I see that you are indeed claiming that SR will be in error when compared with experimental results, I did not realize that you were making that claim. There is a wealth of observational data to the contrary of your claim. Take reciprocity-- experiments have been done with two observers in different frames, and sure enough their results are entirely understandable within the context of having the same concept of their relative velocity. Super accurate clocks on GPS satellites, compared to similar clocks on the ground, are an example of this. If there were a problem at the accuracy level you quote (5e-7), we would know this. Indeed, down to 1e-11 or better relativity works quite well (generalizing SR to include gravity). I'm really not clear on why, if we have v=d/t and v'=d'/t', that you wish to claim d=d' when that does not hold for either t or v. What do you think is so special about d, anyway?