Net Neutrality: To Rebrand or Not To Rebrand

Something interesting is happening on reddit that has the potential to impact the net neutrality debate. A California Congresswoman, Anna Eshoo, is running a contest on reddit to rebrand net neutrality. Here’s the video in which Congresswoman Eshoo outlines the task at hand, and here’s the primary thread where the contest is going on.

So let’s jump in. I have some preliminary thoughts. I spent some time writing about the pros and cons of using reddit for this sort of endeavor, but I worry that’s burying the lede (the writeup is tucked away at the close of this blog post). I’d like to focus instead on questioning Congresswoman Eshoo’s underlying assumption  that net neutrality is the concept that needs rebranding.

I concur with the Congresswoman’s assessment that net neutrality is a difficult topic for people to get engaged with and understand. It lacks the visceral quality of America’s most heated legislative disagreements like gun control, same-sex marriage, or the pro-life/pro-choice debate. These topics are inherently emotional for some folks with easy to understand outcomes (“consequences,” for some). They don’t need a lot of dressing up to ruffle feathers. I can understand why one might jump to the conclusion that the name “Net Neutrality” is at fault. Compared to “Death Tax,” for instance, it’s a dead fish of a term. The discourse surrounding the issue definitely needs amping up. It’s really just a question of approach.

Many of the existing proposals on reddit involve the words “equality,” “access,” or “freedom,” failing to account for some of the associations intrinsic to these terms as they exist within popular political debate. When I hear “Unaltered Universal Internet Access,” for instance, I assume that the issue is legislation impacting citizens ability to access the Internet in general. Now, obviously that’s a component of net neutrality, but it’s not precisely the point. Approaches that focus on freedom similarly conjure up notions of censorship or restriction, which are, again, related but slightly missing the mark. This isn’t to say that such terms don’t belong in the discussion, but elevating this terminology to the forefront dredges up other equally important concerns that are in actuality ancillary to the issue at hand.

“Net neutrality” is a concept that really elegantly encapsulates a specific ideal case scenario: an Internet that treats all data neutrally, regardless of origin, type, or sponsorship. The problem isn’t necessarily the term itself (though I think some redditors have already proposed some interesting alternatives). It’s that the concept is tricky to understand from a lay perspective because it is more or less the status quo. Our most effective analogs–other forms of media–have themselves been disrupted in part thanks to a neutral net, and our short term collective memory makes it challenging for some to think about what tiered access would really entail. What needs to be made more abundantly clear, then, is this alternative. In other words, Congresswoman Eshoo’s efforts should be focused not on re-articulating the concept we need to support, but on codifying the notion of what we ought to attack.

This requires, in part, attending less to outcomes for users and more to the actions of content and service providers. Users will suffer if net neutrality is abolished, of course, but they’re not the ones most actively involved. “Data discrimination,” or the creation of Internet fast lanes, occurs at the behest of ISPs. Those creating the content of the Internet, be they web sites/services like Netflix, Facebook, or Google or advertisers like Coca-Cola, Disney, or Ford, then face the choice of paying for premium treatment or suffering lags or even exclusion by default. Those who aren’t major players risk becoming lost in the shuffle. In short, if ISPs begin employing selective data handling, they stand to profit immensely with all other parties losing out in the long run, especially those content providers who are able to pay for premium services. While such organizations may reap short-term gains by having their sites and services prioritized, they set a nasty precedent  and pave the way for a business model where they become a customer with limited bargaining power or say. The idea of a “fast lane” doesn’t sufficiently capture this; it sounds like a modification to an interstate highway designed to more efficiently handle traffic, as opposed to a profit-driven system of information regulation and potential repression.

I frame the alternative this way with full recognition that the people who stand to lose out most are users, of course. Depending on how “fast lanes” get rolled out, data sponsorship could be executed with varying degrees of transparency, resulting in users choosing to routinely access certain sites without recognizing the economic framework undergirding their decisions. Such a system has phenomenal potential implications for political engineering and manipulation of popular culture (a discussion best reserved, perhaps, for another time). It’s pure speculation to assume that dire consequences are a necessary effect, but thinking about this as a form of censorship provides another interesting framing approach. Unlike governmentally-administered political censorship, it’s economic censorship, where a previously “egalitarian”* system is transformed into a pay-to-play model.

