Gus Gutoski
Gus Gutoski
On hold pending agreement on a proper design pattern for error-handling.
In that case, it seems there is no problem. I'm a bit surprised it works now with only one `.as_ref()`. I wonder whether somebody quietly slipped in a fix without...
> * I have no objection of using this unsafe, since it's very limited scope, and easy to reason about. As long as we add some tests to constantly watching...
Jeb and I briefly discussed enforcing a library-wide `anyhow`-only policy. Anyhow is far more ergonomic when it's used everywhere. Otherwise there's a constant need for boilerplate code to translate among...
True. On the upside, we easily save space for polynomial commitments in the size-critical `Common` struct that is broadcast to all storage nodes. 🙂
Change of plans. # tldr; We will *not* take a `erasure_code_rate` arg. Instead we'll make only a minor tweak. # New proposal Offer two constructors: - `new` simple/generic: any VID...
The goal is to make something analogous to ``` fn VidScheme::is_consistent(commit: &Self::Commit, common: &Self::Common) -> VidResult ```
Fast generation of safe primes would be great for RSA applications.
Obsoleted by #1499
This issue is so low-priority that we'll probably never get to it. Also, we pay a price in complexity to save a few bytes.