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

Re: [RSS-DEV] Why is RSS 2.0 Bad? (Not a Rhetorical Question)

Expand Messages
  • Phil Ringnalda
    ... ... True. What they care about is the way that their page of 50 nested tables takes forever to load, or the way that the formatting in
    Message 1 of 19 , Sep 25, 2002
    • 0 Attachment
      Ben Hammersley wrote:
      > On Thursday, Sep 26, 2002, at 00:27 Europe/London, Ben Hammersley wrote:
      >> . My dad loves RSS, he just
      >> wouldn't recognise it over HTML if he viewed source. The simplicity of
      >> the spec, for him, is meaningless.
      >
      > <much snippage/>
      >

      <more snippage/>

      > Nowadays, you never need see a single bit of HTML to make a web page.
      > With blogging tools you don't even need to consciously make a page. For
      > the vast majority of users, using Frontpage or Dreamweaver or whatever
      > to create their little project, the use of <em> over <b>, or <br/> over
      > <br> may well be transparent, but is utterly moot. Who cares? Not the
      > end user.

      True. What they care about is the way that their page of 50 nested tables
      takes forever to load, or the way that the formatting in their Blogger
      template disappears whenever they post more than one paragraph, or the way
      their geeky friends keep complaining that their quotes and dashes display as
      ? instead.

      > Now, the evolution of RSS has been very fast, and we left the notepad
      > and view source era years ago. RSS feeds are produced automatically by
      > tools created by developers, using templates created by developers, to
      > work on applications created by developers (most likely the same ones),
      > and then and only then do the general userbase get involved. No one
      > creates their RSS feed from scratch in notepad, the way I did sites in
      > 1995, and to insist on simplicity for this reason alone is doing
      > ourselves a disservice.

      The first RSS feed I created was done essentially by hand (in PHP, but the
      code might as well have been typed on a manual typewriter), in my favorite
      text editor at the time: Notepad. That was way back in January of 2002. I'm
      still using the same code today, because no developer has stepped up to
      write a better tool to create RSS from dotcomments. I would have done it as
      RSS 1.0, because after all 1.0 is a better number than 0.91, but I couldn't
      follow the spec well enough to be sure I was doing it right, and I couldn't
      figure out *how* it would be better. A few months later when I wrote a
      script to scrape my Blogger-produced blogs, I'm happy to say that I figured
      out how to do 1.0, though I seem to have taken the advisory limits like 15
      items and 500 characters so much to heart that I created two versions, a
      severely crippled 1.0 and a rich and meaty 0.92. Now, after nine months of
      playing with RSS, I generate most of mine from MovableType templates: a 1.0
      template that I hope I've removed most of the errors and security leaks
      from, a 2.0 template that I based roughly on the default 0.91 template, and
      some comment and individual archive templates that I did myself from scratch
      and copying other people's.

      I am not a developer. I am not a professional. I am an interested amateur
      publisher. I am your audience.

      Phil Ringnalda
    • Ben Hammersley
      ... ... ... ... No, Phil, you *are* a developer. (Sorry to break this to you. :-) You re on this list. You re writing
      Message 2 of 19 , Sep 25, 2002
      • 0 Attachment
        On Thursday, Sep 26, 2002, at 05:51 Europe/London, Phil Ringnalda wrote:

        > Ben Hammersley wrote:
        >> On Thursday, Sep 26, 2002, at 00:27 Europe/London, Ben Hammersley
        >> wrote:
        >>> . My dad loves RSS, he just
        >>> wouldn't recognise it over HTML if he viewed source. The simplicity
        >>> of
        >>> the spec, for him, is meaningless.
        >>
        >> <much snippage/>
        >>
        >
        > <more snippage/>
        <even more snippage/ >

        > The first RSS feed I created was done essentially by hand (in PHP, but
        > the
        > code might as well have been typed on a manual typewriter),
        <snip/>
        > A few months later when I wrote a script to scrape my
        > Blogger-produced blogs,
        <snip/>
        > I hope I've removed most of the errors and security leaks
        > from, a 2.0 template that I based roughly on the default 0.91
        > template, and
        > some comment and individual archive templates that I did myself from
        > scratch
        > and copying other people's.
        >
        > I am not a developer. I am not a professional. I am an interested
        > amateur
        > publisher. I am your audience.
        >
        > Phil Ringnalda

        No, Phil, you *are* a developer. (Sorry to break this to you. :-)
        You're on this list. You're writing scripts to scrape Blogger. You
        contribute to specifications. You're correcting errors in the MT
        template. You're as professional as they come when it comes to this.
        Everyone here is.

        What I'm trying to address is the fundamental difference between RSS
        and HTML - which is that no one other than the member of this list, and
        the sort of people who write their own CMSs need ever look at it. It's
        just not something the average guy would ever have reason to create by
        hand (and not one single person reading this post can claim to be an
        average guy.)

