Wikipedia:Village pump/June 2003 archive 3

User:Wrongbros, made a contribution to List of record labels and it was reverted. Is this Michael? If not, why was it reverted? MB 19:42 10 Jun 2003 (UTC)

Probably because it's doubtful whether we can get independent confirmation that the record label in question exists. See Talk:Messed Up Wreckords. --Camembert

Where did all the articles on U.S. municipalities come from? Did a bunch of people just copy and paste a Census database into the 'pedia? -Smack

See User:Rambot for the lowdown. --rbrwr

Aha! It wasn't a someone, but a something. -Smack

What is NASA's copyright policy? Text on Bell X-1 is copied word for word from [1]. Is that okay with NASA? Kingturtle 11:17 11 Jun 2003 (UTC)

NASA, like all US government agencies, can't claim copyright so the text would probably be in the public domain. --Robert Merkel 14:03 11 Jun 2003 (UTC)

Yep. Note that text from the US govt is often POV, and it's a good idea to crosscheck the material with other sources. Stan 15:37 11 Jun 2003 (UTC)

Im using a computer which doesn't allow a large screen size. Can I either get a no-frames option or have the menu placed on the right of the screen? (or the top) Pizza Puzzle

Wikipedia does not use frames. If you're referring to the quickbar on the left, it can be placed elsewhere (or removed) by setting your preferences. -- Wapcaplet 13:57 11 Jun 2003 (UTC)

The periodic lag is annoying. What are the costs of running the wiki and how much would it cost to improve performance? Pizza Puzzle

There are two dedicated servers, a DB server and a page server, donated by Bomis, together with colo and bandwidth.

The DB server is running at a fairly smooth average 25% load on both CPUs, the page server is bouncing between < 10% load and almost saturating its single CPU. So it looks like the problems is either the page-server CPU, or disk IO fragmentation/contention on the DB (unlikely, it's got plenty of RAM).

Here are some possible routes to making things go faster:

tune the page-server to run more smoothly, using config options
optimise the page-server code to make hot-spots run faster
add an extra CPU and extra RAM to the existing page server
add an extra page server, and do round-robin DNS with a low TTL to balance the load (we can also do automatic cutover if one fails)

The wikitech-l mailing list is a good place to discuss these things.

The Anome 14:04 11 Jun 2003 (UTC)

If 100 wiki users donated $10 im sure we could buy an extra CPU and RAM. Pizza Puzzle

Indeed. There are attempts going on to create a non-profit foundation that can raise funds, and do other things related to Wikipedia and other similar projects. See the wikipedia-l mailing list for more on this. The Anome 14:31 11 Jun 2003 (UTC)

But is there any need to wait for a foundation to be formalized? Why not just create a PayPal account? 209.56.25.161

Heck yeah! I'd donate. -- Wapcaplet 15:07 11 Jun 2003 (UTC)

User:Brion VIBBER, currently our most active developer/administrator, accepts donations; leave him a message on his talk page for details. --Eloquence 15:16 11 Jun 2003 (UTC)

Hey folks, two questions:

I misspelled existence as "existance" when searching for its article, and I discovered that there are many pages with that misspelling in them. I then wondered about what should happen if somebody makes a link to existance. When I first did the search, there was no page for it. I made a redirect page to existence. Is this the right way to handle misspellings? Or is it better for there to be no page at all, so that the person linking to it is more likely to realize that it is spelled incorrectly?
I've noticed that when I sign my articles with the four ~'s, Wikipedia somehow knows to sign my articles as Nelson instead of User:Skyfaller. That's really amazingly smart! How did it figure out to do that?

