The speakers have been asked to discuss major advances expected in the coming decade. For galaxy formation, I will go on record with three promising predictions and one slightly fascetious warning.
The coming years will see the
unification of galaxy formation and evolution.
Until very recently, galaxy formation was studied at z > 2.5 and
galaxy evolution was studied at z < 1 and the period 1 <
z < 2.5 was
referred to as the "redshift desert". But technological advances in
NIR imaging and spectroscopy have made the rest-frame Balmer/4000Å
break and nearby emission lines available for study in distant galaxies.
Development of these "needle-in-a-haystack" techniques now allows
us to successfully find evolved galaxies at z > 2 even though
these objects may be rare at those epochs. Hence we are beginning to
study objects
at z ~ 3 that formed at z > 6 which may turn out to be
much easier than observing z > 6 galaxies directly.
Imaging with the Spitzer satellite is enabling the first studies of
rest-frame near-infrared emission from z > 2 galaxies,
breaking degeneracies between age and dust.
Deep imaging and slitless spectroscopy with the GALEX
satellite are revealing the analogs of Lyman break galaxies
at low redshift
(Burgarella
et al. 2005).
These combined studies may make it possible
to piece together a rough evolutionary sequence,
e.g. DLA 
LAE
LBG
SMG
DRG,
that would form
part of a grand unified model of high-redshift galaxies and AGN.
We will be able to study the interstellar medium in emission at high-redshift. ALMA will enable studies of molecular gas in young galaxies through high-order CO lines. The [CII] 158 micron line, which dominates the cooling of the Cold Neutral Medium phase at both low and high redshift, should be detectable for galaxies with large gas mass or rapid cooling equilibrating the heating due to starbursts. The current set of Early Universe Molecular Emission Line Galaxies consist mostly of quasars and are reviewed (and assigned the questionable TLA "EMG") by Solomon & vanden Bout (2005). Both CO and [CII] have now been detected in z > 6 SDSS quasars, where they provide the best direct probes of the quasar host galaxies (Bertoldi et al. 2003, Walter et al. 2004, Maiolino et al. 2005). Detecting these lines and the sub-millimeter dust continuum from protogalaxies with ALMA will allow us to probe a multivariate mass function of gas mass, molecular mass, dust mass, and stellar mass. Even ALMA sensitivity may only allow detections of the tip of the gas-mass function, but this will provide a complementary set of objects to the tip of the rest-frame-UV and rest-frame-optical luminosity functions currently studied at high redshift, and much can be learned from the intersection and union of these samples.
High-redshift galaxies will be used to constrain dark energy properties. It has recently been shown (Seo & Eisenstein 2003, Linder 2003, Blake & Glazebrook 2003) that the scale of baryon acoustic oscillations provides a "standard rod" that can be measured in the clustering of high-redshift galaxies. The measurement will constrain the dark energy equation-of-state as a function of redshift, w(z), via its influence on the expansion history of the universe. The measurement can be performed at any redshift where the line-of-sight starting at z = 0 is sufficiently influenced by the dark energy, making z = 1 and z = 3 equally acceptable. Of order a million redshifts are needed, and the most likely surveys to accomplish this ambitious goal appear to be HETDEX using the VIRUS instrument under construction for HET and the wide-field multi-fiber spectrograph KAOS proposed for Gemini.
The rapidly increasing sophistication of studies of the high-redshift universe will generate even more jargon. We are already debating proper nomenclature for special categories of DLAs at lower column density (sub-DLAs) and those found in gamma-ray burst afterglows (burst-DLAs or bDLAs). Four-letter object acronyms (FLOAs?) are going to be part of the future.
Acknowledgments.
In terms of organization, comraderie, talks, and facilities,
the 2005 Bash Symposium was a
5
event, which is
inconsistent with gaussian random initial conditions, thereby proving
"intelligent design" by the organizing committees.
I thank the organizers for inviting me to
speak on my favorite topic and the editors for their hard work
assembling this volume. I acknowledge valuable conversations with
Pieter van Dokkum, Priya Natarajan, Jason Tumlinson and Meg Urry
while outlining this talk. I thank the MUSYC Collaboration for allowing
me to show results in preparation.
This material is based upon work supported by the National Science
Foundation under Grant. No. AST-0201667,
an NSF Astronomy and Astrophysics Postdoctoral
Fellowship (AAPF) awarded to E.G.