kou commented on issue #47223: URL: https://github.com/apache/arrow/issues/47223#issuecomment-3153649675
> Do we have any idea where/when the project started using the apache remote name? I think that https://github.com/apache/arrow/pull/626 is the first commit that we use "apache" as the remote name of apache/arrow in apache/arrow. Some other Apache projects also use "apache" as the remote name. For example: * https://github.com/apache/orc/commit/5b9aef1abad247e6b6f83756359e2d204093e580 * https://kyuubi.readthedocs.io/en/v1.9.1/contributing/code/release.html So, "apache" remote name may be common in Apache projects. > I am fine with this change. I am quite used to use `apache` as my remote for any of our arrow related repositories but happy to use upstream for consistency: Wow. FYI: Recent apache/arrow-* repositories such as apache/arrow-js and apache/arrow-swift don't require "apache" remote name. They use separated cloned repositories instead of using "apache" remote in the cloned fork repository. > are there any downsides to using upstream here instead of an explicit remote name? If a release manager has any other repositories that the release manager may push to "upstream" remote directly as a normal operation, the release manager may push to "upstream" remote (apache/arrow) accidentally. I suggest the following for the case: * Use "https://..."; not "git@..." for all "upstream" remotes (Make all "upstream" remotes read-only) * Use separated cloned repositories for apache/arrow and forked apache/arrow -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: github-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
