For the first month or so that I was involved with wikipedia, I found it great fun and very rewarding. Then the response problems started, and the enjoyment I get from it has gradually waned until I am now wondering why I even bother to try and log on. Today, for example, I have attempted to access wikipedia at regular intervals and have never managed to get further than "Recent Changes". Yesterday was the same.
I realise that the software is maintained on a voluntary basis. However, if we can't do any better than this, I think we may as well throw in the towel. What is the point of an encyclopedia that no one is able to read?
Deb
I'm seeing the same problems -- Wikipedia is locked up most of the time,
only responding occasionally at random, making it effectively unusable.
Netcraft reports that is is running Apache/1.3.23 (Unix) PHP/4.0.6
mod_fastcgi/2.2.12 on Linux.
There is now a severe Apache security vulnerability that affects
Apache 1.3 all versions including 1.3.24;
Apache 2.0 all versions up to 2.0.36;
Apache 1.2 all versions.
This looks like something that should be fixed ASAP. And it just might explain some of the problems seen.
-- Neil
On Sunday 30 June 2002 07:33 am, Bryan wrote:
> [Paris] a rather extreme example, but yes, I wouldn't have a problem with
> [[[Paris]] being a disambiguation page] :) At least it makes Paris,
> France's article title more standardized
> with the other city titles when the reader gets to it.
OK, go ahead and fix the hundreds of links to [[Paris]] so that they point to
[[Paris, France]] (which is going to be the new home for this article under
the new city naming convention) and the 10 or so that need to be redirected
to one of the several other meanings of [[Paris]] and then maintain that
disambiguation page by doing the same thing every so often as new
contributors naturally link to THE Paris by simply linking to [[Paris]]. The
whole concept of the disambiguation notice has failed -- people just don't
"go back and fix the link" as requested. Therefore, maintenance shifts to the
person creating these non-articles.
I would prefer however, that since simply saying Paris is almost any context
almost always is unambiguously meant to mean [[Paris, France]], that
[[Paris]] should be a redirect to [[Paris, France]] (giving the French city
redirect priority over the use of the page [[Paris]]).
> I think it's a bad idea to assume some kind of universal context when
> writing articles, because everyone has a slightly different idea of what
> the universal context is.
Well, I guess we should abandon all naming conventions then (which is where
the universal context arises here). If there is no universal context, then
there is no need to naturally disambiguate terms and wikipedia degenerates
into a Jargon file with loads of naming conflicts, misdirected links and an
inordinate amount of the use of tedious piping.
--maveric149
On Sunday 30 June 2002 07:33 am, you wrote:
> The question of what's the "most common usage" is a matter of opinion,
> however. Wikipedia is on the Internet, so naturally there's going to be a
> lot of computer-oriented people accessing it.
OK, then why were there over 50 links to [[virus]] intending to access an
article on the viruses that attack living cells and only 7 that intended to
access computer viruses? The "computer" here naturally disambiguates the term
about viruses that attack computer systems but there is no natural way to
disambiguate the other type of virus. Now each time somebody wants to link
directly to that article about viruses that attack living cells, they have to
write [[virus (biology)|virus]] which is needlessly tedious and can only tend
to_discourage_contributions to those articles.
We need to make linking to articles as easy as possible respecting
<reasonable> ambiguities (50+/7 doesn't cut it).
> I know that when I'm in a
> conversation specifically about computer security, I never say "computer
> worms" or "computer viruses"; I just say "worms" and "viruses," and the
> people I'm talking to know what I'm talking about. Likewise, when I'm
> talking about genetics and I mention "viruses", there's usually no
> ambiguity there either.
Exactly my point. You are talking about these thing in their native context
-- not in the context of a hyperlinked encyclopedia. New contributors get
this concept rather quickly when presented with examples. I don't think it
would be wise to present too many needless examples of parenthetical
disambiguation. As a matter of fact, I am beginning to see the use of
parentheticals when they are not at all needed. For example, one contributor
is making articles on punctuation marks. However, this person at first used
this format: [[exclamation point (punctuation)]] -- as if there was something
else named exclamation points that needed to be differentiated from. New
contributors naturally begin to mimic what they see after their first few
articles.
> I recall briefly debating about whether the planet/Greek god pages should
> be disambiguated or whether the page should belong to the god with a "see
> also:" for the planet, and this was another case where I felt that it was
> far too opinion-dependant deciding whether the Greek god or the planet
> should have "precedence."
I don't have any major issue with the planet/god disambiguation thing -- this
is a case where both things are only known as the same one word term AND
priority really can't be assigned based on amount of usage because the god
usage is much less, but still very significant compared to the planet usage.
Cases like [[Paris]] are less ambiguous with only a small number of
references to the mythical figure and an even smaller number of references to
the other cities by the same name. So priority CAN be assigned.
My whole spin on the parenthetical disambiguation thing, is that it should
only be used in extreme cases where no alternate terms are used. I just HATE
breaking links by directing users to non-articles and I also HATE needless
tedium in creating links (and creating direct links isn't fixed piecemeal as
visitors "drop in by accident" like it was intended -- these links are only
made direct by a systematic campaign by a user or two who have the time and
inclination).
