Wikipedia:Village pump/December 2003 archive 1

From Wikipedia, the free encyclopedia

You have new messages (last change).

Contents

[edit] Copyright vio and reversion, Stub warnings, Search engine

I created a page from text on a web site I am a contributing member of

spent a lot of time setting up the links to other enteries and potential enteries

and someone decided that as on paragraph of the text was the same as the web site that I was breaching copyright and the whole text was deleted.

My points are

  1. Only one paragraph was taken from the web site so why was the whole page trashed? - I had to do the work over to recreate the links
  2. As a contributing member I was not breaking copyright - so would not it have been better for someone to email me first to ask IF I had copyright.
  3. I could not find out how to reinstate the original page.

Another point

I showed my 12 year old daughter the system and encouraged her to enter something - eventually we noticed her school was mentioned but had no entry so she typed in a short entry saying where it was a what type of school it was - just a couple of lines but factual.

Someone then put in a line saying 'THIS IS A STUB' etc. etc. and it just seemed to me to be insensitive and discouraging - given that the information did tell you the status of the school and where it was - may have been short but it was not valueless.

I'm sure many people have made this point - but the absence of an uptodate search engine seems to be a major major flaw in the credibility of the project.

Kevin Flude

As to your copyright issue - the page was not trashed and its content is all still visible. You merely need to make mention you own the copyright on the talk page and everything should be okay.
As to the stubnote thing - this is standard procedure to add a short note as an indication and an invitation for others to expand on an article. I trust the user who added the stubnote meant not to be insensitive or discouraging, but is merely a courtesy to other users. A stubnote is not a judgement of lack of value, but in my eyes an invitation for expansion.
The search engine issue is unfortunate, but I'm not sure how it impacts credibility - considered the Wikipedia runs on donations and support from Bomis, I believe. Dysprosia 09:01, 2 Dec 2003 (UTC)
To re-iterate and expand on Dysprosia's comments:
Sadly the number of people who simply take content off other people's websites and submit it as their own vastly exceeds the number of people who write their content, put it on the web, and then submit to Wikipedia as you did. To avoid legal problems we _HAVE_ to take down material as soon as its legality is questioned - happily in cases like yours when the fact that it is a false alarm is realized the material can be easily recovered. If this has not been done in your case yet, tell me the name of the article and I will do it. I hope this answers your questions 1&2. In response to 3, it is easiest if I point you to Wikipedia:How to revert a page to an earlier version.
Re the stub, I endorse Dysprosia's comments. The "THIS IS A STUB" notice is not a "THIS ARTICLE IS HOPELESS" euphemism. Actually it serves a technical purpose. If you go to the Wikipedia:Find or fix a stub page and click "What links here" you get a list of all the pages with a stub notice... i.e. a list of pages that, in an ideal encyclopedia, would be longer and more complete than they are. Without that notice, the technical trickery would not work.
Re the search engine, thanks to donations, we get a third server dedicated to Wikipedia turned on tonight. (This server alone cost around $6000.) Hopefully search facilities will be expanded as a result of this. Pete/Pcb21 (talk) 14:38, 2 Dec 2003 (UTC)
Perhaps that's not something to rush into.
What's the problem with using Google? I've found it works fine! It's not completely up-to-date, that's true, but that could even be an advantage. When someone is using a search engine in an encyclopedia, IMO they normally want stable content.
The other thing about using Google like this is it may promote our articles in Google when others use Google "native". I don't think Google actually reveals this either way, or even announces when their ranking algorithms change let alone how, rather they keep them a secret to hinder attempts to rig their rankings. But if our visitors from many different IPs use this feature, and then branch to a Wikipedia article, I'd expect Google to notice that. It's not rigging the ratings as such, just providing Google with some valid and accurate evidence which they will happily use IMO. Food for thought? Andrewa 19:57, 2 Dec 2003 (UTC)
The new server has not been purchased for the search engine. It will help to solve the slowness problem that Wikipedia is experiencing since months (even if lately is quite better, but the increase in traffic will get us anyway). If, after this, there is enough spare capacity for the internal search engine, all the better. Alfio 20:37, 2 Dec 2003 (UTC)
Why do you say all the better? That's what was originally suggested in this string, and I expect many agree, as it has generally been assumed that this is the case. But in reading this, it suddenly occurred to me that there are some good reasons for not having one. So, what are the reasons for wanting an in-house search? How does it assist readers? Is it just the vanity thing that other sites have them and we don't want people to think we're not able to?
Are there any problems with using Google apart from this 'credibility' issue? If not, is our credibility really increased by the in-house search engine? Or might it even be increased by being green hat enough to stay with Google?
As for stable content, not being up-to-date doesn't mean it will have the stable version. It would be possible for someone to search just as the page is changed, but it's also possible for Google to cache the page just as the page is changed too. Κσυπ Cyp   20:39, 2 Dec 2003 (UTC)
True. It certainly doesn't guarantee that the version will be stable. I'm not even sure whether it actually improves the probability of getting the content that the reader wants (I once studied such things, but that was more than 20 years ago), but my guess is that it doesn't make this probability any worse, that's all I was saying, probably not very clearly. If this is so, then it's not a valid reason for an in-house search engine.
This discussion has now been taken up on Wikitech-l. Andrewa 00:49, 4 Dec 2003 (UTC)
As for Special:Wantedpages, that's a different issue. If the capacity is there it could be turned back on. Perhaps, have a feature automatically disabling it and any similar processor-hungry frills on the fly whenever performance reaches an unacceptable level? Andrewa 18:21, 3 Dec 2003 (UTC)