--Nelson 15:00 11 Jun 2003 (UTC)

1. I don't know if that's the right way to handle misspellings... of course, the best way is to not have misspellings in the first place! Obviously, if you find misspellings, please correct them. But it's a good idea to have redirects from common misspellings. 2. That is probably because in your user preferences, you put "Nelson" as your nickname. There are some other cool things you can do with your prefs too - check 'em out! -- Wapcaplet 15:07 11 Jun 2003 (UTC)

I just realized that it might be possible that "existance" might be a British spelling of existence. Does anybody know if that is the case? Or is "existance" recognized as a misspelling on both sides of the Atlantic? --Nelson 16:18 11 Jun 2003 (UTC)

No, "existance" is wrong everywhere. -- Tim Starling 02:12 12 Jun 2003 (UTC)

Oh dear, I seem to have created some weird server bug while playing around with the redirect page from "existance" to existence. I put an extra "#" before the #REDIRECT, in order to see whether a search would turn up any more instances of "existance", I thought the redirect page might be messing up my search. Now, if I search for "existance", I get that demented non-working redirect page, and no matter how many times I edit it back to a normal redirect page and save it, when I search for "existance" I get the demented page again. Interestingly enough, when I actually type in the URL for the existance page, it redirects properly, and when I follow the "redirected from" link at the top of existence, it shows a normal redirect page. Help? --Nelson 19:12 11 Jun 2003 (UTC)

Are you using "search" or "go"? --Brion 19:16 11 Jun 2003 (UTC)

Ah, apparently I was hitting "Go", because I was simply hitting Enter instead of clicking the mouse, and it seems that defaults to "Go". Go is still bringing me to the demented page. --Nelson 21:17 11 Jun 2003 (UTC)

The old version is probably still in your browser cache, and because the "Go" URLs are different from standard retrieval URLs, there's a difference in caching behavior. Try clearing your disk/memory cache (completely or just for that page by hitting "Shift+Reload"). --Eloquence 04:39 12 Jun 2003 (UTC)

Pages retrieved via 'go' should not be cacheable, as the result could change to another page either newer or older than the page it first went to. However they may currently be, which would produce the above results. (IE users, note that I believe it's ctrl you need to make reload really reload, while shift for Mozilla/Netscape.) I'm also not sure whether redirects are handled properly. If someone would like to review that code, please do, my eyes water when I get near it now. :) (Look in Article.php for whatever calls checkLastMod() or something to that effect.) --Brion 07:17 12 Jun 2003 (UTC)

Ah, thanks guys, I am using Mozilla, and shift+reload fixed the problem. --Nelson 20:31 12 Jun 2003 (UTC)

Could a developer please update Special:Shortpages? The last update was made on May 13. --mav

Hey folks, sorry about dominating the Village Pump, but I don't know where to put this. There's a problem with numbering that I've noticed in many pages, most recently Paul_Simon_(album), where because the author has put spaces between paragraphs, the numbered list starts over again from one. There is no good way to fix this, because removing the space messes up the formatting, and if you were to manually input the numbers, this would make maintenance more annoying. This is one of those cases where I think HTML-like tags would serve us better, because within the <OL></OL> tags, you can put <br>'s without making the ordered list start over again. However, short of forcing Wikipedians to learn HTML, what can be done for pages like this? Change the code so that you must have two spaces in order to have #-numbered lists start over? That seems like the best solution to me, but I don't know how much work that would take, and how many current articles that would screw up. --Nelson 04:13 12 Jun 2003 (UTC)

It looks like someone already fixed Paul Simon (album), so you can take a look at it to see how it was done. Numbering proceeds normally, even though some extra info is inserted in the middle. Yes, if you leave a blank line in editing, numbering seems to get restarted (since it opens a new <ol> element), but most things like this can be done using Wikipedia's regular formatting tags (#, :, *, etc.) Wikipedia:How to edit a page talks about this a little. (Actually, if you ask me, information this extensive shouldn't be inserted in the middle of lists in the first place - there's probably a better way to organize it to avoid the problem entirely.)

Don't worry too much about anything appearance-related. Semantic accuracy is probably more important than tweaking the spacing between bullet points. Most of that stuff can be tweaked with the global style sheets, I would assume. -- Wapcaplet 11:26 12 Jun 2003 (UTC)

This an S.O.S. to anybody who can help me, and especially to Paul A, who gave me the welcome. I have been smart enough to learn how to use Wikipedia's search engine, edit and create pages, enable links, create this account yesterday night, and successfully login "my" page and this one; but I WAS NOT clever enough to know how to login the rest. Despite my name, my "power" is limited, I'm just a warlock, not a magician... If you can lend me a hand, please write me a message; if I cannot login successfully having been registered I interfere with other users and with Wikipedia too. I'd like to stay, so I'll be grateful for any help.-- The Warlock June 12, 2003, 5:28 (GMT)

