Background: The L1TF vulnerability, also called Foreshadow received a microcode patch, but it came with some new terms that were incompatible with many Linux distributions (such as banning...
Background:
The L1TF vulnerability, also called Foreshadow received a microcode patch, but it came with some new terms that were incompatible with many Linux distributions (such as banning benchmarking, and saying downloading the file to read the terms constituted accepting the terms). Debian put it on hold, Gentoo decided not to release on mirrors, etc.
One doesn't accidentally add a clause to a license that prohibits publishing benchmarks. They tried to sweep it under the rug, it failed, so they chickened out.
One doesn't accidentally add a clause to a license that prohibits publishing benchmarks. They tried to sweep it under the rug, it failed, so they chickened out.
The only way to view the license, was to download the microcode update... Which said that downloading it made the license binding (which would void it under most contract law, I believe, but I'm...
The only way to view the license, was to download the microcode update... Which said that downloading it made the license binding (which would void it under most contract law, I believe, but I'm not entirely familiar with US law).
So, no, they didn't announce it. They just added a couple new sentences and seemed to have hoped that no one would make a fuss - and they were nearly successful. Red Hat, SUSE and others just accepted it, as no problem.
Debian and Gentoo on the other hand had a few issues. (Gentoo restricted it so the license would appear to users, and the update wouldn't appear on any mirrors. Debian just asked Intel to clarify before they would proceed at all.)
It seems like something that would rather be on HN than on The Verge, which is my rule of thumb for that sort of stuff, especially with linking to a Gentoo bug tracker and not an explanatory write-up.
It seems like something that would rather be on HN than on The Verge, which is my rule of thumb for that sort of stuff, especially with linking to a Gentoo bug tracker and not an explanatory write-up.
Background:
The L1TF vulnerability, also called Foreshadow received a microcode patch, but it came with some new terms that were incompatible with many Linux distributions (such as banning benchmarking, and saying downloading the file to read the terms constituted accepting the terms). Debian put it on hold, Gentoo decided not to release on mirrors, etc.
Intel seems to have reversed that decision.
Also discussed here on tildes
Was there an announcement about the change? Was it just a mistake or are they trying to push their luck and backing out due to the negative reactions?
One doesn't accidentally add a clause to a license that prohibits publishing benchmarks. They tried to sweep it under the rug, it failed, so they chickened out.
The only way to view the license, was to download the microcode update... Which said that downloading it made the license binding (which would void it under most contract law, I believe, but I'm not entirely familiar with US law).
So, no, they didn't announce it. They just added a couple new sentences and seemed to have hoped that no one would make a fuss - and they were nearly successful. Red Hat, SUSE and others just accepted it, as no problem.
Debian and Gentoo on the other hand had a few issues. (Gentoo restricted it so the license would appear to users, and the update wouldn't appear on any mirrors. Debian just asked Intel to clarify before they would proceed at all.)
Shouldn't this be in ~comp?
Last one was in tech. That was my only motivation.
~comp is for deep tech; ~tech is for not-explicitly-technical people. Licensing falls within the latter, in my opinion.
It seems like something that would rather be on HN than on The Verge, which is my rule of thumb for that sort of stuff, especially with linking to a Gentoo bug tracker and not an explanatory write-up.