From: lsanger(a)nupedia.com [mailto:firstname.lastname@example.org]
Sent: Tuesday, October 09, 2001 10:13 PM
To: Michael Dill
Cc: Magnus Manske; jwales(a)bomis.com; lsanger(a)nupedia.com
Subject: RE: PHP wikipedia
What we need is a test wiki. I don't much like the idea of using the
German wiki as a test wiki--they've done a huge amount of work already,
and we don't want to mess it up. Maybe we can make a copy of it and put
Go ahead and try the German version for the first
tests. We can probably get Clifford Adams or someone
to help sort out the current perl/CGI data formats. I
beleve that Jimbo might have some information there...
I don't know enough to help with the data conversion,
but if we cant get any support from Clifford or
someone at BOMIS, I will see if I can find a
programmer or two to look at the tarball and help you
with the structure.
Would you help point us in the right direction here.
Thanks to you all for great work!
--- Magnus Manske <Magnus.Manske(a)epost.de> wrote:
> --- Mike Dill <michael_dill(a)yahoo.com> wrote:
> > Yes, call the current PHP script at
> > wikipedia.sourceforge.net
> > I think we need to clear out everything in the
> > database and see how to install a copy of the
> > data. This will allow us to test/debug the script
> > the data conversion. It will probably take a dozen
> > trys at porting it before we get the conversion
> > into a reasonable shape. Then we can ask all the
> > wikipedians to look at all the pages to check for
> > remaining script inconsistancies.
> We not necessarily need to clean the DB. And, should
> we try another
> wikipedia (e.g., the German one) first? It's much
> smaller, so conversion
> won't take that long for the first tries. If that
> works, we can purge the DB
> and install the "real thing". What do you think?
> > If we can get those bugs out, then asking Jimbo to
> > make the transition would be a reasonable
> > Until we can port the current data with some
> > I feel that we should not push it on the world.
> > Lets try a data convertion from the tarball and
> > what comes out. When it seems reasonable to us
> > Larry and Jimbo can decide when it should be
> > implemented.
> I looked at the German tarball, which seems to be
> quite old, and I suppose
> the "real" data is in the "page" sub-sub-...-subdir.
> Do you know if that's
> it? Are just the old versions stored as diffs, or
> the current version as
> well? I'm not that good at perl/CGI, so...
Do You Yahoo!?
Make a great connection at Yahoo! Personals.