-- card: 23305 from stack: in.0
-- bmap block id: 0
-- flags: 0000
-- background id: 3797
-- name: 


-- part contents for background part 1
----- text -----

From: king@entropy.ms.washington.edu (Jim King)

Date: 1 Mar 88 19:31:02 GMT

In article <43633@sun.uucp>, chuq@plaid.Sun.COM (Chuq Von Rospach) writes:
> This is from Dan Shafer on CompuServe, and is a list of Erata from the first
> edition of his HyperTalk programming book.
> 
**followed by list of errata**

Here are a couple of others from the Shafer book:

(1) a typo in the handler on page 118 (i should be x) 
> 
> p.155 - First sentence should be edited to read:  "This command will 
> find the "end" in the words blender and weekend as well as in 
> ending."
> 
(2) another on page 155, the command find "5" will not find all the
numbers with the digit 5 but will find only those which *begin* with 5
(or at least that happened when I tried it) whereas find chars "5"
will find them all.  Amusingly enough, Shafer makes a point of
explaining this quite clearly on the previous page.

This sense that Shafer hasn't read his own book is heightened by the
errors on 140 and 145 (using repeat for instead of repeat with) which
occur right in the middle of an explanation of the differences in
these two kinds of repeat.

I am impressed in a certain sense by Shafer's ability to whip out a
book of this size so fast and I have found some useful information in
it.  At the same time, while some errors are inevitable, it seems that
the scripts in final form could have been extracted (automatically,
with a button!) and checked to see that they worked before shipping
the book out to the unsuspecting multitudes.  While checking the text
would have been a more daunting job, it does seem that Shafer relied a
little too much on what he thought he knew happened instead of trying
it out.

As for the Goodman book, it seems more reliable to me so far, but I
did come across a curious passage on page 555.  He is explaining that
the function the Result contains a non-empty string if a Go or Find
fails.  First (no big deal) the message actually returned by FInd in
the Result is not exactly what he says, but more curiously, the
example at the bottom of the page, which involves typing "the result"
(no quotes) in the message box after some failed Go's and Find's
doesn't show anything although the implication is that there should be
an error message.  On the other hand, when I experimented with this,
it was possible to put the Result in Message inside a script.

HYPERTALK QUESTION:  Why does the Result only seem accessible in a handler and
not from the message box?  Is the Result local in some sense, or is it
cleared by using the message box?

Back to the books:

My guess is that it will be some time before we see a really
insightful book about Hypercard that does not come from someone with
an insider's relationship with the development team (when is Carol
Kaehler's book on scripting due out?) because the author will first
have to do enough Hypertalk programming to get some real insights and
then still will have to write the book.  While the insiders' books are
not quite fair to other potential authors' chances, at least these
books give "the rest of us" much better documentation than the Mac had
in the pre-"phonebook edition" Inside Mac days.



-- part contents for background part 45
----- text -----
Re: Shafer: HyperTalk Programming eratta