On one side of the discussion where issues about what is lost if you bring the social software inside of the firewall. Bob Morgan of University of Washington pointed out that you need critical mass for much of the magic of social software to work. You need many people participating with overlapping interests to form virtual communities. You multiple people to tag the same object to build a meaningful folksonomy. Would you reach critical mass if the population was limited to the enterprise or do you need to be sharing in the wide world for the magic to occur?

An example that I used was Chemical Engineering students using social software out in the world might discover other students at another university. The two groups might work together and learn more both about their projects that the world at large. These forces tend say that we should just use the world-at-large suite of tools.

On the other side is the fact that these systems are out of the control of the university but may be considered by some (especially parents of students) to be under the control of the university. I mentioned a case of an instructor who is using a Wiki on their personal web site for instructional purposes. Students need to go to this site. They may need to participate. This leads to institutional business being conducted at a site that is completely out of the control and governance of the institution. It also leads to institutional information – course content and possibly student performance data (e.g. test data, quizzes and homework data) – being de facto outsourced to an untrusted partner. Who knows what the ISP will do with this information?

The hot discussions around MySpace, Facebook et al show the depth of concern about student privacy and safety and the concerns about the belief that these third party sites are somehow blessed by the University.

Adding to the complexity of these issues is the fact that it is so easy to set-up and use a lot of the Web2.0 / Social Software tools that are out there. You can get a blog set up and running in a couple of minutes. My personal blog is run on Dreamhost. They have one click installs for a variety of software. It would be easy for me to set up blogs, wikis, bulletin boards and galleries for a class. It would take about 15 minutes – tops.

Going back to the earlier example of the faculty member and their wiki – they used their personal web site because their institution doesn’t have any support or offerings for doing wikis in house. Any barriers that the institution erects (including charges) will lead to outsourced solutions. In house solutions may not work as well as the wide world solutions but the institution is mandated to protect student privacy and security. The institution is also legally bound to keep records of performance and it has an interest in keeping an archive of instructional materials.

Is there a balance where students participate in the wide world of social software in a safe and educational way?

The links that I used for the demo are below.

1. UW-Madison’s SOA Migration Strategy – what is it and how do we get one
2. Folksonomy and Web 2.0
3. IT Architecture – What is it and why 3 isn’t enough
4. Identity Management Nouns and Verbs

Note that the Folksonomy slides are from an Internet2 version of the talk and are more inclusive than the slides I used at U-Minn. Actually, I meant to grab these slides not the ones that I used. There are a list of links of the URLs that I used in the Folksonomy demo here:

Links I Use in My Folksonomy Demo

There are three sections:

* Tagging and Multiple Tag browsing
* Folksonomy and Social Discovery
* Cool Apps, REST and RSS

Part 1 – Tagging and Multiple Tags

* http://del.icio.us/
* http://del.icio.us/jimphelps
* http://del.icio.us/jimphelps/folksonomy
* http://del.icio.us/jimphelps/folksonomy%2BConnotea
* http://www.technorati.com/
* http://www.technorati.com/tag/itarchitecture
* http://www.technorati.com/watchlist/add/
* http://www.flickr.com/photos/tags/seattle/
* http://www.youtube.com/

Part 2 – Folksonomy and Social Discovery

* http://del.icio.us/url/fd3ba631e0adbe85e25f3cd519882341
* http://weblog.infoworld.com/udell/2005/06/23.html
* http://blog.pietrosperoni.it/2005/05/28/tagclouds-and-cultural-changes/
* http://cloudalicio.us/tagcloud.php?button=Graph+Tag+Cloud=http://del.icio.us/url/19456e761687204e2bc7857cbc52fe4b

Part 3 – Cool Apps, REST and RSS

* http://www.connotea.org/
* http://www.citeulike.org/
* http://www.infoworld.com/article/06/01/31/74927_HNredhatcommits_1.html
* http://del.icio.us/infoworld
* http://weblog.infoworld.com/udell/iwx/explorer.html#top
* http://www.technorati.com/tag/wag_edu
* http://arch.doit.wisc.edu/jim/
* http://arch.doit.wisc.edu/jim/index.php/tagcloud/
* http://arch.doit.wisc.edu/jim/index.php/category/socialsoftware/

Also, I want to be able to bind email to a calendar event. I have an email request to present at a given time and place. I want to be able to attach that email to the event.

Finally, I should be able to attach, flag, tag events which I do not control but have been invited too. I should be able to add my own notes for myself, my own details, etc. even though I do not control the invite.

* The Web 2.0 movement is about empowering people to publish their own content quickly and easily with a minimum of knowledge. Flickr, Youtube, del.icio.us, Blogger et al allow people to publish content with little knowledge of apache servers, domain name registrations and the rest of the system admin knowledge usually needed to run a web site.

* Social software (which Clay Shirky put as, “the stuff that gets spammed”) allows people share their content with others, to learn who is interested in their content and to learn who has interesting content. It is the “who” part that is of interest. Social software enables the discovery of people of interest similar to how Google enables the discovery of content of interest.

