steam greenlight gama.jpgIt's an exciting time to be a PC indie game developer right now. Valve Software's Steam Greenlight initative, due to launch next month, intends to streamline the submission process to get more indie games onto the platform, in theory allowing more devs to reap the rewards of Steam's enormous ecosystem.

Greenlight is a potential game changer for digital distribution platforms, because it relies heavily on the community. Steam users will be able to submit both released and in-development games to Greenlight, filling their Greenlight game page with trailers, info, screenshots, system requirements and links to press coverage.

Devs can then continue to update the page as frequently as they desire, and once the game is in a playable state, the community can rate your game and give feedback, pushing it towards acceptance on Greenlight.

But while the hype for the upcoming initiative is palpable, there's one element that is quietly being discussed among indies -- with an expected influx of Greenlight candidates, how exactly is Greenlight going to handle discoverability?

We've seen on multiple digital storefronts what happens if discoverability isn't tackled. If your game isn't in a top 25 list, or if, when your game first launches, no one instantly notices it, it is likely doomed to sit at the bottom of a huge pile of titles, festering away, garnering one or two downloads here and there.

Steam's storefront already has a number of options in place that do a fairly OK job of bringing lesser-known titles to light, from its daily deals to its top-sellers lists. Yet hundreds of titles -- most prominently indie games -- still sit in the Steam library and are barely touched, whether they are enjoyable experiences or not.

Valve's Robert Walker recently said at the Develop conference in Brighton that discoverability is a problem that Steam has always had, adding that Valve "[doesn't] think it'll ever be solved."

Jason Holtman, Valve's business development director, added that with Greenlight, he hopes that the community will help solve the discoverability problem through suggestions and a willingness to go a little deeper into the store. "We're hoping you're all going to solve that for us... it's going to be interesting to see what happens," he said.

Greenlight introduces extra scenarios into the discoverability fray too, in that it will be filled to the brim with games that are in a wide variety of playable states. If a developer puts out a playable state too early, then it may well be of poor quality and lose potential fan interest -- yet if the dev waits too long, they run the risk of losing the audience before the publicity push has even begun.

While Valve hasn't yet revealed how it plans to forge ahead with discoverability on Greenlight, there are numerous systems in place on other successful websites that the publisher would do well to consider and apply to its own platform.

Reddit's "Knights of New"

For those who have heard of it but never ventured into its realms, Reddit is a social news website which feeds on user-submitted content. "Redditors" post a link and give it a description, then other users "upvote" or "downvote" the content based on whether they like it or not.

This like or dislike functionality may sounds incredibly simple, yet it does a pretty remarkable job of allowing the good stuff to float to the top and appear on the Reddit front page -- indeed, Reddit receives over 2 billion pageviews a month, so it must be doing something right!

Of course, the question is: how exactly do links get noticed and upvoted from the moment they are submitted, rather than succumbing to popular, already upvoted links? The answer is, for the most part, a group of Redditors who called themselves the "Knights of New".

These people tirelessly go through newly-submitted links, check whether they believe the content to be worth everyone else's time, and then either upvote or downvote accordingly. If enough Knights upvote a particular link such that it is boosted all the way into the top 25 links for a particular section, it will then be spotted by the more casual Redditors, and receive a multitude of upvotes/downvotes as a result.

knights of new.jpgWhy does this group of people bother spending their time doing this? There are a few reasons, from feeling like part of a community to providing what they believe to be a service to other Redditors, but in reality, there's one main reason: karma.

Karma is Reddit's Xbox achievements, PSN trophies, Steam badges. It means absolutely nothing in the real world, other than to make a player feel a sense of accomplishment. If your content is upvoted, you get karma -- however, there is also another way. You can receive karma through having a comment upvoted as well, and this is where the Knights of New come in.

If a Knight spots a link which they are sure is going to explode with popularity on Reddit, they can make sure they are one of the first to post a comment about the link. On Reddit, the first few good comments to be posted on a link are usually the ones that rise to the top and will then be upvoted more often than comments that are late to the party, no matter whether later comments are actually better in quality.

