Comments

Video updated!

Brian Hartman (http://blogs.msdn.com/brianhartman/) just informed me that you have to add a reference to Microsoft.ReportViewer.DataVisualization.dll if you use the chart or gauge visualization. 

 

I will update the screencast when my schedule permits...

 

Thanks Brian!

Here is the response I received:

"The main site feed is all the content that goes to the home page.

 

There are other feeds, like one for your blog, one for all screencasts, one for all media, etc… that would contain this post."

However, I notice that when you navigate to http://channel9.msdn.com/Media/Screencasts/ and click the "Subscribe" icon, you get redirected to the main site feed (http://channel9.msdn.com/Feeds/RSS/).  I'm following up and will report back with how one is supposed to discover the "all screencasts" feed.

-Marc

I'm not sure, but I will try to find out.

-Marc

I actually had WPF pushpins originally, but chose not to go that route.  That's not to say you couldn't.  In fact, I actually use an invisible (Opacity=".01") WPF button over the pushpin rendered by the map.  That's how I chose to implement the ability to hover over the pushpin and get a WPF InfoBox (implemented as a UserControl).  If you really want to have WPF based pushpins, just comment out line 387 in VEMap and comment out line 401 which is the call telling VE to add the pushpin.

I chose not to use WPF pushpins for two reasons.  First, it seemed like overkill to render such a small UI element as a vector graphic since pushpins are typically fixed width and height.  Second, due to the delay in event notification between JavaScript & managed code, the repositioning of the pushpins is a little “jerky.”  So I am a bigger fan of drawing your pushpins with a vector drawing tool, and then saving them as bitmaps.  You get the same visual experience with better overall perf.  Your mileage may vary.

Do you mind explaining why you believe this is a terrible programming practice?  I can guess that you might think the "var" keyword means that "request" is a variant.  Not so.  What you are seeing is a new feature of C# 3.0 called type inference.  The variable "request" is strongly typed.  In this case, "request" is of type WebClient.  Essentially, the way it works is the compiler evaluates what's on the right hand side of the equals sign and ensures that the variable on the left hand side is typed appropriately.

You can see http://msdn2.microsoft.com/en-us/library/bb383973.aspx for more details. 

Personally, I think using the var keyword is a matter of preference, but not a terrible programming practice as you suggest.  The beauty is that the following two lines of code are identical to the compiler:

var request = new WebClient();
WebClient request = new WebClient();

I don't think there is necessarily a right or wrong answer here.  I think it is a bit of a stretch to suggest I have declared request improperly.  It would be more accurate to say that I have declared request in a way that you dislike for valid reasons.  You make some really good points about the side effects of using var.  I appreciate you making these points because it is very important for people to understand the side effects you mention.  Unfortunately, I can't cover every detail about all the technology in use in a screencast/demo.  If I did, the screencast would have been at least twice as long:).

You may be right about laziness.  I find myself using var in a number of scenarios I suspect you would dislike.  Maybe I am using it so much because I’ve done quite a bit of JavaScript/AJAX programming of late and it has become a habit.  Who knows?  I think one could argue a number of technologies such as intellisense, code snippets, and many more have made me a lazier programmer over the years.  I surely know that I can't code in notepad or vi like I used to because of my dependency on both IDE and language features.  Then again, it is also fair to say that the productivity gains outweigh the dependency on the features. 

With the var keyword, it seems the documentation at http://msdn2.microsoft.com/en-us/library/bb383973.aspx would somewhat support your statements since it says “It is recommended to use var only when it is necessary, that is, when the variable will be used to store an anonymous type or a collection of anonymous types.”  However, like any documentation, it is a recommendation.  I still believe the use of var is a matter of preference.  Am I misusing the var keyword?  That’s in the eye of the beholder.  I  know where you stand:). 

The first release is now available.  Details here.

UPDATE:  Folks are experiencing issues sending to the contribution email address.  This has been fixed.  Details here.

This feature is back in ASP.NET 3.5 / Visual Studio 2008.