        Because of this, I'm just saying that to be forever looking to simplify
        the spec to a point where your average consumer can view source and
        learn the spec is pointless and indeed harmful if done at the expense
        of other stuff. Because on the internet, since HTML, the average
        consumer has become a whole load more average.
      • Marcus Campbell
        ... I agree. (Forgive me if I meander or am complete OT, but I fancy a go at posting...) When you talk on the phone you don t care how it works, you just know
        Message 3 of 19 , Sep 26, 2002
        • 0 Attachment
          > Because of this, I'm just saying that to be forever looking to simplify
          > the spec to a point where your average consumer can view source and
          > learn the spec is pointless and indeed harmful if done at the expense
          > of other stuff. Because on the internet, since HTML, the average
          > consumer has become a whole load more average.

          I agree.

          (Forgive me if I meander or am complete OT, but I fancy a go at
          posting...)

          When you talk on the phone you don't care how it works, you just know
          that you type in a number and you get to speak to someone. The actual
          transfer system is irrelevant to the user.

          Now, say you start sending video over the phone lines... do the users
          care that the specs and technology behind it get more complicated? Not
          as long as they only have to rely on dialing a number they won't, they
          just like the fact that they can see the person they're talking to.

          As long as the developers can, and do, make the video phones (and
          networks) then the user is happy. It could be a pixie in a box that
          uses magic pixie dust to get it to work. They shouldn't have to care.

          Similarly developers shouldn't need something that anyone and everyone
          has to understand. Nor should they make the basic format more
          complicated than anything they could _actually_ make use of.
          Developers should be able to make the tools that handle the data that
          creates the system. If they can't, they should be able to use someone
          else's tools. Even developers shouldn't have to resort to pushing raw
          data all the time.

          Making the specs extensible is good - very good - but that doesn't
          mean it's of any of use to argue its merits now or jump right in with
          modules that are only going to be useful some time in the future...
          maybe. Create the killer app. Get the format that works, works well
          and could be extended to work in future. It doesn't need to be
          overly-complicated or overly-theoretical and it doesn't need to be
          overly-"simple" (although I would like to see something without
          obsolete or proprietary bloat). It just needs to work.
        • mof-rss-dev@mfd-consult.dk
          Ben Hammersley wrote on Thu, 26 Sep 2002 02:38:11 +0100 ... I agree with your disagreement. The simplicity of HTML was great for its
          Message 4 of 19 , Sep 26, 2002
          • 0 Attachment
            Ben Hammersley <ben@...> wrote on Thu, 26 Sep 2002 02:38:11 +0100
            in :
            >An anonymous benefactor just privately pulled me up over this, citing
            >the common wisdom that the web was a success because "(HTML) was
            >transparently understandable to people with a minimal technical
            >background". It's a common argument for simplicity uber alles within
            >RSS.
            I agree with your disagreement.

            The "simplicity" of HTML was great for its quick widespread production,
            but as mentioned elsewhere, it's a pain to consume.

            This is, I believe, because simplicity is confused with "optional" and
            other lax definitions, either in the definition or the use.

            I would have thought that a lesson was learned there, that what is
            really needed is *strict* definitions, at least if it is in any way
            meant to be consumed by machines.

            I.e. a <description> element that is not defined as to its contents,
            and a <link> element that can point anywhere, is useless in practice,
            even though it's easy to produce.

            This is why I like vocabularies like the Dublin Core, which provide
            precise definitions for syntax and semantics. They may be somewhat more
            difficult to produce, perhaps needing date format conversion, but they
            are easily used.

            This is also why I dislike RSS 0.9x - it's too loose, it is unusable
            for anything *but* "the display of headlines in a browser for human
            consumption", and why I think that the path of RSS 1.0 is better.

            Why couldn't (X)HTML be used instead of RSS 0.9x?
            <html>
            <head>
            <title>My Blog</title>
            <meta name="DC.description" value="Various ramblings by Me" />
            </head>
            <body>
            <p>
            <a href="http://example.com/">
            <strong>Example "item" title</strong>
            </a>
            Example item description...
            </p>
            <address>me@...</address>
            </body>
            </html>
            And no, this is not a general suggestion for a new "competing"
            format...

            This may be the defining difference between the branches - the possible
            use of the "metadata" as *real* machine readable metadata for use with
            The Semantic Web(tm)?!?


