Quote Originally Posted by John T. Fishel View Post
As far as I can tell, a book list really does no harm. It is only the compiler's view of what is good in the field.
A book list is, by it's nature, a recommendation, but why these books get recommended never seems to be made explicit. How would anyone feel about a list of books "not to read?" - so why feel comfortable with lists "you should read."
These days I limit my recommendations to books because they inform and explain on very specific subjects. EG: I recommend 2-3 books which explain CvC well. Reading them will not ensure you really understand CvC though.
Quote Originally Posted by marct View Post
One of the dangers I have seen with reading lists is the "Cole's Notes" effect where people go off and get synopses rather than actual read them. Toss in a few quotes, vaguely refer to them and, presto-chango, you are now an "expert" in the area.
Concur, or merely harvest them for convenient perspectives taken out of context.
That being said, I like the idea of reading lists, at least as long as they are put together with a specific end in mind and some suggestions on how to tie them together (not necessary in the case of John's work !).
I have no issue with recommending specific books for specific subjects. Indeed I see it as an essential tool in education. I strongly oppose the "read this lot and you'll know about COIN" approach, as exemplified by the CNAS type reading list