Earlier this month (January 2009) I made the prediction that Silverlight penetration would approach Flash penetration (80%+).  It's a bold prediction, especially since it is widely held that as of this month, penetration is in the 25% range.

There was a string of very large scale successes just this month, notably around Obama's inauguration.  These included the Presidential Inaugural Committee choosing Silverlight to broadcast video of the inauguration and CNN creating an unbelievable 3D collage of the inauguration moment using Photosynth & Silverlight.  And that was just January.

Despite my enthusiasm, not everyone is convinced that Silverlight is on it's way up.  This article in Infoworld -- Silverlight adoption hampered by economic crisis -- chief among them.  On one hand, the hypothesis is likely to be bang on, in the sense that nearly everything is likely to be hampered by the economic crisis.  It's sort of stating the obvious. 

But some of the statements in the article are just so opposite to my experiences:

"The UI is considered the last part of the application," said Ryan Peterson, principal and software engineer for Serenity Software, a Harrisburg, Pennsylvania, company that specializes in UI consulting and design. "The mindset has always been and still is: You build the application and then you build the interface. It's a large contributing factor to why people cut that [first]. They think if the application works, we can take care of the interface later."

What?  Maybe the mindset is completely different between Pennsylvania and Toronto, but the above statement doesn't resonate with me or any of my customers.  And while it's true that we do a lot of work in large-scale web properties where User Experience is paramount, we also do a ton of work in the nitty gritty of internal business applications, and User Experience and Interface Design is nearly always at the forefront.

I remain very bullish on the overall growth of Rich Internet Applications (RIAs) in 2009 and I think that Silverlight (and Flash for that matter) have a long rosy future ahead of them.  I'm really looking forward to Mix09 in Las Vegas, where a portion of the web community will come together for the 4th year and discuss, with enthusiasm!, the future of the web.

We use SharePoint a lot at imason.  One of the more minor [but very useful] ways we use it it to access a common set of powerpoint slides for use in the dozens of presentations we create every week.  Here's how it works at imason...

Install the template:

1.  Go to {insert location of template document library}

2. Click on “imason standard powerpoint template”

3. At the prompt, choose to open it in read-only format

4. Save the powerpoint as a template



IMPORTANT:  The file location should be something like C:\Users\{username}\AppData\Roaming\Microsoft\Templates.  If it’s not, change the “Save as type” to “Powerpoint Presentation” and then back to “PowerPoint Template”.  This should set the path properly.

5. Once saved, close the template.

Hook up the Slide Library

1. Open up PowerPoint.

2. Go to “File | New|

3. Then go to “My Templates”

and select “imason standard powerpoint templates”

4. On the Home tab of the ribbon, select “Insert Slide” | “Reuse Slides”

5. This open the Reuse Slides taskbar.  For the slide library select {insert location of your slide library}.  This opens up a super slick slide library where you can grab ‘stock’ slides that imason uses.  Be prepared to be amazed!  Select a slide and wham-O, it’ll be in your deck.

6. One additional trick – select “group by presentation title” – this will organize slides into logical groups like “Company Overview”, “Consulting – UE”, “Consulting – Process” etc…

imason's Ivan Neganov recently presented at the Toronto SharePoint Camp.  In his session, he talked about automated deployments of SharePoint/MOSS-based solutions.  Deployment with SharePoint/MOSS is among the least-understood areas, and it is not trivial.  Most people resort to the first option (the top of the diagram below) - a manual deployment. 

Let's say you're on a project with 10 Iterations and therefore 10 deployments.  Each deployment will take 8 hours to do (definitely not an unrealistic estimate).  So, you'll spend 10*8 = 80 hours deploying.

Now, contrast this to the second option (the bottom of the diagram).  You want to invest, upfront, in automated deployments (this is represented by A).  But how much should you invest?  Well, lets say each automated deployment (little 'a') takes one hour, and you have 8 deployments to do.  So 8*1 = 8 hours spent deploying.  If you compare this back to the first option (80 hours spent), this means you have about 72 hours to invest upfront on building the automated deployment (A).  Considering all the benefits of automation (better quality, fewer errors, repeatable code....) this would seem to be a good investment.

I like how Ivan made the comparison numerical - gives a good sense for how much investment makes sense for automated deployments.

Some of imason's Sharepoint gurus presented this past weekend at the Toronto Sharepoint camp.  Boyan, Jim, Bob and Ivan all presented.  A good review of the day can be found here, including this high-praise for Jim & Boyan.

