Hi all,
We have meta.wikipedia.org to discuss general issues like this "Big
Picture issues".
On Meta it is discussed on http://meta.wikipedia.org/wiki/WikiImages.org
If we discuss this things on this mailinglist, what is then meta for?
Do I have to write every thing two times?
Thanks for helping me to understanding this,
Fantasy :-)
Found and fixed in dev and stable. The linkcache entries weren't being
cleared from memcached, so subsequent visits would pick up the wrong
stored page id number and get very very confused, leading to the
appearence that the move failed or that both the old and new titles
apply to the page.
I've added LinkCache::clearLink() and called it from the move page
functions to keep things clean.
-- brion vibber (brion @ pobox.com)
LDan wrote:
>That's a good idea, but aren't there some big
>technical problems with merging everything (except
for
>the mailing lists)? For example, what if there are
two
>different pictures with the same name? ....
That is a very valid point. And in fact there are many
hundreds of these, but most are exact copies. It would
probably be easier to use the largest image database
(en.wikipedia's images and other media files) as a
base, add then add all the others. For name conflicts
we could ignore anything that is exactly the same size
and use the en.wikipedia version (since it is probably
exactly the same photo). All other name conflicts will
have to be fixed manually. We have already had to do
similar manual work when we disallowed external image
display on en.wikipedia. That wasn't too painful or
time-consuming.
>The same goes for logins.
Most of the duplicate logins will either be from the
same person or from a person who is no longer active
(most of whom just created the account and never made
more than 10 edits). We can make the job easier by
auto-exiring all user accounts that have made fewer
than 10 edits after say 120 days of inactivity. That
should clear many thousands of user names and thus
prevent many conflicts. We could even have a 100 edit
/ one year auto-expiration to clear out even more user
names.
>And why would merging logins fix the problem of
>people not going to the right wiki?
I didn't claim that. But it would reduce the threshold
needed to participate as a logged-in user across
wikis. Not having to deal with separate logins would
tend to encourage more cross-wiki work.
Oh, and the name should not preclude non-image files.
files.wikimedia.org or the slightly odd but more sexy
multimedia.wikimedia.org would work better.
-- Daniel Mayer (aka mav)
__________________________________
Do you Yahoo!?
Exclusive Video Premiere - Britney Spears
http://launch.yahoo.com/promos/britneyspears/
Jimbo:
>Yes, about as odd as having Wikipedia's images at
>images.wikimedia.org, right?
Not in the least. Wikipedia is part of Wikimedia. It
is perfectly logical and the people who would be
surprised probably don't know about the Wikimedia
Foundation yet. That should spark some curiosity and
result is them learning that there is a larger
community creating all sorts of free content.
Images should be a resource that are easily usable by
all Wikimedia projects and language versions. All one
should have to do is upload an image once and then
link to it from any Wikimedia project or language
version. That is why images.wikimedia.org makes sense.
>O.k., I'm not against any of this.
Great! Now we just need to implement it. I volunteer
to help but I don't yet have the knowledge to do so or
the access. I will learn (any links would be helpful).
>Meta, in particular, is a real curiosity these days.
>No one seems to know what what it's for.
The main purpose of meta overlaps with the purpose of
the mailing lists (discuss the Wikimedia projects).
But meta is a wiki and is thus ill-suited to threaded
discussions (and yes I'm breaking threads right now
because I'm writing from work and using webmail).
In the future I would love to see us move away from
the mailing lists and use meta more. Then it would be
more natural to distill the discussion into meta
pages. But that requires MediaWiki features that
simply do not exist yet.
-- mav
__________________________________
Do you Yahoo!?
Exclusive Video Premiere - Britney Spears
http://launch.yahoo.com/promos/britneyspears/
Hello wikitech-l,
I am trying to install the mediawiki-20030829.tar.gz on my
webspace as a readonly mirror of the german cur database.
I already got it managed to insert the 92MB mysqldump
of the german cur_table into my mysql database.
But I can't install the mediawiki. I do have a webspace packet
with 200mb space, PHP 4.2.1, Apache 1.3.23 and MySQL
3.23.51-log on Linux 2.4.9-31 RedHat7.2. I also have phpMyAdmin
2.5.1 (but the mediawiki INSTALL file says: Recommended versions
are: Apache 1.3.27 or later; MySQL 4.0.13 or later; PHP 4.3.2 or later.)
