Retirement rules for GZ subjects that are stars or artifacts

[willettk]

Are there any retirement rules in place for GZ subjects besides the 40 classification limit? If not, would it be possible to retire objects that are marked by several users (say, 5 in a row or 10 overall) as stars/artifacts?

Star/artifact classification count is stored in the subject data, I believe - I'm not sure if that would make it easier to handle.

@vrooje - what are your thoughts?

[vrooje]

I'm all for this in principle, but the minimum star/artifact classification varies a lot with dataset, so we'd have to be careful with the threshold. For example, I noticed in GZ CANDELS that f_artifact ~ 0.4 is common for galaxies that are classifiable but with surface brightnesses near enough to the noise level that the whole image appears speckled. If the typical retirement limit is 40 then that's 16 people who've classified an image of a galaxy as an artifact.

Then again, this is less of a problem with SDSS and I suspect it won't be a big deal with Illustris. Not sure about DEcals.