26 April 2012
I am sometimes asked for technical career advice. A common question these days is whether it is worth learning WPF, or Silverlight – .NET and XAML in general I suppose, or would it be better to learn HTML 5 and JavaScript, or perhaps even Objective C?
This is a challenging question to be sure. How good is your crystal ball? 
XAML appears to be alive and well – WPF, Silverlight, and now WinRT (Windows 8 – and probably Windows Phone 8 and “Xbox 720” and more) all use XAML.
I look at the WinRT usage of XAML as being essentially “Silverlight 6” – it is far closer to Silverlight than WPF, but isn’t exactly like Silverlight either. Assuming success with Windows 8, WinRT will become the new primary client dev target for most smart client development (over the next few years).
The primary competitors are Objective C (if you believe iPads will take over the client space), and HTML 5/JavaScript (if you believe in fairy tales the concept of ‘one technology to rule them all’).
This is where the crystal ball comes into play.
Do you think Apple will displace Microsoft – iPads will replace the use of Windows – as the monopoly client OS?
Do you think the concept of ‘natural monopoly’ that has caused the Windows hegemony over the past 20 years is at an end – that some fundamental economic shift has occurred so companies are now willing to increase their IT budgets as a % of revenue to accommodate multiple client platforms (unlike the past 20 years)? In which case business app developers should expect to support at least iPad and Windows, if not Android, into the future?
Do you think that Windows 8 and WinRT will be strong enough to withstand the iPad onslaught, and that the natural monopoly economic effect remains in place, so Windows will remain the dominant client platform for business apps into the foreseeable future?
These are really the three options, resulting in:
- Objective C slowly overtakes .NET and we ultimately are Apple devs instead of Microsoft devs
- H5/js rules the world as the ‘one technology to rule them all’ and vendors like Microsoft and Apple become entirely irrelevant because we live in a purely open-source world where nobody makes money off any platform technologies, so probably the only hardware/OS left is something like Android running Chrome, because it is a 100% commodity play at that level
- .NET and XAML remain entirely valid, and life generally continues like it is today, with a mix of .NET smart client work and primarily server-based web work with h5/js primarily used to boost the user experience, but not often used to write standalone smart client apps
My crystal ball leans toward option 3 – I don’t think economic realities change much or often, and I struggle to see where IT departments will come up with the increased budget (% of revenue) necessary to build apps for both iPads and Windows over the long term. It will be measurably cheaper (by many, many, many thousands of dollars) for companies to buy employees Win8 tablets rather than building and maintaining both iOS and Windows versions of every business app.
And I don’t believe in the ‘one technology to rule them all’ idea. That hasn’t happened in the entire history of computing, and it is hard to imagine everyone on the planet embracing one monoculture for software development. Especially when it would be counter to the interests of every platform vendor out there (Microsoft, Apple, Google, Oracle, and even IBM).
Still with me? 
To summarize, I think learning XAML is time well spent. Today that’s WPF or Silverlight. There is absolutely no doubt that Silverlight is closer to WinRT than WPF, and people building SL apps today will have an easier time migrating them to WinRT later, whereas most WPF apps will be a pretty big rewrite.
But there’s nothing wrong with focusing yourself on h5/js. If you do so, I suggest doing it in a way that ignores or minimizes all server-side coding. If h5/js does take over the world, it will be used to create pure smart client apps, and if there’s a “web server” involved at all, it will exist purely as a deployment server for the client app. The ‘pure’ h5/js/jquery/etc. world isn’t linked to any vendor – not Microsoft, Apple, or anyone. To me this represents a pretty major career shift, because to truly embrace h5/js as a complete software development platform is so demanding (imo) it won’t leave time to retain .NET or other vendor-specific technology expertise.
For my part, I’m not yet ready to abandon Microsoft for h5/js, because I think Windows 8, WinRT, .NET, and XAML have a rather bright future. A year from now I think a lot of people will be happily using Windows 8 desktops, laptops, and tablets – and hopefully a lot of Windows Phones, and with luck we’ll be looking forward to some cool new Xbox. I live in (I think realistic) hope that my .NET/XAML skills will apply to all these platforms.
What does your crystal ball say?