permaculture

[christophe mckeon gonzalez de leon <chromatophore@gmail.com>]

> Back to square one, a host, ibiblio. Free and very nice. You would have
> a shell account there to do work. Lots of resources to use. I have an
> account there now. I own http://www.ibiblio.org/permaculture. We could
> use that unless our own domain is needed.

that could be good. what's running on the server, is it a linux box?
full install privileges? filesystem access? any guarantees on uptime?

> There's the beginnings of
> Wordpress/BBPress/Mediawiki at that location presently. Maybe this could
> be used also, why not. Mediawiki needs to be reinstalled.

there is also semantic mediawiki

http://semantic-mediawiki.org/wiki/Semantic_MediaWiki

i haven't even had time to even look at it yet. it could be good, but
it might end up being easier to 'roll our own'. nevertheless reading
their manual is on my list because i think they will have solved some
UI problems. they also have i believe some support for collaborative
ontological modelling.

> I own permacultureinfo.com which I will donate. ibiblio would host that for 
> us
> for free except we would have to provide our own external nameserver,
> they provide everything else. If we had a .net, .info or .org  they
> would provide a complete hosting package, free.

i have gaianexus.org, which is what i was planning on using.
gaia: we all know what that means
nexus: a connection or series of connections linking two or more things

> A person wanting to take a complete inventory of his site can use GIS
> devices to locate and record site resources during a walkabout (Garmin
> has some expensive units). Data can be stored in the unit then uploaded
> to a computer (probably USB) then into the appropriate GIS software.

this should definitely be an option for the extremely motivated & well
equiped, but there also needs to be a very simple html interface with
a simple set of base predicates. i was actually thinking about this
last night. it may not be necessary to have coordinates at all. it may
be enough or even better for reasoning purposes for people to be able
to just say:

pond closeTo mulberryTrees
pond farFrom house

this is enough combined with the needs and yields predicates to do
some fairly sophisticated reasoning. it may not be a problem
that these terms are not too strictly defined.

> Maybe this data could be uploaded to the PC OWL (O' Wise One) into that
> global users "home" describing his site; of course as long as that
> dataset contained at least some information that would be useful to
> others, i.e. timeless in nature.

people will be able to enter complete bs, that's a fact of life for
open systems. but, ontologies have semantics for reification, so that
people can make statements about statements, i.e. mollison
thinkItsTotalCodswallopThat (chickens need hippopotamuses) &
mollison hasAuthority high, therefore it's likely that chooks do not
need hippos.  imo, figuring out the UI for such a thing is the hard
part.

> I don't know what format GIS software...

there has been a lot of work done on ontologies for gis, which again i
have not yet looked at properly. probably our needs would be
quite simple here and met by whatever the going standard is. something
like http://www.w3.org/2003/01/geo/. writing triples in this form from
whatever these things spit out should be fairly trivial.

> This is possible!
> What great prospects. We should do this...

i think so too, but i'm getting a bit ahead of myself. there's enough
here for anybody interested to get aquainted with the tools. if you
are interested in knowledge modelling, learn protege. if you are a
developer, learn the java libs and clojure (i'll be using clojure
because it fits the problem perfectly and is a beautiful language, if
you want me to use php you'll have to pay me, a lot). i need to hit
the books for another solid month or two before anything substantial
is begun. what i might look into in the near future is installing a
collaborative ontology editor just for people to start experimenting.
a caveat, ontology modelling is a hard formal process, and there is
study involved.

> As I said last time, I need some online platform to host the
> permaculture and farming resources that I want to develop,
> text and other media.

a widely accepted standard for media metadata

http://en.wikipedia.org/wiki/Dublin_Core

here again, we need not reinvent the wheel

> I am a market farmer with intensely developed
> land. I want to document what goes on here and put that online.
> I am into gardening toolmaking also.

that's great because we will be able to directly see if our ontologies are
capable of modelling your situation. good acid test

> Interesting, Swedish for bumblebee is humlans... nice word.

wonder if this has anything to with the fact they hum?

_c