search

UKZN-Astronomy / corrcal

Python/C code for calibration of quasi-redundant arrays. Different (fixed) algorithm from original corrcal
BSD 2-Clause "Simplified" License
1 stars 0 forks source link

Code License #4

Open piyanatk opened 4 years ago

piyanatk commented 4 years ago

We need to declare the code license. Which licensing should we use?

ronniyjoseph commented 4 years ago

Does UKZN have some kind of Intellectual property guidelines/code license?

piyanatk commented 4 years ago

I am flipping through UKZN IP policy, and basically the ownership of the IP is up to determination from several factors, including whether it was significantly developed using UKZN facilities and funds, or have UKZN name in it.

My personal opinion is that the repo must be open source. Since both Jon and you did not declared any licensing, the current codes fall under GitHub default license, which is open source.

Another thing that I think we want to do is that when we have a mature code, we should transfer it to Radio Astronomy Software Group. This will require significant efforts because they have quite a high standard (e.g., you need test build with travis and etc.)

So maybe we should just declare either BDS-2 or BDS-3 license for the time being?

piyanatk commented 4 years ago

Done in https://github.com/UKZN-Astronomy/corrcal/commit/95162e4660905f36fa34f2f2100b6e50adad8377 commit. I submit a pull request together with directory restructuring.

piyanatk commented 4 years ago

Done in https://github.com/UKZN-Astronomy/corrcal/commit/95162e4660905f36fa34f2f2100b6e50adad8377

piyanatk commented 4 years ago

@sievers 's original repository does not declare a license, so by GitHub terms, his codes are all right reserve, and we cannot make modification. I have submitted a pull request to his original repository to add an MIT license.

piyanatk commented 4 years ago

I do not think my PR will ever be merged to the original sievers/corrcal2 repository. Do you think it is okay to drop this issue, @ronniyjoseph @steven-murray? The current repository has been transferred out of the forges, and I have a hope that the Python version in the pipe will replace the current code.

steven-murray commented 4 years ago

I'm not sure what you mean @piyanatk . Are you wanting to setup this version of the code as the primary one, replacing the original repo? And add a new freer license?

piyanatk commented 4 years ago

@steven-murray This repository was originally forked from @ronniyjoseph's forge of sievers/corrcal2 repository. @sievers' original repository does not declared an opensource license, so technically it is all right reserved him. I have his verbal consent to develop this repository. We declared BSD license and have done many improvements. I requested GitHub to transfer it out of the forked tree, so that we can host it at rasg-affiliates at some point. But all of these legally cannot be done until @sievers declared an opensource license. I submitted a PR to his repository, asked him in-person to approve it. He said, yes, but it has not been approved.

steven-murray commented 4 years ago

Ah OK, I guess it's just a waiting game then. I think dev can continue to get it up to speed for rasg-affiliates in the meantime.