Tomorrow (or Wednessday in Europe) we launch Hydrophobia Prophecy on PSN, and we're delighted to have some great backing from Sony for the launch.

From a design and development point of view, perhaps the most interesting thing about the game is Darknet – our built in developer feedback system which is getting its console debut.

Darknet allows players to pause the game at any time, select Dev Feedback from the menu and then choose contextual feedback comments to upload to our server. When we run through dev builds of the game, we can see icons representing each comment at the exact location the player was in the environment when they dropped it, which builds up a like a heat map of communal player experience which we can then analyse.

See: http://www.youtube.com/watch?v=I7UPa97q320 for more details.

I thought this would be a good opportunity to explain the thinking behind Darknet, and also share some of the fascinating and often surprising results we saw from the initial Steam release.

Before I started to think about Darknet, I had already built survey software into the games website following the original XBLA release. At that point, we were trying to get our heads around the massive divide in opinion about the game. We'd had several great reviews; 9/10 from OXM, 8/10 from IGN, 9/10 from TheSixthAxis and then several terrible reviews like 3/10 from Edge and 4/10 from Eurogamer. The website survey was a way to try and tap into players opinions, and we followed up quickly with a comprehensive patch to fix many of the flaws.

The interesting thing was the response from players; many sent Tweets and messages expressing their delight at having the opportunity to express their opinion and have it taken on board by the developer.

This got me thinking. Why was it that in the age of the Net Generation, when people were contributors to and often creators of online content on sites like Wikipedia and services like Twitter, that mainstream games remained walled off? If Web 2.0 is known as the read-write Internet, why did games remain so read-only?

So after setting out the idea for what became Darknet getting a positive response internally we set about building it. We wanted to expand on the incredibly positive reaction to the website surveys and build a community behind the rebooted Steam version of the game.

We didn't make a big song and dance about Darknet in the game, it was hidden behind an innocent looking pause menu option and otherwise not mentioned, so we didn't expect many people would use it in it's initial, conservative form. We also assumed that the majority of the data would be negative – after all why would people bother to leave a comment if they were enjoying the game, people only reach out to a company when they want to complain, right?

Apparently we were wrong on both counts. When we arrived into the office on launch day, thousands of players had already left data throughout the game and in balance it was positive data that was winning. Over the next few days, the distribution settled at about 2/3 positive data nodes, and that has remained the case since.

We released several updates on Steam, many of which used Darknet data to add checkpoints where it was clear people were getting frustrated, or to balance ammo or iron out obvious difficulty spikes. The response from the community on the forums was ecstatic, and we're seeing the a similar reaction to the Darknet concept from PSN users ahead of the launch this week.

The Net Generation has become accustomed to the Facebook Like Button, the ability annotate YouTube videos and contribute to the accumulated knowledge of mankind via Wikipedia. Mass collaboration and participation is expected, and if games gives their players the opportunity, they will respond with great enthusiasm.

Of course it is still the job of game creators to get things right before release, and pre-launch playtesting sessions will always be important, but why should it end there? Is any game perfect on release? Should we not listen, analyse and continue to learn? More importantly, shouldn't we start to catch up with the Net Generation and deliver them the sorts of collaborative read-write experiences they find elsewhere?