>old message:
1. http://arxiv.org/pdf/astro-ph/0507619v1.pdf
2. http://arxiv.org/pdf/astro-ph/0602519.pdf

Newton gravity might naively be expected to
work well with no need to use full blown general relativity (GR)
for galaxies.  But the first paper claims that is not true!  Actually they
say you DO need to use GR, and they do in a small-G approximate form,
and find much less "dark matter" is needed to explain a bunch of
galaxies. In fact in some cases none needed?

But the second paper claims the first paper had errors but agrees GR is
needed and says GR reduces amount of dark matter needed by 33% vs
Newton.  This however still leaves us needing a huge amount dark
matter.  So the whole "dark matter" problem is not solved, but major
pieces of stupidity allegedly have been revealed.

--just found this paper
http://arxiv.org/pdf/astro-ph/0508377v2.pdf
which claims the original paper was seriously messed up since there was a huge
unphysical singularity they did not notice.

Another paper
http://arxiv.org/abs/astro-ph/0701019
questions more problems.

Another
http://arxiv.org/abs/astro-ph/0601191
found even more problems.

Another:
http://arxiv.org/abs/astro-ph/0510750

So... all in all, sorry, it looks like this whole line of research was
bullshit-dominated.