'Pull' is relevant within the context of obtaining content from
sources that
you regularly receive content from, it is not a generic email replacement
protocol.
there is no reason why it cannot be a generic e-mail replacement
Implementation is relativally simple
We already have two pull protocols, NNTP and RSS. NNTP suffers from the
practical dependence on the broken peer-to-peer flood fill routing model.
Although this is not essential to the protocol one of the responses of
usenet to their spam problem was to demand that isps block external nntp
connections. If you are on comcast you cannot connect to GMANE.
not relevant
RSS is the emergent protocol, it supports richer content than NNTP, is
considerably simpler and has a significant user base even though there are
practically no clients (!)
not relevant
My skepticism here is towards the distinctions made between push and pull.
If you look at the standard email transfer it involves a push SMTP
interaction followed by a pull POP interaction.
POP is unchanged
the purpose of pull is to enable the destination host machine (not the pop
client) to accurately determine the source without the need for structural
changes to the underlying protocol
low cost and effective.
but most importantly transparent to the clients.
In NNTP you have the same
type of interaction but the partitioning is slightly different. In effect
you have the world pushing garbage at your news server and then
pull out the
small quantity that you are interested in.
not relevant. Though I will state that I first came up with the idea to
combat spam on nntp. I then realised its application to e-mail was much more
important
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg