Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!rutgers!cmcl2!lanl!jlg
From: jlg@lanl.gov (Jim Giles)
Newsgroups: comp.lang.misc
Subject: Answers, Chapter 1: TeX (was C's sins... and others)
Message-ID: <66253@lanl.gov>
Date: 19 Oct 90 21:14:58 GMT
Organization: Los Alamos Natl Lab, Los Alamos, N.M.
Lines: 50

> [... All quotes from Dan Bernstein (who else) ...]
> QUESTION 1: Do you see the contradiction in ``I absolutely refuse to
> look at the example you're giving me!'' and ``I am not turning away from
> anything!''?

Had I actually made the first statement, I would have been in
contradiction.  I did not.  I said (and still say) that I will
consider _any_ example that you care to post.  You have refused to
post the example you keep refering to.  However, I _have_ finally
looked over the source of the code you refer to. See below.

>> If there's something in TeX that can't be
>> done with recursive data structures (and the other features I've supported
>> on the net so visibly), then you ought to be able to give a specific
>> example - in C if you like.
>
> Packed array tries are not trivial to implement. I don't see why I
> should bother to type in a presentation just for you, when Knuth has
> already done a much better job.

Well don't bother then.  As it happens, someone _DID_ email the source
code for TeX 'packed array tries' over the weekend.  The news is even
worse for your argument that I expected.  There are _NO_ pointers in
the source code for TeX 'packed array tries'.  The code does everything
through the use of Pascal arrays.  Since arrays are one of the features
that I've "supported on the net so visibly" that I mentioned above, the
TeX code doesn't even meet the criterion.

> QUESTION 2: Why do you refuse to look up packed array tries? Perhaps
> because you're afraid that I'm right?

No, but if I were as cynical as you, I would suspect the reason that
you didn't post the example was that you knew the actual code didn't
support your position.  I am _not_ so cynical.  I'm sure you read the
_description_ of the code (which, no doubt pretended that explicit
pointers were used throughout).  As such, I accept that you argued the
question in good faith.

> QUESTION 3: Can you read? Do you see the word ``packed''? Does it even
> occur to you that there's a difference between a packed array trie and a
> trie (by which people usually mean a list trie or a normal array trie)?

It's not clear to me what kind of packing that pointers allow that
isn't allowed by some combination of the features that I've mentioned.
The example of TeX makes it quite clear that at least one example you
thought important _can_ be done with arrays quite easily.

End of chapter 1.

J. Giles