Re: blogs.msdn.com: RSS consumes too much bandwidth

[Bjoern Hoehrmann <derhoermi@xxxxxxx>]

* Bob Wyman wrote:
>	My hope is that in the process of defining Atom, we'll be able to
>address the bandwidth issue better than it has been addressed in the past.
>There a number of things we can consider. One of them is getting people to
>use push instead of pull (ala: "Atom over XMPP"[2,3]). There must be other
>approaches we can use as well...

Would Atom clients retrieve data they are not interested in? As far as
Atom over HTTP is concerned that would be the only problem we can solve.
So what is the actual problem here?

  * Atom clients or the HTTP servers they access do not make proper use
    of the existing HTTP facilities to reduce bandwidth like e.g.,

      * multi-user caches
      * local caches along with conditional requests
      * HTTP level compression

  * Atom clients retrieve collections of entries rather than individual
    entries even though only individual entries are of interest, e.g.,

      * only the latest entry in an update scenario
      * only the entries the user intends to read

  * Users configure Atom clients so that they know about the latest
    information no later than 15 minutes after publication

  * Atom makes it easier to access major amounts of information and
    thus more users are interested in accessing this information, i.e.,
    the product is too popular and costs more than what the provider
    gains from it

  * anything else?

We can only try to reduce bandwidth *waste* and in order to do that we
would need to know where Atom clients would waste bandwidth and how
much; then we could determine how to mitigate these bandwidth threats,
but a number of such threats are better addressed elsewhere, for example
through proper business models or specialized mass-user services or
protocols.