Hi George [and All],
This morning Sandy, your most invaluable front-liner, posted the following article from EETimes in the Lounge Section of your board. It concerns the delivery of movies over the Internet:
eet.com
What follows, below, is a slightly edited set of my own observations and comments that I posted in a reply message to her, as they relate to one of the oft-unwritten constraints to such deliveries that may, and probably will, ensue. Here I'm talking about the last mile service providers' bandwidth burden looking the "other" way. That is, towards the Internet, as opposed to the area that receives most of the focus in these types of discussions, which is, of course, the local loop looking towards the end user.
I'd appreciate your views on this, along with the views of others here.
-----------------
Hi Sandy,
An interesting article, thanks. It elicits thoughts beyond the obvious obstacles cited. Yes, bandwidth is an issue, and a major one, at that. Also, server farms need to be capitalized and built, and program providers need to bolster the list of titles they offer.
In discussions such as this one it's often the reader's first thought to consider the term "bandwidth" as it relates to access line speeds available to end users.
The author makes this clear in his comparisons of download times required across the spectrum of popular access platforms (he cites DSL and those of slower-speed alternatives, for example), although he doesn't get into the very-high speed variants such as the latest VDSL speeds and FTTH capabilities, capable of tens, if not hundreds of mb/s, and in some cases gigabit speeds.
But even with the very-high-speed alternatives that are now finding their way to the market place, there are other potential gotchas to consider.
Consider the local service provider. In its infancy movie delivery is merely a nuisance drain on upstream bandwidth costs (to and from the public Internet), because downloads are still few and far between. As the art form takes shape, however, and end users become more inclined to make Internet movie delivery their preferred means of title acquisition, the bottle neck will then shift away from access and appear in the transport and uplink spaces, between the local service provider and the Internet proper.
What incentives do local service providers have to ensure that their upstream pipes will handle the onslaught of new demand? In order to make a profit it almost suggests that they will have to be included in the distribution of profits that follow. But this is not their purview, especially if they do not actually provide the servers, or the content or other programming constructs necessary, while providing only local residential access as a horizontal player.
The above speaks to the need for infrastructure stakeholders in the home delivery space to take measures to either partake in a vertical model as opposed to one that is merely access related, or modify the means by which they charge for their services.
As speeds to the home increase, followed by increased traffic loads from upper-layer service providers, the local provider can do one of several things.
1.They can do nothing at all, or only marginally increase their bandwidth to the larger cloud, and allow best-effort to resolve itself, no matter what level of user satisfaction is achieved.
2. Increase bandwidth in the upstream and charge users by traffic type and volumes of traffic consumed.
3. Partner with / or create their own infrastructures and business models to do the home delivery from their own servers or those of their partners.
Does anyone see another alternative that will allow the local service provider the ability to recover their increased costs for bandwidth in support of a mass market Internet-based video delivery model?
FAC
frank@fttx.org |
