Fighting the Terrible Internet Comment

The Internet is a platform with minimal content regulation. This is a mixed blessing - the medium presents near-limitless creative opportunities, as well as the opportunity to squander all of them. It seems like for every bit of novel content made on the Internet, there are ten bits of content that are useless or even toxic.

what no go away all I wanted was a cooking recipe

Finding quality material among the noisy posts of the Internet is a problem that some people might take for granted. It is difficult to imagine a version of popular websites without some degree of "noise" in content. One might argue that the useless contributions are a necessary evil of the open platform.

I would argue that this is not the case. In fact, I assert that stronger top-down regulation could benefit the platform. Today, I want to explore three online website groups - YouTube comments, Reddit, and SomethingAwful - and how their management styles have dictated the quality of their content creation.


YouTube's main features may be its videos, but the diversity of its user base can be best highlighted by its YouTube comments. It is estimated that there are an average of 179 comments to each YouTube video. The YouTube comment system allows for immediate audience feedback - theoretically a very valuable asset.

Unfortunately, YouTube comments have a certain stigma to them - they're completely terrible. They often have nothing to offer that adds to the video content, they can be irrelevant to the topic at hand, and they can be overwhelmingly unproductive.

Stolen from XKCD.

The YouTube comment system has a number of built-in problems that exacerbate this issue. There is a 500 character limit on each contribution, limiting the amount of content per comment. There is a low barrier of entry in order to post a comment to a video, permitting thoughtless commentary. Since the YouTube platform accommodates a wide demographic of viewers, comments could be contributed from children or from individuals with woefully irrelevant life experiences. Since popular videos can rapidly earn hundreds of comments, it becomes difficult to sift through and find the quality posts among the noise.

YouTube's measures for comment regulation are fairly limited. There is a flag system, wherein readers can flag a comment as spam, or vote to 'dislike' a comment. Either action - if done with enough frequency - hides the flagged comment from readers' views. Future readers can opt to view the original flagged comment or keep it hidden. This system is an effective signal to other people about the quality of an upcoming post, but it does not actually offer any incentives to stop making bad content. Spammers are not observably punished beyond the hiding of their comments, and nothing is stopping them from making another YouTube account even if they were punished.

YouTube also offers a 'like' vote for comments to indicate the quality of a post. The most liked comments are also featured on top of the comments list, so that there is easy access to the agreed-upon "good" comments. There is also the option to view comment 'threads', wherein the reader can follow a link to see the message that a newer message was in response to. These tools can help readers navigate YouTube comments to find the good content. However, this system can't necessarily pinpoint the best comments if the rate of comment contribution is going too quickly. Furthermore, likes and response links do not actually do anything to reduce the number of noise posts.

YouTube is not the only platform dealing with this issue. Other websites on the Internet have similar problems with their comments sections. A fundamental problem with comments sections are that they are spaces made to supplement a website's actual content. There is no such thing as a YouTube comment "community", so there is no real sense of reward for peer regulation. Top-down regulation focuses more on hiding bad comments instead of encouraging good comments, and the bad comments don't end up getting very well-hidden.

Let us turn to a more sophisticated community layout, as seen on Reddit.

Reddit is a nice transition point in layout between a comments section and traditional forums, which place emphasis on forum users participating in discourse with one another. Unlike YouTube, there are fewer restrictions on post content, there is such a thing as a Reddit community. This becomes important when we consider the website's high emphasis on peer regulation.

User regulation on Reddit uses an upvote/downvote system, which can be thought of similarly to YouTube likes and dislikes. Highly upvoted contributions rise to the top of the feed while downvoted contributions sink to the bottom. Truly toxic posters get pushed out of sight by consensus, while valuable posts are featured as prominently as possible. This is a more thorough system than YouTube's, which only highlight a few select quality posts.

Top-down regulation on Reddit is also more sophisticated than typical moderation practices in YouTube comments. Reddit's site layout is organized into subreddits, which are each assigned to a particular topic of discussion. Topicality is enforced by moderators, and the only content that is consistently deleted from the website are off-topic posts and explicitly illegal posts. Problem posters are put on what is known as a hellban - they can continue to post on Reddit, but their messages cannot be seen by other people on Reddit. A hellbanned poster can continue to post for an indefinite amount of time before realizing that they are completely invisible.

