Thou shalt have no other gods before the ANSI C standard 1391

Patrick Scheible

That's too generic for me. Let's talk about real-world probabilities. Is there anything else that is at all likely to fail, for real-world platforms, on DJB's list (other than 32-bit int)? If so, how frequent are they? How critical are the security threats that might arise due to them? Can't it often be reasonable to shift the burden to the programmers doing the porting to wacky platforms to deal with these issues?

You seem to have a sweeping view: "All buttumptions are wrong." But I'm not convinced. I want to hear detailed arguments about each of the items. For instance, why should I go to all sorts of contortions to write code that works with a 36-bit CPU, if I'm pretty sure my code will never run on such a CPU? My suspicion is that some of DJB's suggestions may well be pretty reasonable. I think you're tarring them all with the same brush, and that is not persuasive.

Huh? That is non-responsive. I'm talking about what we can buttume when reading and writing code. If you don't like that 64-bit platforms chose to use a 32-bit int, take it out on the manufacturer -- but at least be honest enough to admit to code reviewers and programmers that this is the reality (even if you don't like the reality).