Mark Aisenberg: > Any news on how well, if at all, the ActiveState > packager obfuscates?
It's a little tougher, but I'm close to breaking it. Real Work intervened. :) > Writing an obfuscator is an interesting intellectual > exercise. It's pretty easy to delete comments, > eliminate whitespace, and change local > variable names and subroutine names, of course, Not really, since this involves the well-known challenge of parsing Perl... > I'm surprised that someone hasn't released a module > to auto-obfuscate. Acme::Smirch? :) > I guess it shows that not many > people are using Perl to generate distributable > commercial applications. That's making the tacit assumption that if you're writing a distributable commercial application, you need to hide the source code. Which is a very odd assumption, if you look at it closely - you're obviously not interested in hiding the source from people who are going to abide by your license terms, and people who aren't going to abide by your license are going to break your obfuscation anyway... > Or maybe it shows that people > know that obfuscation is not good security, This, I agree with. > and that > we need a true Perl compiler instead. This suffers from the same fallacy of assumption that your previous statement did. But that won't stop Parrot from having one, because I'm more interested in speed than code hiding. :) -- A witty saying means nothing. -Voltaire
