[tex-k] Bug-report for the TeXbook: Not all non-primitive control-sequences are defined, ultimately, in terms of the primitive ones.

[Laurence.Finston at gmx.net]

I was inclined to dismiss Ulrich's observation, but having given it some thought, I think he is actually correct.  I do believe that Knuth meant the "about 900 control sequences" mentioned further up on page 9 and not all possible macros one could define.  However, plain TeX defines \empty and \space without using any primitive control sequences, so the sentence "All other control sequences are defined, ultimately, in terms of the primitive ones." seems to me to be genuinely incorrect.

If it was changed to "primitive operations" (which would include "typeset a character" and "insert glue") and the "null operation" or "nothing" (for the expansion to nothing of \empty) was explicitly included, then I think it would be correct.

Just my two cents.

> Gesendet: Freitag, 09. Dezember 2022 um 23:10 Uhr
> Von: "Karl Berry" <karl at freefriends.org>
> An: ud.usenetcorrespondence at web.de
> Cc: tex-k at tug.org
> Betreff: Re: [tex-k] Bug-report for the TeXbook: Not all non-primitive control-sequences are defined, ultimately, in terms of the primitive ones.
>
> Hi Ulrich,
> 
>     while explicit non-active
>     character tokens are not considered primitives.
> 
> Although it isn't a primitive control sequence, "typeset a character" is
> certainly a primitive action. I agree the wording is not 100% correct,
> considered as pure language specification.
> 
> If the first part of your quote:
> | About 300 of TeX’s control sequences are called primitive;
> didn't mention "control sequences", it would be more correct, but less
> helpful to actual readers, seems to me.
> 
> I doubt Knuth would want to change anything, but I'll discuss with my
> co-vetters (not soon, as the pile is already large, unfortunately).
> 
> Thanks for the report/suggestion,
> Karl
> 
>