Register?

[icweaver]

Just opening a tracking issue here for getting this much needed package registered.

I guess the question would be where the best home for this would be. Totally happy to keep this in JuliaAstro if that works best, wherever the best discoverability/support would be.

I noticed an old astroalign.py discussion where they were considering using skimages's ransac implementation instead of their own hand-rolled one. I wonder if we should consider something similar on the Julia side, and look into either moving this repo somewhere under JuliaImages, or incorporating it into an existing package there like ImageFeatures.jl, or maybe JuliaStats?

Will ping the folks over in #image-processing to see if we can get their input here too

[cgarling]

Sorry I didn't see this, I forgot to "follow" all activity on this repo when I made it.

My preference is to keep this somewhere I have merge privileges as I've been thinking about implementing some more modern algorithms and would prefer not to wait for external review. I.e., if we moved it to JuliaImages then I would want to be added to that org and given merge privileges for this repo.

I have not seen any interest from that community yet so it seems like this may be staying here.

[icweaver]

I think there's a good possibility that the lack of response just yet is due to #image-processing just being a pretty low traffic channel. My post before last there was already swallowed up by the slackhole, so would you be open to me copying the message to discourse to get more eyeballs?

I can't speak for other orgs, but I imagine there are maintainers at the appropriate orgs that would be open to giving you admin access, even if it's just to your specific repo under their org umbrella.

You've got the final say though. Happy to help support whichever route you feel comfortable with!

[timholy]

This looks like an awesome package, many thanks for contributing it! I'd be happy to give you privileges to move it to JuliaImages if you'd prefer---it's very likely to have many users outside of astronomy. And of course you'd have full maintenance privileges.

[cgarling]

Sure @timholy, I would be happy to move it to JuliaImages.

[timholy]

Strangely my reply seems to have been lost. I think you have to initiate the transfer; I'll accept it and then immediately grant maintainer privileges. I don't think you'll be without privileges for more than a few seconds.

[cgarling]

In order to transfer it looks like I need permissions to create repositories in JuliaImages, so I think I need to be added to the org and given "create repository" permissions. Trying to transfer gives me the error "You don’t have the permission to create public repositories on JuliaImages", see also bullet point 4 of https://docs.github.com/en/repositories/creating-and-managing-repositories/transferring-a-repository.

[timholy]

Sent

[cgarling]

Thanks, looks like the transfer was successful!

[timholy]

Welcome to JuliaImages! It's a very well-designed and engineered package, it's an honor to host here.

If you are interested in any kick-the-tires contributions, I will try to find something to add to an already-great package.

[cgarling]

Thanks for the kind words! If either of you have some spare time I would appreciate help with the readme, I think I'm pretty bad at making them ... typically I just copy an example from the documentation to show in the readme, but it might be nice to have a unique / more interesting example for the readme. A list of use cases/domains where these methods are useful might be nice also (I mostly use them for fitting homography / similarity transformations and other point correspondence models.

Of course, contributions of other algorithms are also welcome but I have limited time for reviewing such contributions at the moment.

[timholy]

@cgarling see #14. I can also submit a doc tweak if you like. (The docs are already very good.)

[cgarling]

Thanks for the help! I think we're good to register now. Any thoughts on debuting at v0.1 or v1.0?