To: Ahda who wrote (17440 ) 12/2/1999 2:32:00 PM From: Jay Lowe Read Replies (2) | Respond to of 29970
>> that may well out last any other area. Indeed. I am not presupposing that @Home offer web-based desktop service in a "closely coupled" way ... by acquisition, internal development, or licensing an existing system to run closely-bound to the @Home wires and boxes. There are lots of ways, technically and organizationally, that web-based service might get to the user via @Home. Some might vote to take the low-ground ... the backbone position. The counter-argument, however, has some interesting meat. It goes like this. @Home should closely-couple the web-based desktop service to their infrastructure because it uniquely solves QoS issues that are a major challenge to these services. In other words, @Home can uniquely enable thin-client, fat-server applications in general. This argument pulls out a key value of @Home's infrastructure ... they can offer scaleable QoS to content providers ... and web-based applications are a specific example of content providers who benefit from enhanced QoS. Ahhaha, Frank? Can you comment on the ability of @Home technology to deploy web processes closer to the user? Can @Home generically deliver distributed server capacity? Notice how web-based desktops and other apps challenge the web-page caching model ... they would tend to thrash net caches pretty well ... they want the server-side processes out as close to the user as possible. Can @Home accomodate this? If so, then the low-ground, backbone model is strengthened. Based on my own technical expertise (?) I think it's simpler in the short run to deploy web-desktop server boxes to @Home POPs as unique entities ... this implies a certain business relationship more intimate than the backbone model allows.