Loading ...
Sorry, an error occurred while loading the content.

410 Gone

Expand Messages
  • Randy Morin
    Sam reveals that nobody is listening to his 410 gone RSS responses. I think it s time to adopt a Feed State spec.
    Message 1 of 9 , Apr 5, 2007
    • 0 Attachment
      Sam reveals that nobody is listening to his 410 gone RSS responses. I
      think it's time to adopt a Feed State spec.
      http://www.kbcafe.com/rss/?guid=20070113172842
      As soon as I get a little time, I want to push for adoption.
    • A. Pagaltzis
      ... There are plenty of readers that *do* respect 410. Why would those that don t respect 410 support an extension element or a different document format?
      Message 2 of 9 , Apr 5, 2007
      • 0 Attachment
        * Randy Morin <randy@...> [2007-04-05 16:30]:
        > Sam reveals that nobody is listening to his 410 gone RSS
        > responses. I think it's time to adopt a Feed State spec.
        > http://www.kbcafe.com/rss/?guid=20070113172842
        > As soon as I get a little time, I want to push for adoption.

        There are plenty of readers that *do* respect 410.

        Why would those that don't respect 410 support an extension
        element or a different document format? They're already ignoring
        a response they don't know to deal with properly. Your spec would
        just be one more thing to blithely ignore.

        Assuming your proposal gets any traction, though, aggregators
        that *do* want to play nice would now have to support yet another
        syndication-related spec. It also seems prudent to expect that
        we'll end up with a range of aggregators which support one but
        not the other.

        On balance, it would seem to me that proposing an additional
        mechanism will not make things any better and will likely make
        them worse.

        Regards,
        --
        Aristotle Pagaltzis // <http://plasmasturm.org/>
      • Bill Kearney
        ... Especially since this is well-plowed territory already. There s definite merit to an in-feed mechanism for redirection or state change. It s not without
        Message 3 of 9 , Apr 5, 2007
        • 0 Attachment
          > On balance, it would seem to me that proposing an additional
          > mechanism will not make things any better and will likely make
          > them worse.

          Especially since this is well-plowed territory already.

          There's definite merit to an "in-feed" mechanism for redirection or state
          change. It's not without problems. But I'd rather have a
          belt-and-suspenders approach for allowing a feed that wants to indicate
          state change rather than leaving it up to server-side configuration efforts.
          What's easier for a feed producer, to edit the httpd.conf and/or setup local
          prefs for the web server, or to edit the XML output? For many
          template-driven sites it's easier to edit the output XML not the server
          directives.

          But hey, history's shown this is a "cat herding" problem.

          =Bill Kearney
        • Randy Morin
          I m not proposing anything new. I m just documenting existing practices. The problem is that there isn t anything where developers can find this type of info.
          Message 4 of 9 , Apr 5, 2007
          • 0 Attachment
            I'm not proposing anything new. I'm just documenting existing
            practices. The problem is that there isn't anything where developers
            can find this type of info. That's why it is being ignored. Cause it
            doesn't exist. By formalizing this stuff, we at least give the
            developers something to ignore or not.

            Randy Charles Morin
            http://www.kbcafe.com/rss/

            --- In rss-public@yahoogroups.com, "A. Pagaltzis" <pagaltzis@...> wrote:
            >
            > * Randy Morin <randy@...> [2007-04-05 16:30]:
            > > Sam reveals that nobody is listening to his 410 gone RSS
            > > responses. I think it's time to adopt a Feed State spec.
            > > http://www.kbcafe.com/rss/?guid=20070113172842
            > > As soon as I get a little time, I want to push for adoption.
            >
            > There are plenty of readers that *do* respect 410.
            >
            > Why would those that don't respect 410 support an extension
            > element or a different document format? They're already ignoring
            > a response they don't know to deal with properly. Your spec would
            > just be one more thing to blithely ignore.
            >
            > Assuming your proposal gets any traction, though, aggregators
            > that *do* want to play nice would now have to support yet another
            > syndication-related spec. It also seems prudent to expect that
            > we'll end up with a range of aggregators which support one but
            > not the other.
            >
            > On balance, it would seem to me that proposing an additional
            > mechanism will not make things any better and will likely make
            > them worse.
            >
            > Regards,
            > --
            > Aristotle Pagaltzis // <http://plasmasturm.org/>
            >
          • A. Pagaltzis
            ... Oh! You mean you just want to document how aggregators should respond to the various HTTP status codes? Now *that*, that is definitely a very good idea.
            Message 5 of 9 , Apr 5, 2007
            • 0 Attachment
              * Randy Morin <randy@...> [2007-04-05 18:30]:
              > I'm not proposing anything new. I'm just documenting existing
              > practices. The problem is that there isn't anything where
              > developers can find this type of info. That's why it is being
              > ignored. Cause it doesn't exist. By formalizing this stuff, we
              > at least give the developers something to ignore or not.

              Oh! You mean you just want to document how aggregators should
              respond to the various HTTP status codes? Now *that*, that is
              definitely a very good idea.

