SHARE
A Field Guide to Textareas of the Internet

Essays on technology, psycho­analysis, philosophy, design, ideology & Slavoj Žižek
ABOUT THE AUTHOR →

Appreciation

August 26, 2012

A Field Guide to Textareas of the Internet

The textarea form field is an essential feature of many web applications, and is obviously crucial in social media and other sites that rely on user-generated content. Textareas are simple, mostly consisting of a bordered rectangle filled with white that users can click into and enter text, but for such a straightforward feature, many different implementations are possible, each one reflecting the designers’ assumptions about what users can and should be doing on the site.

The first decision is whether to use a textarea at all. A general rule of form design says that input fields should be designed in a way that reflects how they are intended to be used. For example, a field for entering a U.S. phone number should be wide enough that a complete phone number is visible, but not so long that users could be confused about what is supposed to be entered. A phone number field should be phone-number sized, reiterating its meanings and purposes at the level of its appearance.

Applying this principle to textareas, we can say that they ought to be used in forms when the designer expects or desires any free-form text input that may extend beyond a single line, such as a multi-paragraph blog comment. From the perspective of the server that processes the input, textareas are indistinguishable from single-line text input fields. A textarea is advantageous solely because of its improved usability and user experience characteristics — when writing, editing or proof-reading a lengthy comment, it’s highly desirable to be able to see what you’ve already written without having to scroll.

Ideally, users would be able to see their complete entry. From this, we may reasonably conclude that textarea design expresses the designer’s sense of the optimal length of a comment. On social media sites, this often translates into site-wide norms about the appropriate length of user-contributed content. Here are some popular social media sites, and the size of their textareas in lines and characters:

Other than YouTube, every site on this list technically allows for longer entries, sometimes over 60,000 characters, so there are no technical constraints in place that prevent these sites from being filled with long essays. Although this does happen on occasion, in general, the length of user contributions on social media sites seem to be shaped by the soft limit of the size of the textarea.

I don’t really have good data to back this up, I’m basing it on my own experience with these sites. I have found data showing that the average Reddit comment has steadily declined from over 450 in 2005 to 150 characters in 2011. This may be a broader trend. For example, in the Silicon Valley startup community, the desire to reinvent email is a minor topic that returns with persistence. The idea recently appeared in Y-Combinator founder Paul Graham’s list of ambitious, billionaire-level problems, where he says:

Email was not designed to be used the way we use it now. Email is not a messaging protocol. It’s a todo list. Or rather, my inbox is a todo list, and email is the way things get onto it. But it is a disastrously bad todo list.

The notion that “email is broken” get its support from the feeling that there is too much of it, which makes it a close relative of the movement to reduce email overload and “reclaim your inbox.” Graham points out that email has changed: originally imagined as electronic letter-writing, it has been distilled down to a medium for coordinating tasks between people. Seen this way, email is no longer (or was it ever?) a writer’s medium, a carrier of expression or thought. Yes, in the narrow sense of the word, we still write, but it is more signaling than writing — a reduction of the word to the minimum.

In this mode of communication, our concerns resemble those of signal processing: speed, efficiency, error correction, noise, protocol negotiation, latency. We don’t ask if someone has time, we ask if they have bandwidth; a person requesting an update may send reply to an email with a single word — ”Ping”; and lengthy messages become a sign of rudeness.

Some messages are compressed to where they can fit in the subject line, and the sequence EOM (end of message) appended to it, a reference to the special control character EOT that’s used in telecommunication transmissions to mark the end of a transmission.

On message boards, after writing a long message, it’s polite to add a one sentence summary and prefixing it with the sequence “tl;dr” — too long; didn’t read. Perhaps the brevity of this statement is intended as a model for the original author to follow, a sign of an emerging social norm that communication should be reduced to the minimum number of characters.

It shouldn’t come as surprise that some of the new email clients that position themselves as “solving email” aim to reduce the size of the reply window to tl;dr size. Two examples come to mind: Fluent.io and Sparrow, both essentially suggesting to users that replies need not be longer than a sentence or two. Of course, it’s possible to display a more traditional email composition window, but the default behavior attests to a design conversation that began with “In most cases, our users don’t need to write more than a few hundred characters…”

Fluent.io takes an even more radical step by reimagining the inbox as something more like a Twitter feed, and instituting an automatic tl;dr mechanism by abbreviating received messages that transgress the boundary, hiding the excess behind a “More” link. This is similar to how Facebook displays only the first 240 characters of a comment. Why only 240? Presumably there is someone at Facebook who said something like, “What could they possibly have to say that they can’t say in a few sentences?”

