Miro, hopefully you received the email with the zip file. It looks like your ISP is having some problems so you may be off-net.
Anyway, I am hoping you can clarify for me the association of the transition cache with a specific transition. Is it by state+target signature or is there some other combination of variables?
I fully appreciate that scripting languages are slower and not of particular interest for embedded systems, but the answer will help me better understand what is being optimized out in the call sequence. I'm still fairly new to the idea of hsm's, so any and all clarification is appreciated!
Once clarified, I will likely implement it in ruby so I can solidify the concepts for myself (and anyone else who cares to examine the code for its educational value). When my copy of the second edition arrives from amazon, I fully intend to rev the code to comply with the latest research you have completed in this space. That code will also be released to you for hosting on the contributors page.
BTW, will QF be released in java like earlier versions or is it C/C++ from here on out?
Thanks for an excellent system, and thanks for keeping it open source!