Hi folks,
I'm planning on moderating a discussion on MediaWiki in the enterprise
(businesses, not the starship) next week at Wikimania (Friday, 8/4), and
I'm hoping to kick off a pre-discussion about that here. My motives
aside, this conversation doesn't need to be about my session to be on
topic...I'm interested in a general conversation about MediaWiki in the
enterprise.
I should start off with a disclaimer to the core MediaWiki devs here:
my knowledge is mainly based on 1.5, and my dev chops are a little rusty
right now. You'll see some half-baked ideas below that I'm planning to,
um, bake a little bit in the next week. I'm also planning on being
around for at least some of Wednesday and all of Thursday of Hacking
Days, so I'm happy to discuss then if asked.
Regardless of whether you're going to Wikimania or the session, I'd like
to hear from you if you have thoughts on the subject. Specifically:
* Users: what sorts of things are you doing?
* Developers: what are the most frequent feature requests from intranet
users? What are the most annoying ones? Which are the best that
haven't been implemented yet?
* Propsective session attendees: I'm really interested in anything you
have to say about what you hope to get out of the conversation.
Here's some things that I see as potential things to discuss:
* Authentication plugins. The authentication interface is very
good...it's gotten a fair workout from a number of angles (particularly,
LDAP). Seems to be reasonably mature.
* Authorization plugins - not much as of MediaWiki 1.5....I'm guessing
not much has changed. I'm planning to possibly discuss ways in which
User.php could be overridden, extended, and/or stabilized as an
extension interface. The idea would be a system where group memberships
are stored in an external database, as well as capability assignments
(e.g. LDAP group "admins" gets the "edit" capability on resource
"MediaWiki: namespace") This is admittedly pretty fuzzy right now.
* More robust read-access controls (and more granular access control in
general). Oft requested, oft rejected. I'm interested in starting a
conversation about what interfaces could be added to the code such that
an ACL-maniac extension to MediaWiki could exist, without a big support
burden on the core team. Before replying to this, please read Brion's
reply to me from November, 2005 on this subject:
http://permalink.gmane.org/gmane.org.wikimedia.mediawiki/8610
* New distribution mechanisms. There's already a Debian package for
MediaWiki, I'm assuming there's an RPM out there. Does a PEAR
distribution make sense? Does it make sense to package MediaWiki
extensions as .deb/.rpm/PEAR thingies?
* Structured data - I know that in my last job, there were several
things we used a wiki in lieu of a database for (e.g. we kept a list of
new licensees of our technology on a wiki). It was nice, because we had
the flexibility to add fields willy-nilly. So, it may be interesting to
discuss the enterprise applicability of projects like
* Training - it's often hard to get someone who is used to mailing
around Word docs as attachments to use a wiki with its
{{{random-adjective}}} syntax instead of WYSIWYG. How do folks deal
with it?
So, that's just some top of my head ideas. Others?
Those are some of my thoughts for now. I'm hoping to get more organized
as Friday draws closer.
For those interested in the session, the details are here:
http://wikimania2006.wikimedia.org/wiki/Proceedings_talk:RL1
The talk page for the page above is a good discussion venue for items
that aren't of general interest on this list.
Looking forward to hearing from you.
Thanks,
Rob
We have just setup a company wiki.
The Versions page reports:
* MediaWiki: 1.7.1
* PHP: 5.1.2 (apache2handler)
* MySQL: 5.0.22-community-nt
We're having a problem which a google/google groups/mediawiki help
search did not resolve.
When a user is logged in they are unable to perform any actions. And
when editing a page one gets this error after clicking "Save Page":
"Sorry! We could not process your edit due to a loss of session data.
Please try again. If it still doesn't work, try logging out and logging
back in."
As a result everyone is anonymously editing pages, which is a big
problem. I've attempted logging out/in. Using Firefox & IE. Clearing
the brower's cache/cookies/passwords. And also making sure that 'Show
preview on first edit' is not checked under Preferences.
Any ideas?
Thanks in advance,
--
Christopher Miller
cmiller(a)creativeengineering.com
(914) 771-5540 x20
> From: elisa.cooper(a)breakingranks.net
>
> I just took a look, and I can't find a button that says "rollback" on
> either the History screen or the Edit screen, though. :-(
It's on neither of those.
Do the "history" tab, THEN you MUST press "Compare selected versions"
before you will see a textual link: "[rollback]" on the right.
:::: Think outside the box! The problem is now "How do you get
different fruit from the same Bush?"
