search

rordenlab / dcm2niix

dcm2nii DICOM to NIfTI converter: compiled versions available from NITRC
https://www.nitrc.org/plugins/mwiki/index.php/dcm2nii:MainPage
Other
899 stars 229 forks source link

Archive old dcm2niix repository? #865

Closed DimitriPapadopoulos closed 2 months ago

DimitriPapadopoulos commented 2 months ago

Describe the bug

How about archiving https://github.com/neurolabusc/dcm2niix, after adding a redirection to this repository? Unless you no longer have the identifiers associated with the neurolabusc organisation.

That would prevent confusion around the reference repositories.

DimitriPapadopoulos commented 2 months ago

I don't think https://github.com/rordenlab/dcm2niix/pull/864#issuecomment-2367784876 addresses the issue at hand.

DimitriPapadopoulos commented 2 months ago

I thought repository neurolabusc/dcm2niix is obsolete and repository rordenlab/dcm2niix is the current one, with the policy by the BIDS maintainers being implemented as different branches development / master in the latter.

Are you saying that the policy by the BIDS maintainers is implemented as different repositories neurolabusc/dcm2niix / rordenlab/dcm2niix?

neurolabusc commented 2 months ago

I do not understand the question. The repository dcm2niix has 228 forks. While at your request I did archive the my own personal fork and include a message in the readme redirecting to the root repository, I do not understand the need for disambiguation. Each fork lists its root fork, I think there are many legitimate active forks to handle unique situations of specific users or explore experimental features. By design, we use a permissive license that encourages others to create their own custom forks. I do not control these forks nor does it seem appropriate to archive these forks that may be appropriate for specific use cases.

DimitriPapadopoulos commented 2 months ago

I know that Git is not a centralised VCS. However, one repository usually plays a more central role than others, and that special repository needs to be identified.

At some point, I had cloned neurolabusc/dcm2niix as my upstream Git repository. Either it was indeed the primary repository at the time, where most of the development happens, or it was my mistake - at the time I think repositories were not as clearly identified as forks of another repository as today. Now, a few years later, I was tempted to keep using neurolabusc/dcm2niix because it was not identified as obsolete. I ended up finding rordenlab/dcm2niix, using the forking information exactly as you write. But it took me some time.

neurolabusc commented 2 months ago

OK, but I archived the neurolabusc repository and clearly redirected to the main repository before I closed this issue, and now you have re-opened this. I think this issue is closed as resolved. Unless you can provide a rationale I think this issue should be closed as resolved.

DimitriPapadopoulos commented 2 months ago

I did not repoen after you archived the neurolabusc repository and clearly redirected to the main repository, but before that.

But most importantly, thank you very much for this great piece of software!