Hopefully someone will get so fed up with it that they just make a powerful replacement from scratch. (Maybe with a TeX export utility.)

I think that, as with standards, the problem isn't no replacement but too many: see Pandoc (http://johnmacfarlane.net/pandoc), Lout (http://savannah.nongnu.org/projects/lout), its (not yet existing) successor Nonpareil (http://sydney.edu.au/engineering/it/~jeff/nonpareil), Skribilo (http://www.nongnu.org/skribilo), and Pantoline (which I can't find with a quick Google search), for example. The problem is that TeX has so much inertia, and such an eco-system behind it, that getting significant investment in any one of them is likely to be difficult to impossible.

I think compilation to TeX is probably always going to be less successful than compilation from TeX—that is, taking the same code and (slightly) improving the processing. See the pdfTeX (http://www.tug.org/applications/pdftex) and LuaTeX (http://www.luatex.org) projects.

As a remedy to too much ambition even when trying to build directly on TeX, see the fate of LaTeX3 (http://tex.stackexchange.com/questions/953/why-is-latex3-tak...).

Alternately, you could rely exclusively on a TeX export utility and make a clean, less verbose language that compiles to TeX. Then existing packages could be either wrapped similarly or inserted as TeX blocks.

on that note, check out the combo https://github.com/softcover/softcover/ + https://github.com/softcover/polytexnic/ which is a very good first start on that approach:

    markdown --> .tex ____ pdflatex --> .pdf
                    \
                     \___ tralics  --> ruby_scripts --> .epub

I think one reason most people wanting a next-gen-TeX project don't aim to compile to TeX is that some of the biggest pain points of TeX are baked into the core. So if you want to improve them, you need to change or replace at least some of the core layout algorithms, not just the front-end input language. For example a big wishlist item for many years has been some kind of improvement on TeX's quite frustrating figure placement, possibly with a more pluggable layout algorithm.