Contradictory feedback from my 2005 PDC talk


I was looking through the feedback from my 2005 PDC talk, and I noticed an interesting contradiction. The written feedback indicated that the first half of my talk (wherein I talked about memory management and "paying your taxes") was more favorably-received than the second half (on user interface issues). On the other hand, nearly all of the questions people asked afterwards were about the user interface issues. "Wow, that thing about parent and owner windows, I think that's why my program is crashing," or "Gosh, synchronized input explains this problem we've been having."

I've been told that "the quality of the questions you get after a talk is the best indicator of a talk's success", so I'm inclined to go with the people who were more interested in the second half than the first half. (Of course, the increased interest in the second half might just mean that I did a worse job in that part and people needed to come up to me to clarify the things I did poorly!)

Maybe my next title should be "Story Time with Raymond". It could be the only 100-level talk at the PDC.

Comments (10)
  1. Nish says:

    Raymond

    It’d be cool if you could give a talk to the VC++ and SDK MVPs during the MVP summit. Some of us couldn’t afford PDC :-(

  2. When I watched the video that Scoble posted from your talk, it looked like the audience was more engaged during the first half. There was a lot of laughter and reaction when you made points. Towards the end of the talk, it was pretty subdued. I think the early parts of the talk integrated humor very effectively while the later parts tended to be unceasingly technical.

    Overall I thought it was a very good talk even though I had read previously almost everything you said. Since there wasn’t much new material, the delivery and style was very important.

  3. Matt Pietrek says:

    During the time when I was teaching & training, I learned that feedback is typically all over the map. The low level detail that thrills one person is "boring" to another. It’s very hard to please everybody when your giving highly technical content, rather than

    "story time". Some well known speakers do "story time" very well, and we love them for just that. :-)

    In the end, the best things I got from my feedback was finding the patterns that indicated where my presentations and style needed work.

  4. Stephen Ritcey says:

    I’m not sure this is contradictory at all.

    In my experience, people ask questions mostly

    to clarify connections they’ve made between

    what they know and what you just told them –

    in other words, from the part of the talk they

    understood best. But a smart audience will realize that the best value was in the stuff they knew less about – but needed more time to digest before they could even pose an intelligent question.

  5. Jamie Curtis says:

    Raymond your talk was interesting for the entire session. Having worked with Windows since version 1.0 I found the entire talk very interesting for me (and I was probably one of the few women in the crowd). You speak slowly and very clearly and your content is well organized and fits together nicely. I think you are a breath of fresh air to the world of Microsoft Developement where obscurity to the inner workings of the operating system are perceived to be "handled" by other layers of software that "do it for you".

  6. MehYam says:

    I don’t think there was much to ask about the first part of your session – what I took away from it was to remember the relationship between cache hits and heap performance, and also the cool tidbit about the CLR being smart about this sort of thing. I’ll bring this up next time pimpl comes up in conversation.

    As for the second part, there were simply a lot more details and ragged ends to the topic that I think evoked more questions in the audience. One thing I wanted to ask (and I’ll ask now): is the combination of thread input synchronization problem and Spy++’s use of hooks is what causes Spy to lock up the shell sometimes when you’ve got an app broken in Visual Studio? It makes sense that if the debuggee gets itself in the chain of synchronization that everything else will stop responding.

    If that’s the case, then part two of the question is why I’m still able to Alt-TAB over to my console window when the desktop’s unresponsive like this (the trick is to bring it up and start killing related processes until the shell undeadlocks itself). This might be an interesting topic for a new blog entry.

  7. zzz says:

    why I’m still able to Alt-TAB over to my console window when the desktop’s unresponsive like this (the trick is to bring it up and start killing related processes until the shell undeadlocks itself).

    Something more insightful than "it’s on another thread/process" could be interesting.

    What I would like to know is; now that the shell people are more heavily promoting all sort of property etc handlers for displaying metadata in Vista, how are these going to be handled in Vista when they get stuck or have hole etc. If there is going to be more extensions to the explorer by third parties, I want them to be manageable from well defined "shell/explorer extension management UI", instead of the current "hope the app that installed it lets you uninstall just the extension, or edit registry". We all know these extensions are going to be full of security holes, bugs and so on, especially since they are not recommending doing them through managed-only interfaces. I want MS to assure us that Vista and its shell is running normally even when you have ton of 3rd party handlers full of most horrendous bugs and bad practises imaginable…

  8. Cheong says:

    IMHO, a talk that can bring more "evolved" question is good indicator, as the ability of stimulating the mind of audiences is a valuable quality for the talk.

  9. MehYam says:

    >>Something more insightful than "it’s on another thread/process" could be interesting.

    That would be incorrect anyway – when the shell locks up in fashion, the *only* processes that remains responsive are console windows. They obey some different set of rules than regular windows do.

  10. Lan says:

    I suspect, unless you had a couple of different Q&A sessions, that the real reason for the second half drawing more questions may be far simpler – the second part was fresher in their minds, so that was what they asked more questions about :)

Comments are closed.

Skip to main content