            Regards,

            Morten Frederiksen
            ---
            A bird in the hand's better than one overhead.
            --
            <URL: http://www.mfd-consult.dk/ >
          • Libby Miller
            ... makes sense to me....looks somewhat like the xhtml profiles Dan Connolly has been using (and I ripped off for events module):
            Message 5 of 19 , Sep 26, 2002
            • 0 Attachment
              >
              > This is also why I dislike RSS 0.9x - it's too loose, it is unusable
              > for anything *but* "the display of headlines in a browser for human
              > consumption", and why I think that the path of RSS 1.0 is better.
              >
              > Why couldn't (X)HTML be used instead of RSS 0.9x?

              makes sense to me....looks somewhat like the xhtml profiles Dan Connolly
              has been using (and I ripped off for events module):

              http://www.w3.org/2000/08/w3c-synd/
              http://www.w3.org/2001/sw/Europe/200207/rsscal/xslt-rss-events.html

              Libby
            • mof-rss-dev@mfd-consult.dk
              Libby Miller wrote on Thu, 26 Sep 2002 17:18:48 +0100 (BST) ... True, that s much along the same lines, except for the fact that
              Message 6 of 19 , Sep 26, 2002
              • 0 Attachment
                Libby Miller <libby.miller@...> wrote on Thu, 26 Sep 2002 17:18:48 +0100 (BST)
                in :
                >> Why couldn't (X)HTML be used instead of RSS 0.9x?
                >makes sense to me....looks somewhat like the xhtml profiles Dan Connolly
                >has been using (and I ripped off for events module):
                True, that's much along the same lines, except for the fact that Your
                profiles are meant (?) to be embedded into a "real" HTML page, whereas
                I was suggesting a "lite" version (probably also suitable for PDAs and
                such).

                Of course, there is no reason as to why the "lite" version couldn't be
                a fragment of the "real" page.


                Regards,

                Morten Frederiksen
                ---
                Adhesive tagline. Lick this XXXX!
                --
                <URL: http://www.mfd-consult.dk/ >
              • Dan Brickley
                ... I think this could be a very profitable area to explore... The XHTML Basic spec s abstract, copied below, calls out some use cases that are close to those
                Message 7 of 19 , Sep 26, 2002
                • 0 Attachment
                  On Thu, 26 Sep 2002 mof-rss-dev@... wrote:

                  > Libby Miller <libby.miller@...> wrote on Thu, 26 Sep 2002 17:18:48 +0100 (BST)
                  > in :
                  > >> Why couldn't (X)HTML be used instead of RSS 0.9x?
                  > >makes sense to me....looks somewhat like the xhtml profiles Dan Connolly
                  > >has been using (and I ripped off for events module):
                  > True, that's much along the same lines, except for the fact that Your
                  > profiles are meant (?) to be embedded into a "real" HTML page, whereas
                  > I was suggesting a "lite" version (probably also suitable for PDAs and
                  > such).
                  >
                  > Of course, there is no reason as to why the "lite" version couldn't be
                  > a fragment of the "real" page.

                  I think this could be a very profitable area to explore...

                  The XHTML Basic spec's abstract, copied below, calls out some use cases
                  that are close to those we see with RSS and content syndication.

                  http://www.w3.org/TR/xhtml-basic/
                  [[
                  Abstract

                  The XHTML Basic document type includes the minimal set of modules required
                  to be an XHTML host language document type, and in addition it includes
                  images, forms, basic tables, and object support. It is designed for Web
                  clients that do not support the full set of XHTML features; for example,
                  Web clients such as mobile phones, PDAs, pagers, and settop boxes. The
                  document type is rich enough for content authoring.

                  XHTML Basic is designed as a common base that may be extended. For
                  example, an event module that is more generic than the traditional HTML 4
                  event system could be added or it could be extended by additional modules
                  from XHTML Modularization such as the Scripting Module. The goal of XHTML
                  Basic is to serve as a common language supported by various kinds of user
                  agents.

                  The document type definition is implemented using XHTML modules as defined
                  in "Modularization of XHTML" [XHTMLMOD].
                  ]]

                  Similarities:

                  * the need for a simple, restricted subset of content (for PDAs etc)
                  * the need for it to be rich enough to support needs of content authors
                  * architecture that supports principled, modular extension



                  Another excerpt:

                  [[
                  1.2. Background and Requirements

                  Information appliances are targeted for particular uses. They support the
                  features they need for the functions they are designed to fulfill. The
                  following are examples of different information appliances:

                  Mobile phones
                  Televisions
                  PDAs
                  Vending machines
                  Pagers
                  Car navigation systems
                  Mobile game machines
                  Digital book readers
                  Smart watches

                  Existing subsets and variants of HTML for these clients include Compact
                  HTML [CHTML], the Wireless Markup Language [WML], and the "HTML 4.0
                  Guidelines for Mobile Access" [GUIDELINES]. The common features found in
                  these document types include:

                  Basic text (including headings, paragraphs, and lists)
                  Hyperlinks and links to related documents
                  Basic forms
                  Basic tables
                  Images
                  Meta information
                  ]]


                  My sense is that there are two main paths ahead for RSS-like formats: a
                  data-oriented path, via RDF and the 'Semantic Web'; and a
                  document-oriented path, via profiles of modular XHTML formats. The third
                  'go it alone' path is one I'm finding less and less attractive. XML
                  *allows* us all to create and promote new file formats; it's easy to slip
                  from that into assuming that such proliferation of content type is a good thing.


                  Dan


                  --
                  mailto:danbri@...
                  http://www.w3.org/People/DanBri/
                Your message has been successfully submitted and would be delivered to recipients shortly.