Re: Semantic: update or remove?

[Stefan Kangas]

<tomas@tuxteam.de> writes:

> I think the problem is that "trust" is far more complex than just
> a one-dimensional Boolean value. So "X trusts LetsEncrypt" without
> any context is practically a vacuous statement.
>
> LetsEncrypt basically checks that some site has control over its
> http endpoint, as resolved by DNS (alternatively, that this site
> has some control over its DNS entry). No less, no more.
>
> Whether this is really, truly WonderWidgits PTY (according to whom?
> Some state institution? The Chamber of Commerce? The WTO? King Charles
> III?)) or not is outside LetsEncrypt's scope.
>
> Caveat TLS Client :-)

You are right of course, but it's important to note that "trust" is
actually quite well-defined in this context.

It means that the relevant software and systems include Let's Encrypt's
root (or an intermediary) in their trust stores, and thus accept
certificates they issue without warnings or errors.