From: Alan Barrett <apb@iafrica.com>
Date: Tue, 18 Feb 1997 08:27:34 -0800 (PST)
To: "Timothy C. May" <tcmay@got.net>
Subject: Re: Anyone have the complete info on CP list alternatives?
In-Reply-To: <v03007800af2f876176ce@[207.167.93.63]>
Message-ID: <Pine.NEB.3.95.970218181653.12473O-100000@apb.iafrica.com>
MIME-Version: 1.0
Content-Type: text/plain


> >Why is it obvious that mail propagation will be faster than news
> >propagation?  News propagation times of small numbers of seconds are not
> >at all uncommon.
> 
> News propagation in seconds to your _local server_, or to _distant_
> servers?

To some (perhaps many, certainly not all) distant servers, provided all
the news servers in the path run appropriate software (designed for low
latency propagation) and have reasonably high bandwidth Internet links. 
Disk to disk delays smaller than 1 second have been measured between news
servers that use software such as nntplink or innfeed to send outgoing
articles immediately.

> The Usenet is thousands of news servers, maybe tens of thousands, and
> news feeds take a while...small articles are mixed in with hundreds of
> megabytes a day of binaries. A percolation process, as opposed to a
> point-to-point process for e-mail. 

Right.  And some of those news links are fast, while others are slow.  If
two sites happen to be connected by fast news links, it's quite possible
for news between those sites to be faster than mail.

--apb (Alan Barrett)