Hi! It is near end of the week now and I still have not worked through the transforms section nor the extends<> section - key features of the library, but only read them once. I am rather busy at present and have to perform this review at night ... Also I have had some trouble installing appropriate compilers (some ICEs with older gccs), gcc-4.3 not buildable on cygwin due to GNU MP build failure (gcc depends on MPFR depends on MP fails to configure). In addition to that I still have questions regarding the chapters I have intensively read so far, some of them will probably disappear once I start playing with the code. With regard to the complexity of proto and the programming power it reveals I would appreciate an extension of the review at least until easter tuesday (2008/03/25) or - if you can afford - until end of march. Of course I understand that this is probably too much to ask for, so if all others feel comfortable with proto already, do not hesitate to finish the review as planned. In that case I vote for "Yes", since at first glance I do not see anything that might hinder a rewrite of Daixtrose's compile time differentiation and simplify feature with proto and because AFAICS Eric addressed the most important issues of an ET library in an elegant and cute way (At present I do not agree with some design decisions, but that is rather a matter of taste). The docs - hard work! good quality! - of course need some extensions and modifications, but I first have to understand the lib before I post critique on the didactic structure of the docs. I think this is normal for a new library and Eric's aim at perfection is high enough to get things fixed even after proto is accepted. Markus Werle