Re: [unrev-II] Fwd: SOAP:Lite, Document clipping with annotation

From: Eric Armstrong (eric.armstrong@eng.sun.com)
Date: Wed May 02 2001 - 15:51:09 PDT

  • Next message: Grant Bowman: "Linux Journal article on BT suit"

    Jack Park wrote:
    >
    > >I am forwarding this post to XML-DEV to this list because, among
    > other
    > >things, it mentions a discussion on IBM's Transcoding Database. I
    > would
    > >think that such a product as that could get a leg up on prototyping
    > an OHS.
    >
    > >Document clipping with annotation
    > >------------------------------------------------------
    > >With WebSphere Transcoding Publisher's XML-based annotation language,
    > you
    > >can identify and extract specific portions of a document, all without
    > >having to touch the HTML source. In this article, we describe how the
    > >annotation language works and provide several examples of how you can
    > use
    > >annotation to tailor Web content for different devices
    > >http://www-106.ibm.com/developerworks/ibm/library/ibm-clip/?open&l=136,t=gr,p=x.sbox
    > >
    An interesting discussion. But:
      a) The "annotation" discussed here is really a clipping mechanism.
      b) The mechanism really consists of xpath identificiations,
         with <keep > or <remove > instructions to do the clipping.
         (I suppose a <note > instruction could be use to actually
          add an annotation.
      c) The xpath mechanism is fairly brittle. Change the
         document structure in any significant way, and all bets are
         off. (On the other hand, the mechanisms shown would allow
         for fairly significant amounts of change before things started
         to break.)
      d) The "debrittling" mechanism is to include comments in the file.
         For us, the comments could be paragraph ids.
         In that case, it would only make sense to apply annotations to
         documents that had been transcoded to have the ids attached.
         Now then, if there were a *versioning* mechanism that made it
         possible to keep consistent IDs after document changes, the
         problem would be solved -- but of course that versioning
         mechanism is, as always, the crux of the problem.

    Community email addresses:
      Post message: unrev-II@onelist.com
      Subscribe: unrev-II-subscribe@onelist.com
      Unsubscribe: unrev-II-unsubscribe@onelist.com
      List owner: unrev-II-owner@onelist.com

    Shortcut URL to this page:
      http://www.onelist.com/community/unrev-II

    Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/



    This archive was generated by hypermail 2b29 : Wed May 02 2001 - 17:49:15 PDT