Alright, it's tomorrow now. Here's my half-baked explanation of what seems to be happening from my perspective.
The two groups I linked to above are both producing software for use in estimating forest behavior.
The first, FMRC, is a USFS office which is more focused on sale-level metrics, and their software is freely available to anybody who wants to use it. It's not truly "Open Source", in that the user is not allowed to modify or redistribute it (in the spirit of the
GNU or
BSD licenses), but that's not really a problem, as the underlying lookup tables are themselves Public Domain, and if one were so inclined, it would be only a matter of time, talent, and money to create a similar program suite from the same background data. That said, the FMRC guys are absolute superstars when it comes to customer support. They're a small group and they work closely together to make sure that their programs work together. There are regional tables for most commercial species grouped together by FS region within the US which, in general, breaks out to separate equations for each species by 16- or 32-foot logs. There are some holes, though; as it happens, nobody has done ANY work on hardwood growth models in R6 AT ALL, which means I have to manually deduct an approximation of the difference between a generic conifer and whatever hardwood I'm cruising. I don't want to go into greater detail on that right now but suffice it to say that I hope somebody does the math for me someday so I don't have to.
The second, FBRI, is a non-profit dedicated to stand-level inventory extrapolated to the landscape level. Their software is a astoundingly powerful suite of macros and lookup tables which push and pull data back and forth between GIS and a relational database. The major licensing difference is that all of their tables are proprietary. Further, through use, they become calibrated to the local ownership being inventoried. They are also a small group of technical wizards who support their product well. The software and what it can do are both pretty intense and I'm still learning but I'm pretty sure this approach will be the way things are done from now on. A major innovation is a feature that generates a local Site Index at a 10-meter resolution based on a whole bunch of mysterious and spooky math.
So: mensuration techniques, software, and data analysis protocols are becoming standardized, which approximates the idea of "Open Source" at the industry level. This is good. It means that over time, timber sale volume estimates will get better and better. We're already coming in at an average error of <50Mbf per million, cruised versus scaled.
However -- local knowledge will always trump sweeping protocol. Nothing beats boots on the ground. The industry needs, now more than ever, to prioritize creating and nurturing expertise. Shutting down and merging offices to save "costs" only serves to squander what expertise we have.