Even if they were, I would expect the adjustments to not be that difficult. This sort of temporal logic already accepts and encompasses many of the issues relativistic time would raise. For instance, computer science already has to handle the possibility that event X will occur but remote site Y literally never hears about it, or hears about it many, many, many multiples of the fundamental speed-of-light latency later, arbitrarily interleaved with any number of other events in the mean time. It would be less of a shock to computer science than it was to physics.
The relevant result from relativity is that you cannot rely on synchronized clocks to construct an unambiguous ordering of events. Latency by itself is not sufficient to cause this, as the system could still wait until the message has been fully distributed, then use the timestamps to reconstruct the timeline.
However, CS does have to deal with this issue for the simple reason that we are already unable to synchronized our clocks to the extent necessary to avoid it (or, at least, it is not worth the engineering effort to do so).