pcplantdb

[Sean Maley <semaley@yahoo.com>]

Watch those synonyms and homonyms.  The usual protocol
is to set up a detailed query tool for each side of
the transformation and one for the transformed data
set.  Data element documentation includes elements
from the transformed database with the transformation
from each data source element.  This can be both
graphical and on a spreadsheet of some sort.  The
implementation is then just a matter of following the
transformation rules and using the query tools to spot
check the data.  For brevity, the approved transformed
database can be used directly, as these are not live
databases requiring a scheduled cut-over date.  Data
cleansing should be performed prior to mapping
analysis.

I did this with Lotus's old Maxcim VMS parts databases
to the IBM Corporate standards.  This can be very
frustrating in surprising ways.  Accepted data values
and active participation are very important to not
waste months of time talking and not doing. 
Fortunately, this is a tiny little data set and very
straight forward current model (judging from the eden
database).

So, ... what's the data model going to look like? 
Does it make sense to see what Heide has for us to
compare notes?  Also, what about HTML scrapping the
other online databases like PLANTS and Hortiplex?

On the modeling, wouldn't it make sense to have things
like a separate table for things like family, genus,
etc?  This helps prevent fabricated data by enforcing
the generally accepted scientific classifications. 
When a new family of plants is found, rare as that
would be, we could have a separate admin page to add
it in.  This alone will make the database smaller and
therefore reduce search times; imagine the table
botanical_names having integer keys instead of
character fields and botanical in-string operations
happening only for the discreet values, rather than
for every plant in the database.

One step at a time...


-Sean.

--- Richard Morris <webmaster@pfaf.org> wrote:

> Heide Hermary wrote:
> 
> > Hi all,
> > As part of our Certificate program in Organic
> Landscape Management we 
> > (Gaia College) developed an interactive plant
> database for our 
> > students.  As one of their assignments our
> students are required to 
> > thoroughly research 5 plants according to the
> following criteria, and 
> > enter the information in the database:
> > NOMENCLATURE: Botanical Name(s), Common Name(s),
> Family
> > DESCRIPTION: Type (woody or herbaceous perennial,
> annual, etc.), Size, 
> > Rate of Growth, Form, Trunk or Stem, Buds,
> Foliage, Flowers, Fruit. 
> > CULTIVARS
> > PROPAGATION METHOD(S)
> > NATIVE HABITAT
> > TOLERANCE TO CULTURAL CONDITIONS
> > RELATIONSHIPS (synergistic / antagonistic) - this
> includes companion 
> > plants, pollinators, animal and microbial
> predators (so-called pests and 
> > diseases) - and yes, some can be both antagonistic
> and synergistic at 
> > the same time.
> > VALUABE ATTRIBUTES (food, fiber, medicine, etc.)
> > REFERENCES
> > 
> > Students are required to upload lots of pictures
> and create links to 
> > on-line reference material, but they are also
> required to describe the 
> > plants (in plain text) for search purposes.
> > 
> > The goal was to eventually make our database
> available to the public - 
> > once we had it "perfected". I can see from the
> discussions that this 
> > planned database will surpass anything we could
> achieve with our limited 
> > technological capabilities and funds. I also think
> it would be desirable 
> > to cooperate and pool our energies on this venture
> for everybody's benefit.
> > 
> > If your database can accommodate our needs for
> identification features 
> > (including searches) I would switch our students -
> and the information 
> > we have collecged so far - over as soon as you are
> ready.
> > 
> > Do you think this would be possible?
> > 
> > Cheers, Heide Hermary
> > Gaia College
> > http://www.gaiacollege.ca
> > 
> 
> I say go for it. Just the sort of thing we need.
> 
> It does open up some interesting technical Q's, on
> how to merge two 
> databases. I suspect this is the sort of thing we
> will have to deal with 
> in the future, so now would be a great time to
> explore the problems of 
> integrations.
> 
> Rich
> 
> _______________________________________________
> pcplantdb mailing list
> pcplantdb@lists.ibiblio.org
> http://lists.ibiblio.org/mailman/listinfo/pcplantdb
> 



                
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - Find what you need with new enhanced search.
http://info.mail.yahoo.com/mail_250