I'm very appreciative of your notes, thank you. I will use google translate to refer to them when I sit down to do this properly. I looked in Sommerfeld and from memory found Lanczos gives a better presentation of the same material, but I'll check it again.
Regarding fluid mechanics, my main concern is that one apparently it's been proven that can't derive Navier-Stokes or any viscous fluid dynamics from an action principle (i.e. friction, a velocity dependent potential), so at best we can do chapter 1 of Landau via action principles + Noether's theorem and no more... However this book
www.amazon.com/Hamilton-Type-Principle-Fluid-Dynamics-Magnetohydrodynamics/dp/3211249648/[/URL]
claims to do it, and apparently explains the flaw in the old approach, page 16 - 17:
[url]http://books.google.ie/books?id=ONGsaXO1VToC&lpg=PA18&ots=3Le8YeDRaS&dq=Angel%20Fierros%20Palacios&pg=PA16#v=onepage&q=Angel%20Fierros%20Palacios&f=false[/url]
While it might be cranky, the only review I can find
[url]http://onlinelibrary.wiley.com/doi/10.1002/andp.200610224/abstract[/url]
gives the book a bad review but doesn't even mention this important issue, it gives out about the book for absolutely ridiculous reasons tbh so it's not a credible source.
The other approach is differential forms, which apparently can derive Navier-Stokes nice enough, just have to find a nice presentation.
Anybody have any thoughts on these, it's be great to read them.