2012/12/26

Content: What's wrong with the Wikipedia Knowledge Model

The Register had a recent piece on Wikipedia: they're rolling in cash and the lucky few are enjoying that gravy train. Talking about this led to a discussion on the strengths and failings of the Wikipedia Content Model vs Classical Encyclopaedias, like Britannica.

Wikipedia has a simplistic Model of Knowledge: There is But One Truth.

If you are an engineer or a programmer, then this is as axiomatic as binary data and logic: true/false, 0/1, just one correct answer to every equation or calculation/test, there is but one truth.

Which is exactly where Wikipedia excels: if you need to look up a technical topic, especially in computing/communications and some engineering, where there is a well defined specification, then you find very good articles, often concise, complete and well explained.

But this is at best an undergraduate view of engineering: incomplete, naive, simplistic and completely without nuance. It's deeply wrong and misleading.

What does Encyclopaedia Britannica provide?
Experts. With a comprehensive knowledge of the field, its history, the current "State of the Art and Practice" and the currently accepted theories with competing ideas and approaches  to problems at 'the edge'. They understand nuance and the many theories nature of the development of science. But more importantly, they bring judgement and an ability to explain their field to others.
Even in objective technical fields, like computing and electronic engineering, there are many competing theories about the field, each with their benefits and limitations, all argued for strongly by proponents and all, even those appearing 'weak' or later proven wrong or limited, offer insight and understanding.

Knowing the whole story, what works, whats been tried, whats been discarded - and why, is as important as knowing the current best theory. And that's for objective fields of knowledge: 'truth' isn't fixed and immutable, but constantly and unpredictably evolving and changing.

The proof of this is the 1902 statement by the newly appointed Commissioner of Patents in the US:
In my opinion, all previous advances in the various lines of invention will appear totally insignificant when compared with those which the present century will witness. I almost wish that I might live my life over again to see the wonders which are at the threshold.
He was proven right, beyond his wildest dreams... The nature of Science is change, "truth" evolves over time.

For more subjective fields, the 'soft sciences', even history and philosophy, the many viewpoints approach of science is vitally important: the discourse is more important than "the" answer.
Or fields requiring judgement, in answering questions about "best", ethics or belief and emotions: the stuff of real life.

Within this Many Theories and Viewpoints model of knowledge, articles with single identified authors are important. Readers may wish to read more by the author or disagree with their viewpoint and avoid their writing.

Contributions from Domain Experts in Encyclopaedias offer the same promise as Undergraduate courses for professionals:
Complete coverage of the field: an A-Z guide of the current state of the art, influenced by the experts viewpoint and approach.
This is the weakness of the "autodidact", or the self-taught: they can easily miss large and important areas of a field. They don't know what they don't know, whilst a "good" undergraduate course carries the promise of knowing what you know and what you don't know: formal education is about removing the "Unknown Unknowns". The self-taught, without feedback from others, can never be certain they've learnt the whole field. [Because they aren't hampered by preconceptions, they can sometimes arrive at startling new answers, but mostly they fall into known fallacies and logic traps.]

To evolve, grow and thrive, Wikipedia needs to embrace single identified authorship, multiple competing views on the same subject and good histories of the subject, reflecting the cut-and-thrust of all lively, developing areas of academic debate.

For Wikipedia to be more than the naive "font of all undergraduate wisdom", it has to embrace the complex, many faceted view of knowledge and discourse in world of graduates and researchers.

No comments: