Skip to Content.
Sympa Menu

pcplantdb - Re: [pcplantdb] Re: new PC plant database

pcplantdb@lists.ibiblio.org

Subject: pcplantdb

List archive

Chronological Thread  
  • From: mIEKAL aND <memexikon@mwt.net>
  • To: pcplantdb@lists.ibiblio.org
  • Subject: Re: [pcplantdb] Re: new PC plant database
  • Date: Thu, 20 Mar 2003 14:48:55 -0600

Do you have a development trial page up for users to input data?

mIEKAL


On Thursday, March 20, 2003, at 02:44 PM, Richard Morris wrote:

Chad Knepp wrote:

Nice one.

I do like the way the advance search has been laid out.
Very clear.

Is source code available?

I would like to announce a new plant database based on data from
PFAF 2000. It's temporarily available (with a few caveats) at:
<http://sandhillfarm.org/pcpdb/>
This is not a well tested product; expect errors and general
brokenness (incorrect data? talk to PFAF).

All we need now is a method to edit the data and the community
can fix the incorect data!

> If it doesn't seem to be
available or goes away suddenly it could be because it 1) exceeded
[cpu/disk/bandwidth] quota 2) crashed 3) was removed because Google is
threatening a lawsuit ;-). The above URL is temporary, but should be
up till the end of March.
Like it? Well there's more. I've improved the simple search engine
to respond well to phrases like "a fast growing shade tolerant vine
hardy to zone 5", yielding results like Virginia Creeper.
Unfortunately these *improvements* are far to CPU intensive to put out
in their unoptimized form and the search still generates more false
positives that I like.

I've been playing with an interesting search metric reciently
given two strings A and B count the number of sub string in
A which match sub strings in B. This seems to be working
remakably well. If the two string are identicle then a large
number of sub strings will match.

It has the property of catching miss spelled words very easily
which is especially important for latin names which often
have different suffixes i.e. end in 'um' or 'us' or 'a'.

Its also good for common names which often have different
punctuation i.e. using a space or a hyphen between two words.
'St. John's wort' can also be wrtten as
'Common St Johnswort', or 'St John-wort'
(this works substring 'John' and 'wort' both match).

Only problem is its pretty slow testing two strings have
order O(len(A)*len(B)).

Since this is both the first major python project and first major
database project I've done, the time spent so far (100-200 hours) has
been extremely fun. Unfortunately, in my current living situation,
those around me are not excited to have me sit in front of the
computer all day hacking just for fun. To summarize in a punny way,
FUNding has run out.

Tell me about it! I dread to think of man hours gone into
pfaf database by me.

I would still like to work on this project and have a number of ideas
to create what I think would be a very valuable asset to the
permaculture community. This post is in part to ask for leads on
foundations, grant programs, etc. that might be interested in
supporting further development of this project.

We are thinking of applying for a grant for DB development.

Here are some ideas about what further development might look like. I
don't have attachment to a particular name but I will be refering to
this project as PCDB (PermaCulture DataBase) to distinguish it from
PCPDB and PFAF.
o Collective Collaboration: The collective knowledge of a group is
always larger than any subset of that group. This is [partly]
why Open Source Software (OSS), like Linux, is growing faster
than it's proprietary counterparts. See
<http://catb.org/~esr/writings/cathedral-bazaar/> for some
theory. PCDB would PRIMARILY aquire and change its information
by handling user inputs/edits/contributions automagically. This
is the essence of a wiki, but wikis [so far] are only capable of
handling very general sorts of information.
o Multi-media, Mutlti-subject: PCDB has no P in it. Why limit this
information to plants? PCDB could easily comprise all of the
sorts of information that make up a good PC book such as theory
in the form of text, images, plant lists, etc... think really
really big book.
o Informal and Alternative Content Formats: The Internet has given
rise to many ways of sharing information that are unique to the
internet. Mailing lists, web logs, wikis, discussion forums,
etc. are some examples. Incorporating some of these content
formats should be quite valuable. For example attaching a
moderated discussion to an article or plant would allow for
others to share their personal experiences.
o Climatic Distinctions: Plants behave differently in different
climates and information for a particular climate doesn't always
apply to another. IMHO, it would be a *good thing* to actively
note these differences and allow users to filter information
based on a particular locale.
o Powerful and Easy: In order for collective collaboration to work,
it has to be easy to use. Good interface design is an important
part. This is why I think that currently PCPDB is somewhat
better than PFAF (sorry Richard).

Yes I agree the pfaf website db interface has got a little bloated.
But it was done well before google came about.
Its all due for a clean up but time is limited.

o Active Content Solicitation: I think actually going out and
collecting existing appropriate data, as opposed to waiting for
user submission, would increase the value rapidly. This would be
similar to assimilating the PFAF data.

Vital. The projects only going to work if people know about
it.

Design wise, I think it should be easy to incorperate tables
from other sources easily.

o Free and Redistributable: This is mostly just my own personal
stance on intellectual property. PCDB would always be free in
the OSS sense. This would require that contributers agree to
copyleft their submissions.

Just one point our license has now changed to the Creatibve Commons
(Atribution Non-commercial, sharealike) liscence. And require the
following to appear:

This work is licensed under a Creative Commons License.
http://creativecommons.org/licenses/by-nc-sa/1.0
(You can copy, distribute, display this works but: Attribution is required, its for Non-Commercial purposes, and it's Share Alike (GNUish/copyleft) i.e. has an identical license.)
We also ask that you let us know (webmaster@pfaf.org) if you link to, redistribute, make a derived work or do anything groovy with this information.

Anyway, putting 500-1000 hours into this project it could be pretty
interesting. I think it would take around 300 hours to get to a
collaborative state allowing others to grow and modify the data
itself, which as I see it, is the most important part.

Minor bugs:

Doing and advanced search on checking for a word in the
culture tab gave the results
http://sandhillfarm.org/cgi-bin/ pcpdb.py?advanced_search=yes&output=Tabled&hit_limit=10&family=&genus=& species=&ssp=&common_name=&cultivar=&zone=&habit=&foliage=&tall=&short= &wide=&narrow=&rating=&cultivation_details=hill&propagation_details=&us es=&known_hazards=

Latin names were not clickable. Clicking on family name gave a cgi
error and clicking on describe gave a blank page.

Also got
File Not Found: /pcpdb_glossary.html

--
Plants for a Future: 7000 useful plants
Web: http://www.pfaf.org/ same as http://www.comp.leeds.ac.uk/pfaf/
Post: 1 Lerryn View, Lerryn, Lostwithiel, Cornwall, PL22 0QJ
Tel: 01208 872 963 / 0845 458 4719
Email: webmaster@pfaf.org
PFAF electronic mailing list http://groups.yahoo.com/group/pfaf


_______________________________________________
pcplantdb mailing list
pcplantdb@lists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/pcplantdb





Archive powered by MHonArc 2.6.24.

Top of Page