pcplantdb

[Chad Knepp <pyg@galatea.org>]

Richard Morris writes:
 > Chad Knepp wrote:
 > > Hello all,
 > > 
 > > Well, I'm currently feeling highly motivated to hack piw and I
 > > actually don't feel like I know what I should be working on now.
 > > Well, I think I should be implementing relationships, but many things
 > > about them are still not clear to me.  I'll send out another email of
 > > questions soon.  In the mean time I can think of a number of other
 > > things to do but I also need some guidance.
 > > 
 > First thing for me is to see dev.permaculture.info upto date.
 > i.e. a warts and all version of where your at.

dev.permaculture.info is current.  The latest changes are edit and
delete of comments.  Look at plant=4433 for some examples or make your
own with the accounts "Test User" and "Test Two" both have password of
"password".  Try to hack other accounts if you can.

 > > One of the things at the top of my list is expanding the level of user
 > > contribution to plants (creating new ones and adding/editing data) and
 > > other elements of the dataset.  This would be done as discussed by my
 > > proposed data model.  I feel that my data/security model has gotten
 > > pretty widespread support and unless I hear serious concerns about it
 > > soon I'm going to suggest that we consider it a decision.
 > 
 > Yes, go for that. Seems like a good thing to do while your energys up.
 > Leave relationships out for a bit.
 > 
 >    One of the
 > > implications though is that the current dataset needs to "be owned".
 > > I suggested that this be an uneditable account under of the name
 > > "PFAF/Ken Fern" (Rich is this name ok?). 
 > 
 > Names fine.
 > 
 >   All of this is pretty
 > > straight forward except for the 6 elements of the plant labeled
 > > Cultural Notes, Propagation Notes, Known Hazards, Edible Uses,
 > > Medicinal Uses, and Other Uses.  My question is should this become one
 > > comment, six, three (Uses, Culture, Hazards), or some other
 > > arrangement. 
 > 
 > I'm keen of having sections of data. So its posible to group
 > all comments on edible uses together. Also so its posible
 > to filter out sections from the display reports.

I'm hoping that a norm will develop of people replying to comments and
keeping a threaded discussion going under the sections Uses, Culture,
etc.

Yes, it's on my personal list of desired features to have preferences
include report customization.  This is a pretty easy feature to add.

 >   While I'm in process of moving these comments how would
 > > folks feel about doing the XMLizing of the [42] style references while
 > > I'm at it.  I would be writing one-off custom scripts to transform the
 > > data anyway and a little regex magic would just be more fun ;-)
 > 
 > Sounds good.
 > 
 > > I'm open to projects/directions other things as well such as
 > > implementing a moderation system (need more discussion though), 
 > 
 > For me moderation is a low priority. It will be some time
 > when we really get sufficient quantity of data to necesitate any form of 
 > moderation. In the slashdot system it started with no moderation,
 > then sysadmin moderation, then user moderation as the level of 
 > contributions grew. I'm inclide to follow a similar path
 > i.e. implementing stuff when we need it.

My take as well.  I wasn't thinking that moderation might not even
make it into the Threshold iteration.  On the other hand moderation
should be pretty easy to implement.

 >  > code
 > > clean up inc. client/server packaging split, locales (more
 > > discussion), somebody explain relationships to me, etc.  Anyway, I'm
 > > putting this out to my advisory board and need some direction.  I
 > > haven't worked since last Tuesday and crazy thing I'm all excited
 > > really feel like doing something.
 > 
 > One thing I'd be very keen to see is mixing in the tanplant data.
 > I know the guy who submitted it would be keen to see his data live on 
 > the web. Its also technically chalanging in how we can mix the two data 
 > sets. This seems to fit well with the comment idea. Think of this as
 > another large batch of comments. (It would be OK to miss out some fields
 > if clashes emerge).

Well, I was hoping that when authorship features where implemented,
that someone [else] would add the data.  Having the server seperated
from the client would allow xml-rpc custom clients to add datasets
such as this.

I think that our dataset should be able to accommodate all pertinent
info, if not we have some schema work to do.

Cheers,
Chad

-- 
Chad Knepp
python -c 'import base64;print base64.decodestring("cHlnQGdhbGF0ZWEub3Jn")'