Activities, Workflows, and Structured Wikis

Gregorio Convertino, who recently joined ASC research area here at PARC, have been looking at how Web2.0 tools like Wikis support workflows within the enterprise. By workflow, we mean activities that are important enough to be documented in the enterprise (either because it is an important client, or that it is an activity that is often repeated.)

For this purpose, we have been doing an overall review of structured Wikis available in the marketplace (either thru open-source, hosted solution, or supported-installation). By "Structured Wiki", we mean wiki engines that are enhanced with lightweight programming features and database functionalities. The focus of our review is primarily, but on only, on the user interface and interesting new functionalities to organize content such as templating and database functions. Important criteria for us are ease of use, power of end-user-programming/organizing functionalities, and licensing.

Over the past few weeks, we have been looking for useful resources on Content Management Systems and Wikis that support structured activity management. On the Wiki side, we have found www.wikimatrix.org to be one of the best guides in understanding this space.

Exemplars of wikis enabling some structure are, for example, TWiki, Xwiki, TikiWiki CMS-Groupware, JSPwiki, MediaWiki, Openrecord (not on Wikimatrix's list). For a comparison matrix, click here. In this example comparison, all of the systems support Page Templates. It's clear that many people are looking for these kinds of functionalities, and we have found some discussion around this on the net. Twiki developers seems to have documented some of their thinking.

But the depth of the discussion so far isn't very deep, because we don't really seem to know yet how much structure is too much structure, and how different enterprise needs are met by each of these solutions. In our work, we are finding it quite difficult to figure out what an Enterprise should implement:

(1) There are so many different flavors of Wikis out there, and they don't always inter-operate well. Choosing one appears to mean that you're stuck with it forever.

(2) Research on end-user templates has not been focused. We have found references in academic literature, but they are pretty sparse. Here is what we have collected so far:

There was the work of Sparrow at PARC.

Di Iorio A. Vitali S., Zacchironiet S. Wiki Content Templating (WWW 2008).

Anslow C. and Riehle D. 2008. Towards end-user programming with wikis. Proceedings of the 4th international workshop on End-user software engineering

Riehle D. 2008. End-User Programming with Application Wikis. In Proceedings of the 2008 International Symposium on Wikis (WikiSym ‘08).

Haake et al. Wiki-Templates. WikiSims 2005

Reinhold. WikiTrails: building context and structure around the content and existing information organization, using trails, or paths, through the Wiki content. SigWeb 2006.

Jochen Rode. 2005. Web Application Development by Nonprogrammers: User-Centered Design of an End-User Web Development Tool. Ph.D. dissertation. Virginia Tech. Click system.

Also, there was of course the concentrated effort on activity-centric computing at IBM, as well as the work done on Co-Scripter (workflows that can be collaboratively built using a wiki-model).

Anyone who can help us understand this area, please get in touch!