[kbenson]

That person is very well known in this community, and in other communities as well.

They are also known for making very specific arguments that people misinterpret and fight over, but the actual intent and literal meaning of the statements is most often correct (IMO).

Whether this is a byproduct of trying to be exacting in the language used that tends to cause people interpretive problems or a specific tactic to expose those that are a combination of careless with their reading and willing to make assumptions rather than ask questions is unknown to me, but that doesn't change how it tends to play out, from my perspective.

In this case, I'll throw you a bone and restate his position as I understand it.

NIST ran the competition in question in a way such that all the judges referred each other, and all are very well known in the cryptographic field, and the suggestion by someone with more common game that they could be bribes in this manner (note not that the NSA would not attempt it, but the implication they would succeed with the people in question) is extremely unlikely, and that DJB would suggest as much knowing his fame may matter to people more than the facts of who these people are, is problematic.

[tptacek]

I'm not sure I'd use the same words, but yeah, the argument I'm refusing to dignify is that NSA could have been successful at bribing a member of one of the PQC teams. Like, what is that bribed person going to do? Look at the teams; they're ridiculously big. It doesn't even make sense. Again: part of my dismissiveness comes from how clear it is that Bernstein is counting on his cheering section not knowing any of this, even though it's a couple of Google searches away.

[throwaway654329]

One trivial example implied by the blog post: Such corruption could be involved in the non-transparent decision making process at NIST.

Regarding Dual_EC: we still lack a lot of information about how this decision was made internally at NIST. That’s a core point: transparency was promised in the wake of discovered sabotage and it hasn’t arrived.

[tptacek]

What do you mean, "how" the decision about Dual EC was made? It's an NSA-designed backdoor. NIST standardized it because NSA told them to. I'm sure NSA told NIST a story about why it was important to standardize it. The Kremlinology isn't interesting: it is NSA's chartered job to break cryptography, and nobody should ever trust them; the only thing NSA can do to improve cryptography is to literally publish secret attacks, and they're not going to do that.

[throwaway654329]

What do I mean? Iran-Contra, Watergate, or a 9/11 report style report, like levels of investigation. Given how widely read the BULLRUN stories were, it’s not credible to suggest the details aren’t important.

The American people deserve to know who picked up the phone or held a meeting to make this happen. Who was present, who at NIST knew what, and so on. Who internally had objections and indeed who set the policy in the first place. What whistleblower protections were in place and why didn’t the IG have involvement in public? Why did we have to learn about this from Snowden?

NSA has a dual mandate, on that I hope we can agree. It’s my understanding that part of their job is to secure things and that part of their job is to break stuff.

NIST has no such dual mandate, heads should roll at NIST. We probably agree that NSA probably won’t be accountable in any meaningful sense, but NIST must be - we are stuck with them. Not trusting them isn’t an option for anyone who files their taxes or banks or does any number of other regulated activities that require using NIST standards.

[MauranKilom]

If that is the case, then what is the explanation for NIST (according to DJB) 1. not communicating their decision process to anywhere near the degree that they vowed to, and 2. stone-walling a FOIA request on the matter?

> Whether this is a byproduct of trying to be exacting in the language used that tends to cause people interpretive problems or a specific tactic to expose those that are a combination of careless with their reading and willing to make assumptions rather than ask questions is unknown to me

Communicating badly and then acting smug when misunderstood is not cleverness (https://xkcd.com/169/).

If many people do not understand the argument being made, it doesn't matter how "exacting" the language is - the writer failed at communicating. I don't have a stake in this, but from afar this thread looks like tptacek making statements so terse as to be vague, and then going "Gotcha! That's not the right interpretation!" when somebody attempts to find some meaning in them.

In short: If standard advice is "you should ask questions to understand my point", you're doing it wrong. This isn't "HN gathers to tease wisdom out of tptacek" - it's on him to be understood by the readers (almost all of which are lurkers!). Unless he doesn't care about that, but only about shouting (what he thinks are) logically consistent statements into the void.

[tptacek]

The explanation for the FOIA process is that public bodies routinely get intransigent about FOIA requests and violate the statutes. Read upthread: I have worked with Bernstein's FOIA attorneys before. Like everyone else, I support the suit, even as I think it's deeply silly for Bernstein to equate it to Bernstein v US.

If you made me guess about why NIST denied his FOIA requests, I'd say that Bernstein probably royally pissed everyone at NIST off before he made those requests, and they denied them because they decided the requests were being made in bad faith.

But they don't get to do that, so they're going to be forced to give up the documents. I'm sure when that happens Bernstein will paint it as an enormous legal victory, but the fact is that these outcomes are absolutely routine.

When we were FOIA'ing the Police General Orders for all the suburbs of Chicago, my own municipality declined to release theirs. I'd already been working with Topic on a (much more important) FOIA case from a friend of mine, so I reached out asking for him to write a nastygram for me. The nastygram cost me money --- but he told me having him sue would not! It was literally cheaper for me to have him sue my town than to have him write a letter, because FOIA suits have fee recovery terms.

I really can't emphasize enough how much suing a public body to force compliance with FOIA is just a normal part of the process. It sucks! But it's utterly routine.

[kbenson]

> If that is the case, then what is the explanation for NIST (according to DJB) 1. not communicating their decision process to anywhere near the degree that they vowed to, and 2. stone-walling a FOIA request on the matter?

Why are you asking me, when I was clear I was just stating my interpretation of his position, and he had already replied to me with even more clarification to his position?

> Communicating badly and then acting smug when misunderstood is not cleverness

I don't disagree. My observations should not be taken as endorsement for a specific type of behavior, if that's indeed what is being done.

That said, while I may dislike how the conversation plays out, I can't ignore that very often he has an intricate and we'll thought out position that is expressed succinctly, and in the few cases where someone treats the conversation with respect and asks clarifying questions rather than makes assumptions the conversation is clear and understanding is quickly reached between most parties.

I'm hesitant to lay the blame all on one side when the other side is the one jumping to conclusions and then refusing to accept their mistake when it's pointed out.