What was physics like?

When I started there were about 24 people on my thesis experiment. It was a follow-on experiment, using almost entirely apparatus from an earlier one. I didn't do any experiment or detector design, though I helped build drift chambers. A lot of the software was there already, but needed redoing, so I worked on simulation and analysis; and of course running the system, where I wound up having to do some data-acquisition development too.

When I ended my career I'd drifted into IT (funding issues) for a big collaboration, but I'd spent most of the intervening time on really big collaborations--so big that the people who designed CMS weren't always still around when it started taking data. There were information silos all over the place--and management knew about that and was trying to make more connections. (It wasn't quite an N! problem, since some of the silos were related and had some communication with each other.)

A much younger researcher got burned by this.

The process of doing research, getting his PhD, and doing more research, was so organized and regimented that he missed out on practice in "ask a question", "talk it out with colleagues", "design a way to answer the question". I'm old enough that I got a little of that, and I've had the opportunity to hang out around some people who were really good at it. This guy felt like he was a cog in the machine: "We want to look at this, so you go run those programs." I grumped that a lot of LHC grad students never got to do anything except develop and run simulations, since the machine wasn't even going to run for a few more years. It hadn't occurred to me that analysis might wind up being plug and play for the students, after the initial development.

He was on Atlas, and I on CMS, so I probably didn't meet him the times I was over there.