When Half Life 2: Episode One was announced, it was with the promise of things will be different this time. Fans, many of whom remembered all too well the bizarre turmoil of delay after delay for Half Life 2, were afraid of waiting another six years for the third segment of the series, and the phrase 'Episode One' conjured with it all manner of ethereal promises: chief among them being "faster releases and cheaper games."
In practice, however, this hasn't been the case– Episode One didn't come out until almost two years later, in 2006, and Episode Two was released a year and a half after that. Episode Three has had no real information about it in quite some time, but even now its timeline is getting a bit long in the tooth– increasingly, it's looking as if Half Life 2: 'Episodes' will have taken the same amount of time to go to shelves as it took for Half Life 2 in the first place, but with a total game playtime a fraction of the length of Half Life 2.
But first, backstory. Valve isn't the first to come up with this concept or suggest this idea. Episodic content has its roots in the old Apogee Software shareware model of the 90s; previously indie game developers, unable to compete with the prowess of big developers with retail connections, would release their titles as shareware– they were free and you were expected to pay for the game if you liked it or wanted technical support (or cheats) for the title. You can deduce that the honor system went about as well as expected, so Apogee brainstormed a better way. Their method, starting with Kingdom of Kroz, was to release the first part of a series (usually a trilogy) as shareware, with the remaining parts available via mail order.
The first part was generally a complete game ending on a cliffhanger that required the purchase of the rest of the series to complete. This model worked out tremendously, to the point that other indie developers and even larger companies utilized it themselves. Apogee released several titles under this model, such as Rise of the Triad and Halloween Harry. They were similar to a trilogy of novels or movies, with the first part being available for free to anyone who wanted it.
Today's version of episodic content was supposed to mean shorter development time (and thus a shorter time between releases), predictable and routine release dates and lower development cost all around. In Valve's case this has failed spectacularly, as each piece of episodic content has been more about showcasing the shiny features of the Source engine than actually getting parts out there for sale– as such, there's been an extended amount of time between each release. There's also the cost to the end user; Episode One was approximately $20.00, and Episode Two was bundled as a part of The Orange Box for $50.00. While the latter included several other games (as well as the critically acclaimed Portal which was tangentially related to the Half Life series) it's unavailable as standalone. One of Episode One's common complaints was its short length of approximately six hours, and while Episode Two tried to improve upon that it didn't make things a lot better. When the original intent was shorter turn around to justify the short length of titles, it stings a bit when you pay the approximate price of a retail release for an experience of just a few hours, knowing you'll be waiting a year or two for the next piece of the puzzle.
Episodic content also has its risks with developers as well. Ritual Entertainment released SiN in 1998, ironically almost at the same time as the original Half Life. When this game had underwhelming sales, Ritual turned to the episodic format with their release of SiN Episodes: Emergence. As one of the first large companies to turn to both the episodic format as well as the internet as a content delivery system, hopes were high but ultimately fell flat and overly ambitious. Originally Ritual intended to release nine games for $20.00 a piece, every six months, with each one lasting four to six hours. In reality they only managed the first one and Ritual as a company was sold to casual games publisher MumboJumbo in 2007, shuttering future plans for the franchise. Or take the case of the alternate-reality game Majestic which was episodic, yet spread out over a monthly subscription as opposed to charging for individual pieces. Started in 2001 and blurring the line between reality and fiction, EA canceled it a year later due to poor performance.
On the other hand, Telltale Games has seen incredible success with this format, reviving two of their older LucasArts franchises in the process. Sam & Max Save the World was a six part episodic series released in 2006/2007 that followed an extremely steady release schedule about a month apart, with Season Two coming at the end of 2007 and the start of 2008. Many people feel that Telltale is the only company to really correctly nail the spirit of episodic content and they're one of the few to turn a significant profit from the business model. MINERVA is a Half Life 2 Source mod developed by Adam Foster that has also released content on a pretty regular basis, with episodes coming out every six months.
In a lot of ways I wonder if DLC is the new incarnation of episodic content; BioWare has promised to support both Dragon Age: Origins and Mass Effect 2 with routine short updates for a few years each, and both Bethesda's Fallout 3 and Gearbox's Borderlands have released several pieces of DLC for their games too. It's increasingly expected of games, whether on PC or console, to offer this future support, and given the ubiquitous and pervasive nature of internet access developers now have the means moreso than ever before. However, DLC still requires a base game in the first place, with all the traditional investment and development such a game entails.
It remains to be seen if future developers will attempt the episodic route or not, and whether or not these companies will go the Telltale route or the Ritual one. Surely there's a benefit towards customers who can look forward to getting a new game every couple months, but only if these releases stick to a predictable schedule and demonstrate reliability.