Yesterday's discussion of User:The Warlock's problem can be found at User talk:200.69.36.106, if anyone's interested in helping out but wants more information about what the problem is. — Paul A 06:59 12 Jun 2003 (UTC)

Do you have cookies blocked or disabled? Because if you do, the software will say "you are now logged in", but it will instantly forget who you are as soon as you go to another page. See HTTP cookie. -- Tim Starling 07:26 12 Jun 2003 (UTC)

Click here for a cookie test and tell us what it tells you. -- Tim Starling 11:03 12 Jun 2003 (UTC)

When I look at "My Contributions" some are followed by the reference "(top)" What does this mean? Thanks! Mkrose

You've done the latest job: top, as in a roof. --Menchi 06:35 12 Jun 2003 (UTC)

Or, to put it another way: if you look at the edit history of the page in question, your contribution will be at the top of the list. — Paul A 07:07 12 Jun 2003 (UTC)

In some articles with chinese wikipedia link, this link work in preview mode, but don't in read mode. ex: Provinces of China, the link is the good one in edit mode too ... 62.212.110.113 11:54 12 Jun 2003 (UTC)

I used to be able to view [diff] in my watchlist, now I can't because I selected "enhanced recent changes" Why cant I have both? Also when I have the menu at the right-the edit box overlaps with it. Pizza Puzzle

Some thoughts (I hope my questions are not academic):

will Wikipedia's collective editing paradigm be suitable for the ever-growing number of articles ? (we hear Wikipedia will hit 1,000,000 mid 2007)
is there a desirable ratio of articles to editors (mind not contributors) and where are we now ?
what is the probability that some newly created article is left unreviewed for months ?
will it be plausible/desirable to have lists of voluntary thematic editors (like Wikipedia:Wikipedians by fields of interest) who backed up by some software (like, for instance, "ancient changes") would take on editing more systematically articles we already have ?

Kpjas 15:19 12 Jun 2003 (UTC)

Those are some pretty deep questions! I don't think anyone has definitive answers to any of them. Since (to my knowledge) nobody has ever used the open-content paradigm on a project of this size (and certainly not one approaching a million articles), it remains to be seen whether it will continue to work as well as it has. I would speculate that as long as the percentage of well-meaning contributors/editors remains high, there is theoretically no limit to the number of articles we may someday have.

There is always a chance that a new article goes unreviewed for quite a while. If nobody notices it as it passes down the Special:Recentchanges list, it could very well sit around for months and not be seen. Though, again, if the percentage of good editors remains high, when someone finally does see it, it's likely to be cleaned up. But you may be right, we could use some kind of systematic way for dealing with this. (I try to skim through Special:Newpages whenever I visit Recentchanges, and catch a lot of nonsense articles that may otherwise go unnoticed, especially during peak editing traffic hours).

I'm sure some of this has been discussed on the various Wikipedia:Mailing lists. Our system has worked pretty well for more than two years, and I'm sure it will continue to do so, as long as we're able to improve and adapt as the situation calls for it. -- Wapcaplet 15:50 12 Jun 2003 (UTC)

"Ancient pages" has only been added a few days ago and we'll have to see how it holds up for the purpose of reviewing past articles. An article will only disappear from that list if it has been edited -- but what if it needs no editing? The list may eventually show many very old, very good articles at which point it might stop being useful. Or we might work through past articles until we get to the User:Rambot imported US census data, and people would stop working through the list at that point because they have nothing to add to articles about 2000 people towns. --Eloquence 16:18 12 Jun 2003 (UTC)

Question: Who actually "runs" Wikipedia? Who provides the server space. Just curious... it seems a little odd to write/edit articles for an organization you know very little about.

