From: ············@gmail.com
Subject: Please reconsider this fabulous book
Date: 
Message-ID: <1132001409.914245.244260@g47g2000cwa.googlegroups.com>
Last time I asked about this book I got discouraging replies that made
me discount it and move on to SICP. I'm now finding myself back to this
book, which I consider a much gentler introduction to SICP, albeit very
faithful to it, as you'll see in the paragraph quoted below.

Almost everything basic and essential is there; lists, trees, etc. Once
the reader moves on to SICP the extra details will all seem somewhat
familiar and much easier.

It's an MIT Press book (just like SICP), by Brian Harvey, a late 1960s
MIT graduate who then moved to Standford and UCB.

>From the preface "A big change since 1984 is that Logo is no longer the
only member of the Lisp family available for home computers. Another
dialect, Scheme, has become popular
in education. Scheme has many virtues in its own right, but its
popularity is also due
in part to the fact that it's the language used in the best computer
science book ever
written: by Harold Abelson and Gerald Jay Sussman with Julie Sussman
(MIT Press/McGraw-Hill, 1985). I have a foot in both camps, since I am
co-author, with Matthew Wright, of (MIT Press, 1994), which is sort of
a Scheme version of the philosophy of this
book.
The main difference between Scheme and Logo is that Scheme is more
consistent
in its use of functional programming style. For example, in Scheme,
every procedure
is what Logo calls an operation-a procedure that returns a computed
value for use by
some other procedure. Instead of writing a program as a sequence of
instructions, as
in Logo, the Scheme programmer writes a single expression whose
complexity takes the
form of composition of functions.
The Scheme approach is definitely more powerful and cleaner for writing
advanced
projects. Its cost is that the Scheme learner must come to terms from
the beginning with
the difficult idea of function as object. Logo is more of a compromise
with the traditional,
sequential programming style. That traditional style is limiting, in
the end, but people
seem to find it more natural at first. My guess is that ultimately,
Logo programmers who
maintain their interest in computing will want to learn Scheme, but
that there's still a
place for Logo as a more informal starting point."

Yes, it uses a version of Logo that he wrote, as to "Why Logo?", I
quote from the preface

"Why Logo?
Logo has been the victim of its own success in the
reputation as a trivial language for babies. Why, then,
of books about serious computer science? Why not
The truth is that Logo is one of the most powerful
for home computers. (In 1984 I said "by far the most
have become larger and Logo finally has some competition.)
language used in the most advanced research projects
in artificial intelligence. Until recently, all of the
trivial, and they tend to underscore the point by strewing
But the cute pictures aren't the whole picture."

In fact, the book is quite lispish, it's a computer science book, about
"formal thinking", not turtle graphics. There's much valuable from the
preface and elsewhere in the text that I could quote, but I'd suggest
you just have a look for yourselves.

I really think it's a fabulous, very fabulous book, it's a gem, I'm
already in love with it, and I really wholeheartedly and sincerely
suggest that you reconsider it and recommend it to the beginning reader
of the lisp literature. It made me fall again in love with the elegance
of the lisp concepts after, I must admit, some disillusionment with how
abstract and remote in immediate practical usefulness it seemed for
someone at my stage. 

http://www.cs.berkeley.edu/~bh/

From: Barry Margolin
Subject: Re: Please reconsider this fabulous book
Date: 
Message-ID: <barmar-D7D245.20364214112005@comcast.dca.giganews.com>
In article <························@g47g2000cwa.googlegroups.com>,
 ············@gmail.com wrote:

> Last time I asked about this book I got discouraging replies that made
> me discount it and move on to SICP. I'm now finding myself back to this
> book, which I consider a much gentler introduction to SICP, albeit very
> faithful to it, as you'll see in the paragraph quoted below.
> 
> Almost everything basic and essential is there; lists, trees, etc. Once
> the reader moves on to SICP the extra details will all seem somewhat
> familiar and much easier.
> 
> It's an MIT Press book (just like SICP), by Brian Harvey, a late 1960s
> MIT graduate who then moved to Standford and UCB.

Your message would make more sense if you mentioned the name of the book 
you're praising.

-- 
Barry Margolin, ······@alum.mit.edu
Arlington, MA
*** PLEASE post questions in newsgroups, not directly to me ***
From: ············@gmail.com
Subject: Re: Please reconsider this fabulous book
Date: 
Message-ID: <1132020419.636899.276610@g14g2000cwa.googlegroups.com>
Barry Margolin wrote:
> In article <························@g47g2000cwa.googlegroups.com>,
>  ············@gmail.com wrote:
>
> > Last time I asked about this book I got discouraging replies that made
> > me discount it and move on to SICP. I'm now finding myself back to this
> > book, which I consider a much gentler introduction to SICP, albeit very
> > faithful to it, as you'll see in the paragraph quoted below.
> >
> > Almost everything basic and essential is there; lists, trees, etc. Once
> > the reader moves on to SICP the extra details will all seem somewhat
> > familiar and much easier.
> >
> > It's an MIT Press book (just like SICP), by Brian Harvey, a late 1960s
> > MIT graduate who then moved to Standford and UCB.
>
> Your message would make more sense if you mentioned the name of the book
> you're praising.
>

It's in the link I provided

"Now FREE for personal use: The second edition of Computer Science Logo
Style, a three-volume series intended mainly for teenagers and their
teachers, or for adults who aren't professional programmers. You can
look at the tables of contents and complete downloadable PDFs and
browsable HTML versions of

