-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi,
Since the DB server is down, I got some spare time looking into the software a
way to have a localized menu for Wikisource and Wikibooks with only one
domain. I have seen no official decision about that, but it seems that what
is projected. First, is someone already working on that? Is there a design
projected about this? I looked through the archives but could not find
anything.
I entered a feature request about this on Sourceforge.
Also I got the software installed and running on my PC.
I suppose that we want that the language changes acccording to users' settings
with a "user_language" field in the user table.
I see a problem with the cached pages. We will need to have a cache for each
language.
I am Ok or what?
Thanks,
Yann
- --
http://www.non-violence.org/ | Site collaboratif sur la non-violence
http://www.forget-me.net/ | Alternatives sur le Net
http://fr.wikipedia.org/ | Encyclopédie libre
http://www.forget-me.net/pro/ | Formations et services Linux
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)
iD8DBQE/7Dykm4KYjQo0y9oRAv3FAJkBk5nc9sYnDe6m6auhjBDF1C69iwCfTVO7
C2ogi7r2O2bBZZmWthGbKmg=
=gRNx
-----END PGP SIGNATURE-----
larousse - single PIII/866
pliny - dual Athlon 2600+
new machine - dual Athlon 2600+
db server - dual opteron mega server
With this setup, we can have larousse act as the load balancer
frontend for the 2 webservers pliny and new, which can be mirror
images of each other. This is much cleaner than any sort of dns-based
setup.
Also, in the short run, it will permit us to just put the intl-wikis
on the new machine and keep en on the existing machine, or something
similar, which should be a pretty reasonable way to load balance as
well while we work out the details of the other architecture.
The PIII/866 is *more* than powerful enough to do iptables load
balancing, if anyone is wondering about that. I've done this before,
and you can handle a ton of traffic with not a lot of hardware,
because packet rewriting is easy.
And once we have this structure in place, one great thing is that it's
very scalable -- to increase capacity we just add extra webservers,
which can go into the mix seamlessly.
--Jimbo
I've altered the rewrite rules for the en/en2 split so that user agents
that don't identify themselves as some variant of "Mozilla" are kept
solely on en.wikipedia.org.
The vast majority of web browsers identify as "Mozilla/XX (something)
blah blah" for historical compatibility reasons. The vast majority of
web search engine indexing spiders, on the other hand, do not. This
should help clean up search indexes so a search limited to
'en.wikipedia.org' can turn up all articles, and internal linking will
be properly taken into account.
This will push a little extra load onto larousse, but not I think too
much.
Just a note: on an overall basis, the traffic for the English wikipedia
is about the same as all the others combined. If we could get the
processor upgrade for larousse actually in place, well that would be
great. Since that doesn't seem to be forthcoming, alternatively, we
could move some of the languages to larousse and put English on pliny
only.
This should help balance things a little more cleanly without yet the
need for IP virtualization. But, that'll require updating DNS for
wikipedia.org, which I don't have access to. Any thoughts on this plan
before we go trying to organize it?
-- brion vibber (brion @ pobox.com)
Just a thought:
At the moment, some articles have disclaimer notices added (in addition
to the general disclaimer covering the whole encyclopedia, of course).
The new MediaWiki namespace has recently improved things by allowing the
text of article disclaimer notices to be standardised and updates
automated.
The process of adding these per-article notices could be made still
easier when the name-value pair infrastructure is added by allowing the
addition of _automatic_ disclaimer messages based on article category
tags. These texts could be sysop-editable, in just the same way as the
MediaWiki tags (or indeed might be implemented as MediaWiki tags).
-- Neil
People on the German wikipedia noticed that a mirror site of the German
wikipedia (wikipedia.t-st.de) has higher Google ratings than the "real"
one, which was attributed to their use of <meta keywords=""> tags.
(Supposedly, Google doesn't care about meta tags, but what do I know?)
I vaguely remember that (at least in Phase II) there were meta-keywords,
generated from the title of the article and all wiki-links. Also, it had
a meta "summary" tag from what the software thought was the first
paragraph of the article.
Could we turn that on again? Might cost some bandwidth, but if it really
gets us higher ratings, we should do this. Maybe for anonymous views
only, since logged-in users probably don't care for meta tags...
Magnus
--- Brion Vibber <brion(a)pobox.com> wrote:
> On Dec 20, 2003, at 08:41, Daniel Mayer wrote:
> > So I guess the needs of Wiktionary, Wikiquote,
> > Wikibooks, and Wikisource (not to mention
> > Wikitravel) are secondary then?
>
> Yes, they are. They're smaller, less popular
> projects, and "the needs of the many outweigh the
> needs of the few".
And minority rights mean next to nothing then? Should
their needs be put permanently on the back burner just
because they are have fewer hits? What about the very
small Wikipedias that get close to no traffic - is
setting more of them up really more important than
internationalizing Wiktionary in German and French?
Some Wikipedias have set up internal dictionaries due
to the wait.
I would love to help but I can't access the server yet
(I know this is on your very long ToDo list that every
body adds to so I haven't bothered you much about it).
Don't get me wrong - I deeply appreciate the work you
do and firmly believe that without you Wikipedia and
the other Wikimedia projects would be in sad shape due
to nearly constant server and software problems.
> That doesn't mean they're _irrelevant_, and I'd
> appreciate an apology for the "chauvinism" comment.
It wasn't directed at you it was directed at the
Wikipedia-specific direction of the code and Jimbo's
apparent approval of that (the code part is not your
fault - or anyone's fault really - it is just the way
things developed).
All I want is for us to move in the direction of
making MediaWiki more usable for all Wikimedia
projects (something we /have/ been doing) - Jimbo's
first comment wasn't in that direction so I challenged
it since what Jimbo says is often seen as policy. But
if you were offended then I apologize. I most
certainly did not want to cause you any grief.
-- Daniel Mayer (aka mav)
__________________________________
Do you Yahoo!?
Protect your identity with Yahoo! Mail AddressGuard
http://antispam.yahoo.com/whatsnewfree
Jimbo wrote:
>I'm primarily supportive of (B). I have no problems with
>MediaWiki being accessible for other users, because if it is,
>then we'll attract more developer talent, which is a good
>thing. But I think our primary focus ought to be support
>of the encyclopedia project.
So I guess the needs of Wiktionary, Wikiquote, Wikibooks, and Wikisource (not
to mention Wikitravel) are secondary then? MediaWiki should serve /all/ of
Wikimedia and also our friends over at Wikitravel. However by doing so it
will become a general use wiki that can be used for many non-Wikimedia
projects.
But please lets stop supporting Wikipedia-centracism/chauvinism and work
toward making MediaWiki maximally useful to all Wikimedia projects.
-- Daniel Mayer (aka mav)