No Net Neutraily? Been there, done that, it sucks.

Net Neutrality is a subject of major discussion amongst the neterati. For good reason. The presence of Net Neutrality guarantees that any network connection — residence, business, cable, DSL, satellite — can send traffic without restriction or different levels of dependability based on protoocol or destination. The absence of NN means exactly the opposite; there may be restrictions based on protocol, destination, or anything else the service provider might decide should be singled out.

And, of course, one of the non-neutral models involves the customer paying extra money to gurantee levels of service for certain protocols or features.

The funny thing is that none of this is new. We have been in a non-neutral world before. And it sucked. More on that in a second.

Why is this such a big deal?

– residential DSL and cable modem companies have oversold their bandwidth such that they cannot meet the contract or marketing promises they have made

– those same service providers are pissed off that commodity generic bandwidth is perfectly excellent for carrying products like Vonage or Skype [telephony] or any of the myrida of multimedia streaming or downloading solutions. They want a cut of that.

But none of this is a new issue. We have been here before. Slightly different form, but same impact on the customer.

Full details after the break…


Back in the mid ’90s, mom and pop ISPs — small regional internet service providers that had an office with at least a T1 and a bank of modems — were the way you connected to the net. I helped run one for a while until it became abundantly clear that the phone companies would crush us as soon as they decided their was neough money in the market (they had the advantage in that they could eliminate two local loops — expensive — by simply providing ISP services out of their central office).

Before that happened, the Internet hit a crisis point as the number of relatively small ISPs exploded across the country. Specifically, all of those ISPs required the IP address space to be subdivided and routed across more and more subnetworks. Unfortunately, the routing tables grew so large that the MAE and CIX central routers could not longer effectively subdivide the network.

The end result was what we called The Great Peering War. That is, you had to make an agreement with the major IP providers — the likes of Sprint, WilTel, and others that owned what was effectively the net backbone — if you wanted your traffic to pass across the internet unfettered. And, of course, the smaller your company the more you had to pay, beg, or promise to get such an agreement.

As well, there was infighting amongst the major providers. Occasionally, one of them would throw a fit and turn off peering with another. If your company had an agreement with, say, Sprint and WilTel got miffed at Sprint and effectively turned them off, your customers wouldn’t be able to see any of the hosts that could only be found by going through WilTel’s network.

Of course, all your customers knew was that they could no longer get to whatever site they wanted to go to and since you — the small ISP — were “their internet”, it must be your fault.

It sucked.

Even in a much simpler ‘net — HTTP, SMTP, FTP, POP, NNTP, and some emerging multimedia protocols were about it — it meant that you never knew what sites or services you could get to. You could not rely upon the net to actually be available.

If we go back to this kind of environment now, it will be worse — much much worse. There are orders of magnitude more sites, services, and protocols that are all integrated into our average desktop computing experience. This isn’t just a case of not being able to watch, say, The Daily Show or not being able to get to some random BBS or Forum.

Worse — and just like it was in 1994 — you won’t be able to pay extra $$ per month to get “the internet without restriction”. Peering and service provisioning is still a complex web of connections and it would be almost impossible to restrict the quality of service without impacting non-customers. Sure, you could place the restrictions at the border routers — at the first line between the ISP and the customer — but that increases the costs of providing premium, high bandwidth, services limited to your network while restricting the same services outside of your particular cloud.

And it requires more equipment spread across more sites. Quite likely more expensive on installation and certainly more expensive to maintain all that equipment.

End result? Restrictions at the peering boundaries, thus reducing the self-healing nature of the internet (part of the country goes dark? Route around it. Not any more, though) and reducing the quality of service of “pass through” traffic.

The customer experience will be awful, worse than in 1994. Apps will mysteriously not work for them, but work fine for everyone they know. Some of their favorite sites may be available one day, but not the next. And it will vastly increase the cost of customer support far beyond ISPs. Any application that makes a network connection will run the risk that said application will not work in certain areas or when the customer obtains internet through a changing set of ISPs.

