We're a few days removed from the most recent Microsoft Professional Developer's Conference (PDC), which means you've had a few days to reflect on what you saw there. Today, what are the real takeaways from this year's PDC for you?
My main take away is that this year was really about battening down the hatches and galvanizing the stuff they introduced last year. For example, last year Microsoft introduced us to Azure for the first time. It also introduced us to the .NET Framework 4, Windows 7, a new version of Silverlight, and other features of that ilk. This year basically reprised those technologies. You got to see how far these technologies have come, in addition to the latest iteration of Silverlight, which has some exciting stuff in it.
Under the surface, PDC is changing in its mission and identity. It's now less of a futures show, and more of a generalist developer show. Its name has always suggested something akin to that, but in practice it has always been forward-looking. In this case, nearly everything discussed appears to have a timeframe of roughly six months until shipping. Some of the stuff ancillary to Azure, such as AppFabric and so on, doesn't have a ship date exactly. But the main focus was very much on the near future, rather than a future that is 18 months out, or longer. With respect to Windows Azure, what type of developer stands to benefit the most from the notion of cloud computing?
Technically, all developers are likely to benefit from cloud computing. That said, you still see a bit of a disconnect between Azure as it is now and Azure as Microsoft would like it to be. We saw one plank in the desired future platform with AppFabric, where you get a Windows Server and an Azure version of this technology. You can see a lot of parity in the designs you can create with the Windows Server versus Azure approaches, but they are rather different environments.
I think you will see Windows Azure come along in stages. Initially, you will see specialist developers who want to target this platform as a niche. Over time, Microsoft's goal is to make the process of scaling up to and down from the cloud seamless.
The economics of this are also important. I do have a sense of what the unit costs are, but I don't yet have a sense of how the total cost for an app of a given scale will work out over a year's span. It does seem that pricing would appeal most to customers who already have datacenter apps. The irony there is such companies already have datacenters, and they might have a proprietary instinct about where parts of their infrastructure run. The cloud is the future, but it isn't the near future. It will take a while for the cloud to become mainstream, and the biggest early adopters of the cloud will not be the key adopters that propel the cloud forward. The folks who will push the concept forward—whose acceptance will signify that the Microsoft's vision of cloud computing has been realized—they don't want to be guinea pigs. When these folks move forward, then you will see the critical mass for the cloud to be adopted in mainstream businesses.
The early adopter period will be characterized by limited market participation. When those more conventional corporate folks take to the cloud, you'll see things move quite quickly. You will also likely see the pricing models go through a couple shifts. Some of these things need to be worked out, and Microsoft had to start somewhere. Microsoft gave over much of its Day 2 keynote at PDC to Silverlight. What do you see as the main role of Silverlight going forward? Who is this technology for?
Silverlight has taken a tremendous stride forward in version 4. What you see in version 4 is that Silverlight is no longer just a Rich Internet Application (RIA) platform, but a client platform in its own right.
Silverlight 4 is for Windows client developer. We already know it's for Web developers interested in richer, non-AJAX, non-HTML development. What we got on Day 2 was a different presentation, including two new significant features. First, we get the ability to go out of browser. Second, once we're out of browser, we get abilities that we might have once have thought were the domain of Windows Presentation Foundation (WPF). The distinctions between Silverlight and WPF melt away. Here's the neat part: You have an app platform that is simultaneously a Windows and cross-platform development tool. The Windows-specific capabilities won't work on Mac, but rather than resigning yourself to a cross-platform subset, you now have the option to jump out and take more complete advantage of Windows if you want to.
Of Brust's 20+ years in software development, 10 involved heavy lifting with Visual Basic and .NET. He is co-author of the the book, Programming Microsoft SQL Server 2008 (Microsot Press, 2008). His company, twentysix New York, is a Microsoft Gold Certified Partner.
Patrick Meader is a freelance editor and writer with more than 16 years experience working for technical magazines.