I think it's a bit disappointing that a 2004 draft cited by at least five articles of Professor Bengt O. Muthen is not available in 2008. The draft sounded extremely interesting; I bet it would make a terrific read.
Well, sometimes research goes in unpredictable directions - we wrote many other things instead. The essential content of that draft is contained in the technical parts of the Muthen-Asparouhov (2008) chapter in the Longitudinal Analysis handbook:
Muthén, B. & Asparouhov, T. (2008). Growth mixture modeling: Analysis with non-Gaussian random effects. In Fitzmaurice, G., Davidian, M., Verbeke, G. & Molenberghs, G. (eds.), Longitudinal Data Analysis, pp. 143-165. Boca Raton: Chapman & Hall/CRC Press.