Has research in computer science at Colleges and Universities become less meaningful?

In Snake-Oil Research in Nature we have a case where Bruce Schneier (who knows something about cryptography) finds a bit of research done in an academic setting that is a little short on relevant expertise.

As an odd aside, note that a number of the articles referenced are in journals that require subscription. Originally the notion was that refereed journals were a concentrated resource. We expect both writers and reviewers to possess an unusual level of expertise in the subject. The resulting reviewed articles should capture the highest order of thought on a presented material. Clearly that is not what we have here.

On reflection - in the past communication within the academic community was generally stronger than communication outside. Researchers and engineers working within companies were seldom encouraged and often prohibited from sharing their results. This is all prior to the widespread availability of the Internet to the general public.

Now the effort needed to share results is near zero. Communication within the academic community is not significantly stronger than outside. In fact, results “shared” within any ghetto that requires a subscription are less available than material simply placed on the web. It would seem we have an inversion.

Does this explain why the “distinguished speakers” I have heard at UCI have been rather disappointing in the past several years (or perhaps just bad timing on my part)?

What does this mean for the role of research in an academic setting?