Yes, I do realize that switching these all to SIMPLE will render point in time recovery as impossible. I made it a point to mention that to the proponents SEVERAL times, and they didn't flinch at all.

Ok, so to recap a bit then, shrinking only the log FILE will not cause fragmentation, which is as I originally thought. My first task then would be to switch all of these databases (nearly a hundred) to SIMPLE to stop the unchecked growth. Then I need to shrink the log files themselves and set the starting size according to how big they might become (if I can ascertain that somehow), setting the autogrowth to a value that minimizes the level of VLF growth to the extent that is possible. Is that about right?

I'm kind of torn here as to whether or not to bother other than to switch to SIMPLE. I would guess at this point that leaving the log files as is after being switched to SIMPLE recovery model would not be harmful as an alternative to screwing with it too much, especially since it will be migrated after a while. I won't be doing anything to SharePoint in terms of indexing, neither creating, altering nor deleting them and leaving the re-indexing to SharePoint itself. I am at least doing weekly integrity checks now and it is being monitored by SQL Monitor so if an index does get a bit squirrelly, I can deal with that at that time.

Thanks folks, much obliged.