Ulf Martin wrote:
I wonder how people (esp. at Pragma) currently deal with this. for projects where we use relatively new features (which evolve) we use frozen trees; Confirm One tree of 2002 ("still-crazy-after-all-these-years"). Another of 2004. Switch to last pdftex/context sometimes second quarter of this year; switch to luatex at the end of next year. Why switch ? Last versions. are better (speed and features);
On 4/14/07, Hans Hagen
with regards to commands and such ... context is just (supposed to be) downward compatible; even kind of obsolete is still there; with regards to different solutions to problems, we often provide control usign low level mode indicators
On average, my macros are not completly portable from one tree to another, but I'm sure that this depend from my poor coding tecnique for 95% . 5% is made by spaces and fonts .
concerning xetex ... keep in mind that there xetex is the moving target (changes/extensions in interface) and to some extend this was true for pdftex as well, but there we could silently adapt
Also remember that Knuth originally intended TeX to be an "eternal" formatting system (thus we have at least the option to expand all macros into plain TeX and keep that as the source file).
plain tex is just a format and unsuitable as expanded format
well, i have some experimental code that dumps the expanded token list into a file; nu fun ... a 50 page moderately complex doc becomes some 25 meg -)
pdf has some sort of compression . Do \pdfcompresslevel=0 \pdfobjcompresslevel=0 make some differences in your 50page document?
Sanjoy has set up an advanced test system ... so anything that you contribute can go in there I will install on my machine.
luigi