It seems to me that the place where the semantic web will have its first success is in areas that already have a tightly controlled vocabulary. For example in biology and chemistry. Are the major scientific organizations such as the American Chemical Society making contributions to the semantic web? Years ago I worked on something we called the Life History Data Bank - for organisms in Texas coastal ecosystems. It took a big effort to keep the vocabulary under control, even with the existance of well worked out systematics.
I haven't heard of "major scientific organizations" doing so, but there could be plenty of efforts I just haven't happened upon yet. There are some efforts to create RDF/OWL ontologies in scientific and engineering fields, including medical terms, but they seem to be worked on at a lower level than "major organizations". Now that SNOMED is said to be openly available, it seems likely (to me, anyway) that it will be translated into OWL.
Keeping the vocabulary "under control" is of course always going to be an issue, whether Semantic Web technologies are used or not. There has been some discussion of when changes are large enough to warrent changing the URI that identifies a term. In OWL, you can say things like "This class is equivalent to that class", and "this property is deprecated". These properties are standard, and of course you can invent your own terms as well to express shades of change and deprecation.
Author of <a href="http://www.amazon.com/exec/obidos/ASIN/1932394206/ref=jranch-20" target="_blank" rel="nofollow">Explorer's Guide to the Semantic Web</a>