-Alex S

Please see Wikipedia:Overview FAQ. --Brion

Hm. Rather than answering the above question, I'd like to add a new one. I just did a Google search on a lesser known contemporary British playwright called Winsome Pinnock. What I got among the first ten hits was http://www.wikipedia.org/w/wiki.phtml?title=Winsome_Pinnock&action=edit, which directly links to the edit page of a non-existing article. Could that sort of thing be avoided? Apparent vandalism in the form of "What the heck ..." etc. is probably due to people completely unfamiliar with wikis being suddenly faced with an edit page. --KF 18:34 12 Jun 2003 (UTC) Ähnliche Seiten

As near as we can tell the only way to keep those out of google is to not exclude edit pages and special pages with the robots.txt file. This would tell google to load up hundreds of thousands of dynamically generated pages and then ignore them. If we tell it not to load them, then it can't see the note in the page telling it not to index the pages. Pretty screwed up system... --Brion 18:44 12 Jun 2003 (UTC)

I had the impression that you could set the robot meta tag in the html to not index? From what you say you have to have a robots.txt saying that those pages should not be ignored by the robot, so that the "no index" robot meta tag can be read? Even if this is the case, it should be implemented. It will look bad when someone finds a blank page b/c it is listed on google. MB 19:36 12 Jun 2003 (UTC)

I'm reluctant to let google load those pages since it would tend to increase server stress. Hypothetically we could recognize googlebot and let it through just enough to serve fake no-index-me pages, or drop the links entirely from google-served pages, but ugh. I hate dirty tricks like that, and it's reeeeal low on my list of priorities. If someone else wants to code up the special cases... --Brion 04:59 13 Jun 2003 (UTC)

Regarding musician entries: if there is a discography listed, should we make separate pages for each album comprised of a track listing? Mick 19:56 12 Jun 2003 (UTC)

No, there are millions of album, 99% are insignificant, except to the fans, whose taste cannot represent Earthlings.

Similarly, not every person mentioned in this encyclopedia deserves an article. For some, a one- or two-sentence description will do, under appropriate section of relevent historical incident.

Yes, you'll find some counterexamples, but they're made by zealous, or possibly overzealous fans.

Note that there are historically and culturally consequential albums which should have an article.

--Menchi 22:16 12 Jun 2003 (UTC)

Mick - I'd say go for it, provided you have something significant to write about each album. Obviously, for famous and influential albums, such as Sgt. Pepper's Lonely Hearts Club Band, an article might just about write itself. Every full-length album by any musician probably has a good story behind it, so I see no reason why you shouldn't write about albums you know and love - provided it's good, encyclopedic information, and not just a review of the album, or a list of personnel who played on the album. (fans' taste cannot represent Earthlings? I won't pretend I understand what that means...) -- Wapcaplet 23:06 12 Jun 2003 (UTC)

It is the summary of what I said, nothing new. But in case you're wondering, it means "what a small group of fans of a particular band like to hear and consider very important in their lives in no way represents what most people on Earth consider to be an essential cultural, social, or historical assets".

If we write every religious affiliation with around or less than 5,000 followers, Wikipedia becomes Wiki-religio-pedia. Likewise, discography or minor CDs leads to Wiki-Amazon.com-pedia.

--Menchi 23:29 12 Jun 2003 (UTC)

Well, yeah, but then again, probably 99% of the entire contents of Wikipedia is of little interest to the Earth's population as a whole. Consider, for example, the stub geographical/census articles imported by Rambot. Discussion of what is or isn't appropriate on Wikipedia is neither here nor there; I just think that if someone wants to write about "minor" CDs, they should feel welcome to do so. Every album is a significant event in a recording artist's life, as well as in that artist's fans' lives, and may have an interesting story behind it. Why not have articles on them? -- Wapcaplet 23:38 12 Jun 2003 (UTC)

Geographical places will always be here, even after the inhabitants are replaced by some other group and names changed. Some CDs may not be so permanent.

But I agree that if there's something interesting and unusual about a CD that distinguish from the regular-Joe CDs, I for one would be interested in learning about it.

--Menchi 23:49 12 Jun 2003 (UTC)

Mailing lists operations

I've been reading and wanting to participate a bit (probably not very actively) in the WP Mailing lists for several months now. But I never actually did, because I never tried a newsgroup or a mailing list before. I finally forced myself into trying it yesterday, but things didn't go smooth, and I couldn't find help or FAQ on the Wikipedia:Mailing lists.

