--On Wednesday, March 16, 2005 11:47 AM -0800 Alfred Reynolds <[EMAIL PROTECTED]> wrote:
A content server should never fail in this way hence the abrupt client behavior when it does. If we can track down which machine does it we can debug the problem and fix it on the server end (that is if it is actually a server problem and not some other networking/client issue).
The word "should" sets off the alarm bells. You guys know about "defensive programming", right? Even if you fix the server end, the client should still get fixed to handle the exception more gracefully.
(Apologies if I seem to have the zealotry of an ex-smoker, but I've made the same mistake in trusting external data (even when I thought I'd caught all the cases) and want to spread the gospel. ;) )
(And I realize that fixing a client updater is similar in nature if not severity to a BIOS update, in that it has potential for support costs from clients too dim to know how to get an update from a website if the regular updater crashes and burns.)
_______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
