Re: [Ag Meinungsfindungstool] (proposal) Metagov as a technical plug-in framework

[Michael Allan <mike AT zelea.com>]

Ronald and Marc, (cc AG Meinungsfindungstool)

I roughly understand the solution (or counter-solution) you propose,
except in one regard.  So I have a question at the end.


Ronald said:
> No, not necessarily [defragmenting the tools]...  This is my
> proposal, how to outbalance the fragmentation of documents...  We
> agree on a set of meta data, that the tools should support per
> document...

This sounds like "defragmented" tools to me.  It's an example of that.
The tools are no longer disconnected fragments because they share two
connections: first the agreed "meta data" standard, and second, the
"consolidated overview" into which they are pushed or pulled. *

I think we both agree (and Marc agrees) that the tools must somehow
(in this meaning) be defragmented.  You offer a specific solution,
while we (Conseo, Thomas and I) do not.  This raises the question of
how your solution relates to what we're proposing in this thread.  The
answer hinges on how you address the problem of fragmented documents
that are *non-compliant* with your agreed meta-data standard.  Either:

  A. All documents must implement the agreed meta-data standard.
     Non-compliant documents are eliminated.  So fragmentation is
     solved by your solution alone.  Or,

  B. Non-compliant documents may implement other, rival solutions.  So
     fragmentation is *not* solved by your solution alone.  You depend
     on the existence of other solutions.

If you prefer A, then your approach to the larger problem is at odds
with and incompatible with what we propose in this thread. **   If you
prefer B, then it is compatible.  Which do you prefer, A or B?


   * Your solution is not entirely new, ofc.  Mark Murphy proposed
     something similar in 2007.  xDebate formats (not yet published)
     is a Semantic HTML (POSH) schemata for federated debate on
     political issues.  We discussed it with him in 2009.  See:
     
http://metagovernment.org/pipermail/start_metagovernment.org/2009-February/001195.html

     The broader context of that discussion was his 2008 essay in
     Rebooting America: The "killer app" of public participation:
     http://rebooting.personaldemocracy.com/node/5499

     Later in 2009, Thomas was chasing the fragmentation problem
     himself and uncovered a different solution, one that nobody ever
     thought of before.  Vote mirroring works on documents (and other
     objects) via their popular support, not their content.  But note:

  ** All we propose is a neutral tool switch that is blind to specific
     solutions.  Like the Internet, it is open to all.

Mike


Ronald Grindle said:
> Hello Mike,
> 
> "So the documents (and popular support) cannot be defragmented without 
> defragmenting the tools."
> 
> No, not necessarily.
> 
> This is my proposal, how to outbalance the fragmentation of documents:
> 
> We agree on a set of meta data, that the tools should support per document, 
> for instance:
> 
> UID
> Title
> Document Type
> Document Status
> Area
> Topic
> Owner
> URL [where to find the document]
> (Author)
> (Keywords)
> 
> UID is a unique ID, to be able to unmistakably identify the document.
> Attributes in brackets () would be optional. The possible entries for 
> "Document Type" and "Document Status" would be pre-defined. The possible 
> entries for "Area" and "Topic" should be fed by a common database (editable 
> by everyone), to avoid "fragmentation by naming".
> 
> To consolidate the data to an overview on the documents there are two 
> solution I can think of:
> 
> 1. All tools report their documents to a central server, that provides an 
> overview of all documents, by calling a common interface. (e.g. a 
> http-request)
> 
> 2. A search tool that crawls through all tools and collects the data. This 
> solution has, at first glance three downsides:
> - The tools must be searchable (like a web page).
> - The documents must provide the meta data proposed above as standardized 
> meta-tags.
> - The tools will be loaded with requests, every time a search is triggered.
> 
> The consolidation of supporting votes could work the same, but is a more 
> touchy matter, as this is a potential field for cheating and therefore 
> requires special measures to ensure trust.
> 
> Best Regards
> Ronald


marc said:
> Hi,
> 
> I strongly agree with Ronald, that there should be a kind of 'standardized 
> meta data' that all tools (plug-ins to the framework) are working on, or at 
> least, are able to convert their data back and forth.
> 
> So what about eDialogus 
> http://www.imc.com.gr/ontologies/eDialogos/consensus/#overview  as a 
> starting point for a unique ontology for discussion systems?
> 
> What do you think?
> 
> Cheers
> Merkbefreiter (marc)
> 
> P.S. Because there is already a 'Mark' on the list, you can safely refer to 
> me as 'Merkbefreiter' ;o)