What does “equinox of date used” mean?

The documentation for an API I often use for quick astronomical modeling and figure drawing says

Positions are given in FK5 heliocentric coordinates in the equinox of the date used.

What does "equinox of the date used" mean. Does it simply mean that the epoch for a value at time $t$ is $t$, rather than some other time (e.g. J2000)?

-

"Equinox of date" means that in accounting for general precession, the equinox (or roughly speaking, the origin of the equatorial and ecliptic coordinate systems) is that for the date for which a computation is performed. Alternatively, some standard equinox (e.g. J2000.0) is used. Catalogs are always referred to a standard equinox whereas ephemerides are (usually) referred to the equinox of date. The difference is significant when pointing a telescope.

-
Oops..forgot to mention epoch. For stars, the epoch is the moment for which the objects' positions are precisely tabulated. Transforming to a new epoch basically means correcting for proper motion, the stars' intrinsic motions relative to each other. The situation is different for solar system objects. In that context, epoch usually means the moment for which orbital elements are precisely computed. This gets rather complicated. PLEASE refer to the Explanatory Supplement or my book. – user11266 Dec 13 '12 at 21:55

It means that the function needs to be given a parameter, a date – in the allowed format(s) – that replaces J2000 as the reference date to define the coordinates. Aladin is another package that has J2000 equinox as the default but the date may be changed.

The FK5 coordinates are defined relatively to some positions and orientations of celestial bodies (especially Earth) at a given moment, and that's what the "equinox of the date" means. The word "used" is separate and means that the user has "used" a value of the parameter, too.

The terms "epoch" and "equinox [of the date]" are related in astronomy but they are not quite the same things (click). In general, the epoch expresses changing quantities but neglects the changing definition of the benchmarks and coordinate systems in time, something that equinox of the date takes into account, too.

The term "equinox" in these discussions about coordinates is used "indirectly" – because the position of the vernal equinox is the conventional place that defines many coordinate systems. But the "equinox of the date" of course means something else than just the "date of the [vernal] equinox".

-
You say that "the function needs to be given a parameter" but there is no other parameter, so in this case does that mean that it just uses the argument $t$ as that parameter? – raxacoricofallapatorius Dec 13 '12 at 21:36
Yup, I am afraid that if there's just one date-like possible argument, it will use it for the equinox of the date, too. – Luboš Motl Dec 13 '12 at 21:46
The qualifiers FK4, FK5, and now FK6 refer to very particular reference frames defined not by Earth's position, but by approximations to an inertial frame defined by distant stars, galaxies, and most recently, quasars and radio objects. – user11266 Dec 13 '12 at 21:52