Re: [Monotone-devel] Re: results of mercurial user survey

[Christof Petig]

Thomas Keller schrieb:
> Bruce Stephens schrieb:
>> I'd want to see some estimates before I found that convincing, though.
>> For example, we know it's going to be slower than the raw network
>> speed because we're taking and sticking data into a database in little
>> chunks.  For the initial pull, we could just copy the whole database,
>> but then that suggests that it might not be the verification as such
>> that's the problem as much as using the database in this way.
> 
> I don't think even an initial pull will be ever equal to a copy. What if
> the db already contains public key certs?
> 
> I'm for introducing a `mtn copy` command which indeed just copies one
> database to another, without any fancy security checking. This would
> also just overwrite all data in the target database and would probably
> be also useful for mirroring or "backups".

This feels _way_ too dangerous to me. Did you never overwrite a file by
mixing up source and destination in a copy? Imagine this on your
complete development history. And resyncing to a backup usually is fast
enough!

But I propose and _second_ an "initial clone" command which works
exactly like a networked "mtn db dump | mtn db load": It should require
a new database to get created in the process and should never works on
an existing one.

I don't know how a command like this fits into a
multi-simultaneous-connection server nicely! Perhaps it should work like
"send all you know about in any order" followed by a real sync.

    Christof

signature.asc
Description: OpenPGP digital signature