Timings

[Jon L White <edsel!jonl@labrea.stanford.edu>]

re: (functionp 'let) -->t
    (functionp 'jillyjobbe) -->t
    Not too useful.   I don't know how you are supposed to tell
    what is a function.  eg. interpreted functions are sone #<interpreted...>..

Lucid implements the CLtL standard -- see p76 -- and is not happy about it;
in the person of Dick Gabriel, we have been active in getting x3J13 to adopt
a more sensible definition of the CL type FUNCTION.  Possibly the other
vendors have simply ignored the craziness of CLtL p76.

re: (delete a b :count most-positive-fixnum) fails.
    There is some artificially small limit on the count allowed.

The limit imposed is not artificial, but the very practical one defined
by the CLtL constant ARRAY-TOTAL-SIZE-LIMIT (see p290); this value,
rather than the accidental placement of the cut between fixnums and bignums,
is the relevant one. [In fact, in Xerox Common Lisp, the reverse situation
holds -- MOST-POSITIVE-FIXNUM is orders of magnitude smaller than 
ARRAY-TOTAL-SIZE-LIMIT!]

re: Also I seem to run out of space very easily while compiling, and later
    while running.  We are just using the default configuration.
    . . . 
    I guess I am spoiled by having worked  almost exclusively with software
    which had all sources and most documentation on line.

I'll have to forward your message to our manager of the Sun Product
release.  I'm not fully sure I know what you mean by "running out of
space", but I suppose it could mean a lack of swapping space on the
disk?  Now, I too am somewhat spoiled in the way you suggest; Lucid
is working hard to get some reasonable on-line documentation.


-- JonL --