Yep, I'm aware that as the title holder (owner) of the data, I am the holder of the original copy rights. But in the Homunculus case in point, it was my own **** fool fault for not watermarking the images. When I posted it, I was naïve and never imagined third parties would be scraping the forums for images. My first inkling was a bloke running his own website about AP who actually asked my permission to use one of my Jupiter red spot animations in an article on Jupiter (with credit). Then I found several revisions of one of my DSO images on astrometry.net, where some presumably good Samaritan was trying to figure out why the system couldn't find a platesolving solution.
The rub is two-fold. First, you don't know what you don't know…who knows how many of my pre-2019 images are out there? Fortunately, I don't stay up at night wondering who's profiting off my mistake of posting free candy. Second, once you find one, what outcome do you want? Chasing down a website owner is a real PITA. If the website owner is ethical and the perpetrator was one of their users, then it's just a simple matter of proving the data is yours (the astrometry.net abuse staff immediately removed the offending images). But the website with my Homunculus? I'd be happy to let them use it for image credit. But that site looks quite dodgy—and I don't care to dissect the website governance, find the page owner, and prove to them my image ownership—all just to have them add credit to my nom de guerre.
The Internet is still a global wild-wild west. So for me, copyright watermarks in the images are like like the KAPU signs in Hawaii—and I would actually prefer people with derivative works watermark theirs—otherwise, the data scrapers are screwing over two intellectuals' data rights! Technically, when I post data I am granting a license. Derivative works are also intellectual property with ownership and copyright protections. But those derivative copyright protections derive from the license that was granted with the original data. I can state the license in the post where I share the data, and I can put the license in a text file zipped in with the data. I'm also looking at how to modify the FITS header to add a copyright and a license header entry.
My preferred licenses are the Creative Commons licenses used heavily by Wikipedia users. Since credit to the originator is really my only embuggerance, I'll probably go with the CC BY 4.0 attribution license. If someone finds a way to make money off my images, I reckon good onya mate!