Crosspost to: foundation-l, wikitech-l
At the end of 2002 on my request the mailing list Announce-l was created.
This because I noticed there was a serious problem whit internal
communication inside Wikipedia. Changes where done to the software,
important decisions where made whitout anybody who is not a mailing list
junky to know about it. Especially the problem to get information form the
English language community to the others.
The idea was to use the mailing list Announce-l for the announcement of
important things that have a impact for all wikipedias in all languages.
This for technical and organisational matters. So at least it could be
possible to inform the non-English wikipdias about what was going on and
maybe even give some feedback.
It has received whit an 100% positive responds but only one (1) reaction.
Again; thank you Daniel Mayer. So it never happened.
We are now 3 years later.
In general, things are a bit better. Aldo I think my original idea would
still be useful.
Now I would like to ask of the following "light edition" of Announce-l can
get support and implemented.
- use Announce-l for the reporting of problems whit the operation of the
services of the Wikimedia projects. Status reports about it and when it
works again.
Not for discussing or for bug rapports, only to inform the community. A
moderated list, low traffic.
This should be done by the technical staff of wikimedia.
Examples;
== search function ==
Some time ago a received by the email system for questions to the Dutch
Wikipedia some questions why the
can not find the new articles the have written.
I know that the search function is not live but it works whit a special
search database. So I asked them to wait until an update. But because it
takes so long for the update I have asked on the lists about it. It seems
that the update system is broken so new articles can not be found.
So now I can inform other users when there are questions about it and can
put up an notice on the search page about this.
When will I work again? I can only test to see.
For something like this Announce-l can be used. When it is known that the
update function for the search system does not work anymore a member of
the technical staf can send an message to Announce-l. And when it works
again or there is news about it again.
== OTRS ==
On a normal day the Dutch Wikipedia receives about 8 emails form visitors.
How the emails are done is an important part of how we can make a good
impression. All emails must be answered fast and professional.
Because I noticed the lack of new emails I start investigation. On
bugzilla no clue, only a ticket a have created so time ago when OTRS was
also not working. On the IRC #wikimedia-servers (or something like that) I
found a notice that OTRS is down. The last message I have received from
OTRS is from Saturday 13 august around 12 hours GMT. I have discovered it
Tuesday around 22:00 hours.
When I know OTRS does not work or will not work I can redirect the
incoming emails so the do not go to OTRS and we can answer them outside
OTRS. But I need to know it does not work to do so.
Also for this Announce-l would be very useful. Send an email "OTRS is
down, emails are in a que" or so and again when it works again.
Or when there will be an upgrade of MediaWiki. To say between X and x we
will upgrade the wiki to version X of MediaWiki. During the upgrade the
database will be locked for about 10 minutes.
Sysops will have after the upgrade have the option to rename account. See
http://some-place_whit_more_info.org
Now new functions are discovered after an upgrade.
If, and only if, Announce-l is always used for this type of things this
list can become very useful for those wikipedians who like to inform there
community and visitors about what i going on whit there wiki.
--
Contact: walter AT wikipedia.be
Ook een artikeltje schrijven? WikipediaNL, de vrije GNU/FDL encyclopedie
http://www.wikipedia.be
There was a bug in our db.php configuration file which caused all the
load selectors in the database server list to be set to NULL. This
confused the selection code, and everything loaded up samuel even when
there was slight lag on it.
I've fixed this, and it seems to distributing load a bit more equitably
and not pausing on short lags so much.
While tracking this down, I added in the support for secondary debug log
files that we've talked about from time to time. You can add additional
log files for debug groups in $wgDebugLogGroups, calls to wfDebugLog()
take a group name and will log to that file if it's set, or pass through
to wfDebug() otherwise.
This should make it easier to temporarily log particular issues on the
production boxen to help in tracking down problems.
-- brion vibber (brion @ pobox.com)
On 17/08/05, Juliano Ravasi Ferraz <ml(a)juliano.info> wrote:
> Rowan Collins wrote:
> > No, according to that link "When Google sees the attribute
> > rel="nofollow" on hyperlinks, those links won't get any credit when we
> > rank websites in our search results."
>
> According to the link, right before the sentence you quoted:
>
> "...but you can also instruct Googlebot not to crawl individual links by
> adding rel="nofollow" to a hyperlink."
This aspect of the "standard" is really ambiguous in all the
announcements I've seen [I just read the original Google, Yahoo! and
MSN blog entries], and many many people are stating with conviction
that it *doesn't* stop the page being spidered. After all, the stated
purpose is to mark an link to another site as "untrusted", not to mark
an in-site link as not suitable for crawling.
But this may be a usable side-effect for URLs which won't be linked to
anywhere else on the Web (one thing that's very clear from what I've
read is that no *negative* rank or *delisting* is interpretted from
the link). I guess the only way to find out is to add it to a few
links of this sort and see if they do in fact get spidered...
--
Rowan Collins BSc
[IMSoP]
On 8/17/05, Juliano Ravasi Ferraz <ml(a)juliano.info> wrote:
> According to the link, right before the sentence you quoted:
>
> "...but you can also instruct Googlebot not to crawl individual links by
> adding rel="nofollow" to a hyperlink."
Still, there are other search engines that don't respect the
attribute, and so the point is moot, no?
Hello.
I'm going to translate MediaWiki to Basque (the language of the Basque Country,
also known as Euskara).
I want to know if there is any similar project started, in order not
to make it twice.
Thank you very much,
Aitor
On 16/08/05, Juliano Ravasi Ferraz <ml(a)juliano.info> wrote:
> > [I wrote:]
> > IIRC rel="nofollow" is badly named - it doesn't stop anything from
> > following the link, only from using it to calculate page rankings.
> > Unless I'm wrong about that, it's irrelevant for the current
> > discussion.
>
> According to http://www.google.com/webmasters/bot.html#www , a
> rel="nofollow" causes google not to crawl through that link.
No, according to that link "When Google sees the attribute
rel="nofollow" on hyperlinks, those links won't get any credit when we
rank websites in our search results." That particular page doesn't
make clear whether or not the page will in fact be added to the list
of pages to crawl, but other discussions of the topic have assumed
that it will. So, despite the name, the link *is* followed, it's just
not used for ranking.
--
Rowan Collins BSc
[IMSoP]
---------- Forwarded message ----------
From: Xin Zhang <cyferzhang(a)gmail.com>
Date: Aug 16, 2005 11:19 PM
Subject: problem with wikipedia.org log in
To: zh.wikipedia(a)gmail.com
Hello, Dear Admin,
I forget my password at wikipedia.org and the system didn't send a new
password to me. So could you help me regain my user name?
I have a user name of "Cyfer" at wikipedia.org, and I contributed some
items in the polymer and related areas. In most case I post from
129.2.52.139. Hope this could help.
Xin