Details

    • Type: Wish Wish
    • Status: Closed
    • Priority: Minor Minor
    • Resolution: Fixed
    • Affects Version/s: None
    • Fix Version/s: 2.0-alpha1
    • Component/s: None
    • Labels:
      None

      Description

      Logging this as a bug so that it eventually gets resolved.

      From log4j-dev thread "Future development of log4j 2.0"

      On 20/02/2010 12:42 AM, Christian Grobmeier wrote:
      there might be other discussions which could
      lead to frustrations on several sides. I think on discussions like
      having author tags and such.

      On Feb 19, 2010, at 5:57 PM, Ceki Gülcü wrote:
      Author tags is a way of recognizing contributors. Recognition is a
      very important aspect of oss.

      Author tags have been discouraged by the board, http://www.apache.org/foundation/records/minutes/2004/board_minutes_2004_02_18.txt (section D)

      This statement by Dirk-Wlllem van Gulik, then president of the ASF, expands on the rationale:
      http://marc.info/?l=xml-cocoon-dev&m=107787986409413&w=2

      Don't think we need to tackle the issue until there is code someone wants to tag. I'm not planning on tagging any code that I initially develop.

        Activity

        Hide
        Ralph Goers added a comment -

        Closing since there have been no other comments since this was marked resolved.

        Show
        Ralph Goers added a comment - Closing since there have been no other comments since this was marked resolved.
        Hide
        Ralph Goers added a comment -

        Log4j 2 does not contain author tags.

        Show
        Ralph Goers added a comment - Log4j 2 does not contain author tags.
        Hide
        Gary Gregory added a comment -

        +1 to remove author tags.

        Show
        Gary Gregory added a comment - +1 to remove author tags.
        Hide
        Christian Grobmeier added a comment -

        Thank you Ralph.

        I am +1 on removing all author tags from the code. There is a team list which is imho enough. The question "who wrote the class" is not easy to answer. Classes become refactored. Who wrote it then, when it just contains the old name? Does one deserve an author tag when he fixed two typos? And so on. For me author tags are ego driven and in my whole career I never had a need for them.

        Instead, a contributors list on the website is fine. In addition there is of course need to name Ralph as the innovator of log4j2. A good place is the projects history. Without his super powers we would not discuss this now.

        Speaking personally, I really don't need author tags telling peopl e what I did in code. I mean there is no place what I did for this project OUTSIDE the code, and OSS does also live of blogposts, print articles, cvonference speakers and so on.

        Ralph removed the tags in 2010, I am closing this issue. If we need more discussions, please lets do that on the mailing list (and reopen this one, if needed)

        Show
        Christian Grobmeier added a comment - Thank you Ralph. I am +1 on removing all author tags from the code. There is a team list which is imho enough. The question "who wrote the class" is not easy to answer. Classes become refactored. Who wrote it then, when it just contains the old name? Does one deserve an author tag when he fixed two typos? And so on. For me author tags are ego driven and in my whole career I never had a need for them. Instead, a contributors list on the website is fine. In addition there is of course need to name Ralph as the innovator of log4j2. A good place is the projects history. Without his super powers we would not discuss this now. Speaking personally, I really don't need author tags telling peopl e what I did in code. I mean there is no place what I did for this project OUTSIDE the code, and OSS does also live of blogposts, print articles, cvonference speakers and so on. Ralph removed the tags in 2010, I am closing this issue. If we need more discussions, please lets do that on the mailing list (and reopen this one, if needed)
        Hide
        Ralph Goers added a comment -

        I have removed author tags for the classes in my experimental branch.

        Show
        Ralph Goers added a comment - I have removed author tags for the classes in my experimental branch.
        Hide
        Ceki Gulcu added a comment - - edited

        Let's say that author tags remains a question awaiting consensus, with the board leaving the door open for author tags to be adopted or rejected by each individual PMC. With regards to people contacting the author directly upon reading the author's name in a tag, I can confirm that it occurs, but the phenomenon is entirely marginal, at least from my perspective. Receiving an email once every six months (my experience) cannot be considered an issue. Of course, your experience may vary.

