The problem with emerging technologies, concepts and methodologies is figuring out how much they can deliver versus the level of hype that generates expectations. It is nice to be able to postpone this kind of judgment call as long as possible - at least until there are sufficient documented results out there to provide a fair degree of certainty about any conclusions. However, this is not possible when something comes along that begins to generate very high expectations across a broad spectrum of the data world and beyond. There is an obligation to form some kind of judgment so people have a basic idea of what they may be getting into. Wikis have risen in the collective consciousness beyond the level where they can be safely ignored and are at the point where they need serious consideration. Suddenly they are almost the default answer for any form of knowledge management. You now encounter otherwise reasonable data professionals who are looking for any vehicle to turn into a wiki project. But just how real are these expectations, and how well founded is the enthusiasm? A wiki, as everyone is now probably well aware, is a collaborative Web site oriented to providing knowledge in some domain. Anyone can enter information, change or comment on anyone else’s contributions. There is a minimum amount of governance, usually only to prevent offensive input. The concept is wonderful, but all great content requires review and/or validation and organization. How do wikis perform in these areas? The Internet, particularly the blogosphere, has been successful in providing review. Dan Rather found this out when he hosted a “60 Minutes” piece that was critical of George W. Bush’s stint in the Air National Guard. Bloggers came out of the woodwork to provide some very negative feedback on the reliability of the evidence that had been presented. The resulting brouhaha led to Dan departing CBS a little earlier than anticipated and somewhat under a cloud. This clearly demonstrated power of collective wisdom to review facts is one thing for millions of people for whom politics is a passion, perhaps even a blood sport. But it is not so easily transferred to the inside of a corporation. There is no pool of potential reviewers, including genuine experts, simply waiting to catch wiki contributors’ mistakes. There may be a few experts, but are they trolling a growing set of wikis every day just to see what they can catch? How would they even know where to look? Surely if they had that much time on their hands, they would be downsizing potential. If facts in a wiki are not reviewed by subject matter experts (SMEs), it is even less likely that someone who is not an SME is going to take the trouble to validate them. Yet we all know from systems and data analysis how important validation is. There is a set of tricks for validating what we understand from a single user, such as asking the same question at different times, asking the inverse of a question, providing counterintuitive examples and so on. It is repetitive and it takes effort. Wikis simply avoid this - they are not set up for validation by analysts who are not SMEs. In regular data and systems analysis we have recognized the analyst function for decades. Why should we think that wikis can magically bypass it? There is another feature of wikis that is likely to make them dear to the hearts of data administration (DA) units. This is the fact that they rely on SMEs to update them, i.e., people who do not work in a DA unit. Uncovering hitherto unrecognized responsibilities and obligations in the rest of the enterprise is a core competency for some data administrators. Simply setting up a wiki and announcing that SMEs are expected to contribute to it, presumably during their long periods of spare time, and quite possibly for nothing in return, is unlikely to contribute to the betterment of the perception of DA in any organization. Lectures about what everyone else can do for the greater good of the enterprise from a unit that does not create services and infrastructure that demonstrably add value are likely to go unheeded. Perhaps the greatest problem with wikis is that they can be anarchic. They have a starting point that simply does not model the domain of knowledge they are trying to address. The idea that this will evolve over time is unrealistic, but currently seems to be an article of faith. If it were true, we would not need to do data modeling either; individual decisions about how to manage data would eventually default to an optimal architecture across the enterprise. Of course, we know that is not true, so it seems a little odd for any DA unit to embrace wikis where lack of prespecified structure to organize content is a fundamental principle. If the need for structure is recognized, then the role of wikis has to be questioned. The obvious alternative is to build a “proper” knowledge management system, with understood requirements, a correct design and relevant functionality targeted to the needs of a particular audience. Yet wiki aficionados should not despair. Every silver bullet has a silver lining. XML was going to make data exchange a no-brainer and eventually did become a useful technology, albeit falling far short of the hyped expectations. Perhaps wikis, by showing that the promise of knowledge management can be fulfilled, will hasten serious efforts to adopt it. There should be no doubt that all enterprises really do need more and better knowledge management, and kudos to wikis if they help bring it about.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access