Galaxy (and cluster) formation in the first few billion years of the universe is a hot topic among astronomers and cosmologists. The recent annual AAS meeting was a chance for many researchers to share their findings (and for some to grab some limelight, to enhance their chances of a job or funding).
The current model envisions that galaxies formed from the bottom up, by the growth of larger and larger clumps through collision - the hierarchy model.
The results presented were somewhat of a surprise, but the sample size is still very small, and lots of details have to be filled in.
In any case, I think the Sky and Telescope report is more balanced; a snippet: "All of the too-early galaxies and clusters that astronomers are finding do not, in any case, pose a crisis for cosmology — at least not yet. There's no reason to think the reigning Cold Dark Matter model of cosmic structure formation — a foundation of today's cosmology — is far out of whack. Galaxies and everything else made of normal matter are only a small part of the picture, and the brightest galaxies, the ones that show up over great distances, are surely only the tip of the iceberg. Most of the mass of the universe is in the form of unknown, nonbaryonic dark matter (matter not composed of atoms), which outweighs all normal matter by five or six times. What's apparently missing is a proper grasp of how gas gets funneled along the streams and clumps of dark matter into the pools that become galaxies, and how and when large numbers of the brightest stars begin to shine within the biggest of these. In today's universe, normal matter and dark matter are closely tied to each other — where there's one there's the other, at least on large scales — but early in cosmic history this probably wasn't so."
http://skyandtelescope.com/news/article_1149_1.asp
The nature of 'dark matter' continues to be frustratingly elusive. There's been some good research, and the big picture is self-consistent. However, without some real 'dark matter' in our hands to prod and poke, its relationship to the Standard Model, SST/SMT, LQG, MOND, or anything else is indirect. Who knows, maybe there's a subtle tautology lurking in all those equations?
