[MLton] SML documentation tool(s)

[Stephen Weeks]

> >  * portability across SML implementations (too hard)
> 
> I guess you have a good cause to say this, but in some cases I don't see the
> hardness here. For example, what kind of incompatibility is met when
> developing a red-black tree implementation? Of course, when doing something
> more system specific, I can imagine the difficulties.

To be clear, I don't mean that we should go out of our way to be
nonportable.  I agree, red-black trees probably would be portable.  I
simply mean that it's not worth our (the MLton community's) time
porting our work to other SML implementations or going to the great
effort that the SML basis library did to allow multiple conforming
implementations (e.g. with different integer sizes).  The benefit is
just not worth the cost.