Windows Phone 7 and Silverlight Slider Binding Bug

It took me about 4-5 hours today to finally recognize why one my custom controls behaved in a strange way. It is a control derived from the Silverlight slider control and I encountered strange behavior when binding the Maximum and Minimum properties. After countless checks to make sure, that my ViewModel logic was flawless I finally discovered that … the order of attributes in the xaml code *does* matter when binding these exact properties.

It turns out, instead of this

<Slider Name="slider" Grid.Column="1" Minimum="{Binding MinValue}" Maximum="{Binding MaxValue}" />

it should be this:

<Slider Name="slider" Grid.Column="1" Maximum="{Binding MaxValue}" Minimum="{Binding MinValue}" />

(Source: http://social.msdn.microsoft.com/Forums/en-US/silverlightgeneral/thread/5a08a3c7-66d3-4ee0-aa29-43bb8d186ab7)

Other people have noticed the same: http://dotnetbyexample.blogspot.com/2009/08/bind-silverlight-3-slider-value-minimum.html


Als grosser Kindle Fan erzaehlen mir auch Leute staendig von dem tollen Gefuehl ein Buch aus Papier in der Hand zu halten… passend dazu gerade gelesen:

Neben dem wirtschaftlichen haben die Verlage noch ein anderes, ein emotionales Problem. Ein Insider formuliert es brutal: “In den Verlagen sitzen oben oft alte Menschen, die sehr konservativ sind.” Positiver könnte man formulieren: Büchermenschen lieben Bücher, interessieren sich für Papierqualität, Bindung und Typografie.

Quelle: http://www.spiegel.de/netzwelt/gadgets/0,1518,791813,00.html

Data is everything!

As we all know data is everything in optimizing and improving web pages. Until last week Impera lacked the functionality to display data over time. Even though we were able to show some snapshots of certain metrics (in German):


there was no continuous tracking and measuring enabled. On Friday I finally wrote some background scripts which save certain metrics at fixed time intervals. These can then be displayed using the amazing jQuery flot graph library as beautifully formatted graphs over time. Here you can see the average number of online players per hour (GMT+1 timezone and mostly German users from GMT+1):


I think the results are as expected with a peak during the day and a valley after 2 a.m. However, this data is only aggregated over a timespan of two and a half days. I expect the following graph to be more of interest when a few weeks of data have been captured:


Even with this little data it can be inferred, that Impera players get up earlier on Mondays than on weekends (what a surprise!).

There are a few other graphs available but more data is needed for them to be of any interest.