Skip to Content.
Sympa Menu

pcplantdb - [pcplantdb] Website, RDF

pcplantdb@lists.ibiblio.org

Subject: pcplantdb

List archive

Chronological Thread  
  • From: Bear K <bear@ursine-design.com>
  • To: Permaculture Plant Database <pcplantdb@lists.ibiblio.org>
  • Subject: [pcplantdb] Website, RDF
  • Date: Thu, 16 Dec 2004 09:43:47 -0800

Hi Folks,

On the website front: I've rolled in a few changes to the mock up. See what you think, and what other things you'd like to see. Sorry, John S, it's still a thick header design. I still haven't made many changes to the title typography, I'll try to put out some other options shortly. And please excuse me if my understanding rhizosphere shapes is lacking.

I was looking at ways an advanced (Step 2) front-end could communicate, and started looking at RDF.
RDF helps to provide more meta-information on web pages, helping to create the "Semantic Web" <- (Thesaurus Steph?)

Here's a quote from a site:
"RDF (Resource Description Framework) is a model for storing graphs of information. Given a set of resources, each resource being a thing, such as a person, a song, a web page or a bookmark, RDF is used to create relationships between these resources. "

Here's a reasonable intro: http://xulplanet.com/tutorials/mozsdk/rdfstart.php


What draws me to this?
• You can create your own custom information descriptions and structures.
• This might provide a great why to distribute information (a la L. London).
• There seem to be tools that can automatically create RBDMS schemas from RDF data, opensource and written in C which interfaces to most common languages (Python...). See: http://139.91.183.30:9090/RDF/ <-- (Techie folks look here)
• Front-ends like TouchGraph have already been designed to handle this information.
• Schema seems to be easily changeable.
• Information is handled in a similar way to OO languages.
• The format is XML for folks who like that.
• Imagine folks providing TrackBacks for references to plants. Then you can see what garden bloggers have done and are saying about a plant.

RDF is composed of triples, forming a relationship (between objects, or literal information [plant height | pfaf_usefulness_score ]). A triple is composed of (subject) -> (relationship | predicate) -> (object | target)

<http://permaculture.info/Plant/Cyphomandra-betacea> ---> family ---> "Cyphomandra"
<http://permaculture.info/Plant/Cyphomandra-betacea> ---> seeAlso ---> <http://plants.usda.gov/search?id=233908902939444598439>
<http://permaculture.info/Plant/Cyphomandra-betacea> ---> possibleLocation ---> <http://permaculture.info/Microclimates/WestWall>
<http://permaculture.info/Plant/Cyphomandra-betacea> ---> possibleLocation ---> <http://permaculture.info/Microclimates/Woodland>

This isn't in RDF syntax, and I'm not including much of the functionality in this demo, as I'm still learning what it could do.
Notice how objects can use the permaculture.info site. Functionally this serves as a unique ID for an object, anything with that same URI would be the same object.

It seems like there's cool thing that can happen:
If a web page adds some RDF information:
<http://permaculture.info/Plant/Cyphomandra-betacea> ---> uses --> <http:userhomepage.com/recipes/TamarilloSalsa>
That information can be rolled into the DB for that object. To properly work, the information would need to fit into what we define in our Plant schema. There's more information out there on how to judge trustworthiness, etc.

Cheers,
Bear


Archive powered by MHonArc 2.6.24.

Top of Page