On Tue, May 14, 2013 at 8:59 PM, Harry Chesley <
che...@acm.org> wrote:
> Aha. We force it to use long polling all the time -- our application
> isn't sensitive to efficiency (relatively few users, light load), and
> we're running on Tomcat, so we decided to avoid WebSocket compatibility
> and testing issues. But that's something we should definitely be aware
> of for the future.
>
> As always, your answer is prompt and informative. CometD and us CometD
> users are very lucky to have you responding to list postings. (Having
> been on your side of things, I know you only hear from people when they
> have issues, so wanted to take the opportunity to mention that.)