The Internet most of us experience is not the World of Ends suggested by the end-to-end system design concepts around which the Net was originally architected and built.

Instead we have something that is faster-than-dialup, and faster-than-it-used-to-be; but is not The Net. Instead it is the part of the Net that's left in a pipe that's optimized for television, for one-way few-to-many "content delivery" and for locking users into client roles, while servers labor somewhere else.

I just had FTTH (fiber to the home) installed. And, while it's way cool in some ways, it's also uncool in the way it prevents far more business than it generates for itself. It would be great if the carriers made it easy for businesses to grow on the Net, and then suppoted those businesses with services that helped those businesses thrive and grow. But the carriers would rather serve "content" to mass quantities of "consumers" while chaging prohibitive prices for "business-class" services. Hey, it's a mass media mentality, and they have every right to it.

But...

With the rise in network-intensive gaming, of bittorrent traffic, of the need to share big files (e.g. photos and videos), heavy users of the Net will inevitably chafe loudly at legacy asymmetries in the Net subset provided by carriers, at least here in the U.S.

So some of us (mostly on email list backchannels) have been thinking about how we need to rejigger this thing somehow. Either we work with the carriers, or we work around them.

I favor the latter, mostly because the flywheels in carrier methods and mentalities are too large and biassed to spin forever right where they are. They'll move eventually; but it will take competition to do it. (Let's leave "net neutrality" and other legislative options out of this.) That's what I'm suggesting here.

Among the workarounds I've long fantasized about is what some of us call "fenceline" build-out, from house to house and apartment to apartment, finding their way to aggregated buying power sufficient to pay for connections along poles to backbones, where the cost of connectivity could be split enough ways to become affordable to everybody.

I don't think that will happen without some business selling the service that does the connecting and the support after the connecting is done. That business would provide pure Net connectivity, rather than the customary crippled sort (e.g. port 80 and 25 blockages) we get from the carriers. it would provide that connectivity over cabling (preferably fiber), wireless (something more sturdy than wi-fi) or both. Let's call that business a 'netco' — something that works as an alternative to the telco and the cableco.

To succeed the netco (or the netco industry and its companions) would need to characterize what they provide as pure internet, and what the cablecos and telcos provide as a crippled sort. You know how the anti-abortion people characterized their cause as "right to life"? Changed the whole game. (Not saying that move was right or wrong, just that it was effective. It was galvanaizing.)

Anyway, I'm thinking out loud here about this because the Linux movement grew to a large degree around Web server deployment. As the Net grew, Linux grew. And vice versa. In fact, the Net and Linux grew to the point where, frankly, we won.

That leaves us at a long Now What? moment. Too long, in fact. Linux search queries are down. Apache market share is falling. We need a fresh issue to drive the Linux conversation into a new old cause: enabling true freedom for connected users. Is this that cause?

Here are some questions...

1) Can we see a new business — a netco — be built to work around carriers that insist on providing only crippled Internet? (How? When? By whom?)

2) What would be some clever characterizations of pure uncrippled Internet?

3) Who would be on this new business's side at the backbone level? Surely not the telcos and cablecos, but how about Level 3 and NLR? How about Google?

There are models for such a company, by the way. One is Indienet.dk in Denmark, which I wrote about in the January 2007 issue of Linux Journal. Denmark is different, yes; but it's good to know a model exists.

Meanwhile, think about what will happen when ordinary customers — those consumers who are now also producers — get fed up with assymetry and Net crippling here in North America. It will happen, and when it does there will be a lot of business in filling demand for uncrippled connectivity.

Or so it seems to me. How about you?

Suggested reading: Everything by Bob Frankston.