I foresee a very vigorous, not to say rancorous, debate over whether J2000 is kept in perpetuity or whether we move to J2050 within this decade.
Historically celestial coordinates have been defined in terms of the Earth’s mean equator and orbital plane at a particular epoch but we now have the ICRF which is based on VLBI measurements of objects in the distant universe. The ICRF is supposed to be completely decoupled from the Earth’s rotation and its motion but its coordinate axes are (almost) aligned with J2000 for convenience. Adopting a new epoch, such as J2050, would break that close alignment so I can’t see it happening.
Given that (most) people seem happy to decouple civil time from the Earth’s rotation I don’t think that it is a particularly big deal to do the same with equatorial coordinates. Anyone who needs to can precess a position to any date they want and the benefits of having a fixed celestial coordinate system significantly outweigh the drawbacks.
It so happens that, as VLBI measurements get more precise, it seems that there are still some dependencies on the Earth’s motion in ICRF (secular changes in aberration due to our motion around the galactic centre for instance) so we haven’t quite managed to eliminate our Earth-centred view of the universe from the celestial coordinate system as yet.