> Only the co-ordinated time is loosey goosey because it is based on an approximation of the length of second holding the duration constant, rather than the length of a second being based on being exactly 1/86400 of a day (and thus variable in length). In other words, multiplying a difference in Julian day numbers by 86400 will give you seconds, but they will be [UT1 seconds](https://en.wikipedia.org/wiki/UT1#Versions), tied to the rotation of the Earth, not [SI seconds](https://en.wikipedia.org/wiki/Second), tied to the atomic clocks. Since the difference here is only a few leap seconds, it's probably not important. So it goes.