(Is a patent that has expired still counted as a "patented discovery"? This changes the meaning quite a bit)
Why is it "bad" if patents are the reason things are held up, but "not bad" if it's a common trade secret or proprietary application? Didn't Google hold back the entire industry by not making their secret sauce for large scale management of IT infrastructure broadly available?
IMO, the patent problem is a scoping problem -- we allow patents that aren't appropriate. The concept of patents makes it possible to make large capital investments to actually invent things. We need that imo.
The Google infrastructure would be a good example. Amazon and Facebook have built similar facilities.
In the pharmaceutical case, generic manufacturers are often able to start producing a medicine on the day the patent expires. In the Google/trade secret case, the world may forever lose valuable tools and techniques that are kept proprietary once, but cease being useful to Google.
Whether it "slows progress" has got to be an empirical question with results that vary from field to field. The "with patents" case in software forces not only rediscovery but reinvention and engineering around patents (which exists in the pharma case as well).
Now, a software patent system that involved filing a copy of the source code with the patent? That would be a great way of preserving tools in the public record. But that's not how it works at the moment.
Isn't that how the patent system is supposed to work?