[Osr-101] [Fwd: RE: Implementing Neutron into Xopus]

Michael Wechner michael.wechner at wyona.com
Thu Feb 8 17:03:37 CET 2007 

Hi

Here some discussions with Laurens and Lon from Xopus re editing of 
"resources" within a context.

I am not sure yet how we can best cover such cases, but hope that we can 
find some consensus soon.

It seems like many "clients" would prefer something liks this (e.g. 
TinyMCE) and also has some similarities with "PDF forms".

Cheers

Michael

-------- Original Message --------
Subject: 	RE: Implementing Neutron into Xopus
Date: 	Tue, 6 Feb 2007 12:42:41 +0100
From: 	Laurens van den Oever <laurens at xopus.com>
To: 	Lon Boonen <Lon at q42.nl>, Michael Wechner <michael.wechner at wyona.com>
CC: 	Donat Grimm <donat.grimm at wyona.com>, Andreas Wuest 
<awuest at student.ethz.ch>, Jasper Van der Kolk <jasper at xopus.com>
References: 	<45C7B569.8010708 at wyona.com> 
<646876D1454BE6468B01FDAEB1FC07816AB86D at DS01.q42.local>



Hi,

Another approach would be:

<html>
  <head>
    <title>Hello World</title>
    ...
  </head>
  <body>
  ...
  <div id="calendar">
    <link rel="neutron-introspection" type="application/neutron+xml"
href="edit-calendar.xml"/>
    ...
  </div>
  ...
  <div id="mainArticle">
    <link rel="neutron-introspection" type="application/neutron+xml"
href="edit-article.xml"/>
    ..
  </div>
</html

That would make editing regions of the page a lot simpler.

BTW: Folksonomies are very popular these days. But I think we already
have a tagging method that offers far better information about content:
XML. The only problem with XML is that Google doesn't index it. 

The reason for that is that there is no XML linking standard that is
supported by all browsers. So there are virtually no sites that serve
XML instead of HTML, so Google has no reason to index XML.

I think there may be a role for Neutron here. If Google would understand
Neutron, it would be able to retrieve the source content for a page
instead of the content and navigation, ads, summaries from other pages,
etc. Then it would then be able to focus on the content only AND Google
would be able to use the tag names to identify the nature of the content
using the brute force methods they already use for HTML.

So I think it would be valuable to keep this in mind when standardizing
something like Neutron.

Best,

Laurens van den Oever
CEO
Xopus

laurens at xopus.com
http://xopus.com
+31 70 346 70 53

Waldorpstraat 17F
2521 CA Den Haag
The Netherlands

-----Original Message-----
From: Lon Boonen 
Sent: dinsdag 6 februari 2007 11:35
To: Michael Wechner
Cc: Laurens van den Oever; Donat Grimm; Andreas Wuest
Subject: RE: Implementing Neutron into Xopus

Hi Michael,

maybe something like this?

<introspection xmlns="http://www.wyona.org/neutron/1.0">
  <edit mime-type="application/xml" name="Calendar" id="calendar">
    <open url="calendar.xml" method="GET"/>
    <save url="calendar.xml?yanel.resource.usecase=save" method="POST"/>
    <validate url="calendar.xsd" type="XSD" />
    <views>
      <view name="wysiwyg">
        <transform xsl="calendar2xhtml.xsl"/>
      </view>
      <view name="xml-view">
        <transform xsl="treeview.xsl"/>
      </view>
    </views>
  </edit>
  <edit mime-type="application/xml" name="Article" id="mainArticle">
    <open url="article.xml" method="GET"/>
    <save url="article.xml?yanel.resource.usecase=save" method="POST"/>
    <validate url="article.xsd" type="XSD" />
    <views>
      <view name="wysiwyg">
        <transform xsl="article2xhtml.xsl"/>
        <transform xsl="postprocessor.xsl"/>
      </view>
      <view name="xml-view">
        <transform xsl="treeview.xsl"/>
      </view>
    </views>
  </edit>
</introspection>


The page belonging to this:

