All things related to multicast

[Chris Rapier <>]

Leonard Giuliano wrote:
> On Thu, 13 May 2004, Bill Owens wrote:
>
> -) On Thu, May 13, 2004 at 04:24:36PM -0400, Richard Mavrogeanes wrote:
> -) > My two cents:
> -) >
> -) > Multicast is not used more often because:
> -) >
> -) > 1. The default condition of virtually all routers is to disable 
> multicast.
> -) . . .
> -)
> -) All true, and I certainly know that multicast is not an easy thing to
> -) configure, manage and repair (or even understand!) but I'm
> -) intentionally preaching the choir here. I wasn't asking why Joe Public
> -) doesn't use multicast more often, I was asking why *we* don't use it
> -) more often.
> -)
>
> I would echo earlier sentiment that just providing simple voice teleconf
> isn't worth the hassle of using mcast or even IP.  Or put another way, if
> you are sitting at your desk, will you use your cell phone or your land
> line?  

I know my comments aren't specific to this group but in a more generalized sense 
I have to say that the simple easy way isn't the way a lot of people do things. 
They take a simple method (say, writing down a phone number in an address book) 
and make it complicated (say, buying expensive hardware and learning to write in 
a new alphabet) but the perceived (and occasionally real) advantages of it make 
it intriguing enough to get the early adopters to use it - from there more 
people will buy into it until it becomes the default method. Its some ways its 
like fashion driving technological adoption. Whats good about this is that as it 
becomes more widely adopted novel uses and applications are developed which may 
or may not have been in the designers original intent.

I think multicast is a lot like that. The trick is to find that one application 
that draws in the early adopters. Once the multicast is setup and being 
maintained for one purpose the others (like conference calling) will just be a 
matter of course because the infrastructure is already there (regardless of its 
real benefits). I honestly think that application will be something like 
internet radio. A lot of people I know have at least though "man, I'd be a much 
better DJ than this person". If the application existed they might be more 
willing to put their money where their mouth is. The difficult part of this 
application isn't the DJ stuff though. Its building a 'radio' application that 
doesn't suck and is appealing to the kids (especially the collegiate linux kids) 
(and before anyone says "Vic and Vat already exist" - they aren't good enough - 
protocol maybe, interface no way). If the kids adopt this in college and start 
to really want it the pressure they exert on the market as they enter the real 
world will push more ISPs to offer multicast.

At that point things like conference calls, video streaming, and new 
applications will start to be adopted by the wider market and we'll see some 
real progress.

Just my thoughts on it.