I didn't port the tth subroutines from dangerousMacros to PGcore and the new PGbasicmacros.pl during the (partially completed) upgrade of the PG fundamentals from perl 4 to object oriented perl 5. I didn't port them for the reasons mentioned by Arnie.
I can probably do the port (I'll have to look at how much of the old infrastructure will have to be moved over) but I won't be able to get to this until
some time next week or next weekend. Porting tth to maintain backward compatibility won't do any immediate harm, but itdoes allow users to live in the past :-) and adds a bit to the overhead of maintaining the system.
It would be worth it to me to see if the user would consider MathJax or jsMath as an alternative -- to my knowledge both of these are superior to tth in every feature. If not I'll see what can be done about maintaining tth for a little while longer, but I'd be interested in the reasons for needing tth so that I can communicate this back to the MathJax and jsMath folk; perhaps they can plug any gaps.
tth was a great protocol for the beginning of math on the web -- very much in the spirit later extended by MathJax and jsMath and orthogonal to the LaTeXtoHTML and "images" paradigm. But I'm pretty sure it is no longer updated or maintained and may cease to work on modern browsers at some point.