Thoughts on the Google/Verizon net neutrality proposal…

I’ve had a few days to sit and think about the proposal that Google and Verizon have published, and I think it is time to write some of it down. If nothing else, in the future it will provide me perspective in future years, as August 12th rolls around each year, this post will circulate back to the top and I can see how prescient I was.

First of all, if you haven’t read it, you probably should. It’s only a couple of pages, so surf on over and give it a read. I’ll still be here when you get back.

Verizon-Google Legislative Framework Proposal

Okay, did it leave you with an uneasy feeling? It did me. Most of the people I respect seem to think that I should feel uneasy, but a select few seem to think that I’m overreacting. I am not sure which is right, but I’ll let you know what my specific qualms are, and then follow it with a few general comments.

First of all, the section on Consumer Protections qualifies that broadband Internet service providers should not prevent users from accessing “lawful content of their choice” or “running lawful applications and using lawful services of their choice”. IANAL, but I’m a little troubled by this concept of “lawful content” or “lawful applications”. Lots of people are whining about this provision, presumably because it prevents them from downloading copyrighted material on the Internet. I’m a bit concerned by this, because it gives the ISP a convenient excuse for nearly any modification of internet access: “we are trying to prevent trafficking of copyrighted materials”.

The section on Network Management is similarly a bit strange. I agree that ISPs need to be able to efficiently run their networks in an efficient manner, but I am a bit concerned at the idea that “any technically sound practice” can be used, without any guidance as to the underlying principles that must be followed. A technically sound practice might include simply disconnecting the top 1% of the bandwidth users to mitigate connection problems. It might even be agreed to by the majority of service providers, but it isn’t necessarily the right choice for consumers. I think it is entirely reasonable to allow ISPs to limit the total bandwidth or capacity, or to price their service according to bandwidth used, but not to pick and choose which services or which users are “managed” out side of those basic constraints.

But the section that really disturbs me is the “differentiated services”. I wonder really what those differentiated services are, and they were very coy in their proposal. Some people might think that delivering VOIP over the internet is a “differentiated service”. Or running an FTP server. Or downloading Netflix. Applications to your mobile phone. All of these things were “distinguishable in scope and purpose from broadband Internet” at some point, but now are part and parcel of what people use the Internet for every day. I cannot help but think that this is an attempt for ISPs to carve out special exceptions for neutrality on new services so that they have a preferred track to sell such services and prioritize them over the more “traditional” Internet services. This is bad for competition, and bad for consumers.

I’m also troubled by the desire for an exemption for wireless networks. There is simply no rationale given, merely the assertion that they have “unique technical and operational characteristics”, so the requirement of transparency would be required for operation over the wireless networks. I can imagine that the cellular phone networks are very happy with this provision, since they already run some of the least consumer-friendly and most lucrative networks available, and this provision is essentially carte-blanche to permit or deny any service on their networks that they like. They would have the power to be the kingmakers of the wireless internet, purely by allowing or denying companies access to consumers.

Okay, those are pretty specifically the parts that annoy me. In the end, we give up network neutrality on wireless networks, we grant the ISPs the power to differentiate their networks by selling differentiated products and we have to deal with consumer agencies who are asked to fill a law enforcement role (or who can at least hide behind it as an excuse for nearly arbitrary changes in service). We also agree to allow the industry players to decide what practices can be used to solve their networking problems. And in return consumers get….? Greater transparency, okay. And the ability to ask the FCC to investigate on a case by case basis any abuses, but without any additional rulemaking power.

I’m underwhelmed. I think you should be too.

Google is partnering with Verizon on this proposal, and I think we as consumers should be skeptical of this. After all, the wireless industry engages in all sorts of activities which are terrible for consumers, and which are antithetical to the principles that allowed the vibrant Internet growth of the last decade or more. Wireless companies uniformly try to lock consumers into long term contracts, with high penalties for early cancellation, and often include excessive activation fees for account changes. Consumers often have no protection against accidental overages, or must pay (again) excessive charges to prevent overages. Consumers pay high fees for SMS messaging, despite the nearly neglible costs associated with sending SMS because of the incredibly low amount of information sent. Consumers have to deal with phones which are locked by vendors to prevent a consumer from leaving a particular network and going to a competitor. Consumers cannot get any information about the delivered call quality, or the percentage of dropped calls.

The Verizon-Google proposal talks about the “increased competition” in the wireless arena, but even though there are probably three or four carriers that might reasonably be used in an area, often the effective number is only two or even one major carrier. When you combine this with the long contracts and high penalties for switching, this presumed “competition” simply vanishes.

Consider AT&T’s most recent filing with the SEC. As part of that filing, they had to list the risk associated with the possibility of losing the relatively lucrative exclusivity agreement with Apple and the iPhone. They downplayed this risk in a number of ways, including statements like:

As these exclusivity arrangements end, we expect to continue to offer such handsets (based on historical industry practice), and we believe our service plan offerings will help to retain our customers by providing incentives not to move to a new carrier. As noted above, more than 80 percent of our postpaid subscribers are on Family Talk® Plans and business plans that would involve moving the whole group to a new carrier. Moreover, the vast majority of our postpaid subscribers (including Family Talk® Plan users) are allowed to accumulate unused minutes (known as rollover minutes), a feature that is currently not offered by other major post-paid carriers in the United States, and users would lose these minutes if they switched carriers. As is common in the industry, most of our phones are designed to work only with our wireless technology, requiring customers who desire to move to a new carrier with a different technology to purchase a new device. In addition, many of our handsets would not work or would lose some functionality if they were used on another carrier’s network (even a carrier using GSM technology), requiring the customer to acquire another handset.

The cynic in me reads this as “the majority of our customers won’t switch because our industry has made it painful and costly to switch”.

Okay, so what is my answer? I think that we need clear separation between carriers and service providers. Carriers should not be allowed to sell services or differentiate traffic with respect to source or application. They should be required to be transparent. They should be permitted to sell Internet bandwidth in ways which allow customers to pay for the bandwidth and latency that they need. Consumers should have portability of devices, and be protected against absurd overages and upgrade fees. And the future Googles and Facebooks of the world shouldn’t be stifled by Verizon and AT&T.

Okay, back to work.

HOWTO make a simple HTTP server…

It’s occasionally useful to have a basic HTTP running to serve the files in a directory. You might want to fetch some MP3 or movie files from one machine, or in my case, a PDF document that I’m hacking on my Linux box while ssh’ed in from my Windows laptop. In the past, I have used Jef Poskanzer’s mini-httpd, but there is an even simpler way, especially if you already have Python installed.

Just “cd towhatever directory”, and then type “python -m SimpleHTTPServer”. Voila! An HTTP server running, bound to port 8000 on that machine.

Very useful.