>>13356578The object might change, as you said. The ruler might change, you the observer might change. You might or might not be aware of those things changing, but they would affect the property you are presuming to measure in imperceptible ways. You can only define the measurement that you took in relation to the parameters you know to the extent that you know them. So when reporting your measurement (if you want to be clear how close it might be to absolute truth it might be) you'd report the (for example) time, temperature, atmospheric pressure, and any other parameter that might influence. Then you have the conversation all over again about how accurate any one of those is.
It's a minor point really and practically irrelevant most of the time. But y'know, there it is.
