I had the same issue and came up with the same solution a while back myself. So far, I havn't found a way to fool proof the users individual settings, leaving us to assume we have to work around things like caching with a little slight of hand. As long as it's the XML document (e.g., one not bookmarkable) this is perfectly legit.
One way to make it seem less kludgy is to use something meaningful for the random string, like the server time (URL encoded of course, in seconds, ala NotesDateTime object) then at least there is some potentially useful value, giving it an air of legitimacy, in your appended value and it's still relatively unique.
Nick,
I had the same issue and came up with the same solution a while back myself. So far, I havn't found a way to fool proof the users individual settings, leaving us to assume we have to work around things like caching with a little slight of hand. As long as it's the XML document (e.g., one not bookmarkable) this is perfectly legit.
One way to make it seem less kludgy is to use something meaningful for the random string, like the server time (URL encoded of course, in seconds, ala NotesDateTime object) then at least there is some potentially useful value, giving it an air of legitimacy, in your appended value and it's still relatively unique.