Jim Schwartz and Boyan Tsolov showed how they have managed to stretch InfoPath and SharePoint beyond the typical limits. I cannot describe how exciting it was to see Jim and Boyan actually reproduce many of the methods they used to implement a highly successful InfoPath project in front of us all, with full explanations and helpful suggestions. You can find the material here: http://www.imason.com/blogs/jim_schwartz/archive/2009/01/25/advanced-infopath-development-with-sharepoint.aspx and I highly recommend giving it a review. If everyone at imason is as friendly and intelligent as these two guys then I expect to see them be pretty darned successful and am rather envious.

Well done guys and a great job by the organizers!

Predictions are a risky business, but they're fun to read and fun to make nonetheless.  I generally find I'm less wrong when I make fewer predictions, so here are my best 4 for 2009.

Fellow RD Barry Gervin makes some great 2009 predictions himself (couldn't agree more with the prediction about the death of optical media) as does Guy Barette with his classic French Canadian humour.  You can also catch other Regional Directors' predictions at The Region.

1.  The reign of "Data Tier == Database" is over... finally.

For 10 years, "data tier" has been synonymous with database.  Sure, clever architects have created fancy schemes that hid the database behind a COM object / .Net Object / WSDL / WCF or what have you, but behind it all was a database.  More often than not, it was the database that was:

• The scalability bottleneck
• The most difficult thing to scale (scale up, not scale out)
• The most difficult thing to troubleshoot
• The most expensive thing to scale

Thankfully, this era is coming to an end.  What will replace this yeoman in the data tier?  Search Platforms. The market leaders are Autonomy/Verity, Endeca and FAST.  This market has already consolidated with Microsoft's purchase of FAST last year.  Endeca, I believe, is looking for a buyer.  There are other smaller players, of course, like Google and Microsoft's Search Server and Sharepoint Search line.   Search Platforms will stand side-by-side with Database Platforms in the data tier.  In classic CRUD (Create, Read, Update, Delete), the Search Platform will take on Read, and for now the database will keep Create|Update|Delete. 

This is the biggest change we'll see in Enterprise Architecture this year, and probably for this decade.

2. Silverlight will support desktop applications... finally

Both Adobe (with AIR) and Sun (with JavaFX), provide a rich application platform that is not dependent on the browser.  I pressed Scott Guthrie to support desktop application with Silverlight as a core scenario back in March, 2008.  It wasn't on the table then, but I think it'll come to Silverlight nonetheless, hopefully in version 3.  Sure you can work around the limitation as we did with The Weather Network (Jim provides the details here), but the workaround is not without its drawbacks. 

Microsoft will see that the future of the web is in rich applications that run both inside and outside the browser, and provide outside-the-browser support in Silverlight.  This doesn't mean WPF is dead; quite the contrary.  There's a rosy future for WPF - there's huge potential in the touch/surface applications which will likely need more graphic horsepower than it makes sense to pack into Silverlight and the touch/surface applications don't need the cross-platform support that Silverlight provides.

3. Silverlight desktop penetration will approach Flash (80+%)

We only recently got the first information about Silverlight penetration coming in at 25%.  By far, penetration on the desktop (or lack thereof), is the biggest barrier for consumer sites to build applications on Silverlight.  But the momentum is huge, and I predict that by the end of 2009 we'll see Silverlight penetration approaching the range of Flash (80+%).  If Microsoft can do this in 2009, it stands to dominate the RIA market in 2010.  If it can't, Flash stands to reign supreme though 2010 if not longer.

This is a major undertaking.  It is widely held that Flash is the most installed software in the world so even getting into the 80% ballpark will be a major success.

4.  Windows Mobile as we know it is over... finally

Windows Mobile has not delivered.  By most accounts, the iPhone (v1) has surpassed WinMo (v6) in market share already.  If there's one thing Microsoft can't stomach, it's losing to a new entrant to a market.  Ballmer and Co., are going to change directions (read:  major disruption).  How might this happen:

1. A Microsoft-branded phone.
2. A Zune phone.
3. Blowup the WinMo team
4. Microsoft Buys HTC (or another phone hardware company)
5. Toss the long-held strategy of device-independent operating system (ala PC)
6. Microsoft exiting the mobile market

In this list I think 1. and 2. are exceedingly unlikely as this would alienate Microsoft's partner channel around mobile.  For the same reason, this eliminates option 4.  I think 6. is also exceedingly unlikely, because being relevant on Mobile is a cornerstone in Ray Ozzie's strategy.  But it's possible they switch strategies wrt "relevance" by betting on Silverlight Mobile vs Windows Mobile.  Unlikely, but possible.

This leaves item 3 (blowup the team) and item 5. (abandon 'device-independence') as the best contenders.

Like I said, making predictions is fraught with peril.  But it is fun.