NNTP: I can read all message of all lists on NNTP, but when I posted 2 messages there in two different lists, none showed up. Only a little pink square appeared beside the replied message's icon, indicating that a reply was sent, however...
I also tried the newsgroup ab.test, and everything works, including replies. So my mail client works.
E-mail: I know how to post new message using Hotmail, but how to reply to another person's old message? I can't even see old messages posted this morning just a few hours before my subscription succeeded.
How do I send test messages? I really don't know how this mailing list thing work.

--Menchi 22:28 12 Jun 2003 (UTC)

1. I've found posting messages via GMANE (i.e. NNTP) to be a bit dicey. The one thing you really have to know is that GMANE will send an "autoauthorizer" message to the "from" address, which you have to reply to before the post will be forwarded to the list. You also have to use a "from" address which is subscribed to the mailing list, otherwise it will be delayed for a spam-check.

2. To reply to a hotmail message, you click on the message and then the "reply" button. If there's too many messages and the one you want to reply to is off the page, there are "previous" and "next" links to click on, at the bottom of the list. I would strongly recommend setting up folders for each of the mailing lists, and creating filters to send the posts to the correct folder. Note that wikien-l + wikitech-l will fill up your 2 MB limit in less than a week, so you have to regularly clean up.

-- Tim Starling 01:35 13 Jun 2003 (UTC)

2. So I can't reply to old messages that were posted before my registration?

Thanks for the tips and the warning. --Menchi 02:54 13 Jun 2003 (UTC)

No, you can't. -- Tim Starling 03:05 13 Jun 2003 (UTC)

Well, you can of course write a response to an old post, but you'll have to paste in the title and quote text yourself and the message won't be threaded properly in some mail readers. But that's a minor annoyance; if it's the only way you can reply to something you want to reply to, by all means do so. --Brion 04:59 13 Jun 2003 (UTC)

Hey, I'm a newbie here. I just wanted to mention that there will probably a flood of newbies, as this site was just featured on the show The Screen Savers on Tech TV in the USA. I was really curious, and decided to check it out, and that's how I found out. :) ManicGypsy

Three pages have been temporarily protected as a result of the influx: Martha Stewart, Monkeypox, and Gregory Peck. -- Notheruser 02:09 13 Jun 2003 (UTC)

Well, the Gregory Peck page was just too easy...

The pages are no longer protected. -- Notheruser 03:31 13 Jun 2003 (UTC)

And as a result, they are once again under attack. -- Zoe

I have re-protected Martha Stewart for now. Evercat 17:46 13 Jun 2003 (UTC)

Seems to be unprotected again. Luckily everyone seems to be keeping a close eye on it to keep it from being vandalized. ManicGypsy 03:24 14 Jun 2003 (UTC)

Thanks to all of you who helped me in solving my problem, it's fixed now. The Warlock.-- June 13, 2003, 6:37 (GMT)

A friend of mine told me today that the Wikipedia "doesn't work" because he and a friend of his created a fake religion called Fieldism and it was still there. Needless to say, I deleted it, but this concerns me a bit. I think we should create a way for users to verify articles. To verify articles, you must have been around for a certain amount of time, or made at least x contributions, or something like that. And you shouldn't be able to verify articles you created. what do you guys think? Could someone post this to the mailing list? MB

Recently I have been thinking of how to ensure the quality of articles in wikipedia. I have no doubt about wikipedia will reach 1 million articles soon or later (aside from technical problems we might have), but the quality is questionable. It is true that as has the number grown, the quality of the majority of articles have improved too. Surely there are a lot of good articles (see Brilliant prose). Alas, there are significant number of articles that are written poorly, contain factual mistakes or even hoax mentioned above. We should not excuse them by saying we have far more good articles. People that are unfamiliar with wikipedia will reach such article, they might be disappointed or discard wikipedia as a pot of craps. However, we also should not get rid of articles that look fake, are written poorly now but can be expanded. That is the last thing we want. Fieldism might not show up in google but can turn out to be true. There are a lot of topics that never show up in google but trus thus should be covered by wikipedia thoroughtly. We need a better mechanism google-check.

