atyy said:
How about thus 10^500 number that pops in the popular media all the time? I'm a biologist and fudge my models all the time, but even that sounds too much.
This is the number of possible string models that one can build, many of them having nothing to do with the real world.
How many possible models could you build, as a biologist, if you weren't concerned with matching data? (An ACTUAL count isn't necessary, just ballpark it :) ).
MTd2 said:
Lisi gets the SM, however, it comes with aditional particles for each of the 3 generaltions within it. That is, the Standard Model is not a subgroup of his theory, but it is subset of his E8, and those particles are called by Jacques Distler calls "spurious particles".
The chiral problem happens when one tries to find the standard model in such a representation that should give SM as we know it today, that is, without aditional particles. And the lack of chirality is not the only problem to this approach, the other one it is that in this case, one can only get on generation.
So, I really don't see any problem with Garrett's model yet, considering that these aditional particles were not yet studied properly in his model.
There are lots of problems with Lisi's model. The biggest one is chirality---the SM is chiral, and it doesn't look like Lisi can get chiral fermions out. If that is the case, then it is just wrong. I am no expert in his work, though, and will stand corrected.
Secondly, Lisi doesn't quantize gravity. Lisi gets an SO(4) out of E8, which is different from actually quantizing gravity. So Lisi's theory shows how to embed SM x GR into E8. While this is beautiful and impressive, it is different from quantizing gravity.
Also, grand unifying theories, like SU(5), contrary to popular belief are not falsified yet, since the proton decay half life was ruled out until 10^(34) years. But given that the upper threshold for proton decays in several GUTs are around 10^(38) years to 10^(42)years.
The minimal models are all dead, for various reasons. One has to dress the models up with large higgs representations, which are ugly and poorly motivated. See these papers:
http://xxx.lanl.gov/abs/hep-ph/0108104,
http://xxx.lanl.gov/abs/hep-ph/0007213
Also note that most of the people who started working on Grand Unified Theories in the 1980's now work on string theory.
Nevermind that the works with Wan's brainding might (already did, but didnt bring the picture together yet) bring Standard Model, and possibly all redundancy of string theory, to LQG. For these reason, I would rather study LQG for the sake of introducing a new mathematical theory, that is might works with the real world, as I believe it will, than sticking with a string monoculture. Indeed, the progress is slow, but it is getting faster lately, as more people want to see something different.
If you're talking about Smolin's last paper, he predicts, pretty clearly, a fourth SM generation, which is ruled out by observation.
Either way, this seems engineered. There are several ways that chiral matter and non-Abelian gauge symmetries appear in string theory, "for free". Take Type II string theories. You build a theory of open strings, and you realize that these open string endpoint have to live on hypersurfaces called D-branes, which are required for the consistency of the theory. Then you examine configurations of D branes and find that you can stack them on top of each other and get a U(N) theory out. So there you go---a non-Abelian gauge theory for free, given by things that are already appearing in the theory in the first place.
MTd2 said:
Alain Connes hasn't given up his theory. After a few weeks his theory was ruled out, he came up with something new and even will release a new book on this... Wow, without experiments, high energy physics will become literaly feudal properties.
Alain Connes predicted a higgs a 160-170 GeV, and Fermi Lab ruled a higgs out at that mass earlier this year. But, it's always possible to fix your model by adding stuff.