        Having one's name on a team list both for committers and contributors, is a different way of providing acknowledgment. The team list answers the question "who is working on project foobar?". An author tag answers a different question. "Who wrote java class org.foo.bar.SoylentGreen?". Users will stumble upon authors tags while perusing source code. One does not stumble upon team lists as causally or as often.

        As Ralph observes, information about authorship for a given piece of code can be obtained via SVN. But this involves an active command while authors tags placed at the top of each class get as many eye-ball hits as there are readers of that class.

        Incidentally, in french copyright law, the author name can never be removed even by the copyright holder. According to french law, one can own copyrighted material but only the author can claim authorship and the publisher can never deny attribution. This is called right of attribution. Contrary to copyright, right of attribution is perpetual. One might pooh pooh the french for various reasons (as I often do) but the distinction they are making between proprietary rights and moral rights is quite interesting. For more details about this topic see http://en.wikipedia.org/wiki/French_copyright_law

        Show
        Ceki Gulcu added a comment - - edited Let's say that author tags remains a question awaiting consensus, with the board leaving the door open for author tags to be adopted or rejected by each individual PMC. With regards to people contacting the author directly upon reading the author's name in a tag, I can confirm that it occurs, but the phenomenon is entirely marginal, at least from my perspective. Receiving an email once every six months (my experience) cannot be considered an issue. Of course, your experience may vary. Having one's name on a team list both for committers and contributors, is a different way of providing acknowledgment. The team list answers the question "who is working on project foobar?". An author tag answers a different question. "Who wrote java class org.foo.bar.SoylentGreen?". Users will stumble upon authors tags while perusing source code. One does not stumble upon team lists as causally or as often. As Ralph observes, information about authorship for a given piece of code can be obtained via SVN. But this involves an active command while authors tags placed at the top of each class get as many eye-ball hits as there are readers of that class. Incidentally, in french copyright law, the author name can never be removed even by the copyright holder. According to french law, one can own copyrighted material but only the author can claim authorship and the publisher can never deny attribution. This is called right of attribution. Contrary to copyright, right of attribution is perpetual. One might pooh pooh the french for various reasons (as I often do) but the distinction they are making between proprietary rights and moral rights is quite interesting. For more details about this topic see http://en.wikipedia.org/wiki/French_copyright_law
        Hide
        Ralph Goers added a comment -

        While author tags have been discouraged by the board it is up to each PMC to determine whether or not to use them. Personally, I don't like author tags. They encourage people looking at the code to email the individuals directly instead of contacting the mailing list. While some may desire this as a part of their business model, I find it particularly annoying when people email me off list and always respond by pointing them back to the list.

        I also think that having one's name on a "team" list, both for committers and contributors, is a far more effective way of providing acknowledgement. In addition, anyone can find out who the actually contributed to what through a few simple svn commands. Again, it would be far more effective to provide a web interface to the source code so people can look at the history and do diffs right from the log4j web site than using author tags.

        Show
        Ralph Goers added a comment - While author tags have been discouraged by the board it is up to each PMC to determine whether or not to use them. Personally, I don't like author tags. They encourage people looking at the code to email the individuals directly instead of contacting the mailing list. While some may desire this as a part of their business model, I find it particularly annoying when people email me off list and always respond by pointing them back to the list. I also think that having one's name on a "team" list, both for committers and contributors, is a far more effective way of providing acknowledgement. In addition, anyone can find out who the actually contributed to what through a few simple svn commands. Again, it would be far more effective to provide a web interface to the source code so people can look at the history and do diffs right from the log4j web site than using author tags.

          People

          • Assignee:
            Ralph Goers
            Reporter:
            Curt Arnold
          • Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Development