    * Symbolic Computing, a Logo programming text that concentrates on
natural language processing rather than the graphics most people
associate with Logo.
    * Advanced Techniques, in which discussions of more advanced Logo
features alternate with sample projects using those features, with
commentary on the structure and style of each.
    * Beyond Programming, brief introductions to six college-level
computer science topics. "
> --
> Barry Margolin, ······@alum.mit.edu
> Arlington, MA
> *** PLEASE post questions in newsgroups, not directly to me ***
From: ············@gmail.com
Subject: Re: Please reconsider this fabulous book
Date: 
Message-ID: <1132020940.419037.68880@g49g2000cwa.googlegroups.com>
············@gmail.com wrote:
> Barry Margolin wrote:
> > In article <························@g47g2000cwa.googlegroups.com>,
> >  ············@gmail.com wrote:
> >
> > > Last time I asked about this book I got discouraging replies that made
> > > me discount it and move on to SICP. I'm now finding myself back to this
> > > book, which I consider a much gentler introduction to SICP, albeit very
> > > faithful to it, as you'll see in the paragraph quoted below.
> > >
> > > Almost everything basic and essential is there; lists, trees, etc. Once
> > > the reader moves on to SICP the extra details will all seem somewhat
> > > familiar and much easier.
> > >
> > > It's an MIT Press book (just like SICP), by Brian Harvey, a late 1960s
> > > MIT graduate who then moved to Standford and UCB.
> >
> > Your message would make more sense if you mentioned the name of the book
> > you're praising.
> >
>
> It's in the link I provided
>
> "Now FREE for personal use: The second edition of Computer Science Logo
> Style, a three-volume series intended mainly for teenagers and their
> teachers, or for adults who aren't professional programmers. You can
> look at the tables of contents and complete downloadable PDFs and
> browsable HTML versions of
>
>     * Symbolic Computing, a Logo programming text that concentrates on
> natural language processing rather than the graphics most people
> associate with Logo.
>     * Advanced Techniques, in which discussions of more advanced Logo
> features alternate with sample projects using those features, with
> commentary on the structure and style of each.
>     * Beyond Programming, brief introductions to six college-level
> computer science topics. "
> > --
> > Barry Margolin, ······@alum.mit.edu
> > Arlington, MA
> > *** PLEASE post questions in newsgroups, not directly to me ***

Here it is
http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=3987
From: Barry Margolin
Subject: Re: Please reconsider this fabulous book
Date: 
Message-ID: <barmar-9095C6.20450215112005@comcast.dca.giganews.com>
In article <························@g14g2000cwa.googlegroups.com>,
 ············@gmail.com wrote:

> Barry Margolin wrote:
> > Your message would make more sense if you mentioned the name of the book
> > you're praising.
> >
> 
> It's in the link I provided
> 
> "Now FREE for personal use: The second edition of Computer Science Logo
> Style, a three-volume series intended mainly for teenagers and their
> teachers

The link was to someone's personal home page.  I thought you might have 
been referring to that book, but there's so much stuff on his web page 
that it was not obvious.

-- 
Barry Margolin, ······@alum.mit.edu
Arlington, MA
*** PLEASE post questions in newsgroups, not directly to me ***
From: ·············@gmail.com
Subject: Re: Please reconsider this fabulous book
Date: 
Message-ID: <1132033039.860347.48180@g43g2000cwa.googlegroups.com>
············@gmail.com wrote:
> Last time I asked about this book I got discouraging replies that made
> me discount it and move on to SICP. I'm now finding myself back to this
> book, which I consider a much gentler introduction to SICP, albeit very
> faithful to it, as you'll see in the paragraph quoted below.


Simply Scheme, also by Brian Harvey, seems like a better option for a
gentle introduction to SICP than the Logo books you mentioned.
From: ············@gmail.com
Subject: Re: Please reconsider this fabulous book
Date: 
Message-ID: <1132093670.558965.259630@f14g2000cwb.googlegroups.com>
·············@gmail.com wrote:
> ············@gmail.com wrote:
> > Last time I asked about this book I got discouraging replies that made
> > me discount it and move on to SICP. I'm now finding myself back to this
> > book, which I consider a much gentler introduction to SICP, albeit very
> > faithful to it, as you'll see in the paragraph quoted below.
>
>
> Simply Scheme, also by Brian Harvey, seems like a better option for a
> gentle introduction to SICP than the Logo books you mentioned.

I think the logo book would be a nice intro to the scheme book, and the
scheme book in turn would be a nice intro to SICP.

The logo book seems pretty basic, easy to digest for someone new to
lisp concepts, it takes its time in exaplaining things, it introduces a
lot of the essential stuff such as lists, trees, functions, recursion,
tower of hanoi, etc. It's also written with much tenderness and
patience.

I think it's fabulous for the beginner. Perhaps not for someone who
already has scheme in his pocket, but for someone new to the lisp world
and literature.

It's really more or less the same essentials of the stuff in SICP, just
much more informal and kinder for the first-time reader.
From: Michael Sullivan
Subject: Re: Please reconsider this fabulous book
Date: 
Message-ID: <1h62x47.7sj44eddj6luN%use-reply-to@spambegone.null>
·············@gmail.com <·············@gmail.com> wrote:

> Simply Scheme, also by Brian Harvey, seems like a better option for a
> gentle introduction to SICP than the Logo books you mentioned.

I really liked HDTP, which covers a lot of the same ground as SICP but
with a gentler start, less hairy math and more practical problems for
non-number crunchers.  Very clear writing and contains a great
exposition of what's important in good program style.  SICP assumes you
are either a hacker or abstract/discrete math geek when you start.  HDTP
looks like it's aimed at a wider audience, but following it to the end
still gets you to a place the typical programmer fears to tread.


Michael

[*] _How To Design Programs_, Felleisen, Findler, Flatt & Krishnamurthi