# Extra-academic pre-commercial research organizations seem less relevant than in the past If you want to do pre-commercial research, it’s almost impossible to avoid academic institutions. [[Modern academia is a nebulous institution characterized by some combination of labs with PIs being judged on papers and labor being done by grad students]]. This has not always been the case. In the middle of the 20th century, there were research-based companies like [[Bolt Beranek and Newman]] and large industrial research labs. Before that there were gentleman scientists and independent inventors. (This trajectory is well-tracked by [[aroraChangingStructureAmerican2020]]) A thing that I’m trying to wrap my head around is the fact that while to some extent those parallel systems still exist today, it feels *different*, like they aren’t as relevant or do things that spill out into the broader world somehow. People in [[BAE Systems]] and [[Draper]] still do research; [[Boston Dynamics]] was for at least a decade primarily a DARPA consultant shop; [[SRI International]] and [[Xerox PARC]] and heck even [[Bell Labs]] still exist; most large companies still do have R&D departments. So why does it feel so different now? Some (not necessarily mutually exclusive) possibilities: - These places are no longer attractive places to work for the best people. (This, of course, is coupled to the theory that there is a long tail of generic “talent” that produces the majority of innovation) [[When ‘talent’ isn’t code for ‘specialized training’ it means the role or industry has not been systematized]]. There are a number of reasons for *that* explanation. Some of them include: - Better options either in terms of prestige (academia) or pay (tech salaries). - People no longer have the same freedom over what they work on at those places. (Which is then downstream of more project-based funding with lower margins) - The sorts of things that those organizations find it profitable/important to work on are no longer the sort of things that the best people want to work on. (You then have to ask *why* on that front). - These places don’t have the flexibility or timescales to work on things that are majorly impactful. In order for this explanation to be correct, there would need to be some characteristic timescale and/or scope flexibility that was critical for broadly impactful work. Inflexibility and shorter timescales could be downstream of tighter restrictions from their funding sources (primarily the government and single large corporations) on how their funding is spent (which in turn is downstream from transparency and activist voters and investors); it could be a cultural shift that organizations need to justify what they’re working on and how it supports a funder’s goals more specifically. - There’s less of an overlap between the things these places are working on and broader impact. It could be that the areas that these organizations work on or the type of work have stayed roughly the same but impactful work now looks different or is in different disciplines. (But then you need to ask *why* there and why would academic organizations be more able to shift). - Broader frictions on getting innovations out into the world mean that these places (which do have more of a mandate than academia to focus on invention/innovation) have taken more of a relative hit. These frictions could come from everything from the cost of commercialization, consumer appetite for new technology, return expectations from investors, and more. - You’re just seeing declining marginal productivity of R&D across the board coupled with these organizations being less good at marketing/less in the limelight than academic organizations. - These places are simply not getting as much money compared to academic organizations as they were in the past and relative funding is directly tied to relative impact. - It’s literally just feels. This misperception could be downstream of some idealized picture of the past or the human brain’s inability to grapple with extremely high-variance processes that just happened to yield results very early in a time series. (I suspect this one is not the case because there is compelling empirical evidence that the structure of research has changed; it would be surprising if changing structure didn’t have an effect on which institutions were more or less relevant.) ### Related - [[Academia has developed a monopoly on pre- and non-commercial research]] - [[Extra-academic Lab Spaces]]