I like the Pull Request model and try to encourage folks to use it always. Using it for feedback sounds like a great way to go. If every contribution where made through a pull request, and committers always applied all contributions by merging the pull requests, there would be no need for this file.
As for editing individual files and using @author, I would prefer to avoid that path as comments in a file about its history can lead to extra "noise" in the form of more history to correct past history. Furthermore, using commit IDs as references helps to clarify just what was done; they definitely do not belong inside an individual source file that is a functional part of the working software. I see a single text file, in which small comments can be added, as very lightweight - easy to use, and right "on point" (i.e. in the code base to which the update was made). And, if all goes well, it never gets used because the commit messages should suffice. We could do the same thing with a page on the ActiveMQ website, but updating that website is harder in my experience. It also splits the history into two places (one in GIT and one on the website). As a developer applying a contribution, I'm already in GIT, so making a second edit there is "at the right place and the right time." -- View this message in context: http://activemq.2283324.n4.nabble.com/Re-activemq-git-commit-Add-MERIT-txt-and-give-thanks-to-Mark-Frazier-for-his-contribution-tp4692815p4692830.html Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.
