By 1903, Falconer Madan, one of the most notable of all Bodleian librarians, was probably sick and tired of explaining why his catalogs gave more details on some books than on others. His answer he christened the "degressive principle"; it called for "varying a description according to the difference of the period treated or of the importance of the work to be described."  This is hard to fault since it "makes use of the experience of [one's] predecessors." As for the term "degressive" (not digressive, or moving out, but degressive, moving down, descending, dismounting), I am told that it is often used in studies of medieval land taxation. Clearly it is not a term that leaps quickly to mind. Madan perhaps found it useful in intimidating his learned Oxford colleagues, especially those who loved to complain about libraries but whose minds did not always leap quickly when library practices were involved.
Madan saw "four forms for the description of a book, showing by degressive changes what details may fairly be omitted in short descriptions and what appear to be essential." These forms were, namely: a "full description," another with only "a certain amount of fullness and detail," a "short description," and a "minimum description." Helped by a team of eminent bibliographers, he proposed a scheme of seventy-eight practices, which he illustrated with examples. These he revised in 1923,  now with three levels instead of four: early, middle (beginning in 1557), and modern (beginning in 1800). Their books were, he proposed, "important," "interesting," and "ordinary."
Madan's plan reflects the rise of modern descriptive bibliography. It also underlies the more recent concept of cataloging levels, which permits polemical tracts and children's books to appear alongside scholarly treatises; detailed rare-book citations next to brief, routine ones; and music, maps, and films mixed in with them. The concern is for interfiling within a single system. How can citations fit together, complex and detailed ones next to simple ones, the important with the trivial? At the technical level of statements and sequences, Madan and his teams worked out practices that their successors have assumed and built on up to today. Making adaptable systems is the easy part. Any logical mind can do it. Today computers are making it even easier.
It is in selection that the game gets hard. What is important; what gets favored, and why? In Madan's day, arbitrary decisions were not questioned when they came from authoritative sources (like Madan). Early books were all important, later ones all interesting, modern ones all ordinary, because Madan said they were. Today selection is no longer so simple, and even harder is explaining why. Bibliographers today may never hear of the degressive principle, and if they do they may wish they hadn't. Those of them with great and deserved authority (Fredson Bowers and C. Thomas Tanselle come to mind) usually introduce the term by suggesting that others may be defining it differently--and they usually do so with conspicuous respect. The principle itself has a very long history, probably beginning with the monastic stenographers whose degressive principles were revealed by God. The principle is very much alive today. It cannot go away, and I don't think it should.
A degressive principle of sorts is applied to music, as we see it as important, interesting, or ordinary because of its sound in performance or its artistic conception. The principle as applied to the library materials of music, whether in the form of notation or recordings, is another matter. Notation is for performers, recordings for the general public, all of whom were once seen as "ordinary" at best. Another level--"oblivion"--thus needs to be introduced, for the contemptible or superfluous. It is reserved for what ought to fall outside the system entirely, what should be discarded or at best go to an arrearage pile. This, of course, was the practice in libraries for too long. …