The results of this system are not particularly compelling. One study has asserted that the majority of Reddit is comprised of lo-fi content, mostly consisting of memes and simple images. Since Reddit is compartmentalized in ways where subreddits can have differing criteria for topical posts, some subreddits do end up demanding higher quality contributions than others. Of course, on the other side of the bell curve from the higher quality content, you've got issues with racism, sexism, and pedophilia. Since creating new accounts on Reddit is about as simple as creating a new YouTube account, even hellbanned posters can find their way out of their punishment.

Stolen from ModernPrimate. So how's it feel to know your cat pictures share a website with this?

Reddit allows a lot of noisy content (and some very reprehensible content) to stay in circulation, despite the upvote/downvote system. If the poor content is centered around a theme, then someone can make a subreddit where the poor content is tolerated. One can argue that this is a phenomenon that extends beyond Reddit, and that if Reddit were to put harder blocks on bad content, then the bad content would simply emerge in another corner of the Internet. One might fear that an online space with too many regulations against bad content - even noise content - could not flourish, exactly for that reason.

To explore a hard counter to that fear, we can turn to SomethingAwful.

The SomethingAwful forums, having been around for over ten years, have had time to refine their rules and their community. Prominently featured in its rule set is the following quote:

Before you post: Before posting, please ask yourself the following question: "Am I making a post which is either funny, informative, or interesting on any level?"
If you can answer "yes" to this, then please post. If you cannot, then refrain from posting. If you post anyway, the mods will probably gas your thread, automatically awarding you a 15-minute probation. ~SA Rules

You must pay $10 in order to make an account on SomethingAwful. This immediately places a value on one's account beyond forum utility - you don't want to post poorly on your account, because you might lose your money's worth. That is, unless you decide to pay $10 again. But chances are, you have less disposable income than the moderators have patience.

Punishments typically take the form of probations, wherein the offending poster is forbidden from posting on the forum for a set amount of time. More serious offenses receive bans, wherein the offending poster must pay ten more dollars if they want access to their account again. In both cases, the offending message is preserved in its original context, with an added note that the user was probated/banned for this post. This is in stark difference to the previous systems - SomethingAwful shames its offenders, while Reddit and YouTube try to sweep them under the rug.

The result of this system has been a community where quality and effort is rewarded and praised. SomethingAwful users include the creator of the original Slenderman mythos, and countless photoshopped pictures that have since made their way around the Internet. The age and scope of the forums make it so that lots of the beloved aspects of online nerdiness can be found on, if not originally traced to, SomethingAwful -  4chan founder moot had his start on the website, for example. Since everyone who posts is implicitly satisfied with the rules, there is a healthy presence among posters to police themselves.

This does come at a cost, primarily being SomethingAwful's significantly lower traffic rating as compared to YouTube or Reddit. The forums are certainly not inclusive of all prospective members, nor do they try to be. They are an example of what high standards for their user base can produce. SomethingAwful's payment model is unique to SomethingAwful, as far as I've seen.

This is also where things tend to get hairy about SomethingAwful's protocol - how effectively can one generalize the model to the greater Internet, without crossing the lines into some serious censorship?

The SomethingAwful model carries its own backlash by being exclusionary, and there are no measures in place to prevent the bad posts/noise posts to emerge elsewhere online. Should we expect higher-quality content to naturally remain at a premium, lest we begin enforcing some catch-all laws about online expression?

It's possible, but there are already net-wide patterns emerging that run contrary to this. Public shaming - one of the same forces that help contribute to online social justice - might be a solution that helps keep the online population in check. Websites like EncyclopediaDramatica purely exist to expose (and make fun of) stupid things happening on the Internet. There are YouTube channels, many tumblrs, and other online records (yes, SomethingAwful will go out of its way to shame itself too!) that exist solely to point out the absurdity of some online communities. Such practices signal to readers what sort of behavior is acceptable online, as well as provide a consequence for behaving poorly online.

Shame culture can be useful online if channeled properly, but there is still a distinct lack of top-down regulation. SomethingAwful's model may be effective, but it helps that the website is a comedy website. Other examples of forums expecting high standards from its posters can include the Atheism+ forums, and politically loaded forums like Democratic Underground and Free Republic. It becomes very tricky to find objective arbiters of quality control, outside of enforcement of basic sentence structure.

As the Internet continues to mature, we will have to address the way we communicate with one another online. It's simply more efficient for a reader to go through websites with minimal noise content. Perhaps there will always be a space online where low-effort noise contributions are permissible, but online environments that encourage better communication and deeper thought may be likely to produce better things for the Internet as a whole. The best we can hope to do is to demonstrate the benefits of more discerning online environments, and maybe these spaces will marginalize the rest.