...

Hey, I just got Appl. Crypt. 2nd Ed. yesterday. Meanwhile with the 1st Ed. and the code I had in my CDs I could very well manage with most things, protocols, algorithms... Of course, the version of Crypto++ I have there is outdated now, but the algorithms are still valid, and I still have much more algorithms there than the current version has, and I can always plug in or adapt a new one should I need to.

Would you like to suggest some of your favorite algorithms not in the current version of Crypto++ for inclusion in the next version?

Wei Dai

Point taken. Well, as it is now, it is good enough for my current needs. And it's a great work you've done. What I was referring to instead was actually what I don't want or expect in a "good" library: simple, extraneous, old, maybe breakable algorithms, things no longer useful... they make a great material for studying and learning, and sometimes to demonstrate some colleagues how simple some approaches are. These materials, that I wouldn't use in a decent application are still usefull for me from time to time, and are nice to have at hand. Personally, I would prefer some algorithms not to be in Crypto++ less someone use them confident that if they are there, they must necessarily be good, although they might be useful for learning purposes. Nevertheless, as soon as I find something I think interesting to add, I'll let you know, or even contribute it myself. And thanks again for this fine piece of code! jr