On 2009/12/15 07:36:39, robertvawter wrote: > > Is it better to have this fail at runtime than to require people to update > their > > custom schedulers (by making these abstract)?
> Do you think it's worth breaking source-compatibility for this class between 2.0 > and the next 2.x release? I'm fine with it, but would like to get sign-off > before making a breaking change. It seems likely to be a breaking change anyway, just at runtime instead of compile time. As we add more uses of this facility in our widgets, it seems unlikely that anyone wouldn't run into it. Also, it seems unlikely anyone has implemented a subclass of this, so better to fix it now. http://gwt-code-reviews.appspot.com/119803 -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