* Folksonomies enable the loose classification of content. This classification is done by people and is not a rigid hierarchical taxonomy but a loose metadata based classification scheme. This classification scheme is great for informal content such as blog entries or URLs and content that may be interpreted differently by different social groups.

Knitting these three things together, we get a system which enables the easy publishing of content, discovery of the people behind the content and of others interested in similar content and a simple way to classify information content. Employing these systems inside of an enterprise should allow for much better communication and collaboration with the added benefit of social discovery. Jon Udell uses scripts to analyze del.icio.us data to discover people with similar interests and content that they are reading that he has missed (Jon Udell: Collaborative filtering with del.icio.us).

In the early days of System Operations and enterprise computing, we were working on Data Management. The big question was, “how do we store all of this data and then recall it when we need it?” Huge tape libraries with colored number codes were built and technicians fetched and threaded tapes. We then advanced to Information Management where the issue changed from data management, which is raw and uninterpreted, into, “How do we find the actual information inside all of the this data?” It was no longer good enough to find the file by file name, we wanted to find the file that had the information we needed. This was followed by the Knowledge Management movement. The enterprise wanted to find the patterns that were present in the information to glean a higher level understanding of the enterprise.

Enterprises that take advantage of the Web 2.0 / Social Software / Folksonomy movement will move on to next level: Social Mining. This will enable users to mine the informal content within the enterprise and find others with similar interests and the content that those others find interesting. As one person I work with put it, “it is virtual mentoring”.

Thoughts?

- JJP

From Kevin’s blog,

>Today I have donated some code and examples to this site. While none of it should be considered production level code, I believe it’s a good starting ground for a number of possible projects that can be worked on here.
>
a) Xythos and Eclipse — setup guide for running Xythos and the default shipping Tomcat within Eclipse (great for debugging).
>
b) Xythos and Konfabulator — A JavaScript library that can access Xythos (Upload, directory listing, move, copy, etc). I can only imaging the widgets that can be created, here hopefully is the first step, the layer to talk to the storage web service.
>
c) Xythos Pictures — A small Servlet that generates thumnails and picture albulms from uploaded pictures. The most interesting part to me is that picture name and description are stored as meta-data. If you don’t like the way the album looks, create your own. The meta-data is there for you to work with. It’s stored with the Object and NOT in the application!!!!

Kevin and I had a long conversation (and Xythos asked me to present) discussing the role of Folksonomies in Enterprise Information management.

I am very excited by the news that Xythos is starting up this open source resource for Xythos development. I have high hopes that this effort will really drive the vision and expand the options for using Xythos’s products in creative ways within the Enterprise. I hope that tagging and social “browsing and mining” (similar to Jon Udell’s work on Collaborative Filtering) will find its way into the product through open source efforts and product development. I know that Ed Miller (CEO and President of Xythos), Kevin and others at the company get the idea and see the power of tagging and social networking.

As Ed said during our meetings, “Xythos is infrastructure not an application… What we want to do with this site is demonstrate that Xythos is a powerful piece of infrastructure that can enable a variety of applications.”

Kudos to Xythos and I look forward to watching and participating in this effort.

- Jim

EDUCAUSE has just announced a tag for their upcoming conference. It looks like the will aggregate the various RSS feeds into a single page.

connect.educause.edu | Technology In Academia — Connect @ EDUCAUSE
Tagging the Annual Conference

If you’re blogging the conference or uploading photos from the show to a site that supports tagging (ala Flickr), please use EDUCAUSE_ANNUAL to classify your entries. We’ll be displaying aggregate information in the right-hand column at the url below:

http://connect.educause.edu/folksonomy/educause_annual

>TagCloud is an automated Folksonomy tool. Essentially, TagCloud searches any number of RSS feed you specify…

I keep running into the term “Automated Folksonomy tool”. To me, this is like “Fresh Frozen” or “Hand Made by a Robot”. The point of a “Folksonomy” is that there are people – real live human flesh and blood – adding terms which provide social value and content evaluation. Folksonomies aren’t about extracting metadata. Folksonomies are about people, who have similar social knowledge, judging content and adding metadata based on their personal assessment. The term “cool” or “gnarly” or “awesome hack” have social value that is not extractable by a bot crawling RSS feeds. If my friends in my social circle tag something as “cool” or “awesome” then I will want to look at that object. It is their judgement that I value and the tag is a way for them to share their judgement with others.

TagCloud may be “cool” or an “awesome hack” but please spare me on the automated folksonomy stuff. Automated Folksonomy is a pile of Fresh Frozen Hand Made by a Robot stuff.

From Clay Shirky’s Blog:

>This piece is based on two talks I gave in the spring of 2005 — one at the O’Reilly ETech conference in March, entitled “Ontology Is OverRated”, and one at the IMCExpo in April entitled “Folksonomies & Tags: The rise of user-developed classification.” The written version is a heavily edited concatenation of those two talks.

Clay Shirky has posted a new version of his Ontology is Overrated talk.

{JJP}