Any and all subjects re volunteer computing are interesting to me. My personal efforts and projects focus on access to justice.

I'm interested in attending.

Yep egg on face ... fooled again

New BOINC Client released 7.20.0 Windows 64-bit
Release notes at https://boinc.berkeley.edu/wiki/Release_Notes

Download from standard BOINC download page.

Error code: SSL_ERROR_RX_RECORD_TOO_LONG

That's the error message that appears when I try to attach a new computer to the project, and also when I try to report it by logging in to the message board. Hoping somebody will see this and let them know.

He isn't presenting that. That blog is just commenting on the inspiration anyway.
I don't see any comments by him, or any of the Rosetta staff on the present status, let alone future direction, of Rosetta@Home.

Not that the project is distinct from the software they are using in-house.

However, I do have long years of experience. Things changed drastically last year, due to the advent of the AI/machine learning techniques.
They do most of their stuff in-house, and send only the 4.20 Rosettas that they receive from outside sources (Robetta I think) out to the crunchers.
The other work is the Pythons, which are from a single researcher. They are the ones that really need the support, and aren't getting it.
I think we can assume that it is not a widely-supported project within Rosetta now, and probably won't be.

The other thing that AV programs can do is check their company's database of known programs and their reputation. New programs don't yet have a reputation, and haven't been though a full evaluation in the AV company's labs. So they can panic: the fuss usually dies down in a few days.

Checking that ca-bundle.crt is still serviceable is on the Release Manager's checklist for new versions (I put it there), and there's supposed to be a new version to coincide with the release of Windows 11 on October 5.

When they make it out of testing CPDN OpenIFS tasks will use open source code.

Numberfields has also published their code:
https://numberfields.asu.edu/NumberFields/forum_thread.php?id=488&postid=3063#3063

Had to look ESXi up to see what it was.

Is this something to do with the virtualisation? I know that benchmarks under WINE are I think about a quarter of what might be expected though you can get an accurate benchmark there by telling BOINC to use only one CPU, then manually running benchmarks from the manager after which you can let it use however many cpus it wants. I will have to look at my VB to see if it suffers the same issue.

performance will always be (much) slower in a VM than when ran natively.