The New York Times (registration required) reported yesterday that Verizon Wireless refused to allow an opt-in text messaging program from an abortion rights advocacy group. Verizon, to its credit, quickly reversed course, blaming the incident on an “incorrect interpretation of a dusty internal policy.” The reversal came as a relief, though not a great surprise. From the moment the story started drawing significant news coverage, it was hard to imagine that Verizon Wireless would stand by its initial decision. Abortion is a controversial topic, but the messages from Naral Pro-Choice America would have gone only to subscribers who signed up for them. And while Verizon may want to keep hate speech and spam off its network, there’s no reason to think Naral’s messages would come close to either category.

Anyway, Naral will now be allowed to run its text-messaging program. And Verizon will surely take a careful look at its policies and where they are intended to apply. Verizon Wireless reportedly had told Naral that its policy barred message programs that “promote an agenda” or “may be seen as controversial.” In contrast, the later-released Verizon statement spoke approvingly of text messaging being “harnessed by organizations and individuals communicating their diverse opinions” and being used “to communicate broadly.” CDT would like to see Verizon Wireless adopt an express policy reflecting those views, and make that policy public. Text messaging is a popular and effective means of communication, and there would be no good reason to allow its use for purely commercial interaction while sharply limiting it for other legitimate realms of discourse.

The bigger question, though, is how this incident bears on the ongoing “Internet neutrality” debate. I suspect both sides will say it supports their position.

Most obviously, neutrality proponents can cite this as an example of discrimination by a network operator. The network in this case was a mobile phone network, but it is not hard to imagine analogous scenarios involving the Internet. The incident highlights the wide latitude that the current legal framework gives carriers to decide when and how to treat certain traffic, based on the identity of the sender or the subject matter of the message. For text messaging, as for email or other Internet communications, network operators aren’t subject to common carrier requirements — so Verizon Wireless was and continues to be free to reject messaging efforts by Naral or anyone else.

Opponents of government-imposed neutrality policies can argue, however, that the Verizon Wireless incident illustrates the ability of the marketplace to control discriminatory behavior. The argument would go like this. Sure, once in a while some carrier — whether through error or by design — may take some action that discriminates against certain content. But the flap that Verizon Wireless quickly faced shows that such actions aren’t tenable. If the content is something that at least some subscribers actually want, publicity and related market pressures will force a quick reversal.

I think there are a few problems with that argument. First, some communications could be time-sensitive — think of a text message campaign right before an election or a big vote in Congress. What if a carrier were to block or delay a particular message campaign, perhaps citing some aspect of its terms of use, and then later bow to pressure and allow the messages only after the main damage has been done?

Perhaps more importantly, the Internet is different from mobile phone networks. Naral had to get permission from Verizon Wireless in order to initiate the text-messaging program to Verizon subscribers. That meant that when Naral was rejected, it could go to the New York Times with its Verizon correspondence in hand. In other words, there was a clear paper trail, as well as an entity capable of taking the story public.

Fortunately, the Internet doesn’t require any kind of analogous permission to communicate with an ISP’s subscribers — that’s a big part of why the Internet has been such a hotbed for innovation. But it also means that there may be no paper trail. A small advocacy group or company may distribute information online via its website, or via an RSS feed, or via email updates. If an ISP were to impair any of those channels of communication, there wouldn’t be any “rejection letter” for the group or company to waive around. Indeed, there might be no easy way for the group or company to find out, much less to prove that a purposeful action by the ISP was the source of the problem.

This potential lack of transparency could be even more serious on the subscriber’s end. If a particular organization’s RSS feed or email update feature isn’t working right, how would the subscriber know the cause? Many would likely assume that the fault lies with the entity offering the update feature (Naral, for example) or perhaps some software compatibility problem. If in fact the cause were an ISP decision to block or degrade that particular content, there wouldn’t necessarily be a way for the subscriber to find out and potentially complain.

As we’ve said before, CDT believes carefully crafted rules could be better than relying on each instance of discrimination making it to the New York Times. But whatever one thinks about the wisdom of enacting neutrality rules for Internet service, the need for transparency ought to be plain. To the extent that an ISP treats traffic differently based upon the content or source, clear public disclosure should be required. And disclosure ought to be specific concerning what traffic is prioritized, de-prioritized, blocked, etc. As the Verizon Wireless-Naral incident illustrates, a generic policy against agenda-driven or “controversial” content is too open ended to provide a clear sense of how it will be applied.

This entry was posted on Friday, September 28th, 2007 at 3:21 pm and is filed under CDT, Internet Neutrality. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.