What is the current status of Tsallis entropy in entropy modeling?

[apeiron]

Anyone up with what is happening with Tsallis entropy?

I read intriguing but opaque accounts of his ideas a few years back but heard little since. Can anyone give a simple explanation where he fits in entropy modelling at the moment and how he is being received?

 

[xlines]

Anyone up with what is happening with Tsallis entropy?

I read intriguing but opaque accounts of his ideas a few years back but heard little since. Can anyone give a simple explanation where he fits in entropy modelling at the moment and how he is being received?

Well, as far as I understand and remember, that line of thought failed. Many objections were raised, in particular q-generalisation of entropy does not satisfy Second Law for q != 1 nor can it be defined for two systems with different temperature, or something like that - I can't remember exactly, but I have special issue of EPJ dedicated to nonextensive statistics, maybe there is some discussion there ... I'll be back =)

 

[apeiron]

That might explain why things went quiet.

I could not follow Tsallis's own exposition clearly enough to even begin to see if it was flawed. But the general principle of switching from a gaussian static system perspective to a powerlaw expanding system one seems to be an important one in statistical mechanics.

It is an idea whose time may be ripe. For instance, there is the very entertaining Nassim Nicholas Taleb books on the economics of "mediocristan" and "extremistan". There is the rise of scalefree network models.

I had the impression that Tsallis (and Renyi?) were working on a scalefree model of entropy, so to speak.

Normal ideal gas type entropy models have a single scale (a single temperature, a gaussian average of microstates).

So what would be the story for entropy with microstates self-similar over all scales? A powerlaw expression of temperatures.