Anyway, so what is the damn solution? Actually I am not sure about it yet. Nupedia has a good mechanism to ensure the quality of articles, however, it has been bogged out given its bureaucratic system. We don't want to bureaucratic approval system to wikipedia, which is why I have not had such yet despite debates and needs of some kind of peer-review. RecentChanges and My watchlist are our effective peer-review system, which is simple to handle but have worked most of the time thus far.

But we probably need more than them to improve the quality of articles further more. You can come up with other ideas, but what I am thinking now is to put a primary editor of an article to each article or somewhere else. This is not against the wiki-way which we believe in. Wikipedia is a collaborative documenting where anyone can edit any page. This is why any article has no author. We don't want that articles become territorirized by someone who cares about them, saying like this is my article, don't touch it, if you want to do something, go to your articles. This is written well already and your contribution only makes the article poor because your point view conflicts with that is now in the article.

My idea is not this but make explicit credits about who is responsible for articles. If you see a sentence like Dogs are considered a God during Edo Period in Japan, most of you probably wonder is it true and might go to remove that, doubting about accuracy. But if you see, the article with a credit of one who is specialized in Japanese history and culture, you can be sure. There is another problem with this that any article can be expanded later, not unlikely after the article is reviewed by professionals. This should be covered by a way like if some points are added after review, that portions should be noted as not reviewed yet in some ways, preferably by one who actually wrote them.

Reviews should be done by those who can claim their credibility in the similar way done in academic peer-review journals. They are like Ph.D or native people. If we don't want to clutter actual articles, we can use the space at the top of a talkpage. This is actually happening already in some ways. Many wikipedians already show their profession or degrees and put their primary articles to their user page, or sometimes people who doubt about ask those who probably know. This is a nice habit we already have. We do not have to discard this, but we can keep this kind of efforts in formal ways.

-- Taku 20:26 13 Jun 2003 (UTC)

Eh. I wouldn't worry about it too much. It seems to me that Wikipedia will always have a relatively high concentration of well-meaning and conscientious editors who generally will not let total nonsense articles stay around. The sorts of people who would post nonsense are probably quite likely to get bored and go away after a short time; the ones who stay are usually the ones most interested in improving Wikipedia.

As for determining what is nonsense, well, Google obviously can't answer all questions, but if some topic doesn't show up in a Google search at all (or is very poorly represented), that topic tends to be highly suspect. We've all seen the discussions (boy, that's the nice word for it) over articles like Neutrosophy. I would be very surprised if your friend's Fieldism article had remained for long; someone was bound to run across it sooner or later and find out it was garbage.

We probably could use a good system of at least getting reliable Wikipedians to look at all articles at least once. In practice, this seems to happen anyway, but a more systematic way of doing it wouldn't hurt. I don't think this will become much of a problem until the article-to-reliable-editor ratio gets much higher, though. The bigger Wikipedia gets, the more reliable editors we'll gain. I think if it's worked for this long, for 100K+ articles, it will continue to work for some time to come. -- Wapcaplet 22:49 13 Jun 2003 (UTC)

Yeah. "Fieldism" was spotted and thrown out. Vandals (who probably see themselves as "practical jokers" or "testing the system") are not really that useful as critics -- it is, in my opinion, like complaining that a city isn't clean after you have urinated in the street (as a test, you understand, as a test). Karada 23:02 13 Jun 2003 (UTC)

I've always been quite impressed with the security of Wikipedia. I was talking with someone last night about Wikipedia and he told me that the security was lapse, because he had been able to vandalise one of the pages, tea bag. I went to see what he had done, but it had already been reverted (less than 60 seconds). I told him that was our security. CGS 09:05 14 Jun 2003 (UTC).

Heh. What security? Did you tell him that's the whole point? It's very easy to paint graffiti on Wikipedia, but all the walls are made of teflon... -- Wapcaplet 15:43 14 Jun 2003 (UTC)

One suggestion I have is simply to cite sources of information. Sometimes people just use sources in their own heads and that is fine, but in other situations they really should cite the source of the information. I've done a lot of work with geographical locations (cities, states, and counties) and citing them using Geographic references for one example. It helps critics of the articles to know that the information can be verified by checking the sources. Sometimes this requires someone to work slower and doing the boring part of adding citations even though it is not as fun as writing the article itself. -- Ram-Man