[Zope3-dev] Re: [ZODB-Dev] ZODB4 project plan

Nick Pavlica nick.pavlica@echostar.com
Wed, 27 Nov 2002 16:12:50 -0700


My 2 Cents!

  In theory, Zope versions give "Web Application" developers and=20
administrators allot of power, however in practice the transactional=20
undo/history, copy, paste, and rename functions provide an extream amount=
 of=20
functionality in maintaining applications in the Zope space (other=20
application servers could learn from this).  While Zope is my favorite Py=
thon=20
application to date, the need for good object oriented databases is growi=
ng=20
rapidly.  If ZODB evolves into a "Quality" general purpose database, Zope=
=20
will certainly benefit because the time and energy spent improving the=20
database will increase the performance, scalability, and reliability of Z=
ope,=20
and ZODB based applications.  I feel that the ZODB should become the next=
=20
killer Python application, and could provide host of new opportunities fo=
r=20
the Zope and Python communities.  There are some features that I would li=
ke=20
to have in any database, and hope that in time the ZODB will accomplish t=
hem. =20
All of these features are achievable, and have been done in OO databases =
like=20
Objectivity.  Their db architecture is definitely worth reviewing, it may=
=20
spawn some new ideas in the ZODB development.

ZODB Wishlist:
- Distributed
- Replication
- Data Consistency/Integrity (ACID)
- Multiple Storages (physical and or logical)
- Scaleable (TB +)
- Highest Possible Performance
- Easy to use, maintain, and develop for.=20
- The ability to Query objects.


On Tuesday 26 November 2002 1:22 pm, Guido van Rossum wrote:
> > I've just posted a very rough ZODB4 project plan in the Wiki:
> >     http://www.zope.org/Wikis/ZODB/ZODB4
> >
> > I've delayed posting this draft because I wanted to post a better
> > draft.  I've stopped delaying, because I probably won't get much time
> > to make it better until I start working on the individual tasks.  But
> > I'll be happy to answer questions where it is unclear and add items
> > that people think are important.
>
> An additional issue that came up in a meeting here at PythonLabs is
> the fate of "versions".  Basically, we recommend against using them
> with Zope.  So they are only there for backwards compatibility or
> perhaps for people using ZODB outside Zope.  Removing versions will
> remove complexity to the code all over the place: the version argument
> to some APIs can be removed, the ZEO cache implementation and wire
> protocol become simpler, and so on.
>
> Is there anyone who would regret losing versions in ZODB4?
>
> --Guido van Rossum (home page: http://www.python.org/~guido/)
>
> _______________________________________________
> For more information about ZODB, see the ZODB Wiki:
> http://www.zope.org/Wikis/ZODB/
>
> ZODB-Dev mailing list  -  ZODB-Dev@zope.org
> http://lists.zope.org/mailman/listinfo/zodb-dev

--=20
Nick Pavlica
EchoStar Communications
CAS-Engineering
(307)633-5237