At the Game Developers Conference Online in Austin during the second week in October, a dozen hopeful young entrepreneurs approached our booth selling versions of the same ambitious vision. “We’re building a Massively Multiplayer Online Game. It’s going to be the next World of Warcraft, the next Call of Duty. It’s gonna be huge,” they said. These hopeful game mavens were seeking insight on exactly how you build an MMO infrastructure. Not surprisingly, many of them were woefully unprepared. They didn’t know what a load balancer was. They had no idea about data transport costs between data centers. They hadn’t really thought about the impact of hardware at cloud providers on the user experience of their game customers (hint: old servers usually mean unhappy or lost customers).
A few years ago we would have shaken our heads, wished them good luck, and figured we would probably never see that game released. Today there is a far greater chance that these folks can bring the next MMO to market. What’s changed? The arrival of specialized Platform-as-a-Service (PaaS) offerings. In a nutshell, a PaaS allows an entrepreneur to focus on building their cloud application, avoiding upfront capital expenses through outsourced management of their IT infrastructure to a third party.
This is a level up from the cloud itself, which provides outsourced compute power in a more raw fashion. PaaS providers offer database as a service (MongoLab, MongoHQ) or runtime environment as a service (Heroku, Nodejitsu, AppFog), for example. A PaaS can also deliver even more advanced capabilities. StackMob, for example, puts in place a suite of PaaS offerings that radically streamline mobile application development, launch and hosting by providing in one integrated package an environment to code up, host, and run in the cloud mobile applications.
What’s more, a PaaS can even be taken to the point where it eliminates almost all technology skill requirements. GameSalad allows designers to quickly design and publish game applications that even include animation purely using visual design tools. Eliminating technological complexity serves to eliminate a critical barrier to innovation and new company foundation. We’ve all met someone who felt they had a great idea for a new application of some sort. “And I’m looking for a developer or a CTO,” is inevitably the next thing out of their mouth after they pitch the idea.
If that same great idea could be built far more easily and quickly with a far smaller tech team and nominal capital expenses, then, logically the cost of bringing that idea to market drops considerably. This is the value of the PaaS. And its not just a value to guys with no tech chops. Someone that is a very solid designer and front-end coder may be highly technical but may not be comfortable at all with designing and managing a NoSQL database.
Even highly skilled developers with both front-end and server-side chops usually turn to network engineers for help with load balancing, DNS and other infrastructure aspects that are critical to ensuring a cloud-based application runs fast and clean on any device anywhere in the world. Dennis TK, founder of Foursquare, is fond of explaining that as soon as he got funded, he hired an ex-Googler to completely recode his app to keep it from breaking. If Dennis had been building Foursquare in the present, he could have probably built a more reliable, faster app by leveraging a far more robust PaaS ecosystem to remove many of the software development and infrastructure management requirements that probably caused Foursquare to be so buggy and break all the time in its initial inception.
How many more kids like Dennis are out there with whizbang innovations that could be the next WoW, the next great service for medical records delivery, or the next amazing tool for crowdsourced scientific problem solving? We don’t know but we are a lot more likely to find out in a new era of PaaS-fueled cloud innovation.
Lisa Petrucci is the VP of Global Marketing at Joyent. She started out as a network engineer for Lotus and has held senior sales, marketing and business development roles in enterprise computing companies for the past two decades at companies including IBM and SixApart.
Image courtesy of Flickr user ansik.
Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.
- Infrastructure Q1: IaaS Comes Down to Earth; Big Data Takes Flight
- Infrastructure Q3: OpenStack and flash step into the spotlight
- Infrastructure Overview, Q2 2010
alt=''
border='0'
/>
Android users have about 250,000 apps to choose from but most have little use for any of them outside the top 50. That’s according to new data from Nielsen’s Smartphone Analytics, a new initiative that analyzes data from on-device meters.
The growing picture here is that Android has a real issue in making sure that more developers can thrive in the Android Market. If you’re a developer and you can’t afford to pay for marketing or you don’t have some amazing viral hit, it looks like it’s very hard to get your app used and that makes it hard to crack profitability on Android for all but the biggest apps. If you’re not being used that often, you can’t expect to garner that much in advertising or in-app purchases.
Google has done a lot of work to improve the app experience in Android Market but it’s got to do a better job of aiding in discovery. And Android developers need to keep pushing the quality of apps in Android Market so people take note of not just the top apps. With developers increasingly getting their revenue from freemium apps, which now generates 2/3 of the revenue in the top 100 games on iOS, it’s even more important for developers to get their apps used and to have longer term engagement with users. That’s something Google still needs to keep working on with its developers to ensure that the success of Android Market is not only limited to a few top publishers.