But this is a self-fulfilling prophecy. If the designers at Facebook believe users don’t really have anything to say, they won’t build tools that make that very easy, which will in turn discourage users from posting longer comments. Social media sites and forums often have economic reasons for this: a site-wide norm of short messages means more page views and more ad impressions, and creates the appearance of a healthy, active community that everyone wants to be part of. It also lowers the barriers for new users to join the site, because short messages don’t demand a lot of work or commitment.

For some partisans of Silicon Valley, the expectation that users should think about and value what they write online is criticized as elitist, racist and technophobic, belonging to an era where white men wrote with quills and inkwells. We’re told that the triviality of online communication signals the liberation of the people from these ancient standards, so we should find tweets reporting on today’s breakfast menu deeply significant.

This is persuasive only if you believe that the venture capitalists of Sand Hill Road are servants of the people and simply deliver to us the products that we have demanded, a view that too credulously regards the free market as a neutral delivery mechanism of social preferences. In reality, social media sites have significant ability to shape the communication of its users in ways that maximize profitability, and for the most part, this means as demanding as much participation as possible to ensure that graphs of engagement presented to venture capitalists go up and to the right.

Maybe I’m exaggerating a bit. Designers are well aware that not all participation is good participation. Aside from trolls, flamewars and spam, the sheer volume of user-generated comments can overwhelm a site and make it impossible to find anything worth reading. The short-term goal of increasing user engagement has to be balanced against a long-range goal of ensuring that visitors find worthwhile content. So in reality, the game of revenue maximization involves a trade-off between quantity and quality.

How far can you increase quantity before the attendant decrease in quality starts to hurt page views? One widely-used method of addressing this problem is building filtering and ranking mechanisms so that high-quality content is easy to access and low quality content is relatively invisible. On Facebook, heavily liked and commented status updates are more likely to show up on your news feed. Twitter’s search feature defaults to filtering by “Top,” showing highly retweeted and favorited tweets. Many social bookmarking sites like Reddit encourage users to vote on comments, which translate into points called karma accruing to the authors. Karma has two functions: first, to increase or decrease the visibility of a comment to viewers; and second, because karma is publicly visible, it also signifies the reputation or status of the author.

So the desire to increase one’s social media status is encouraged on many social media sites. Democratic-egalitarian rhetoric notwithstanding, the inability for those with low karma scores to make their voices heard is believed to be a feature of a healthy community, not a problem to be overcome.

Last year, Facebook released a commenting plugin that would allow third-party websites to restrict comments to users who were logged in to Facebook, so that their comment could be tied to their real name and Facebook identity. To address the problem of quality, comments are sorted by “Social ranking.” The algorithm that calculates this index are not public, but is described in the following way: “Comments are ordered to show users the most relevant comments from friends, friends of friends, and the most liked or active discussion threads…” Anecdotally, it seems to heavily favor users with many friends, in addition to activity that generates lots of likes and comments. In other words, your social rank is approximates the economic benefit you bring to Facebook.

Disqus, a blog commenting plugin used on this website and many others, offers spam controls and administrator and community moderation tools, and bills itself as “elevating the discussion.” This is a positive spin on the now widely-accepted view that comments are a cesspool filled with spam, trolls, flamewars, bigotry, violence and so on, an observation made famous by ex-Google CEO Eric Schmidt.

As social media has gone mainstream, some quarters of the hacker community have reacted with disgust. In their minds, they build—with the best intentions!—nice, democratic social media sites that are going to destroy the gatekeepers and empower us all to express ourselves and unlock our creativity. But some bad apples have befouled this beautiful nest that they’ve carefully designed, by exploiting the anonymity of the internet, an explanation that became widely known after being coined the Greater Internet Fuckward Theory in the web comic strip Penny Arcade.

But the hacker community has to reinvent its own history and present to blame mainstream users for the decline of online standards. Peer moderation systems to eliminate bad behavior were developed in the mid-90s, a time when the web was dominated by technology enthusiasts. One well-known implementation was on the influential “news for nerds” website Slashdot.org. And the Greater Internet Fuckward Theory was originally formulated to explain the behavior of hardcore online gamers, a group that was and is dominated by white male geeks. Recent events have shown that they continue to be among the worst offenders.

The sense that this behavior comes from outside, not from “people like us” is entirely a false one. If anything, the internet is a cesspool because its social software transmits the values and norms of the people who built it. While we should be careful to avoid lumping in every individual that identifies with this group, at the same time it cannot be denied that it does have a dark, repellent streak. The business models that they profit from only accelerate these problems.

The hacker community, unable to face its own inner demons and the problems of capitalism, has externalized them, and turned on the masses they once pretended to liberate.