Spontaneous linking is one of the reasons why wikipedia has been a success --
we should do everything reasonable to preserve this by limiting the use of
parenthetical disambiguation and by also limiting the creation of non-article
list disambiguation pages.
--maveric149
What I have done in order to get in, since I have to be online for hours
anyway, is to open up two browsers. In my case Netscape and Explorer. I put
Explorer (which I don't use at work) on Wiki; then go ahead and work using
Netscape.
When I get in the other browser is useful for looking stuff up and external
sites. Easy to check when there is a lull in things to see if the Wiki page
has come up. Only problem is sometimes have to abandon some edits in the
middle.
Good to see this finally being talked about.
Fred Bauder
As an experiment which wouldn't hurt much (certainly not as much as
the current slowdowns), I second the plan to disable the user
contributions feature for now.
Jason or Jimbo, the quickest and dirtiest way to do that would be to
remove/rename the special_contributions.php file.
As mentioned earlier, it would also be good to remove/rename
special_asksql.php because the developer passwords have been
compromised. It is even possible that some of the direct sql queries
are slowing down the site.
Axel
Since the passwords are out, it might be advisable to disable all
direct SQL access to the database for now; anything else a sysop or
developer can do is logged and so doesn't pose a big problem.
Simply removing special_asksql.php should do the job.
Axel
On Saturday 29 June 2002 12:01 pm, Bryan Derksen wrote:
> Candy worms, sure, but I can easily imagine writing an article about
> something computery and referring simply to "worms" within it because the
> "computer" context is already clear to the reader of the article. Why make
> the editor check every link to be sure that it doesn't lead to some
> completely irrelevant article that just happens to have a name in common
> with what you really wanted?
Why would [[computer worm]] be anything else? Why wouldn't [[worm]] be about
the most common usage? This is why we have naming conventions that aim to
naturally disambiguate terms from each other using the least complex naming
scheme.
The context here is a hypertext, cross-linked encyclopedia -- not some
limited jargon file on computers. Using the word "worm" by itself is jargon
when you mean "computer worm" (which <is> the term used in news reporting and
anywhere else where context must be established by the term itself -- such as
in a hyperlinked encyclopedia).
If we can't assume at least some intelligence from our contributors to
naturally disambiguate terms then we might as well turn [[Paris]] into a
non-article disambiguation page to catch those pitiful souls who expect to
link to an article on Paris, Texas by typing [[Paris]] -- just because the
article they were writing in was in the context of Texas history.
There are higher level contexts we are dealing with here.
--maveric149
On Saturday 29 June 2002 12:01 pm, you wrote:
> During a routine check for uploaded junk, I saw two gzipped SQL files. I
> downloaded one (the other was huge) and found that it contained my
> password, as well as other user names. This means that any of our junk
> uploaders (and junk was uploaded) could have found this file, read it, and
> log in as anyone, including a sysop (the file also tells who is a sysop).
> Please change your passwords immediately to prevent this.
>
> phma
I would like to, but I can't even access the main page right now. :(
I'm sure impersonators will have similar difficulty -- just nobody get mad if
user:maveric149 goes on a rampage against disambiguation pages.
It isn't me, honest. ;)
Just kidding.
--maveric149
On Saturday 29 June 2002 12:01 pm, you wrote:
> Changing all links to always bypass the disambiguation page is nice and
> convenient, but it's not absolutely necessary. If it were, then why have
> the disambiguation page at all?
The reason is to catch any future honest mistakes for true disambiguation
issues. Such as this conundrum: "Where might the article on [[Mars]] be in
the in context of a hyperlinked encyclopedia? Now "planet" is not a part of
its name in any context, so it would be silly to find it at [[planet Mars]].
Hum, how about I just type in [[Mars]], hit preview and see where it takes
me."
The bingo! They find out that an article about the planet lives at [[Mars
(planet)]].
This is different when alternate two word terms are actually a part of a
title that <naturally> disambiguates a term. Such as <computer worm> -- here
the addition of the word "computer" is totally valid and widely used. We can
reasonably expect a person to think: "Hum, I'm witting here in an on-line,
cross-linked encyclopedia. Where might an article about computer worms be....
How about [[computer worm]]. Yep there it is." Besides, there is a link to
[[computer worm]] at [[worm]].
If you like, we can move this to the top in a short row stating: Other uses:
[[computer worm]]. Then the obvious disambiguation you want is there and an
actual article is at [[worm]] that is about creepy, crawly, slimy worms and
we don't have to place that article at [[worm (biology)]] which can only be
linked to through pipes.
In fact, it might be a good idea to have a very short row list like this in
other places in order to prevent the subvertion of a the common use of a
single word term for non-article disambiguation pages just because other
terms also share the same single word within narrow contexts (but are also
widely known by two or more word equivalent). I would suggest having this at
a -1 font and be limited to a row with maybe a line separating it from the
actual article. This shouldn't be any more obtrusive than having language
links for articles AND also acts as a disambiguation page without turning it
into a non-article list.
I could be very happy with this.
---maveric149