4 ms·
Stream of consciousness post here. For your first bullet, people should be aware of mt_rand()'s superiority to rand(). For your fourth bullet, people should be
by carbocation 17y ago
Stream of consciousness post here. For your first bullet, people should be aware of mt_rand()'s superiority to rand(). For your fourth bullet, people should be aware of hmac. (Am suggesting that awareness of these should be a necessity, though not sufficient for some purposes.)
Edit: Just read your powerpoint - it's a wonderful overview, with brief code samples. People getting started with this broad topic should definitely give it a read. And now I'm sad I missed your live talk a few weeks ago.
- mattyb 17y agoFor the first bullet, I usually just do the following: dd if=/dev/urandom ibs=1000 count=1 2>/dev/null | tr -dc '0-9' | fold -w[number of digits you need] For the fourth, I agree with wbond: use bcrypt when possible. Also, slide 27 should probably say 'fingerprint' in the second section.