<html>
  <head>
    <link rel="neutron-introspection" type="application/neutron+xml"
href="introspection-proposal.xml"/>
    <title>Hello World</title>
    ...
  </head>
  <body>
  ...
  <div id="calendar">
    ...
  </div>
  ...
  <div id="mainArticle">
    ..
  </div>
</html>


Multiple edit-nodes refer to html-elements on page using id-attribute
Validate-node refers to schema or dtd or relaxNG, validate node can be
extended in the future (with children)
Views-node contains one or more view-nodes.
View-node contains one or more transforms/xincludes/CSS-links (or in the
future also other stuff)


groetjes Lon


-------------------------------------------------
Q42:   http://www.q42.nl
Xopus: http://xopus.com
http://maps.google.com/?t=h&ll=52.0692,4.3232  

> -----Original Message-----
> From: Michael Wechner [mailto:michael.wechner at wyona.com] 
> Sent: Monday, February 05, 2007 23:53
> To: Lon Boonen
> Cc: Laurens van den Oever; Donat Grimm; Andreas Wuest
> Subject: Re: Implementing Neutron into Xopus
> 
> Lon Boonen wrote:
> 
> >Hi Michael and others,
> >
> >Xopus can edit one or more parts of a page.
> >We almost never edit the whole page, since usually the whole 
> page also
> >holds stuff like navigation, header, footer... Stuff that 
> has nothing to
> >do with the article being edited.
> >
> >So.
> >This is how Xopus does it:
> >
> >Inside the page there are one or more div-elements like this
> >(html-brackets replaced by []):
> >
> >    [div xopus="true" autostart="true"]
> >      Loading Article...
> >      [xml]
> >        [config version="1.0" 
> xmlns="http://www.xopus.com/xmlns/config"]
> >          [pipeline xml="xml/simplearticle.xml"
> >xsd="xsd/SchemaRepository/SimpleArticle.xsd"]
> >            [view name="WYSIWYG View"]
> >              [transform xsl="xsl/SimpleArticle_Content.xsl" /]
> >            [/view]
> >            [view name="Tags View"]
> >              [transform xsl="xsl/tagson.xsl" /]
> >            [/view]
> >            [view name="XML View"]
> >              [treeTransform /]
> >            [/view]
> >          [/pipeline]
> >          [import src="config/config.xml" /]
> >        [/config]
> >      [/xml]
> >    [/div]
> >
> >As you can see the div has an xopus-attribute and an autostart
> >attribute.
> >Inside is a pipeline definition, telling Xopus what XML, XSD 
> and XSLT to
> >use.
> >There can be several views inside a pipeline definition. Like in this
> >case wysiwyg, tags-on and xml-view.
> >Each view can consist of several XSLT-transformations and/or
> >xinclude-resolves.
> >Each div (canvas) can also have a config file associated to it.
> >
> >It would be optimal if Neutron could support all of this.
> >We shouldn't restrict ourselves to a page that is completely 
> editable.
> >You must be able to define editable part(s) on a page.
> >Multiple views might not be necessary. But multiple 
> transformations per
> >view are necessary.
> >And of course a link to the validation schema and config file.
> >
> >Can you tell me if Neutron can support this?
> >  
> >
> 
> not as you describe it, but I will discuss it with Andreas 
> tomorrow and 
> try to make a suggestion how we can "neutralize" it ;-)
> 
> Cheers
> 
> Michael
> 
> >And if so, how?
> >
> >
> >greetings Lon
> >
> >-------------------------------------------------
> >Q42:   http://www.q42.nl
> >Xopus: http://xopus.com
> >http://maps.google.com/?t=h&ll=52.0692,4.3232  
> >
> >  
> >
> 
> 
> -- 
> Michael Wechner
> Wyona      -   Open Source Content Management   -    Apache Lenya
> http://www.wyona.com                      http://lenya.apache.org
> michael.wechner at wyona.com                        michi at apache.org
> +41 44 272 91 61
> 
> 



-- 
Michael Wechner
Wyona      -   Open Source Content Management   -    Apache Lenya
http://www.wyona.com                      http://lenya.apache.org
michael.wechner at wyona.com                        michi at apache.org
+41 44 272 91 61

More information about the Osr-101 mailing list