Permaculture Information Web

[Stephanie Gerson <sgerson AT stanfordalumni.org>]

What if there's different, and perhaps complimentary, information in entries
about the same plant/relationship - but because they get different ratings,
they aren't necessarily read together?

For example, entry 1 and 2 are about the same plant/relationship.  Entry 1 has
information X, entry 2 has information Y.  Information X and Y are
complimentary, but because entry 1 gets a higher rating than entry 2, readers
only read information X and miss information Y.  Or is someone (i.e. users)
going to take responsibility for condensing information, so that bits and
peices from different entries can be woven together?

Although this model is successful from the vantage point of ensuring quality,
Chad mentioned that it will end up fragmenting the content...

Also, I've been thinking about filtering.  There is already such an abundance
of information - and when it gets overabundant, it becomes useless.  Can
entries be deleted by users/moderators - i.e. if they get a low enough rating
or are not used for a given amount of time?  How to present only relevant and
useful information, vs. storing all information for the sake of storage?  How
to allow information, when it becomes obsolete, to go extinct?

some change I found in my pocket
*Stephanie

------ Original Message ------
Received: 02:32 PM PST, 03/25/2005
From: Chad Knepp <pyg AT galatea.org>
To: Permaculture Plant Database <pcplantdb AT lists.ibiblio.org>,
piw AT lists.ibiblio.orgCc: 
Subject: Re: [pcplantdb] data/security model proposal

Stephanie Gerson writes:
 > > Richard Morris wrote:
 > > 
 > > The disadvantage is that we'll end up with
 > > more fragmented data. There won't be the (current) definitive version of
 > > the dataset for a particular plant. Instead there will be
 > > a loose collection of datachunks.
 > 
 > HMM.  DO USERS TAKE RESPONSIBILITY FOR SCRUBBING THROUGH AND DOING SOME
 > TIDYING UP?  (DO THEY DO THAT WITH WIKIPEDIA?)  BECAUSE I FEEL LIKE IT
WOULD
 > BE NICE FOR THE DATA TO BE COHESIVE...
 > *S

This is an important point to understand and something I responded to
in part earlier but could stand more clarification.

My proposed data model would allow for multiple versions of data for
any element in the database.  Each version would be submitted, edited,
and possibly deleted by its author/owner.  Although these authors
would not be able to edit or delete information by other authors, they
would have the power to judge the quality of other authors
information.  By stating that this information is either good/helpful
or not (moderation), these multiple different versions of the data
become rated as to their quality.  Upon display, the versions of data
that are judged to be more correct (higher moderation score) are give
display priority, placing them nearer the top of the page, etc.  Low
scoring (or possibly negative scoring) may not display at all without
following an additional link.

This is very different from Wikipedia and other models that have a
notion of definitive/objective information and only one "correct"
entry for any element.

I think that the model I'm proposing is a good choice for the sort of
data we are handling because it's nice to have a second (and third)
opinion about plant characteristics, uses, and culture.  I think this
will make even more sense when/if we include more abstract elements as
well. 

-- 
Chad Knepp
python -c 'import base64;print base64.decodestring("cHlnQGdhbGF0ZWEub3Jn")'
_______________________________________________
pcplantdb mailing list
pcplantdb AT lists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/pcplantdb




+++++++++++++++
Stephanie Gerson
sgerson AT stanfordalumni.org
(c) 415.871.5683


____________________________________________________________________