> Even if the google servers dont provide time that
> is correct, its good enough to run testcases again.
> Products however of course shouldnt use it.
So, it seems that he wants to run out-of-the-box with some timeservers so that the developer/compile/testing machine doesn't have to specify them manually for tests. On the other hand he wants to require vendors to add a custom configuration so that they abide by the rules of, e.g. pool.ntp.org, by registering a vendor pool and configure systemd.
The only sane approach for me seems to be to discover the correct ntp-server in the scripts running the testcases, and leave the ntp-client in systemd unconfigured and untested if that's not possible.
Forcing developers to have a properly configured system is a much saner approach, because it affects way less people than potentially millions of users to whom a broken/wrong timesyncd.conf leaks caused by some snafu during configuration of the systemd timesyncd when packaging.
I love how in one sentence he accuses Poettering of being "convinced he has all the answers for everyone", and in another he states anyone who approves of systemd must be "naive or lazy or just plain clueless developers" lured by "the promise of making their lives easier".
The only sane approach for me seems to be to discover the correct ntp-server in the scripts running the testcases, and leave the ntp-client in systemd unconfigured and untested if that's not possible.
Forcing developers to have a properly configured system is a much saner approach, because it affects way less people than potentially millions of users to whom a broken/wrong timesyncd.conf leaks caused by some snafu during configuration of the systemd timesyncd when packaging.
EDIT: Typos, wording.