We should add a couple of FAQs based on this query: 1) "Why has my material been deleted" 2) "Why has my page been marked as a 'stub'?". -- Tarquin 13:23, 5 Dec 2003 (UTC)

Yes, good idea. Andrewa 08:56, 6 Dec 2003 (UTC)
Uh, to find out what the stub thing is all about, one clicks on the link stub shown at that very article which is marked so. If the author doesn't want to click on that, he is not gonna go about to find an FAQ and discover where exactly his question would be answered. --Menchi 08:58, 6 Dec 2003 (UTC)

[edit] Internet Connection

Just curious, what type of Internet connection does Wikipedia use? What is the bandwidth?
66.32.17.177 09:09, 3 Dec 2003 (UTC)

There's a statistics page for the Wikipedia (here), which say that for the last month, the peak throughput was 13650806 KiB/day ~= 1264 Kbps, or less than enough to saturate a T1.
The computers that host the Wikipedia (there are now three of them; huzzah) are all located in Bomis' data-centre, and Jimbo claims (semi-frequently, on the mailing lists) that the Wikipedia's traffic gets lost in the noise, though there is suspicion that he's just trying to be polite and modest about his expenditure on the project.
For this to be so (Wikipedia traffic < 10% of total), Bomis probably has a DS3 or better. Sorry I couldn't be more definite.
James F. (talk) 17:06, 3 Dec 2003 (UTC)


[edit] Link to redirect?

Is there a way to make a wikilink that will go to a redirect page, without redirection? Or must I use a regular HTML link?
Thanks, Tualha 15:39, 3 Dec 2003 (UTC)

AFAIK, you have to use a regular HTML link. Why do you want to do this, by the way? Pete/Pcb21 (talk) 15:43, 3 Dec 2003 (UTC)
For a reference in Wikipedia:Redirects for deletion. It would save people having to go back. If there is a wiki way to do it, I'll note it in the page guidelines. If you have to use HTML, it's less trouble to go back. Tualha 15:47, 3 Dec 2003 (UTC)


[edit] Anonymous edits

This cannot possibly be the first time this has been suggested, but I wonder if it might be better if anonymous IPs are only allowed to access WP read-only. Looking at the vandalism alerts, the vast majority are anonymous IPs. I suspect what happens is that a casual user browses in by chance, finds (to their surprise, probably) that they are able to change the content, and without much thought, does so. Usually this will be something facetious or silly, simply because this is what happens when people are presented with an unexpected opportunity. Those who are more considerate and understand what WP is about will be more than willing to register, it's not as if it takes much effort or costs anything. The ability to edit anonymously was probably very important when WP was first started, simply to get the ball rolling. It's rolling very nicely now, so this feature is no longer a blessing but a liability. I'd bet that implementing this would cut petty vandalism by 90% overnight, without having any significant effect on WPs growth. Thoughts? GRAHAMUK 07:03, 5 Dec 2003 (UTC)

