[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
- To: RMS at MIT-AI, (BUG LMMAN) at MIT-AI RMS@MIT-AI 05/22/78 04:31:44
- To: (BUG LMMAN) at MIT-AI
- From: DLW@MIT-AI 05/23/78 01:54:34
1) Backquote's usefulness is not particularly restricted
to macros. I think that it should be documented where
LIST and CONS are documented, or where regular quote is documented,
or some such place, instead of under macros.
Although it is not restricted to macros, 99% of the use of it
can be found there.
2) The documentation on macros follows the mathematical
order, defining everything before it is used. This makes it
unmotivating for the newcomer, who will get turned off
by the ugly macros that don't use ` or defmacro, and
will give up without realizing that macros are really easy to use.
I hope you aren't seriously suggesting that we document DEFMACRO
before documenting simple macros. I cannot hink of a more confusing thing
to do. I do not consider the LMMAN an advertisment for Lisp;
it is a reference manual.
3) While SETF is a macro, it is so fundamental that
people shouldn't have to have read about macros to have heard
about it. The action of SETF is no more mysterious to a beginner
than the action of SETQ.
My answer to this is my real answer to all three parts of this
letter: the macros section is not a little obscure section that very
few people will read; it is a fundamental part of the manual that
each and every user is going to have to read. Macros are much
more important in the Lisp Machine system than they traditionally
have been in Maclisp; we use them a lot, and we have lots of installed
system macros where Maclisp has none. I think the present orginization
is OK for now. When we have lots more text written, I intend
to look at the stuff and really figure out the organization then,
including such issues as the policy on what gets to be a "Chapter",
"Section" or "Subsection".