Magazines and Periodicals

[KGill]

(1) that a pilot might be nice

Completely agreed.

(2) we should reuse as much of the structure as possible, and go from there (we can certainly make additional changes down the road).

Well, I'm not sure which proposal your post is 'aimed' at (if not all of them), but I realize mine is probably the biggest leap from 'normal' practice so I'd just like to point out that the main part of the system proposed probably wouldn't require any technical changes on your end, just a new system of categories. (The page to browse full issues would need some technical changes, of course, although the hypothetical new FTE template at least could be created by any of us, and perhaps there would just be some medium-sized adjustments to the Add File page other than that.) The CW is already equipped to handle category intersects with maximum ease, so maybe only minor tweaks would be needed there In fact, it might be worth considering combining my proposal with Philip's and creating a system where one could browse by year of publication and publisher in addition to the methods we already have now - for any work, not just magazines/periodicals. This would not exactly be a light undertaking, but would it perhaps be worth thinking about?

[imslp]

My advice would be to, instead of designing a system from scratch:

1. Come up with the closest approximation to what you want with the current system only (i.e. the composer category and the work pages), and create a small pilot.
2. Come up with a list of improvements ranked from the most important to the least important, and we can decide where we want to draw the line. Improvements should be backwards compatible wherever possible (e.g. instead of replacing the People category, perhaps adding another category).

I think this procedure is best, because depending on how many items there are we can adjust how many improvements we will make. For example, if there are less than 100 such magazines/periodicals it would not be efficient to spend lots of time making improvements, whereas things would be different if there were 1000+.