About Vance | CALL resources main page & Site Index | Vance's papers and presentations
Back to the home page of this presentation | Forward to RSS and PULL technologies

Mastering competencies for collaboration and aggregation in distributed learning networks

by Vance Stevens
Department of Computing, The Petroleum Institute, Abu Dhabi, UAE

Aggregation of knowledge in distributed learning networks

This section explains the two terms in the title, aggregation and distributed learning networks, and how the network tends to mediate and stabilize construction of the knowledge-base as knowledge flows throughout the network.

Distributed Learning Networks and Connectionism

A distributed learning network contains information at its various repositories, or nodes, each populated by network users/members. There is a difference between knowledge and information. Knowledge derives from the effective and focussed processing of information. Aggregation refers to pulling selected and filtered content from a network into one place and accumulating it where it can efficiently (without being drowned out by noise from irrelevant matter on the network) be acted on in that centralized location and processed into knowledge useful to the many different users of a network.

The devices used by information processing mechanisms on the network as we see it emerging today act mainly on tags. Tags are metadata that users give to information to try to characterize its content. Aggregation either works on tags or on subscriptions to RSS feeds. Either way, users query the sum total of content on the network and pull information to them filtered through metadata such as the tags assigned to content they are examining or RSS feeds working off scripts which describe the data so that it can be retrieved selectively.

Stephen Downes has made many talks describing characteristics of distributed learning networks (e.g. 2007). George Siemens characterizes relations within the network as connectivism. In one salient remark, he states that "The pipe is more important than the content within the pipe." (Siemens, 2005); that is the functioning of the network has more bearing on knowledge distribution within the network than the strength of that knowledge itself (.

Aggregation and folksonomies

To some the proliferation of information and content over the Internet appears to be a massive tangle which presents daunting challenges to pick through. Google searches for example, can yield hundreds of millions of hits. However there are emerging systems whereby creators of that data as well the millions of people who access it have opportunities to describe it in simple terms known collectively as folksonomies. Tools such as aggregators (e.g. Bloglines, Google Reader) are available to query the web based on these folksonomies; e.g. tags and other user-created data stored at numerous social networking sites, and return results for users following in the footsteps of the original taggers, who are themselves leaving tags and other signposts for those yet to come, so that the Web becomes navigable thanks to a means by which everyone using it can leave directions in such a way that later users will benefit from the meta-knowledge (as described by Weinberger in hs Future of Education presentation, 2007). Thus the pipe itself becomes of prime importance to the knowledge passing within.

These tools have potential for language teachers in helping their students to develop voices which can be added to the welter of information already on the Internet, but which, if tagged systematically, can be heard by others tuning in to shared frequencies. Later in this presentation we will examine a method through which contact with peers and other interested parties can be effected by taking advantage of such characteristics inherent in the emerging Web 2.0.

Nowadays the pipes and mechanisms for distributing information and knowledge within a network, and with it the ability to cross-fertilize adjacent networks, are being developed and expanded at an unprecedented rate. This has had the effect of subverting the power structures that existed when communication was essentially done in print media, which could be controlled by publishers and other authorities with agendas possibly different from those seeking publication (which sometimes results in those wishing to be published altering their views to align more closely with those of the arbiters of what gets published).

According to Dieu and Stevens (2007)

Contrary to traditional media like radio and television, which distribute their messages one-way from centralized static locations (from a sender to a receiver), social media are two-way, distributed, and part of a shifting internet-wide social network (peer to peer). This enables individuals to communicate their own viewpoints and negotiate meaning with many others, creating their own content and constructing their own "spaces" to network according to their own choices, not controlled by media owners.

Now users are gradually taking control of media via a plethora of Web 2.0 sites such as blogs, YouTube, Flickr, etc, as described in Time's most recent person of the year cover story (Grossman, 2006). Normal people now have easy and free access to media through which to put forward their voices, and many are taking advantage of this opportunity and circumventing traditional print publishers, with little compromise to quality control, for example as in the case of Wikipedia, where quality is maintained though a finely balanced system of near instant response to vandalism and bias by users. This has given rise to the accelerating emergence of the long tail, that segment of niche enthusiasts who previously had little or no outlet for their output and convictions (Anderson, 2004).


The Long Tail blog: http://longtail.typepad.com/the_long_tail/

The Question of Quality in Web 2.0

In anticipation of a down-side, questions have been raised about quality, especially by those with least experience with the system. When you buy a book you can be reasonably assured of quality-control. It's been through a quality control process. True, it's also been through a filtering process, a process that can border on censorship, or that will introduce some other form of bias. But granted, it will be hard to avoid bias in anything you read (though it's somewhat galling to have to pay for bias).

