At National Lab, like at a university, the PI gets to decide who is an author. My inclination, which is how things worked in my branch at National Lab is to list everyone who contributed to the work as an author. This means that technicians are authors, not just listed in the acknowledgements. This is different from the definition of an inventor for patent purposes (an inventor has to have made an intellectual contribution, which may or may not be true of a technician, depending on the tech and the lab). My collaborator (from a different branch at National Lab) prefers to use the inventor rules for authorship, and relegates everyone else (even some of the people who actually generated the data) to acknowledgments.
Thus the argument--I want my technician listed as an author. Collaborator wants my technician in the acknowledgements only. My technician took data that is critical to the paper, which I think earns authorship. My technician did this under my direction, and followed my protocols without changes, which my collaborator thinks means no intellectual contribution (and for a patent, they would be right). This was a joint project, so we are co-PIs, and I can't just override their objections. I am hoping to end this disagreement quickly (and amicably, but since I am no longer at National Lab, I will fight pretty hard for my tech, especially since collaborator will have limited opportunity to screw me over in the future).
The broader question of who is an author is interesting to me as a new TT prof. I've been in labs where authorship was treated as a prize to be handed out. All morality aside, this leads to awful group dynamics. But making everyone in the group an author is just as bad--why work when you can expand the CV for nothing? I've also been in groups where the bar to authorship is pretty high, and this can lead to reluctance to help out a groupmate with something time consuming. I've been fortunate to work mostly with people who are upfront about authorship rules, and don't renege at the end. I've heard many horror stories, though.
I am in a sub-field where collaboration is the norm, since our experiments require multiple disparate areas of expertise. A typical paper in my research area will have 3 to 10 co-authors. My own philosophy is to give proper credit as much as possible, both for ethical reasons, and because I find that people are far more willing to work with me when they know they will be properly credited. In my group, I keep track of what everyone is working on, and I also encourage my students to let me know when someone has made a substantial contribution that might be invisible to me as the PI so we give proper credit where credit is due. After leaving National Lab, I see how truly important this is, since I need to trust that my former colleagues will do the right thing with the projects I left behind. In my previous collaborations, each PI submitted a list of contributors in order of importance, and that combined list became the author list. I see now that I will need to discuss this issue right up front, since fighting about it at the endpoint is a real drain.