I do not have the mysql root password, access to the php.ini
nor access to a shell. The name and the user of the database as
well as the directory on the webspace
("DOCUMENT_ROOT = /www/htdocs/lexion/") are "lexion".
This is what I did:
uploaded the phase3 directory from the gzip package to my
webspace. Renamed those 2 files to LocalSettings.php and
AdminSettings.php. In the localsettings I entered:
$IP = "/www/htdocs/lexion/wiki";
...
$wgServer = "http://www.mydomain.org";
$wgArticlePath = "{$wgScript}?title=$1";
$wgEmergencyContact = "wikiadmin(a)mydomain.org";
...
$wgDBserver = "localhost";
$wgDBuser = "lexion";
$wgDBname = "lexion";
$wgDBpassword = "well,mypass...";
$wgDBsqlpassword = "well,mypass...";
...
$wgReadOnly = true;
In the adminsettings I also entered my mysql user/pass.
Than I added a .htaccess file with:
"AddType application/x-httpd-php .php .phtml .php3 .php4"
So that the phtml files are also parsed with php.
Than I opened http://www.mydomain.org/phase3/install.php and
got:
Directory "/www/htdocs/lexion/wiki" created. Directory
"/www/htdocs/lexion/wiki/upload" created. Directory
"/www/htdocs/lexion/wiki/style" created. Directory
"/www/htdocs/lexion/wiki/upload/tmp" created. Copying files... * * *
Warning! This script will completely erase the existing database
"lexion" and all its contents. Are you sure you want to do this? (yes/no)
I can't answer the questions, since I didn't run it with a
shell. And it would be great if I could keep the lexion
database with the cur table, it was hard enough to upload and
import that 92megs...
If I know access the www.mydomain.org/wiki/wiki.phtml I get a
error 503. (Yeahh I know, install isn't finished...)
Please let me know if you know anything that could help me. Is
it possible with the conditions mentioned above to install that
script? What could I do?
Thank you very much in advance!
--
Best regards,
Freerk mailto:freerk@gmx.net
Opps! Last post was to wrong list. Sorry.
Jimbo wrote:
>And images.wikimedia.org would be a project
>unto itself, so to speak, where images are
>collected (perhaps automatically) and copied
>from the other projects, and given neutral
>captions, and entered into a big image library.
>The rules for inclusion in *this* would be very
>strict, not relying on fair use. All images
>would have to be public domain or GNU FDL or
>something compatible.
I like that idea! But just replace
"images.wikimedia.org" with either
"files.wikimedia.org" or "multimedia.wikimedia.org"
(since we want more than just images), and replace
"project" with "sub-project" (since its resources
would be used by all projects - like meta). This also
preserves the per wiki upload utility which would be
oddly disabled for Wikimedia projects under my plan
but would still be needed by non-Wikimedia projects
that use MediaWiki (thus there would be little
Wikimedia developer interest to maintain that
functionality). But there would need to be a prominent
mention and link to the multimedia file database on
every Wikimedia project's upload special page.
Something like "You can use this page to upload
multimedia files from your harddrive. But you may also
want to visit [our multimedia file database] to see if
a similar file is already there."
Then when the much-talked-about upload form is
enacted, the software could automatically copy
anything tagged as PD or under a free content license
to the Wikimedia database. Anything marked as "fair
use" or "special permission" would just reside on the
wiki's own upload space.
>And then if people needed an image for something,
>they could just copy it from there.
Why copy when you can export! :) There could be an
"Export this <filetype> to the [<language>]
[<Wikimedia project>]" as <file name>.<ext>" (it would
also be neat if it showed everywhere in Wikimedialand
where it is being used). Oh and anything exported to a
different file name should not create a copy of the
same file at the Wikimedia file database under a
different name. Category tags can also be used here as
well (that would help with searching).
-- Daniel Mayer (aka mav)
__________________________________
Do you Yahoo!?
Exclusive Video Premiere - Britney Spears
http://launch.yahoo.com/promos/britneyspears/
Jimbo wrote:
>Hmm, well, it is going to be usable by all Wikimedia
>projects no matter what we call it.
Anything which is usable by all Wikimedia project
needs to be at a project-neutral domain. Would it not
be odd to have Wikipedia's images at
images.wikibooks.org?
>>Which reminds me: Why hasn't this list been moved to
>>wikitech-l at wikimedia.org yet? Wikipedia-l has...
