Many companies, unfortunately, go out of their way to not produce meaningful science in the sense of publicly available (trade secrets) or publicly usable (patents) knowledge. Biotech especially.
My point was only that (by the definition I think makes the most sense here) they are doing science to the exact extent they give up the intellectual-property monopoly, whatever that extent happens to be.
That does not mean that dissemination of new information has to work like it does now. The Republic of Letters was very different but quite successful. But if you have taken no effort to make your discovery available to others, I’d say you’re not advancing the cause of human learning; if you have taken effort to ensure it is not available, like what many patents aim for nowadays, I’d say you’re working against it.
On the other hand, those who at the same time were experts in, for instance, making shoes, made efforts to keep secret their techniques.