On Monday 12 Sep 2011 Kevin Krammer wrote: > > Either I'm misunderstanding you, or that is unbelievable. From the > > earliest days of basic coding you could set a timer so that if no > > response was obtained by then the loop exited. Are you saying that 30 > > years on this is not possible? > > Ah, but the timer won't fire because the process is executing the other > stuff. For any kind of interrupt to work one needs at least a second > execution context. Even then cancelling an operation might not be possible > if it wasn't designed for that or is blocking in a call to some external > code, e.g. a file read. > > Things can usually be done more asynchronously than they are done now, but > it is way more complicated which is why initial implementations of things > are almost always synchronous. > > The case of something blocking due to networking is definitely fixable and > should have been asynchronous from the start. I recommend letting the > respective developer know about this limitation. That was my feeling, too. Thanks for the explanation. Anne
Attachment:
signature.asc
Description: This is a digitally signed message part.
___________________________________________________ This message is from the kde mailing list. Account management: https://mail.kde.org/mailman/listinfo/kde. Archives: http://lists.kde.org/. More info: http://www.kde.org/faq.html.
