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

Re: RSS Profile 1.13 Published

Expand Messages
  • Randy Morin
    If we call this an error, I ll propose a change to the spec to allow extension attribute. Please try and be civil instead of trying to create an issue. This is
    Message 1 of 15 , Oct 3, 2006
    View Source
    • 0 Attachment
      If we call this an error, I'll propose a change to the spec to allow
      extension attribute. Please try and be civil instead of trying to
      create an issue. This is an obvious compromise.
      Thanks,

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

      --- In rss-public@yahoogroups.com, Sam Ruby <rubys@...> wrote:
      >
      > Randy Morin wrote:
      > > This is exactly what I was afraid of. Please treat these as warnings,
      > > not errors. The schema attribute is harmless.
      >
      > From a coding perspective, the difference between coding a check for a
      > warning and coding a check for an error is inconsequential.
      >
      > I'll hold off for a few days, but in the end, I would rather code to
      > what the profile actually says, and not based on input I see in an
      > individual email or weblog posting.
      >
      > While I would agree that the schema attribute is relatively harmless,
      > "The spec, on the other hand, if read literally, is clear on the
      matter."
      >
      > - Sam Ruby
      >
    • Mark Woodman
      As a neutral third party on this matter, I would like to observer that everyone contributing to this thread has been quite civil and void of emotion. Giving
      Message 2 of 15 , Oct 3, 2006
      View Source
      • 0 Attachment
        As a neutral third party on this matter, I would like to observer that everyone contributing to this thread has been quite civil and void of emotion.   

        Giving all the benefit of the doubt, I see nothing that suggests somebody trying to create an issue.   We all want to be faithful to the spec and its intent, so this has been a productive discussion thus far.  Keep it up, guys.

      • Sam Ruby
        ... I emphatically did *NOT* create this issue. When this was first brought to my attention in 2005, there was no obvious process for resolving these issues,
        Message 3 of 15 , Oct 3, 2006
        View Source
        • 0 Attachment
          Randy Morin wrote:
          > If we call this an error, I'll propose a change to the spec to allow
          > extension attribute. Please try and be civil instead of trying to
          > create an issue. This is an obvious compromise.

          I emphatically did *NOT* create this issue.

          When this was first brought to my attention in 2005, there was no
          obvious process for resolving these issues, so I avoided creating an
          issue by removing an error message, and creating a unit test case:

          http://feedvalidator.org/testcases/rss/must/unknown_attribute_in_unknown_namespace.xml

          When *YOU* said that "an extension attribute would contradict the spec",
          I asked you for clarification.

          When the clarification was ultimately codified into profile text, I
          coded up -- but did not deploy -- the check in question. This caused
          one unit test case to fail. Researching further (basically by googling
          on terms found in that test case), I found references, and presented
          them to this mailing list.

          I've offered to hold back on deployment and to conform to whatever the
          profile ultimately says.

          If anybody can find any point in this process where I "created" an issue
          or was anything other than "civil", please let me know.

          - - -

          Meanwhile,

          (1) I do believe that the spec is clear, and
          (2) I do believe that the RSS 2.0 roadmap is clear

          So, why can't schemas be defined as extension elements instead of
          extension attributes?

          - Sam Ruby
        • Randy Morin
          Sorry Dude, didn t mean to imply you were already uncivil. You weren t and I apologize. I only meant to head off the issue. I m looking for a compromise here.
          Message 4 of 15 , Oct 3, 2006
          View Source
          • 0 Attachment
            Sorry Dude, didn't mean to imply you were already uncivil. You weren't
            and I apologize. I only meant to head off the issue. I'm looking for a
            compromise here. Obviously, this is another area where the spec is
            dumb. I want to avoid the issue. I sit between Atom and Dave everyday.
            Give me some slack please.
            Thanks,