  • User:Snoyes did an informal survey and found about a third of edits by anons were vandalism. The usual counter-argument to your suggestion is that "vandals will then register, making them harder to remove". This argument probably overestimates the resolve of most vandals, who tend to quickly leave, even without being banned meaning many would be unlikely to register. I think that if registered-only editing is implemented, then anons should still be able to edit talk pages. User:Adam Carr has suggested even stricter restrictions. Maximus Rex 07:10, 5 Dec 2003 (UTC)
  • It might be useful to show the IP of even registered users when making edits. This would help to identify users with multiple IDs (used to support their own views in talk pages) as well as the ISPs of vandals with user accounts. Anjouli 08:03, 5 Dec 2003 (UTC)
    • Showing the whole IP isn't a great idea. If nothing else, it enables internet-based attacks on specific users. For example (in my evil spammer persona, below) I might conduct by wicked spamming while simultaneously DoS-ing as many developers as I could. Showing all but the tail octet (e.g. 192.168.22.???) would be okay, however. Lots of sites do this. -- Finlay McWalter 02:09, 6 Dec 2003 (UTC)
    • I think it would be a violation of privacy. Just because other sites do it does not mean the wikipedia should do so too. One of the benefits of registering is that you're not showing your IP. If it were abtracted (or encrypted, hashed, whatever) I would be OK with it. In fact I would rather even the IP's of anons were not shown. People should also have a look at the Draft privacy policy and help improve it. Dori | Talk 04:02, Dec 6, 2003 (UTC)
  • (slightly offtopic) A while ago, after reading a rather scary article about spammers were starting to spam into blogs, I sat down and figured out how a spammer might successfully spam wikipedia. Among the conclusions I came to was that I (in my hypothetical evil spammer persona) would only use signed-in users, never anons. This is because any regular admin can block an anon, but it takes a developer (and generally some discussion) to block a signed in user. Armed with a few dozen IPs (trivial for a professional spammer) and a few hundred automatically-created accounts, a smart spammer could easily flood us with so much crap that it would be a full-time job blocking and reverting them. To date we've been lucky that 99% of our vandals only want to put "Zaphod is l33t" into a couple of pages and then go play Mariocart. So I'm not expressing an opinion as to whether anons should be blocked, but be aware that such a move affords no protection against the smart, determined vandal or spammer (from whom we have greatly more to fear). -- Finlay McWalter 01:58, 6 Dec 2003 (UTC)
    • My understanding is that "regular admins" can block users, though the interface does not make it obvious how to, but only developers (or rather only one specific developer) should make that decision. --Pakaran 02:04, 6 Dec 2003 (UTC)
      • Or at least it doesn't make it obvious enough for me to see at first glance :( --Pakaran 05:06, 6 Dec 2003 (UTC)
    • Actually, the presence of anonymous IPs makes it easier to spot vandalism. When looking for vandalism, I usually just scan the IPs and ignore the user names. Kingturtle 09:12, 6 Dec 2003 (UTC)


[edit] CVS Blame

Is it possible to do a 'CVS Blame' on an article? Jahs 17:39, 6 Dec 2003 (UTC)

What does that mean? RickK 19:59, 6 Dec 2003 (UTC)

No, it's not possible. ("CVS blame" is really the "cvs annotate" command which shows who last modified each and every line, very useful tool, actually, and not just for blaming people for problems.) Daniel Quinlan 20:58, Dec 6, 2003 (UTC)
Sorry, should have explained for non-geeks. :) So it's not possible at the moment, but is there a technical reason against it? WikiPedia keeps the full history (AFAIK) so in theory it'd be doable.
Well, it's a somewhat expensive operation and non-trivial to implement. In addition, because Wikipedia is paragraph-based rather than line-based, I doubt it would be quite as useful. However, I agree it would be useful. It's open source, so you can always implement it and contribute the feature. Daniel Quinlan 00:52, Dec 7, 2003 (UTC)
Well, see: http://www.research.ibm.com/history/ --Brion 06:51, 7 Dec 2003 (UTC)
Mind-blowing! Thanks for the pointer. So when do we get the open-source version?? Oh,well, anyway, there's another fine thing here, thanks to IBM's old pratcie of publishing research topics: some shmuck can't come along and patent the whole idea, now that it's published. Dandrake 19:12, Dec 7, 2003 (UTC)
To make feature requests, see wikipedia:bug reports. To contribute, start by reading wikitech-l. Martin 04:29, 7 Dec 2003 (UTC)

[edit] linking to a French Wiki image

can someone tell me how to link to an image that's in the French wikipedia space (or tell me if this is a silly thing to do)? thanks. seglea 20:56, 7 Dec 2003 (UTC)

It would be best if you re-uploaded it on the english wikipedia. The image may change afterwards, it could be deleted, etc. Anyway, currently the only way to do an inter-wiki image link is to use a full html link, like this: flag of the United States. Alfio 21:14, 7 Dec 2003 (UTC)

[edit] How-to to Wikibooks?

