Koranteng Ofosu-Amaah straps you into his intellectual rollercoaster with his encyclical on the Low End Theory of Networks. It’s a summary of all the forces tearing the telecom industry apart, but I must warn you it’s not for beginners: zero to 80mph in two seconds. It took me about 5 years to get to be able to understand all what he’s writing about, and a whole train journey to digest the content, so don’t panic if it takes you a week to follow all the links and get the rest of the context sorted in your mind.
Here’s what I regard as the crux, and it’s right up at the front:
Reed, Saltzer and Clark described the End-to-End principle quite simply in terms of systems design. The heart of the matter is this oft-overlooked sentenceFunctions placed at low levels of a system may [Martin’s empasis] be redundant or of little value when compared with the cost of providing them at that low level.They knew when they were writing that this notion had wider applicability than the telecommunication networks that were their initial focus hence they labeled their work “end-to-end arguments in system design”.
One little word. “May”. It’s a world of pain and opportunity. We don’t really know how broad the applicability of the end-to-end principle above really is. For example, consider peer-to-peer (P2P) file distribution in a mesh network. It might be a good idea for intermediate nodes to cache fragments for re-transmission as routes decay and change due the constantly changing network topology. That would mean them being aware of the nature of the bits — “file transfer”.
I came across an already well-known example earlier today of the limits of the end-to-end principle and Internet Protocol. It kind of assumes that transmission is free, which it clearly isn’t. So services like email filtering can in principle be done at the network edge, but by then the user has already paid the cost of reception. The damage is done before your spam filter gets a look-in.
Another quotable quote: (well, the least I can do is to return the favour…)
Those advocating what they call “network neutrality” are simply being cute. They are implying, for quite pragmatic and rhetorical reasons (read effectiveness of lobbying), that there are no costs to neutrality; that neutrality is value neutral. It’s a nice trick as far as framing a debate goes, but it is a trick nevertheless and it should be discounted accordingly.
I’d very much say the same thing. The idea of “user-driven network architecture” is probably closer to what most neutrality advocates really want. Any prioritisation or tiering in a competitive (or user-owned) network environment is being done because that’s what users demand, not because it lets a duopoly provider price discriminate away the space under the demand curve.
PS - No politics on Telepocalypse, but as a bit of fun, imagine Rumsfeld’s taxonomy of knowledge being remixed, and you can have one of the following four opinions on the Iraq expedition: “An honourable success”, “a dishonourable success”, “an honourable catastrophe”, and “a dishonourable catastrophe”. Not saying which one I think (if nothing else because I can’t make my mind up…)
Posted by Martin Geddes at 10:21 PMTrackBack URL for this entry:
http://www.telepocalypse.net/cgi-sys/cgiwrap/mgeddes/MT/mt-tb.cgi/749.
OK. I'm a pretty smart guy, but I have no idea from the above post why network neutrality is a bad thing. I have an open mind; I'd most like for someone to explain, in simple terms, why this is so, and assume that I'm not a computer whiz.
For example: Since it looks like there won't be "network neutrality", why is this good? Is my blog or my business website going to be unavailable to some people?
Etc, etc, etc.
Posted by: at June 29, 2006 01:53 AM