With all the protocol comparisons, people have overlooked the obvious. Ask 
yourself a serious question, how many web developers (or even small app 
developers) actually have a clue about SMTP, NNTP or even (though slowly 
growing) XMPP? Compare that to HTTP. A developer's pet cat probably knows all 
about HTTP about now ;). Pubsubhubbub has the benefit of being so simple and 
obvious, that implementing it is exceptionally easier - irrespective of the 
technical advantages/disadvantages of other alternatives.

 Pádraic Brady

http://blog.astrumfutura.com
http://www.survivethedeepend.com
OpenID Europe Foundation Irish Representative





________________________________
From: Thomas Lord <[email protected]>
To: [email protected]
Sent: Fri, February 19, 2010 9:28:39 AM
Subject: Re: [pubsubhubbub] educate me

On Fri, 2010-02-19 at 03:52 -0500, Bob Wyman wrote:
> If you're looking for protocols that should be considered as
> alternatives to PSHB, you'd probably do better to look at NNTP
> (Network News Transfer Protocol) rather than SMTP. However, it is
> important to realize that in most applications, even though NNTP based
> systems appear to be "pushing" data, this is only appearance. What is
> really going on in most cases is that down-stream servers are polling
> up-stream servers for data.


The historically validated way to combined NNTP's polling with
push is to bridge NNTP to, guess what, SMTP.  You know, like
getting digests of an NNTP feed from Google groups, though
you could make those digests machine readable if you have
anti-polling religion.

> Additionally, it is important to understand that the NNTP
> protocol is not particularly well suited to handle the case of
> hundreds of millions of "groups." Much of NNTP consists of 
> commands to list and get metadata about the supported groups
> or articles in groups. That stuff just isn't reasonable to provide
> when you've got hundreds of millions of items in the database... 

Aren't you overlooking "wildmat"?   I am not saying that
wildmat is a complete solution to the "hundreds of millions of
'groups'" problem, only that it is a long way there.  Why not
incrementally improve it rather than blowing it up and starting
from scratch?

>  A better protocol than NNTP would be XMPP.

That is the kind of statement that really sends chills
down my spine because of the layering errors implicit in
it.  It is a bit like someone saying "A better protocol than
FTP would be TCP" - to which one can only say, in essence, 
wtf is this person trying to say?


> PSHB isn't a protocol casually designed 

Nobody has argued otherwise.

Regards,
-t

Reply via email to