During winter conferences, which take place between mid-February and the end of March in La Thuile, Lake Louise, and other fashionable places close to ski resorts, experimentalists gather to show off their latest results. The same ritual repeats during the summer in a few more varied locations around the world.
The two months preceding these events are really work-intensive, as hundreds of different data analyses must converge in synchrony. Imagine how miserable your life can become if you are the convener of a physics group and you have the responsibility to bring 30 different results to completion within the allotted time: each of them requires a half dozen presentations to vet, hours of review meetings to sit in, the polishing of public documentation, etcetera.
The question is, what is the reason for this 6-months cadence? Of course there is no real justification in terms of synergy with the running of experimental apparata (LHC, for instance, has a yearly data-taking schedule). The simple answer is that it has become a habit. Nowadays large collaborations cannot fail to provide updates on their most important physics investigations every six months, lest they give the impression that there is some problem with their data, or that they are sleeping on the job.
There is also another reason why it is hard to move away from this cadence today. If two experiments search for new phenomena and have roughly the same amount of data to look at, and their sensitivity to those physics processes is similar, then it is very important that they publish their results without delay: once the result of the competitor is out, showing that the sought physics processes has been ruled out "for new particle mass below Mx", e.g., the importance of your own result, also roughly excluding the same parameter space, decreases significanly. This means fewer citations in all forthcoming articles discussing the same physics model, for instance.
I find the whole thing a bit questionable, although I cannot really say I have a proposal for a significantly better working model. My lack of satisfaction stems from the fact that usually the experiments try to publish in scientific journals the analyses as they were shown in the conferences, or with minimal modification; what was shown as "preliminary" at the conference has better be confirmed in the scientific article, to avoid giving the impression that whatever is preliminary is not to be trusted in full. This has a negative impact, in potential, with the architecture of the analyses, as the experimentalists working at them know they have to work around the clock in order for their result to make it to the conference. Hence they may feel implicitly discouraged from improving their analysis setup or provide significant innovations, which make convergence within the deadline less certain. "Turning the crank" and spewing out well-controlled and understood results, but maybe not so cutting-edge ones, becomes a hardly avoidable choice.
Add to that the fact that often, once a large collaboration publishes a result on some physics measurement based on a given set of data, they become unwilling to allow another publication based on the same data and search. The reason for this is that there is always new data to work on, and if the management of a big experiment allows the working force to disperse on refining old results, the whole machinery grinds to a halt.
Mind you, I am not criticizing anybody here: although I do believe there is an issue, it is very difficult to steer a large experiment away from this equilibrium point. What I would like to know, if you have a minute to spare for the comments thread, is if you share my concerns or if you rather believe that the way the large HEP experiments operate in this respect is totally sound... Thank you for your opinion!
If you wish to have an insider view of particle physics in large experimental collaborations, you should consider reading my recent book "Anomaly! Collider physics and the search for new phenomena at Fermilab". More information is available at the World Scientific site of the book.