One classic debate on this issue is whether tis better to consult a published encyclopedia or Wikipedia.. Cost and convenience are two important factors (free and at your fingertips for anyone with a computer connected to the Internet). Wikipedia for obvious reasons carries more current articles - the new Pope was named in Wikipedia the moment the smoke left the chimney (according to Stephen Downes, who himself had tried to be first), and you can look up almost anything mentioned here and find it in Wikipedia. So currency is better with Wikipedia, and also scope (there are something like 750,000 articles in Wikipedia). Also the words in those articles are machine-searchable in Wikipedia, not so in any printed work, where we must rely on a relatively crude index. So on four counts (cost, currency, retrieval, and scope) Wikipedia seems preferable, and there remains only the integrity and bias of that information. You now have to pay $30 to read the article in Nature Magazine comparing accuracy in Wikipedia and Encyclopedia Britanica, which found them not far apart on that score, although considering that some colleges have banned their students from citing Wikipedia articles in reports, there would appear to be some concern (or are they just trying to force the students to use something besides Wikipedia as a definitive reference on every question? ).

Regarding quality, much depends on which nano-second you look. One valid concern with using wikibooks in classrooms is that teachers would need a locked down copy of what they saw one day and intended to show to students the next (or risk big surprises in class). There is no way to lock-down Wikipedia. Articles can change one moment to the next. But in general changes tend to be well mediated. In practice the wisdom of crowds prevails (look it up: http://en.wikipedia.org/wiki/The_Wisdom_of_Crowds. What we see with Wikipedia is that strong biases tend to be expunged, and people are as a result careful with what they write in order to avoid attracting modification. Even cases of vandalism or spam tend to be dealt with quickly, either by humans or bots that continually troll Wikipedia.

For an excellent example of how wikis undergo change over time, and how a distributed network operates to mediate both vandalism and radicalism, check out the fascinating screencast of the evolution of the article in Wikipedia on 'Heavy metal umlaut": http://weblog.infoworld.com/udell/gems/umlaut.html .

References

Anderson, Chris. (2004). The Long Tail. Wired Magazine, Issue 12.10 (October 2004). Retrieved August 20, 2007 from: http://www.wired.com/wired/archive/12.10/tail.html.

Dieu, Barbara, and Vance Stevens. (2007), Pedagogical affordances of syndication, aggregation, and mash-up of content on the Web. TESL-EJ, Volume 11, Number 1. Retrieved August 20, 2007 from: http://tesl-ej.org/ej41/int.html

Downes, Stephen. (2007). Personal Learning the Web 2.0 Way. Keynote presentation at WiAOC 2007, http://wiaoc.org. Slides: http://www.slideshare.net/Downes/personal-learning-the-web-20-way Audio Part 1: http://streamarchives.net/node/84 Audio Part 2: http://streamarchives.net/node/83.

Grossman, Lev. (2006). Time's Person of the Year: You. Time Magazine, Wednesday, Dec. 13, 2006. http://tinyurl.com/3chjkb.

Siemens, George. (2005a). Connectivism: A Learning Theory for the Digital Age George . International Journal of Instructional Technology and Distance Learning 2, (1). Retrieved August 20, 2007 from http://www.itdl.org/Journal/Jan_05/article01.htm.

Siemens, George. (2005b). Connectivism: Learning as Network-Creation. Retrieved October 4, 2006 from: http://www.learningcircuits.org/2005/nov2005/seimens.htm.

Weinberger, David. (2007). Knowledge beyond Authority. Presentation given at the Future of Education online conference. Retrieved June 30, 2007 from:
http://ltc.umanitoba.ca/foe-2007/David_Weinberger/

David Weinberger and Andrew Keen debate Weinberger's thesis in his most recent book, All Things Miscellaneous, posted by Kevin Werbach July 9, 2007 in Conversations Hub: Supernova 2007. Retrieved August 20, 2007 from: http://conversationhub.com/2007/07/09/video-david-weinberger-and-andrew-keen/. For the direct video link http://conversationhub.com/podpress_trac/web/259/0/sn-weinberger-keen.mp4

Back to the home page of this presentation | Forward to RSS and PULL technologies


For comments, suggestions, or further information on this page, contact Vance Stevens, page webmaster.

Last updated: September 3, 2007

Copyright 2007 by Vance Stevens
under Creative Commons License: http://creativecommons.org/licenses/by-nc-sa/2.5/