Hence -- ironically -- the poor discoverability of Reddit comments is what causes these Reddit Knights to scour for new great links, and therefore solve the website's overall link discoverability issue.

How, then, can this be carried over to Greenlight? For starters, Greenlight needs to have a rating system, which it no doubt will have. Either a 5-star rating system or even an upvote/downvote system similar to Reddit's would work wonders, especially with the Knights of New model. A screengrab of the system posted on the official Greenlight community page (as shown below) suggests that Greenlight will indeed use an upvote/downvote system, although this hasn't been officially detailed as of yet.

greenlight1.jpgSteam can then look to reward those people who take the time to rate new projects with Steam badges, Steam achievements and accolades to make them feel special -- however, Steam is also in the position where it can in fact offer more than simply a sense of achievement. Steam could also offer free games and DLC to those people who are willing to sift through the crap and find the top quality titles.

Once these early raters -- or "Greenmoonlighters", if you will -- then boost new games up the popular charts, more casual Greenlight users are more likely to spot them, check them out, and then also give them a rating.

Now, that's not to say that the Reddit system is flawless. A lot of reposts, memes, spam and badly sourced links still get through to the front page, while true original content and relevant news is left behind.

This can be down to some Knights simply not realizing that a link is not original, or a Knight just upvoting a link on the off-chance that it gets big and will therefore earn them some sweet, sweet comment karma.

However, many of these issues would potentially be solved by themselves in the Greenlight version of the system. Reddit links only stay on the front page for less than a day thanks to the huge influx of links submitted, meaning that links come and go very quickly.

Greenlight games, on the other hand, would stick around in the top list for a while longer as there simply won't be as many submissions as Reddit. Therefore, those games which are more worthy of any clones or the like which pop up will no doubt rise higher and knock the bad stuff down.

greenlight2.jpgThere's also the factor that is how vocal the gaming community can be. If a clone or bad game were to appear on the front page of Steamlight, social media like Facebook and Twitter would no doubt go ballistic, with hundreds of people retweeting and sharing the news that there is a blight on Steamlight.

This would then lead to an influx of downvotes, removing all badness from the page and only leaving the good projects. At least, in principle - whether it would all go off without a hitch can only really be seen by putting the idea into practice.

Rate this

As mentioned previously, there will no doubt be a user rating system in place -- most likely an upvote/downvote system, judging by the official information and screenshots given up to this point.

However, another very popular rating system used to great effect is the 5-star system. It's incredibly important that rating systems like these are carefully designed to consider numerous factors, from the users who only every vote 1 or 5 stars, and those who spread on social media to mass rate a particular game, whether it be good ratings or bad ratings.

When it comes to ratings, Valve would do well to learn from the Kongregate and Newgrounds rating systems. Both of these browser game websites feature a similar system that has proven to help bring the good games to the front page and leave the badly designed games at the bottom.

To begin with, when a game is submitted, it is classed as "under judgment" until it has had enough ratings, at which point the rating is revealed and the game is slotted into its rightful place in the charts.

This is done for two reasons: so that users are not influenced by other ratings when making their rating, and also so that a game is given a chance with multiple users, rather than receiving a couple of 1 stars and instantly being ignored by anyone else.

greenlight3.jpgIt works incredibly well too - Kongregate currently has around 60,000 games, and yet only 1 percent of those games has a rating over 4 out of 5 stars, while only 14 percent have a score that is greater than 3 stars. The front page of Kongregate regularly features these top rated games, and will cycle through them such that, if your game is highly-rated, you will be featured on the front page numerous times.

Now, combine the "under judgment" system with Reddit's Knights of New, whereby Knights cannot see the ratings that have already been given (this sort of already happens on Reddit, in that you can't see the rating on a new post until you click through to it), and you have a pretty nifty system for discovering great new games, and leaving the not-so-good in the dust.

Could all this work for Greenlight? Perhaps -- and it would no doubt need a tweak or ten to keep it in line with the existing Steam framework. One thing is for certain though - discoverability is going to be one of the biggest challenges that Greenlight faces, as does any new and existing platform, and the above potential solutions give just a small insight into how deep the problem can truly go.

[Mike Rose wrote this article, which originally appeared on Gamasutra.]