I'm not sure this discussion captures all the nuances of the debate, but it's a good introduction, even if it is written from the point of view of e-learning in large enterprises. Still. One thing it says is this: "Online content puts the largest demands on servers and pipelines. Once that content is pushed out to local management via localized content servers, two problems are solved. Content no longer flows from a single central server through restricting pipelines, and control of content is where it should be, at the local level." This bothers me. Not because what it says is false. But if your e-learning content is plugging up your servers and pipelines, then you need to reconsider your content (or your pipelines). Would the web work if, before using it, we had to download all the content to a local machine? Of course not. Sure, there is caching on the web, but there's no real sense of 'local' content management. So where's the flaw? It's this, I think: in the presumption that the content is all coming from a centralized server. That's what creates the load. But why would you do that? It would be like accessing the web from some huge 'web central,' which would be ridiculous. That's my problem with this article: both options offered are flawed. That's what's wrong with contemporary e-learning: those are the two option we're being given by all the vendors. Why is this so hard for people to understand? Am I the only person in the world who sees this? We have a functioning model of a massively distributed information system that works - the web - and yet vendors and consumers of e-learning continue to propose and implement centralized architectures. It boggles the mind. It just boggles the mind.
Today: 2 Total: 1097 [Direct link] [Share]