>I don't really object, but just note that these moves
>have been done without my conscious knowledge or
>support or opposition. I'm not really clear on what
>the point is, actually.
Because we are all part of Wikimedia. In an offlist
post to me you also added your support to moving meta
to the wikimedia.org domain. Moving mail.wikipedia.org
is a natural extension of that and it would then be
logical to depreciate @wikipedia.org for
@wikimedia.org.
>It does make sense to me that stuff having to do
>primarily with wikipedia, such as wikipedia-l for
>example, should be on wikipedia.org, while stuff
>having to do with the foundation as a whole,
>encompassing more than just the flagship
>encyclopedia, should be on wikimedia, like the legal
>mailing list, etc.
Wikipedia is part of Wikimedia. To help with a
cultural shift to become less Wikipedia-centric,
moving the Wikipedia-related mailing lists to the
wikimedial.org domain makes perfect sense. Otherwise
each project will want to have their own mail.<project
name>.org subdomain. That means we will have more
subdomains to set-up and maintain and the different
projects will be more isolated from each other than
need be.
We are /all/ one super-community here - Wikimedia -
thus we should all share things such as mailing list
domains, image space, meta, log-ins etc. This will
tend to bind us all together, instead of being
separate projects which happen to be on the same
server. People from one project should be reminded
where it makes sense that there are other Wikimedia
projects. This will encourage cross-polination of
ideas and also help to make sure projects do not
duplicate work (no need for textbooks in Wikipedia, no
need for dictionary entries in Wikibooks etc).
>Is there any reason why we should move *all* the
>mailing lists to wikimedia?
See above. If the language-specific lists are also
moved then they should be renamed in this format:
WikipediaXx-l. Brion has already stated that this
should not pose technical problems and should in fact
be fairly easy to do. It would also be confusing to
have a mix of @wikipedia.org and @wikimedia.org
mailing lists laying around - they are too similar and
it is not apparent that they both work. Best to have
just one and it is best to have that be at a
project-neutral domain name.
Oh, it would also be nice to have "A [Wikimedia]
project" under the logo of every Wikipedia project.
[Wikimedia] would be a link to the Foundation's home
page. This would be another thing we could do to help
enforce the idea that we are all together and also
help direct the energies of newbies to the right
projects.
Big picture issues.
-- Daniel Mayer (aka mav - hopeless INTJ mastermind)
__________________________________
Do you Yahoo!?
Exclusive Video Premiere - Britney Spears
http://launch.yahoo.com/promos/britneyspears/
I wrote:
>Oh, it would also be nice to have "A [Wikimedia]
>project" under the logo of every Wikipedia project.
>[Wikimedia] would be a link to the Foundation's home
>page. ...
Grr. Replace "every Wikipedia project" with "every
Wikimedia project." Yes, even I am still struggling
with my Wikipedia-centratism.
-- Daniel Mayer (aka mav)
__________________________________
Do you Yahoo!?
Exclusive Video Premiere - Britney Spears
http://launch.yahoo.com/promos/britneyspears/
Jason wrote:
>Why not just serve the images from
>images.wikipedia.org (which I imagine would resolve
>to the database machine for the moment). Am I
>missing some key point that makes something fancier
>necessary?
images.wikimedia.org would be better since it would be
usable by all Wikimedia projects.
Which reminds me: Why hasn't this list been moved to
wikitech-l(a)wikimedia.org yet? Wikipedia-l has been...
For that matter mail.wikipedia.org,
download.wikipedia.org, test.wikipedia.org and
meta.wikipedia.org should also be moved to the
wikimedia.org domain.
And before anybody mentions that I could move it,
please consider pointing me to some tutorials on how
to do it and then give me the necessary developer
access. I wouldn't mind doing some of the more mundane
development tasks to free up some time for the real
developers, but I really should know what I'm doing
first. ;)
-- mav
__________________________________
Do you Yahoo!?
Exclusive Video Premiere - Britney Spears
http://launch.yahoo.com/promos/britneyspears/
It doesn't matter what address we use, but the mailing lists should NOT
be supported by the same server as the website. Because when the web
server goes down, then we can't talk to each other.
By the way, if Maveric wants to take on the heavy duties of
administering the mailing lists, we should probably accept his help.
He's got the personality and aptitude for it, and he's well liked.
(Grumpy Old) Uncle Ed Poor