As has been pointed out by posters on this DL, I've now "resigned"
twice from the project. On the first occasion I said I was leaving but
committed to finish off my forum / wiki work first as long as we could
avoid a ping-pong of personal criticisms / hostile posting on the
ooo-dev DL. The second time I dropped this commitment because it became
clear to me that we were again slipping into personal attacks. Even so,
what I don't want to happen is for my decision to leave to cause a
"denial of service" because of my being a "single point of failure" for
some activities. However, I won't discuss this thread further on this
DL. I leave this to TJ et al. I will reply to constructive personal
emails, and am happy for any recipients to quote that content here.
In the case of the forums, the bulk of the setup work had already been
completed and so I have been working with Drew Jensen in the background
to transfer their support and administration to him should the project
decide to proceed with this migration. I have pretty much documented
these tasks.
The wiki is a somewhat different issue, because this work wasn't so far
along, and so I wanted to make some statement on the DL on this. We
have a stable build of MW 1.15.5 (albeit not in svn) running on
ooo-wiki.apache.org with a stale but recent clone of the production wiki
on it, as the only material content updates have been by a user
Hohenheim to some of the OOoES NL pages. I have also activated TJ's
account and upgraded this to admin / bureaucrat, and I am willing to
offer TJ the same occasional support that I am currently providing to
Drew whilst he gets up to speed.
Whatever the project's long term plans for the wiki, it will remain
provide a useful reference resource from which to migrate content if the
Oracle service is shut down. IIRC, a couple of contributors /
committers have asked for a copy of the schema and content so that they
can work on it and on any data migration. So making this available seems
to be a sensible way ahead since the vast majority of the content is
already publicly accessible and could used / scraped under the terms of
the PDL, etc. I will place this on the usOOo site. So please expect to
see a file
http://user.services.openoffice.org/__backup/d27f1bb7c71872c4d2e6eb04d4fc6006/README
posted next at the weekend which will explain how to download this content.
The issue that I need to be careful about (really on behalf of Oracle /
the old OOo projects) is that some data elements are not already
publicly accessible and which could in my view be deemed to be personal
data falling under EU Data Protection legislation, so I would need to be
a little careful about how I expunge these from any copy for
developers. As far as I can see these could all be mangled to avoid
such issues:
*user_password*
This is already stored in highly mangled (MD5 including salt) format so
is not readily attacked apart from brute force guessing. What I will do
is to replace 3 internal letters by 3 pseudo random letters to make such
this attack more impractical. (see note 1)
*user_real_name
*This field is not made publicly available, and is only used for content
attribution. I will therefore blank this in the dump. However I will
keep a copy and make this accessible to any PPMC member who warrants
that this will be used for the purposes of content attribution.
*user_email*
This field is not publicly available. I will replace it by
strcat(md5(user_email), '@dummy.com') (see note 1)
*watchlist
*Remove all watchlist entries. Users can always recreate their own list.
*Notes:
*
1. Part of my reason for doing this was a point that Rob made: that
Apache should require users to explicitly acknowledge some form of ToU
as part of re-registering to recover any existing account. I have
already started to develop a MediaWiki extension form on my VM MW 1.17.0
version which would allow any user to "reconnect" to his/her account if
either (i) they knew the existing password, or (ii) knew the existing
email address and could issue email confirmation from that address. Any
competent PHP/MW developer could easily do this with the above data
element mangling.
2. I will delete all accounts that have made no contributions to the
Main or User namespaces.
3. I am still concerned about this whole issue of content attribution.
IIRC, under the old OCA contributors only vested joint rights to
Sun/Oracle and free licence on any patents. The base copyright and
patents were maintained by the originator. The PDL is even tighter of
contributor rights retention. So far we have discarded all audit of
contribution in svn and are currently considering doing likewise for
documentation content, dropping contribution audit trails and defaulting
to blanket Apache copyright. As some third parties might take a
different view on this, I propose to keep a full private archive of the
D/B myself as an independent audit reference.
With these changes, I would be happy to put a copy of the wiki dump on
one of the old OOo servers so that it could be pulled by any ooo-dev
developers.
