Prev: COMPILE-FILE tertiary value, failure-p
Next: Some thoughts on extensible sequences
From: ccc31807 on 30 Jul 2010 09:55
I've now been through what seem to be the traditional textbooks for
Common Lisp, Wilensky 'CommonLISPcraft', Winston and Horn 'Lisp, 3rd',
and Touretzky 'Gentle Introduction.' If I were asked to recommend
introductory texts for a new Lisper, this would be my evaluation.

I think the best is Touretzky. It's probably the shallowest of the
books, but it is gentle, thorough, and has lots and lots of exercises
with the answers in the back of the book. By the time you finish the
book, you will have written some substantial scripts and have a good
understanding of why things work the way that they do.

Second best IMO is Wilensky. It is deeper than Touretzky, still has
lots of exercises, and the answer manual can be had separately. The
problem with Wilensky is that it's the most out of date of the three.
I especially like his pedagogical methodology.

Third IMO is W&H. This is still a good book, but not particularly for
someone brand new to Lisp because of the depth and bias toward AI. It
certainly is good as a supplement because it contains a summary of
Lisp functions, and the last half of the book is meaty.

Seibel's 'Practical Common Lisp' is an attractive book, and much more
suggestive than these three, but it isn't a text. It was my first Lisp
book, and deserves its place on the shelf, but ultimately isn't
filling. It's like a rich dessert -- well worth eating but somehow in
the end leaving the reader wanting something more substantial. THIS
ISN'T A CRITICISM. Seibel pretty much says the same thing in his
introduction: he deliberately covers a number of substantial projects
as an illustration of the uses of Lisp and wants to leave the reader
wanting more.

Finally, Graham 'ANSI Common Lisp' is NOT a book for the complete
beginner. It might be good for a class, or for a tutorial when working
one on one, but it leaves too much out that needs to be explained to
the beginner.

I'm writing this looking forward to Barski's book, scheduled for
publication September 15, 2010. I've been through 'Casting SPELS' and
liked his approach. I think that it probably deserves mention to new
Lispers (or old ones for that matter) but it's premature at this
point.

I'm also writing this disappointed at the delay of Levine's O'Reilly
project. I think that the O'Reilly books are the top of the line for
tech books, and believe that this particular one would have taken top
honors of the CL books available excepting maybe PIAP.

CC.
From: Rupert Swarbrick on 30 Jul 2010 10:37
ccc31807 <cartercc(a)gmail.com> writes:
> Seibel's 'Practical Common Lisp' is an attractive book, and much more
> suggestive than these three, but it isn't a text. It was my first Lisp
> book, and deserves its place on the shelf, but ultimately isn't
> filling. It's like a rich dessert -- well worth eating but somehow in
> the end leaving the reader wanting something more substantial. THIS
> ISN'T A CRITICISM. Seibel pretty much says the same thing in his
> introduction: he deliberately covers a number of substantial projects
> as an illustration of the uses of Lisp and wants to leave the reader
> wanting more.

Isn't the idea that one reads PCL and then supplements it with the
hyperspec for the actual reference bit?

To be fair, I starting with a mixture of emacs lisp, the hyperspec and
On Lisp, about 3 years ago. I'm not suggesting that this was the easiest
or most efficient approach. Oh, and I read(/deciphered) quite a few
hunks of free libraries (at least drakma and cl-ppcre if I remember
correctly). Oh, and you need to have something(s) you're trying to
write, otherwise you'll never do anything.

But, yeah, I would suggest that people trying to learn any new
programming language should give up trying to find "the perfect
introduction to <language>" and instead spend their time learning the
darn thing. I'm pretty certain that a book can only be _so_ good:
there's always some work required on the student's part.

Er, this was supposed to be a reply to your post, but maybe I can go for
slightly-related-rant instead. Incidentally, my comments apply to
learning other subjects (especially maths) too. Get two or more texts in
electronic or book or whatever form and then read bits of all of
them. You'll never read any of them "properly" cover to cover, so don't
expect to. Then _do_ something. By the time you've satisfied yourself
that you _did_ it, you'll understand the theory you needed.

Rupert
From: gary.schiltz on 30 Jul 2010 20:27
Personally, I can't imagine a better introduction to Common
Lisp than Seibel's Practical Common Lisp. Not a textbook per se,
but compelling and, as the name implies, practical.
From: Kenneth Tilton on 30 Jul 2010 22:28
gary.schiltz wrote:
> Personally, I can't imagine a better introduction to Common
> Lisp than Seibel's Practical Common Lisp. Not a textbook per se,
> but compelling and, as the name implies, practical.

Agreed. Practical and tho not a textbook, technically solid. We here on
c.l.lisp saw Peter digging deep into the language to get his book right,
and it shows in the final product.

Otherwise, I think Graham nailed it with ANSI CL, and Touretsky always
gets favorable reviews here.

kt

--
http://www.stuckonalgebra.com
"The best Algebra tutorial program I have seen... in a class by itself."
Macworld
From: Captain Obvious on 31 Jul 2010 14:09
What's about PAIP?
 |  Next  |  Last
Pages: 1 2 3
Prev: COMPILE-FILE tertiary value, failure-p
Next: Some thoughts on extensible sequences