One day and i'm done.  More or less.  All of the KMDF and UMDF talks were front-loaded into the conference so we were finished "talking" by 4:15. 

Eliyas and I gave our talk this morning on what's new in KMDF and UMDF.  It seemed to go pretty well, though it's definately one of the least technical talks i've given on the topic.  It's also my least rehearsed - i didn't start running through the slides until I got into my hotel room last night.  Hopefully it didn't show too badly :).  I did a quick poll of the room and most of the attendees were aware of UMDF and KMDF & a good number are using KMDF.  UMDF adoption is sparser (which isn't surprising) but actually the number of hands was better than I was expecting.

I got to poke my head into Landy's Q&A session on the memory manager.  Landy always manages to pack a room and it's easy to understand why. 

The KMDF chalk talk was pretty lively.  Folks have been playing with KMDF for quite a while so there were a number of good questions.  The UMDF chalk-talk had a large number of "newbies" which we honestly weren't expecting.  I think Praveen did a pretty good job of adapting to that - we walked thorugh the skeleton sample briefly then showed debugging some things with the debugger extensions to walk through your driver once it's up.  Shefalli had a little time at the end to run through her slides on testing your UMDF drivers and the tools that are available for that.

Now that Nar's done talking about the I/O manager changes it's time to go back to the hotel.  I'm ready for a nap.

-p