Hmmm...given the participants, it would seem they are going to prematurely claim a victory for a Cosmological Constant, without acknowledging the correct answer, which is that DE does not exist.

A bold assertion, Kea. I admire your strength of conviction, but, am skeptical. I think the case for dark energy is strong, if not compelling. If DE is tossed out of the 'equation of state' [so to speak], cosmology has a ponderable mass of a mountain to climb. We live in interesting times.

The web of physics is woven with very long threads so it's hard to reject one part of it without setting yourself up for a multi-year crusade. Eventually you'll end up tilting at wind generators.

Varying-speed-of-light cosmologies predicted the apparent acceleration before it was ever observed. It also predicted precisely the result just seen. I suggest people start considering all the alternatives before they decide to pass judgement. Come on, does anybody seriously believe that a fudge factor in GR has any explanatory power? It's unrelativistic. Einstein must be turning in his grave.

I think a lot of folks are missing the point of the article. For the purposes of the study they're describing, it doesn't much matter what the reason for the acceleration is, they're just parameterizing the behavior of "dark energy". If dark energy is really just a manifestation of modified gravity, then the result is still relevant. They're likely working with this parameterization:

[tex]P=w(z)\rho[/tex]

In other words, above is the equation of state (or effective equation of state) of the "fluid" plugged into Einstein's equation. To first order, it can be expanded as

[tex]w(z)=w(z=0)+\frac{dw}{dz} z[/tex]

For a cosmological constant, w=-1 and dw/dz=0. It appears that their measurements are consistent with a cosmological constant, though at least three things should be kept in mind:

1) The result doesn't rule out any modified gravity theory that mimics a cosmological constant.
2) The error bars on dw/dz are probably quite large.
3) Any supernova measurement is potentially subject to a large number of systematic errors.

Nonetheless, it is an interesting result. Whatever the dark energy turns out to be, it seems to behaving much like a cosmological constant. A great deal of money is currently being spent to check this statement to ever higher precision.

SpaceTiger is on the right track. 'Dark Energy' is actually a catch-all phrase that describes a host of models including modified gravity. There are plenty of 'dark energy' models that actually do not contain any new exotic energy. It's just a somewhat misleading collective name.

For more detailed info on how the equation of state parametrization includes the generic set of models see http://arxiv.org/abs/astro-ph/0305286" [Broken].

For instance you could phrase a varying c theory in terms of the Dark Energy equation of state w(z). It seems odd but all it says is that such a theory is degenerate with the existence of dark energy with that equation of state model.

This degeneracy can be broken by non-SN methods such as growth measures, so the theories can be still be distinguished at some level. The fact that growth measure are consistent with Cosmological constant (or dark energy with similar properties) and cold dark matter universe but not so consistent with the varying c theories is the reason the physical dark energy picture is preferred (though it's far from a closed case).

Did these varying [itex]c[/itex] analyses you refer to properly take into account adjusted luminosities, or do they assume the same standard candles? Which theories have been so tested?

Kea, your insights are sharp, as usual. My main objection to variable c is it fails to explain cosmological redshift without wrecking almost every other independent distance indicator model.

The basic premise is, if I understand correctly, that c increases over time - thus implying the permittivity and permissivity of empty space has 'relaxed' since the big bang. Would not 'alpha' also vary? If all the fundamental constants of nature conspire to create such an illusion, this conjecture cannot be falsified. That does not mean it is wrong, merely irrelevant.

I'm not quite sure I understand what you mean. If SN1A are not standard candles then we'll have to simply throw them out as being useful cosmological probes entirely.

I'm not sure what you mean by 'adjusted luminosities'? The apparent luminosity you expected an (assumed standard) SN1A to have at a given redshift depends on the cosmological model, with dark energy parametrized by the equation of state w. As I said, a varying c theory will be degenerate with a given dark energy model (i.e. with the equation of state matched to achieve this degeneracy), in terms of the expected luminosity-redshift relation from SN1A observations.

To the question of which theories have been tested then, ALL varying c theories are constrained through the current constraints of the dark energy equation of state, since they can be mapped to degenerate dark energy models. I don't know the specifics of which c models exist, what there equivalent equation of state is and therefore which are ruled out and which are allowed.

Remember the the parameter(s) w(z) are a mathematical convenience. If dark energy is a real exotic energy then it refers to the physical equation of state that it has. In other 'dark energy' theories, which by that I mean 'theories that explain the SN1A results' then the parameter(s) w(z) don't have the same physical meaning, but the effect of the theory, be in modified gravity, varying speed of light etc, can be represented with these parameters.

I strongly encourage you to read the Linder & Jenkins paper I linked in my previous post. Not all of it, just the first 2 sections (less than 2 pages). That should explain in detail what I mean.

That all being said, varying speed of light theories fail to explain the observed cosmic structure, so there is a good reason they are disfavored. Plus the limits of variations to the fine structure constant pretty tightly constrain any variation in the speed of light.

The theory I have in mind has an effective Barrett-Crane spin foam description with the correct equation of state, which follows from the Kashaev volume conjecture. But in the more mathematical version of this M-theory (via twistor Strings) the Riofrio cosmology should work, because it is based on the correct Machian principles. Of course there are still standard candles...that is the point, after all...but the real luminosity depends on the cosmology. I somehow doubt the Riofrio cosmology has been properly checked by professionals. It should be, because she gets a fit to the WMAP power spectrum, whereas the Standard scenario doesn't do too well on this count.

Oh, and by the way, c is slowing down, not speeding up.

Regards the standard candles, note that SN1a aren't all just assumed to be the same - the original luminosity is more accurately estimated from a model which depends on the colour of the supernovae.

This announcement is interesting because, by looking to deeper redshifts, it should be able to eliminate most of the likely systematic errors (eg. possible grey dust just making distant candles look dull) from the SN1a data that initially produced the acceleration discovery.

A readable (and authorative) explanation, by Brian Schmidt, is "Measuring global curvature and cosmic acceleration with supernovae", Class. Quantum Grav., 19, 3487 (2002).