Full error message: No generic method 'Sum' on type 'System.Linq.Enumerable' is compatible with the supplied type arguments and arguments. No type arguments should be provided if the method is non-generic."

Showing in design time, working at run time. 

Expecting the popup to be kept open, its zorder is still on top of the grid it is linked on, but not on top of other applications whose zorder is closer to the user than the application window containing the grid.

The reason for the performance issues is the size of the ItemsSource for the combo column and the internal lookup logic.

A new IsLightweightModeEnabled property of GridViewComboBoxColumn will be introduced with R2 2016. Its default value is False. 
When set to True, the new lightened lookup logic is triggered, thus the performance of the column is significantly increased.

I want to get ItemsSource by retrieving a list from a dictionary, so I need 2 parameters, but I can't use MultiBinding to do this!

Update: Before our change the GridViewGroupFooterCell content was right aligned and the GridViewFooterCell content was left aligned. We decided to unify them by setting default left alignment to both. This can be easily modified by changing the Column FooterTextAlignment property to Right or by using the following implicit styles: 

<Style TargetType="telerik:GridViewGroupFooterCell" BasedOn="{StaticResource GridViewGroupFooterCellStyle}">
    <Setter Property="HorizontalContentAlignment" Value="Right"/>
</Style>
<Style TargetType="telerik:GridViewFooterCell" BasedOn="{StaticResource GridViewFooterCellStyle}">
    <Setter Property="TextAlignment" Value="Right"/>
</Style>

The fix is available in R3 2017 of UI for WPF.

There is a mistake in the text shown in the GroupPanel of RadGridview in french.
it should be something like "Glissez un en-tête de colonne et déposez-le ici.."

The problem should be resolved with lib version 2015.3.1019.

It should be resolved with lib version 2015.1.504.

When Unicode characters are copy pasted from grids via CTRL + C and then pasted into Excel or Word (for example) as a HTML table, it seems like for every Unicode character a certain amount of bytes / characters are cut off from the end result.