Thanks for the insight. I had considered posting some example collisions - there are examples on the net. Ron Rivest predicted collisions for MD4 many years before it was released, which was 23 years ago and the creation of collisions for fixed-length function outputs have been known for many hundreds of years.
Yes - there are groups who assert that MD4/5 are broken, with the availability of modern high-speed computers. MD4/5 are still heavily used as the digest in many systems and protocols so I guess we live with what's out there in the real world. We have to be realistic about what we want to change and what's already in circulation
What you say about finding MD4 collisions is interesting. When you say they "can" be found in less than one second do you mean "it has been shown to be possible to do it in less than 1 second" or "you will find a collision in less than 1 second". I didn't know that, if it's the former. Could you post some example code that would do it? - and I'll add that factoid in to the video. It'd certainly make it more interesting showing a demo spewing out MD4 collisions at 60 per minute! That'd be incredible!
"Amateurs shouldn't be writing about cryptography" - I guess you can please all of the people some of the time, or some of the people all of the time - but you can't please all of the people all of the time
Thanks for the expert insightful observations on MD4 collisions.