The eBay Way

Onward With Web 2.0
We recently finished version 4 of our architecture, and we think it's revolutionary. The architecture supports a new presentation-tier technology that lets engineers more efficiently program global code that incorporates a number of Web 2.0 constructs, such as dynamic pages, Ajax, JavaScript, CSS, Flash, and back-end coding. Engineers can develop integrated end-to-end applications from one development environment in Java. We also have made significant investments in developer productivity tools that are completely integrated with the Eclipse IDE.

One of our main new technology initiatives involves developing a platform for creating efficiencies for my development team, especially in the area of scale. Previously, for example, when we built something for 200 million users, we would have to go through a specific set of processes--some not always appropriate for small experiments or for an application that might be around only for a couple of months. On the other hand, as those short-term applications or small experiments grew more popular, we would have to manually scale or rewrite them to get to the next level. Now, developers will be able to use the same platform, whether they're building a large application for 200 million users or one for a couple of hundred.

We are one of the world's first and largest Web services development platforms. While we already have hundreds of public APIs, we are building a new innovation platform that includes a service-oriented architecture component and will feature security, management, analytics, and monitoring. Projects will move easily from one level of maturity to the next, and the system will be able to scale and be pulled into a core system--very similar to a hub-and-spoke model.

How does all this impact our customers? By creating richer eBay experiences more quickly. Also, we will let some third-party developers work on this platform, adding to the system's ongoing innovation and growth.

We want to avoid the paradox that often occurs when a team of builders works on the large core systems of a business, but as the system grows, resources become scarce and the systems become more rigid. We're hoping our platform will enable our team to experiment without restricting their investments.

Today, our team members usually write experimental code in the Web app framework Ruby on Rails or in a PHP prototype. As they try to take it mainstream, they generally have to rewrite it in another application toolset to scale and offer better security. We'd like to change that so the platform will be as efficient as any rapid prototyping and built with Web 2.0 technologies, yet highly scalable. We've started development and will start testing this year. We expect to be finished next year.

Innovating for a community of our size and maintaining the reliability that's expected is challenging, to say the least. Our business and IT leaders understand that to build a platform strategy, we must continue to create more infrastructure, and separate the infrastructure from our applications so we can remain nimble as a business. Despite the complexity, it's critical that IT is transparent to our internal business customers and that we don't burden our business units or our 233 million registered users with worries about availability, reliability, scalability, and security. That has to be woven into our day-to-day process. And it's what the millions of customers who make their living on eBay every day are counting on us to do.

James Barrese is VP of systems and architecture at eBay ([email protected]). For further discussion of eBay's innovative architecture, go to our new blog, CIOs Uncensored.

Continue to the sidebar:
Tips For On-Demand Innovation From eBay