When the OED says, with British restraint, that a use is avoided by many writers, it obviously means good writers, although it is too polite to come right out and say that

Since our last meeting, the cumbersome machinery of moving paper from one place to another has coughed up two weighty lexicons I had hoped to write about in the December issue. They are The Random House Dictionary of the English Language, second edition, unabridged, and the compact edition of the Oxford English Dictionary, Volume III, which is described as a supplement to the OED, Volumes I -IV. I already had the two-volume compact edition, and I don't know what the hell all these numbers mean, so I think I'll talk about the first one. For those who have not seen them, however, I should explain that the compact OED editions utilize an optical reduction technique that enables the printing of four pages of the original on each page of these cumbersome volumes, which are best studied with the aid of a magnifying glass. The two main volumes and the near supplement comprise more than 20,000 pages. and, since I can afford neither the money nor the shelf spate to have the real thing on hand, I am grateful for this publishing gimmick, however awkward it may be to use.
The nest edition from Random House (it calls itself RHD-II for short) is a welcome addition to my collection. It is, for now, the newest big, serious dictionary around. It is the first new entry in its class since 1966, when RHD-I was published, and boasts "over 315,000 entries in 2500 pages" and "over 50,000 nevi words and 75,000 new meanings." (I am quoting here from the dust jacket, and if you are put off by the missing comma or the misuse of over, please do not judge this book by its cover.) It weights in at a respectable 10.5 pounds on my kitchen scales, but is a slender volume in comparison with my beloved Webster's Second (600,000 entries in 3,200 pages and not weighable on said scales). Its type is easier on tired old eyes. It skimps on etymology.
The first word I looked for, of course, was hopefully. Although what I found put me in a snit, it also introduced me to a welcome feature: "-Usage" notes appended to many controversial definitions.
Although some strongly object to its use as a sentence modifier, HOPEFULLY meaning "it is hoped (that)" has been in use since the 1930's and is fully standard in all varieties of speech and writing: Hopefully, tensions between the ft.-O nations wilt trace. This use of HOPEFULLY is parallel to that of certainly, curlously, frankly, regrettably, and other sentence modifiers.
Setting aside the fact that I have dug in my heels for life on the matter of hopefully, I was discouraged by a couple of other things in that paragraph: the use of ease as an intransitive verb and the apostrophe in l930's, not to mention the manner in which the writer is begging the question by saying that the use is parallel to the other four adverbs he lists. But there a something almost puckish about the whole thing, and I decided not to heave the book out the window just yet. Instead, I turned to the new supplement to the OED and found an added meaning for hopefully:
It is hoped (that); let us hope. (Cf. G. hoffentlich it is to be hoped.) orlg. U.S. (Avoided by many writers.)
As everyone knows, the OED is basically an historical dictionary. When it prescribes, it does so with British restraint. If it says that a use is avoided by many writers, it obviously means good writers, although it is too polite to come right out and say that.
I would not characterize RHD-II as a prescriptive dictionary, but it is a tad closer to being one than is the rigorously descriptive Webster's Third. It admits, however reluctantly, that it might be worthwhile to preserve the distinction between imply and infer. It is quite conscientious about sticking such labels as slang and informal in the proper places.
On the other hand, so to speak, it defines little finger as "the ringer farthest from the thumb, the smallest of the five lingers." It gives several definitions for thumb, none of which is finger. It defines finger as "any of the terminal members of the hand, asp. one other than the thumb." I have just looked at one of my own hands, and my little finger is the shortest of my four fingers, but longer than my thumb, albeit thinner, and I am not willing to cut off a thumb and a pinky and do a water-displacement test just to find out which is "smaller." I am satisfied simply to know that while one of my ringers may be a thumb, my thumb is not a ringer, but I still am lacking one more finger for any little ringer to be smaller than.
Such nonsense aside (although the nonsense was inspired by my stumbling over what I hope was atypical imprecision). I consider RHD-II a good investment as the most up-to-date and comprehensive available source of information on how the language is being used, for better or for worse. I am resigned to the improbability of my living to see the production of a dictionary containing both that information and valid guidelines to how the language should be used.