagging behavior.

Folksonomy refers to how people organize their personally collected information into categories. Think of how you file documents in directories based on the project they are for. People are now applying the term to how they organize bookmarks in online bookmarking systems or photo sharing systems.

A lot of the power of folksonomy comes from being able to look across how a number of people are tagging an item and finding items tagged by other people with the same tag you are using. However, a major issue with folksonomy is that the tags you use are relatively undefined. You yourself may even forget what they mean and there is frequently tag drift as tags are used slightly differently by different people. Today, Eric Meyer gave an interesting talk on the idea of emergent semantics and suggested a mechanism that may help a bit with tag drift.

Continue reading "Emergent Semantics and Folksonomy"

Bud posted this on March 13, 2005 | Permalink | Comments (0) | TrackBack (0)

Steve Rubel is spreading his meme in del.icio.us

Steve Rubel has done a brilliant job of making sharing itself into the self-interested activity people seem to need to contribute to folksonomy. Everyone who doubted you could use del.icio.us to spread your meme should look at what Steve is doing.

Recall this post about using del.icio.us to spread your meme that Steve Rubel was so kind to just pick up and publicize about a month ago. Well, Steve has become a practitioner. He is asking people to indicate their desire to have him link to their blogs by posting the link under the del.icio.us tag “micropersuasion”.

Continue reading "Steve Rubel is spreading his meme in del.icio.us"

Bud posted this on March 7, 2005 | Permalink | Comments (0) | TrackBack (1)

Maya and the Information Commons

Maya is aiming to provide an information commons for community construction.

The Community Engine

Creating Effective Information Communities

folksonomy

Getting people to tag the Internet and then figuring out how to use that information