rws@EXPO.LCS.MIT.EDU (Bob Scheifler) (01/14/89)
I was
under the impression that the protocol had not changed between 11.2 and
11.3 in any way. So how can there be any problems?
Actually, there were minor protocol changes between R2 and R3 (see the file
doc/Protocol/CHANGES in R3), but it's pretty hard to believe that any of these
changes would cause real problems. Ever hear of "bugs"? There are bugs in R2
servers that some R2 clients either depended on or didn't exercise; there are
bugs in R2 clients that have been corrected. An example (in the other
direction) is that the R2 xterm had bogus code that causes it to paint wierd
against and R3 server with real ink metrics. An R3->R4 example is that
clients (like the vanilla R3 xterm) that attempt to set invalid bits in the
do-not-propagate event mask will fail against an R4 server.
I don't think we have any compiled list of what bugs might be exercised
when running R3 clients against R3 servers. If you have specific examples of
problems you want to describe, we can try to figure out what's causing them.