No thanks. Give me a neutral net. I have tried the alternative and do not want to go through that hell again. Charge everyone a few extra $$/month or drop the bandwidth guarantees to something that can be met.

Just don’t break the damned ‘net.



7 Responses to “No Net Neutraily? Been there, done that, it sucks.”

  1. Phil says:

    Net neutrality is important; there’s no doubt in my mind that its the right place to be. What concerns me, though, are the implications of legislating it. Regardless of the intentions, ceding that authority to the government will give them more, by implication, and I bet we’ll see just what the legislature thinks that is hidden in and attached to any bill with a “net neutrality” moniker.

  2. Joe says:

    Nice points.

    When it comes to network neutrality, it’s even worse in the wireless world.

    Here’s an excellent paper by Tim Wu who’s a law professor at Columbia:
    http://www.newamerica.net/files/WorkingPaper17_WirelessNetNeutrality_Wu.pdf

    – Joe

  3. GadgetGadget.info - Gadgets on the web » No Net Neutraily? Been there, done that, it sucks. says:

    […] charliemarks wrote an interesting post today!.Here’s a quick excerptNet Neutrality is a subject of major discussion amongst the neterati. For good reason. The presence of Net Neutrality guarantees that any network connection — residence, business, cable, DSL, satellite — can send traffic without … […]

  4. Bob Frank says:

    Yes, but this is how it will be marketed: http://i7.tinypic.com/5z6vt4n.jpg

  5. Imp says:

    Nice article! I wish I could agree. In some cases Network Neutrality does seem nothing more than a question of provisioning, but in others it genuinely has to do with keeping users safe.

    Universities filter known malware packets. ISPs drop known phishing email. This is also not new: in the early 1990s universities with full netnews feeds dropped a usenet group devoted to child pr0n. These seem like reasonable — good, even — things. I suspect that statistics on dropped spam and botnet traffic would show that if we moved to a purely neutral Internet it would auger into the ground. This suggests an equally reasonable stance might be don’t move to NN; don’t break the ‘net.

    Yet it is a very quick hop to filtering out merely undesirable content. Personally, I want all of my spam and will sift through it myself. If I fall for a phishing attack that’s my own damn fault. But I would hate to see that standard applied to my mother. Especially when I get the phone call to rebuild her system. And there’s not much I can do for her when she has her social security number, birth date, etc. out in the wild.

    I am not arguing with any particular point you raised, but rather with the notion this is something simple. Rather than dropping into pro- and anti-NN camps, I would rather see discussion about how to get the advantages of NN without also incurring unacceptable costs, a world where ISPs and telcos are gatekeepers with the power to decide who speaks and listens. If the geeks can’t figure it out, the wonks sure won’t.

  6. HOTI Dave says:

    Interesting post — definitely a different take than the usual. You’ve nailed one key point — the Internet is not a one big network, it’s a network of networks. And because of that, the conundrum of net neutrality is that while the *policy* has always been one of neutrality, the fact of *infrastructure* has been much different.

    Now, I work on net neutrality for the Hands Off coalition in Washington, which opposes new laws in this area, and does have some ISPs in our membership. From this perspective and the experience I’ve picked up, it’s overwhelmingly clear to me that there is no appetite for the kinds of wars you describe. The lone case of blocking that supporters of new NN regulations can name, is when Madison River blocked Vonage briefly in 2005. The FCC stepped in, told them to knock it off, and they did. What everyone realizes today that they didn’t really get a decade ago is that if you have the pipes, you want all the best content on it. And if you have the content, you want it on all the pipes.

    The point of tiering isn’t to play blocking games with content owners or ISPs, it’s to offer even faster speeds for content, and see to it that those who are using more bandwidth are paying their way — and not say, being subsidized by your grandma who only ever checks her e-mail.

  7. bbum’s weblog-o-mat » Blog Archive » AT&T Blocks 4chan? says:

    […] (I remember the last time the big providers went against net neutrality in the early 90s. It sucked.) […]

Leave a Reply

Line and paragraph breaks automatic.
XHTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>