One thing of value Shirkey's essay led to was this piece by Richard Gabriel entitled "Worse is Better". He makes the important point that simplicity is a more important value than correctness/consistency/completeness: though all those are important, simplicity has more survival value. Though this comes from long ago (not much discussion about Lisp, Scheme, and CLOS any more, alas), the basic point still holds true.

Shirkey has written a strong piece against the Semantic Web. Tim Bray has what seems to me like a more reasonable response, that fingers the key issue (pay attention here):  

• the real value comes from getting metadata
• getting metadata is not cheap

Ergo (in the syllogist spirit) getting the real value will take a lot of work. The Web Semanticians i know (i work with some) tend to under-estimate how hard it is to get the meta-data, and that's certainly one reason it hasn't caught on faster.

This is all related to the SemANT project, which is simply an elaborate plan for metadata on Scripture. Doing this kind of meta-data manually takes a lot of work: but i can't see any other way.