This is going to be a quick post on using the PivotViewer (both the v1 and the SL5 version). Every once and a while you realize that you do something out of habit and forget to mention it to anyone, and that was the case here.
Silverlight 5 has just been released and it’s a couple of weeks before Christmas, so what would be than to start a new blog series on the “12 Days of Silverlight”? This series is an introduction to some of the new Silverlight 5 features that I find interesting. This is by no means an exhaustive list, but it’s some of my favorites.
So without further rambling….
On the first day of Silverlight, the team delivered to me… native multi-window support.
One of the features that prevented many projects from using the original PivotViewer was the static nature of the collection. Once a collection was loaded it was not possible to modify any of the trading cards within that collection. This limited the potential workflows that PivotViewer could be used to implement. The Silverlight 5 PivotViewer changes all of that. PivotViewer now responds to changes within the collection as well as changes to the individual objects themselves. The best thing about this new feature is that we don’t have to make any changes to the PivotViewer to take advantage of this.
Continuing with the series on PivotViewer, the next new feature to look at is the concept of semantic zoom. “So what is semantic zoom” you ask? Semantic zoom is showing different visual information at different zoom levels. In the case of PivotViewer, the trading card image will change based on how large the trading card is on the screen.
I have had several people ask me why you would want to change the trading card as the size changed. If you think of the normal PivotViewer interaction, you begin with seeing a lot of items on the screen at a relatively small size. The amount of information that you can provide at that level is very small. In fact, typically you are only able to provide one or possibly two pieces of information effectively (where most of the time it is just one). Once you start filtering the data, the number of visible cards gets smaller and, as a result, the size of each card gets larger. This gives you the opportunity to effectively display more information to the user. Once the user selects a trading card then majority of the PivotViewer screen real estate is filled with the selected item. This gives you the best chance of displaying detail information to the user. As you can begin to see, as the user gets closer to the individual trading card, you can effectively show them more information. This is semantic zoom.
With the release of Silverlight 5 just around the corner, I thought I would start a series on the new PivotViewer and how to get up and running with it. We will start the series off by exploring how to create a client side collection.
One of the most anticipated features of the new PivotViewer is the ability to create client-side collections. PivotViewer accomplishes this by utilizing the traditional ItemsSource property for data and a form of XAML data templates to define the trading card. It is interesting to note that the fundamental core of PivotViewer is still using the DeepZoom technology. If you are thinking that DeepZoom only handles images and not XAML, you would be correct. Under the hood, PivotViewer is rendering the data templates to images and then adding those images to DeepZoom. Remembering that little tidbit will help later down the road.
Today I released a new lesson on the PivotViewer Lessons (http://pivotviewerlessons.codeplex.com) project. A commonly asked question I receive is “how do I change the font in the filter pane?” The latest lesson shows an example of how to tackle that specific problem.
You can also take a look at the new release here : http://pivotviewer.championds.com
The new lesson demonstrates how to change the FontFamily for the entire filter pane or just specific regions (ie Category Names).
Detailed posts on each of the lessons and a few more challenging lessons on in the works. There is also the new PivotViewer project coming.
As you might have guessed, I am pretty excited about the next version of the PivotViewer. We received the first glimpse of it during the keynote of the Silverlight Firestarter last December (http://www.silverlight.net/news/events/firestarter/). During MIX last week, Nick Kramer gave a talk on “Advanced Features in Silverlight 5” where we saw the new PivotViewer in a bit more detail. If you haven’t seen Nick’s talk yet, you can watch it online here: http://channel9.msdn.com/events/MIX/MIX11/MED12
Well I just made it back from MIX 11 and it was truly an amazing time. If you were unable to go, make sure you check out http://live.visitmix.com for links to all of the sessions and keynotes. I meet a ton of great people this week and got to reconnect with some old friends.
So I know that this topic is nothing new, however, this is still one of the most common deployment issues that people ask me about. That being the case, I decided to add a post on the topic to demonstrate my approach.
I love the work the Silverlight team has put in to generating client stubs for web services. It was not that long ago that I was having to write out all of those stubs and I’m glad to be done with it. However, there is one piece of the puzzle that seams to be missing from the solution: smooth deployment of the Silverlight application and the WCF services.
This past weekend I gave a talk at Houston Tech Fest 2010 about the Silverlight PivotViewer. For a couple of demos in my talk, I used a collection generated for the MIX 10 video sessions that are online. I decided to move that collection to our labs server and thought I would talk about it a little here first.
MIX 10 Collection : http://labs.championds.com/MIX10/MIX10Collection.cxml