[racket] datum-intern-literal and lists?

At Mon, 25 Jun 2012 17:11:48 -0400, Danny Yoo wrote:
> > But as to measurements, what were you running that led you to try to
> > optimize this function?
> 
> I hacked a personal copy of scribble/run so it wrapped the computation
> with a call to the statistical profiler.  I then ran it on some of my
> own scribble files (as well as the Racket Guide and Reference) and
> looked at the generated profiles.

That's great! I've done that a lot, but my sense is that there are
still many improvements to find that way.

How much improvement in overall run time (as, say, just a ballpark
percentage) did you get from the changes?