Skip to Content.
Sympa Menu

pcplantdb - Re: [pcplantdb] checkin

pcplantdb@lists.ibiblio.org

Subject: pcplantdb

List archive

Chronological Thread  
  • From: Richard Morris <webmaster@pfaf.org>
  • To: PCPLANTDB <pcplantdb@lists.ibiblio.org>
  • Subject: Re: [pcplantdb] checkin
  • Date: Thu, 21 Oct 2004 00:30:03 +0100

(somehow I've lost the original post so I may be out of cotext).

Lawrence F. London, Jr. wrote:
Chad Knepp wrote:

Options:

o Keep developing HG in Zope as planned. This is a doable option
although I'm currently not very excited about it. I also am no
longer sure that developing in Zope will be in anyway better or
faster than alternate methods.

Somehow I've got the same feeling about Zope. It is too complex a system to really do the quite adventious system we are trying to put together.
It seems good for doing fairly stratforward CMS systems but when it comes to more involved objects it seems awkard.
So much time needs to go into getting things to work the zope way.

End of the day we will get the best results, if you work in the environment you feel happiest in.

o Develop from Eden. Look into ORMs and/or take a week off to write
one. In my current frame of mind this sounds like what I would
like to do, although it means we wasted 33 hours or about 5% of
our current funding.

From a puerly personal viewpoint I prefer this option. I feel I am nowhere near Chads level in understanding Zope this makes it hard for me to do a good job of helping Chad at the code level. I'd feel easier working with systems I know more about.

Looking at ORM's seems good. I'd actually consider doing it by hand,
possibly the best way to actually understand the mapping.

Nothing was wasted. I think this is the best new path to take. My vote is yes
for Develop from Eden.

Even if we no longer use Zope I think we have made great progress. I feel we know much more at a code level the type of system we need.
In particular I feel that we know better what the Objects in the OO system need to be. From my experience this is the hardest part of getting an OO system together, it takes a lot of messing about and a lot going into the waste basket before you actually arrive at the correct set of objects, its only then that the real coding can begin.

o Do something even more radical like just use Medusa (ZServer) and
ZODB together getting rid of all the Zope cruft. Part of the
reason I was so sold on Zope is that I really liked the
development of ack! in ZODB. It might be interesting to integrate
ZODB with Twisted <twistedmatrix.com>.

Hum, better than Zope I think. Is the ZODB the best object database about, is it easy to use to store our own objects with our own indexing systems and relations between the objects? Theres a host of other object databases about, I found a good site which compared object databases and their capabilities, apachee have an intresting one.

Form a brief look at twisted I'd say cool, it seems to provide a lot of the stuff we need. However worth looking at
http://twistedmatrix.com/documents/current/howto/tutorial/index
The code there seems nice and understandable. More a library than
the Zope type of system (a good thing imo).

It might be worth examining DIVMOD http://www.divmod.org/Home/Projects/Nevow/
which seems to be the prefered twisted "Web Application Construction Kit".

Interesting times!

Rich

--
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






Archive powered by MHonArc 2.6.24.

Top of Page