              Regards,
              --
              Aristotle Pagaltzis // <http://plasmasturm.org/>
            • Sam Ruby
              ... Yahoo doesn t (document) handling 410, but do support a variety of feed redirection techniques: http://publisher.yahoo.com/rss_guide/faq.php Feedburner s
              Message 6 of 9 , Apr 5, 2007
              • 0 Attachment
                Randy Morin wrote:
                > I'm not proposing anything new. I'm just documenting existing
                > practices. The problem is that there isn't anything where developers
                > can find this type of info. That's why it is being ignored. Cause it
                > doesn't exist. By formalizing this stuff, we at least give the
                > developers something to ignore or not.

                Yahoo doesn't (document) handling 410, but do support a variety of feed
                redirection techniques:

                http://publisher.yahoo.com/rss_guide/faq.php

                Feedburner's techniques are also reasonably sane:

                http://blogs.feedburner.com/feedburner/archives/001251.html

                Just mentioning these as potential input.

                One area that particularly needs clarity is how one should handle
                switching from one format to another. (Say, from Atom 1.0 to RSS 2.0,
                or vice versa). Nick, at least, suggests that his reader will
                transparently handle either format equally, as long as one keeps the
                atom:id / rss2.0 GUID the same:

                http://nick.typepad.com/blog/2006/05/pick_a_format_a.html

                - Sam Ruby
              • Clinton Gallagher
                I concur, the effort to emphasize handling response codes is best expended on informing the ignorant how to best manage this concern. Furthermore, if there was
                Message 7 of 9 , Apr 6, 2007
                • 0 Attachment
                  I concur, the effort to emphasize handling response codes is best expended
                  on informing the ignorant how to best manage this concern. Furthermore, if
                  there was a method to trap and report noncompliance to the public domain the
                  tide would rise and syndication would float a lot more boats. That's fer
                  sure...

                  <%= Clinton Gallagher
                  <%# 9-11 Was An Inside Job
                • James Holderness
                  ... I think that s a fantastic resource and it would be of great benefit to aggregator developers and feed producers if it were more easily accessible. I m not
                  Message 8 of 9 , Apr 7, 2007
                  • 0 Attachment
                    Randy Morin wrote:
                    > Sam reveals that nobody is listening to his 410 gone RSS responses. I
                    > think it's time to adopt a Feed State spec.
                    > http://www.kbcafe.com/rss/?guid=20070113172842
                    > As soon as I get a little time, I want to push for adoption.

                    I think that's a fantastic resource and it would be of great benefit to
                    aggregator developers and feed producers if it were more easily accessible.
                    I'm not sure I necessarily agree with all the SHOULDs, MUSTs and CANs, but
                    that's just nitpicking. Overall I'd be very much in favour of the board
                    adopting something like this.

                    That said, I think a good part of the reason for aggregators not supporting
                    things like 410 is that users typically don't care. If you're an aggregator
                    developer and you have to choose between implementing some new, super-cool
                    widget, or adding 410 support, which one do you think will impress your
                    users more? I doubt permanent redirects and 410 responses are on the top of
                    anyone's most-requested-features list.

                    Regards
                    James
                  • Greger
                    On Sat, 7 Apr 2007 12:01:05 +0100, James Holderness wrote ... I think that the rss specs(different) should do nothing at all to provide fo rfeed states. the
                    Message 9 of 9 , Apr 7, 2007
                    • 0 Attachment
                      On Sat, 7 Apr 2007 12:01:05 +0100, James Holderness wrote
                      > Randy Morin wrote:
                      > > Sam reveals that nobody is listening to his 410 gone RSS responses. I
                      > > think it's time to adopt a Feed State spec.
                      > > http://www.kbcafe.com/rss/?guid=20070113172842
                      > > As soon as I get a little time, I want to push for adoption.
                      >
                      > I think that's a fantastic resource and it would be of great benefit to
                      > aggregator developers and feed producers if it were more easily accessible.
                      > I'm not sure I necessarily agree with all the SHOULDs, MUSTs and CANs, but
                      > that's just nitpicking. Overall I'd be very much in favour of the board
                      > adopting something like this.
                      >
                      > That said, I think a good part of the reason for aggregators not supporting
                      > things like 410 is that users typically don't care. If you're an aggregator
                      > developer and you have to choose between implementing some new, super-cool
                      > widget, or adding 410 support, which one do you think will impress your
                      > users more? I doubt permanent redirects and 410 responses are on the top of
                      > anyone's most-requested-features list.
                      >
                      > Regards
                      > James
                      >
                      I think that the rss specs(different) should do nothing at all to provide fo rfeed states. the transport is over http and will be so, http does the job of telling if it is there or not, if ithas moved or not. it has nothing to do with the feeds internals.
                      my 2c
                      >

                      --
                      http://www.gregerhaga.net/
                      http://hack-space.biz/




                      [Non-text portions of this message have been removed]
                    Your message has been successfully submitted and would be delivered to recipients shortly.