Peter Seale's weblog

Saturday, February 09, 2008 1:47:05 AM UTC 

I've just finished reading this entry titled Why Microsoft Will Win and Dominate the Web linked off of DotNetKicks.com. The author of this blog post claims that, as Microsoft succeeded in the 80s by creating a better developer platform (the original Windows was a better developer platform than the Mac), Microsoft will again succeed by creating a better platform. This time Microsoft's platform is Silverlight, and it is poised against the web.

Let's be very clear about this:

• Silverlight 2.0 is in an early alpha/unstable state.
• Silverlight 2.0 does not emit standard (or even quirky) HTML. As such,
  • it cannot be indexed by search engines, and
  • it will not work on mobile devices. We'll get to the "mobile devices" bit later.
• Last time I checked, Silverlight 2.0 had less than 5 controls to use. E.g. textbox, button, dropdown control. I'm fuzzy on the specifics; there might have been as few as 0 (zero) controls. No databinding.
• The last four times I tried it, Silverlight 1.0 would not install on Firefox on Windows. I don't know why either, but I do know it doesn't bode well for the future.
• The MSDN home for Silverlight reference material lists a grand total of about 30 pages (as of 2008-02-08).

The checklist represents Silverlight 2.0, today. In my opinion, this is the only legitimate Silverlight you are allowed to talk about. And let me tell you, the Silverlight above, as of today, isn't going to dominate anything.

So when I read a description of Silverlight 2.0 that looks something like:

• A platform on which fantastic applications can be built.
• The best development toolset provided for the web.

If you compare their list with my list, let's just say they don't match up one for one.

Where did they go wrong?

How could we veer so far from each other, and still talk about the same product? Am I overly pessimistic? Yes. Yes, I am.

Magical Fairyland Silverlight

What has happened is that this guy allowed his enthusiasm to get away from him. Which is fine; it's good to be enthusiastic. However, what's not okay is that, in his enthusiasm, he mistook Silverlight with "magical fairyland Silverlight." Instead of describing Silverlight's impact on the future web, he got a little turned around, and perhaps, just perhaps, a little carried away. What he ended up describing was not the future Silverlight platform, but a mirage.

With this context in mind, let's compare the two lists again:

Silverlight 2.0, as observed by Peter Seale, 2008-02-08:

• Silverlight 2.0 is in an early alpha/unstable state.
• Silverlight 2.0 does not emit standard (or even quirky) HTML. As such,
  • it cannot be indexed by search engines, and
  • it will not work on mobile devices. We'll get to the "mobile devices" bit later.
• Last time I checked, Silverlight 2.0 had less than 5 controls to use. E.g. textbox, button, dropdown control. I'm fuzzy on the specifics; there might have been as few as 0 (zero) controls. No databinding.
• The last four times I tried, Silverlight 1.0 would not install on Firefox on Windows. I don't know why either, but I do know it doesn't bode well for the future.
• The MSDN home for Silverlight reference material looks like it's a grand total of about 30 pages (as of 2008-02-08).

Magical fairyland Silverlight:

• A platform on which fantastic applications can be built.
• The best development toolset provided for the web.

Now it makes sense!

The Golden Rule

PLEASE stop comparing unreleased, FUTURE versions of your product with the CURRENT crop of competitors.

In this case, it's especially important to note that magical fairyland Silverlight (or anything moderately resembling it) won't be released anytime soon, because in the future, by the time something resembling magical fairyland Silverlight actually arrives, it might already be too late! Consider the following emerging markets:

• sub-$200 laptops
• 3G cell phones
• iPhone
• iPod
• BlackBerry
• Google's Android
• TV set-top browsing devices (e.g. gaming consoles, DVR)

Do all these things run Silverlight? By the time we're all checking our GMail from our iPhone on the bus, or checking our "Microsoft Live Passport Flickr For The Web" from our SmartPhone, IM'ing from our iPod, or emailing a Google spreadsheet on our android phone, will those devices be Silverlight compatible? Will search engines update themselves to index Silverlight content by then, or will our Silverlight apps remain invisible to Google search? Like so many other technologies have done in the past, will this platform sink as well? I don't know. BUT NEITHER DOES THAT OTHER GUY!

This happens all the time

Don't get me wrong; this sort of thing happens all the time. .NET guys compare .NET 3.5 to Java 1.1. Java guys compare Java 7 to .NET 1.1. Windows guys like to complain how expensive Macs are; Apple guys like to complain about how insecure Windows is. Linux guys complain about IIS security. SharePoint guys like to put on the rose-colored glasses when it comes to describing the SharePoint pie chart of features.

It's everywhere if you look for it.

I guess I just had to write this up, on behalf of all the other times I've seen this sort of wishful thinking happen. If you take anything away from this post, let it be "don't get caught up in the potential of a product without taking a hard look at what it is, today." There, that wasn't so bad.