:::: Jan Steinman <http://www.Bytesmiths.com/Item/99AN10>
We have recently started an all architecture wiki at Archiplanet.org.
As part of other sites we also offer commercial image licensing of
user submitted images. We would like to be able to do this with our
wiki as well.
Ideally people would upload a large, several megapixel, image. An
untouched version of this file would be stored outside the public
wiki. A version scaled down to about 1,000 pixels across with a small
watermark in the corner would be added to the wiki. For making
smaller versions for pages that are bigger than 250 pixels across the
raw original would be used and a watermark added. For thumbnails 250
pixels or less, no watermark would be added.
I'm new to Mediawiki, but I think I can do at least some of this
using hooks that exist. Are there any extensions or other features
that would aid in this process at all?
The most important bit is storing an unmarked full size version out
of public view and a marked smaller version as part of the wiki.
Thanks,
Mike
I want to put a default category for new pages as the "logged-in user
drafts"
[[Category: Username Drafts]]
so as to keep 'work in progress' seperate from articles that are complete.
Is there a keyword or a simple tweak to get current username?
thanks
- imti
Hi folks,
I've been looking at the namespace manager in wikidata, as well as
looking at how things are in the trunk. I'm eager to have improved
namespace facilities, since it makes my life easier in creating
extensions. It may be easier to discuss these ideas face-to-face later
this week, but I figured it'd be better to compose my thoughts as a
starting point.
Suggestions:
* Change $wgNamespace to a static variable, accessed by new global
function: wfGetNamespace($nsnumber). This makes it much easier to
enforce the edict "Do NOT alter the $wgNamespaces object" (especially if
you threaten to return a clone from wfGetNamespace in future versions,
performance be damned)
* Create mNamespace member variable for the Title and Article objects
which is merely a reference to the corresponding namespace object (viva
PHP5).
* Add utility function for getting the default name given a constant.
Change this:
global $wgNamespace;
$wgNamespaces[NS_TALK]->getDefaultName()
...to one of the following, as appropriate:
wfGetDefaultNSName(NS_TALK)
...or (preferable when applicable):
$this->mNamespace->getDefaultName()
* (down the road) Create new objects which extend the Article object,
and remove some/all of the namespace conditionals from the Article
object altogether (e.g. UserTalkPage object for User_talk:, SpecialPage
for "Special:", and SystemPage or MediaWikiPage for "MediaWiki:")
* Consolidate "new Article" instantiation in a class factory. It may
be easiest to make this a method of the Title object, so that rather
than creating an article like this:
$article = new Article($title)
...you create it using:
$article = $title->newArticle();
(actually, this doesn't need to wait for a new namespace manager...I may
just submit a patch for this)
Those are some thoughts for now. Looking forward to seeing everyone at
Wikimania.
Rob
"Esta mensagem do SERVIÇO FEDERAL DE PROCESSAMENTO DE DADOS (SERPRO), empresa pública federal regida pelo disposto na Lei Federal nº 5.615, é enviada exclusivamente a seu destinatário e pode conter informações confidenciais, protegidas por sigilo profissional. Sua utilização desautorizada é ilegal e sujeita o infrator às penas da lei. Se você a recebeu indevidamente, queira, por gentileza, reenviá-la ao emitente, esclarecendo o equívoco."
"This message from SERVIÇO FEDERAL DE PROCESSAMENTO DE DADOS (SERPRO) -- a government company established under Brazilian law (5.615/70) -- is directed exclusively to its addressee and may contain confidential data, protected under professional secrecy rules. Its unauthorized use is illegal and may subject the transgressor to the law's penalties. If you're not the addressee, please send it back, elucidating the failure."
---------- Forwarded message ----------
From: Erik Moeller <eloquence(a)gmail.com>
Date: Jul 31, 2006 4:45 PM
Subject: Wiki expert in Nigeria needed
To: Wikimedia Foundation Mailing List <foundation-l(a)wikimedia.org>,
wikipedia-l(a)wikimedia.org
A friend of mine, Wayne Mackintosh of the Commonwealth of Learning
(col.org), is looking for a local wiki expert in Nigeria to help with
wiki skills development "to a group of folk with low technical
experience." The project "is aimed at developing free content for
health professionals." The wiki expert would be paid, of course.
Anyone know someone who might qualify, or someone who might know someone?
Thanks,
Erik
Hi,
If you care to take a look at
http://www.startux.org/index.php/Apache
you'll see that there are lines that doesn't wrap inside the box.
Why is that and how can i solve this issue ?
Any help would be appreciated.
Warm Regards,
Mário Gamito