Am 02.01.2011 um 21:07 schrieb Joseph Wright:

> If the idea is that you can use LaTeX3 without needing to know all of the underlying TeX

still it would be great if the behavior of the L3 macros were formally defined in TeX terms: I often experience that I cannot use certain L3 macros because it is not documented whether they expand to, say, an <internal integer> or an <integer denotation>. \dimexpr ... \relax is guaranteed by the e-TeX manual to be an <internal integer>, but what \int_eval:n does is undocumented—in fact, it expands to an <integer denotation> without trailing space, making things like

\x = \int_eval:n { 1 + 1 } 1

possible. 2e's counters and length were designed to make such effects impossible, but L3 reintroduces them :-(
\dim_eval:n, on the contrary, expands to an <internal dimen>. I think that is the right choice because it is faster and leads to fewer problems. I think a formal description like the following would be nice:

"Formal definition: \int_eval:n is a non-long, non-protected, non-outer macro. The token sequence \int_eval:n <left brace> <integer expression> <right brace> expands after an undefined number of expansion steps to a token sequence that represents an <internal integer>."
Personally I would benefit from such guarantees because I could use more from L3. At the moment, my packages use a mixture of 2e, L3 and TeX primitives, and until L3 can replace every bit of TeX and 2e, the TeX speak won't disappear.