(Separately, here’s a brief and purely semantic argument. Neutrality, in and of itself, is not necessarily compelling. It becomes a compelling (or repelling) term purely through juxtaposition. Defining the alternative to a neutral standpoint is critical to positioning it as positive or negative; whether it constitutes inaction when action is needed, for instance, or else equates to fair treatment in the face of inequality is subject to how non-neutrality is depicted. From a verbal standpoint, we’ve got our work cut out for us.)

TL; DR: Net neutrality doesn’t need rebranding. The opposition does. The current reality of a neutral net doesn’t inspire fear or alarm, but the alternative sure might. And when it comes to impacting discourse and generating grassroots “viral” spread, the more emotion you can involve (particularly high-arousal emotions like surprise or outrage) the better.** I used the words “data discrimination” earlier; these appear several times throughout the thread and are, in my opinion, a step in the right direction. Through making clear exactly what’s at stake when ISPs start treating some data differently, hopefully sound minds will prevail.

—Footnotes
*This is also subject to debate; see Matthew Hindman’s “The Myth of Digital Democracy” and the growing body of research on search engine gatekeeping for further reference. I actually don’t buy for a second that the Internet is an entirely equal playing field, but it is more equal in its present state than in a tiered alternative.

**See Jonah Berger’s body of research on word of mouth and “virality.”

–reddit Pros and Cons–

reddit is a social news site, comprised of individual posts, links to outside content, and commentary by millions of users. The site operates using an upvoting and downvoting system; each user account is able to distribute one upvote or one downvote to any comment or discussion. In theory, this system crowdsources content evaluation and elevates the best content fairly democratically.

There are problems with assuming that this is what’s organically occurring. As with virtually any process, the system can be gamed; users can create multiple accounts or write bots that amplify their upvoting and downvoting power. More importantly, however, the actual architecture of the site isn’t necessarily ideally suited for the evaluation of content en mass in a neutral, agnostic way.

Most redditors are using free accounts. When they open new discussions, the top 200 comments are the only ones to automatically load.* In threads where there are, say, thousands of comments, users have to manually load additional commentary, presuming they care enough to read that long. There are also a number of sorting options that users can employ; the default is the “best” algorithm, which takes into account not only comment score but also the number of people who’ve interacted with the comment, displaying relevant information with higher fidelity. Users can also opt to sort by the most controversial comments, the comments with the highest or lowest raw scores, or simply the newest or oldest comments. Each of these has pros and cons, but suffice to say that each provides a different prism through which the limited amount of comments a user sees is presented.

There are many cues beyond the raw content itself that can influence judgements of whether something is bad or good. These judgements, called heuristics, allow people to assess information more quickly and can impact perception and opinion quite a bit. One of the more well-documented cues and heuristics that may be at play within reddit is comment score, producing what researchers from Hebrew University call “Social Influence Bias.”  Basically, comments with high scores are more likely to aggregate further high scores; given that the default sorting system uses comment scores to determine visibility, this results in a rich-get-richer effect for a select number of already-endorsed comments and the consequent obscuring of new comments from view. We can attribute this to any number of factors. Endorsement, for instance, may play a role; we give some credence to the opinions of our friends and community, and may subconsciously conclude that a high number of upvotes is an indicator that a comment is “good.” This is just one potential factor impacting information assessment; verbal presentation, argument articulation, recency of posting, overall site reputation, and dozens of other factors may also play a role. The point very broadly is that our opinions are influenced by a number of things and that our judgements do not occur in a vacuum.

Congresswoman Eshoo’s post is in “contest mode,” meaning that comments are randomized, scores are hidden, and replies to parent comments are collapsed by default. This eliminates many of the concerns stated above. There’s a more neutral approach taken to comment exposure by virtue of randomization. Social influence bias is eliminated through this randomization and the elimination of social cues. So, in theory, the environment is ideally suited for unfettered idea review.

The potential consequence of this is the sacrifice of what researchers call “ecological validity.” Basically, it’s all fine and good to derive experimental results in a laboratory setting, but you need to be sure that you haven’t done so at the expense of mimicking reality. I think this randomization model is probably good for contests in general, but maybe not for something oriented around swaying public opinion. In so far as our opinions are influenced by our peers, it may be useful to assess idea popularity in a space that puts these opinions on display. Seeing which catchphrases gain social traction on reddit may have been a potential indicator of their virality potential in other settings.

—Footnotes
*They can also opt to show 500 comments (the number of comments displayed increases for users with paid accounts).

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s