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



            --- In rss-public@yahoogroups.com, Sam Ruby <rubys@...> wrote:
            >
            > Randy Morin wrote:
            > > If we call this an error, I'll propose a change to the spec to allow
            > > extension attribute. Please try and be civil instead of trying to
            > > create an issue. This is an obvious compromise.
            >
            > I emphatically did *NOT* create this issue.
            >
            > When this was first brought to my attention in 2005, there was no
            > obvious process for resolving these issues, so I avoided creating an
            > issue by removing an error message, and creating a unit test case:
            >
            >
            http://feedvalidator.org/testcases/rss/must/unknown_attribute_in_unknown_namespace.xml
            >
            > When *YOU* said that "an extension attribute would contradict the
            spec",
            > I asked you for clarification.
            >
            > When the clarification was ultimately codified into profile text, I
            > coded up -- but did not deploy -- the check in question. This caused
            > one unit test case to fail. Researching further (basically by googling
            > on terms found in that test case), I found references, and presented
            > them to this mailing list.
            >
            > I've offered to hold back on deployment and to conform to whatever the
            > profile ultimately says.
            >
            > If anybody can find any point in this process where I "created" an
            issue
            > or was anything other than "civil", please let me know.
            >
            > - - -
            >
            > Meanwhile,
            >
            > (1) I do believe that the spec is clear, and
            > (2) I do believe that the RSS 2.0 roadmap is clear
            >
            > So, why can't schemas be defined as extension elements instead of
            > extension attributes?
            >
            > - Sam Ruby
            >
          • Sam Ruby
            ... Cool. For now, I ve temporarily changed the severity level to a warning, but left the text of the message alone and committed and deployed the result.
            Message 5 of 15 , Oct 4, 2006
            View Source
            • 0 Attachment
              Randy Morin wrote:
              > Sorry Dude, didn't mean to imply you were already uncivil. You weren't
              > and I apologize. I only meant to head off the issue. I'm looking for a
              > compromise here. Obviously, this is another area where the spec is
              > dumb. I want to avoid the issue. I sit between Atom and Dave everyday.
              > Give me some slack please.

              Cool.

              For now, I've temporarily changed the severity level to a warning, but
              left the text of the message alone and committed and deployed the result.

              http://tinyurl.com/h47xj

              As to the future: if the profile changes, I plan to track to the
              profile. If I hear nothing else and the profile remains as is, I'll
              update the Feed Validator to match the profile at some point after the
              next substantial revision unrelated to this change.

              It is worth noting that to date I have found no otherwise-valid feeds
              that would be marked as invalid given this restriction. I fully expect
              the number of feeds affected to be small.

              - Sam Ruby
            • Rogers Cadenhead
              My reply appears to have been eaten. I m not figuring out something about this: What schema stuff is adversely affected by a prohibition against namespace
              Message 6 of 15 , Oct 4, 2006
              View Source
              • 0 Attachment
                My reply appears to have been eaten. I'm not figuring out something about this: What schema stuff is adversely affected by a prohibition against namespace attributes in core elements?
              • Randy Morin
                Examine the document element. I ve removed extra stuffing to make it more obvious.
                Message 7 of 15 , Oct 4, 2006
                View Source
                • 0 Attachment
                  Examine the document element. I've removed extra stuffing to make it
                  more obvious.

                  <rss xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
                  xsi:noNamespaceSchemaLocation="http://www.thearchitect.co.uk/schemas/rss
                  -2_0.xsd" version="2.0">

                  The noNamespaceSchemaLocation attribute is used to specify an XML
                  Schema location for an XML vocabulatory like RSS that is not in a
                  namespace. If your XML parser validated, then it could use this
                  attribute to find the schema to validate the XML against.

                  Hope this clarifies,

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

                  --- In rss-public@yahoogroups.com, "Rogers Cadenhead" <cadenhead@...>
                  wrote:
                  >
                  > My reply appears to have been eaten. I'm not figuring out something
                  about
                  > this: What schema stuff is adversely affected by a prohibition against
                  > namespace attributes in core elements?
                  >
                • Sam Ruby
                  ... In all of my years of working with XML schemas, I was not aware of this particular attribute. But after a quick scan of the web, I came across a w3c
                  Message 8 of 15 , Oct 4, 2006
                  View Source
                  • 0 Attachment
                    Rogers Cadenhead wrote:
                    > My reply appears to have been eaten. I'm not figuring out something
                    > about this: What schema stuff is adversely affected by a prohibition
                    > against namespace attributes in core elements?

                    In all of my years of working with XML schemas, I was not aware of this
                    particular attribute. But after a quick scan of the web, I came across
                    a w3c schema validator:

                    http://www.w3.org/2001/03/webdata/xsv

                    Trying this on the document in question results in the following:

                    http://tinyurl.com/q2l77

                    Looking at the output, I observe two things:

                    1) For this to work, EVERY namespace must resolve to either a schema
                    document or a RDDL document. This particular document referenced
                    five namespaces (atom, feedburner, itunes, geo, mediaRSS), NONE of
                    which resolve to either a scheme document or a RDDL document.

                    2) The schema that Jorgen originally authored doesn't appear to handle
                    email addresses that end in comments. This is a fairly common
                    usage for author tags, in fact there even is an example of such
                    in the RSS 2.0 specification.

                    - Sam Ruby
                  Your message has been successfully submitted and would be delivered to recipients shortly.