On Mon, Apr 30, 2001 at 07:05:43PM -0500, Brandon wrote:
> Yes, but I already have such an e-mail client, which is why I'm asking how
> this is useful.
firstly, your email client is just one example, not everyone might wish
to use it (some might desire a think-cash based solution), and email is
just one example. Another would be a client which supported submission
to an in-freenet key index. Conventionally, such a client couldn't
terminate before it had found the end of the stack which might take
several minutes - this would be intolerable to many users and thus this
kind of solution would be essential.
> But the index API doesn't necessarily need to be non-blocking as the
> Python e-mail daemon could be run in the background and connected to by a
> short-lived client.
Wouldn't it be better to have just one daemon, which runs as part of the
node, rather than creating a new daemon for every service you want to
offer?
> In the case of an e-mail daemon, this approach doesn't make all that much
> sense. If you were going to send e-mail via a short-lived program such as
> a command line client then it would be good to get rid of the daemon and
> just have the client and the node. However, the preferred way to send
> e-mail is via an SMTP daemon so that it can integrate with current mail
> readers. This must run as a daemon anyway because you don't want to have
> to start it up every time you want to send e-mail. Since it's running as a
> daemon you don't gain anything with this new API.
Personally I think that email over Freenet without an anti-flooding
mechanism such as think cash, is extremly risky. We have already seen
submission mechanisms being flooded (see Snarfoo), and this situation
would only get worse. As far as I can see, it would be difficult to
write a think cash client which used smtp.
> So basically this proposal is useful when you *don't* want to run a
> daemon, and e-mail is a bad example since you do. So can you give an
> example of an application for which you wouldn't run a daemon but would
> use this functionality?
Many. Key submission. Weblogs on Freenet. Think-cash email systems.
And 101 things I haven't thought of. And given that this system has
other uses, it could do most of the work currently done by your email
daemon (and thus benefit from improvements such as DBRs and algorithm
improvements automatically).
Ian.
PGP signature
