obliterate functionality?

[Giorgos Keramidas]

On Wed, 19 Mar 2008 18:54:55 -0700 (PDT), cowwoc <cowwoc at bbs.darktech.org> wrote:
> Dirkjan Ochtman wrote:
>> Have you seen Linus' talk on Distributed Version Control Systems?
>> While he is not always nice about it, he has some interesting things
>> to say on why it is good if all repository mirrors are created equal.
>
> Let's be clear here: repository mirrors are repository clients are two
> different things. For one thing, mirrors to be equal to one another in
> permission whereas clients tend to have less control.

I don't really `get' the distinction you are trying to make about
`mirrors' vs. `clients'.  In the world of Mercurial there is no `client'
concept.  Anyone who can `clone' a tree is a self-sufficient, fully
independent `mirror' of the original tree.

> One would expect mirrors with read/write access to fall under the
> guise of the same trusted organization, or else how do you prevent
> someone from checking in random junk into the FreeBSD repository using
> one of the mirrors? Point being, if all mirrors are equal, then they
> should respect each other's obliterate commands.

Mirrors are usually one-way in the FreeBSD world.  They can `read' but
they cannot (and don't really need to) `write' to the main CVS tree.
This means that we are trying to solve the wrong problem, if we spend
too much time worrying about `read-write mirrors'.

-- Giorgos