On 11/10/18 10:23 AM, Ted Lemon wrote:
> I think the issue with XMPP is that it's not actually a good solution
> for the problem we're trying to solve with it. There are way too many
> moving parts, and the moving parts don't actually add value. Whenever
> I've tried to climb the XMPP learning curve, I've fairly quickly
> realized that I don't have time to do it as a yak shave, which is what
> it always is, and so I give up and use something else. Obstacles
> abound: the base protocol doesn't actually do what we need,
By "we" do you mean the IETF? If so, an I-D defining requirements for an
IETF chat service would be welcome.
One of the reasons we use Jabber today is that we need to archive the
discussions ("Note Well") and we can do that server-side with standard
XMPP servers. It's not at clear to me how we'd do that with, say, Slack.
It could be that Meetecho can serve all our needs, but we need to define
them first.
> so we need
> extra add-ons, and then there needs to be a server,
Well, XMPP is a client-server technology, so yes.
> because federation
> is mandatory,
False.
> and so on. Better tools wouldn't hurt, but they'd be a
> bit of a band-aid over the basic problem that it's not scoped to address
> the right use case.
See above. The I-D submission window is open. :-)
Peter