I think that that page should be deleted, and all of the articles that it links to should be integrated into Wikibooks. I got a vote of consent on talk:How-to, but I want a little more discussion before I undertake such a significant change.-Smack

Are you proposing one or more "How-to" books for Wikibooks using as starters these articles at Wikipedia? If so, sounds like a good idea. How-to articles consolidated into one or more texts would probably be more suitable as a Wikibooks project IMHO - 24.94.82.245 02:59, 2 Dec 2003 (UTC)- Marshman 03:18, 2 Dec 2003 (UTC)
I have no doubt that that is a good idea. What I meant (sorry I wasn't clear the first time) is that the Wikipedia articles (or sections of articles, as the case may be) should be deleted. -Smack 07:15, 2 Dec 2003 (UTC)
Would it be possible to leave the how-to pages but replace the contents of each one with a link (maybe even a redirect ?) to the relevant part of Wikibooks ? Articles that link to the how-to pages would then not be left with broken links. -- Gandalf61 10:10, Dec 2, 2003 (UTC)
That's a sensible idea. Is it possible to create redirects like this ? Smack might I suggest you start with the cookery pages, as It is clear where you could put them in wikibooks. Drop me a line on my talk page when you are ready to start and I'll give you a hand. theresa knott 11:07, 2 Dec 2003 (UTC)
Yes, it is possible to make interwiki redirects. Unfortunately, such a redirect seems to omit the familar Redirected from line. User:Smack/interwiki redirect test -Smack 02:18, 4 Dec 2003 (UTC)
It is possible, but a lot of people see it as a bad thing, because of the difficulty in editing the redirect and the confusion that being at another site may cause. See m:Redirected user pages considered harmful. Angela 02:23, 4 Dec 2003 (UTC)
Point taken. So would it be wise to delete the how-to content altogether? -Smack 04:00, 4 Dec 2003 (UTC)
If you just delete, you lose the edit history. Therfore, before deleteing you need to work out a list of authors for each article, and post their names in the edit summary of the page when you move it to wikibooks. ie {moved from wikipedia -authors are , name, name , name .....} So that we comply with the copyleft licence. Does this make sense ? theresa knott 11:13, 10 Dec 2003 (UTC)
The most sensible thing would be to transfer the page with the complete editing history intact, wouldn't it? Note that Special:Export was specifically designed for this sort of thing, though the requisite import facility isn't done just yet. --Brion 12:01, 10 Dec 2003 (UTC)

[edit] Wiki Farm for mediaWiki

I am trying to set up a special purpose wiki where only members of a special interest group can edit. I really like the MediaWiki that Wikipedia uses, and Wikipedia is an excellent example to show people how a Wiki works. Does anyone know of a Wiki farm that uses MediaWiki? I mean a host that runs MediaWiki that will allow me to set up my own Wiki, probably for a fee. pstudier 17:31, 2003 Dec 9 (UTC)

I don't know what a wiki farm is, but have a look at these sites Wikipedia:Sites using MediaWiki and see if you find what you're looking for. Also check out meta:MediaWiki. Dori | Talk 17:51, Dec 9, 2003 (UTC)
I'm not aware of any wiki farms using MediaWiki. Maybe you should try asking Jimbo Wales to set one up on our servers. -- Tim Starling 23:54, Dec 9, 2003 (UTC)

[edit] A proposal on categorization

This is a little idea of mine that I have written up. Please read and comment. Zocky 03:44, 9 Dec 2003 (UTC)

That's a very good idea. The only problem with it is that it has already been suggested, discussed, developed and implemented, and now only awaits some sort of decision on whether or not it should be released on the general public. Except it uses [[Category:xxx]], not [[cat:xxx]]. -- Tim Starling 03:47, Dec 9, 2003 (UTC)
Really? Well, it being the sensible solution, it's no wonder. I tried looking for it, but couldn't find anything of worth (and I did try suggesting something like this on meta quite some time ago). Please provide links. Zocky
It was mainly discussed on the mailing lists wikipedia-l and wikitech-l. See Eloquence's proposal from wikitech-l in February 2003 and Magnus Manske's completion note in July. -- Tim Starling 04:05, Dec 9, 2003 (UTC)
Heh, the same thing :) Is there anything in my text that's new, or should I just trash the whole thing? Zocky 04:13, 9 Dec 2003 (UTC)
I finally found where I tried proposing it first! It was on sourceforge: [[1]], back in January, but it didn't get any comments until last week :) Turns out many people suggested it. Must mean it's a good idea. Zocky 04:29, 9 Dec 2003 (UTC)
So what's holding up a trial? That's the only way we'll see whether it works, ie (a) sits comfortably on existing culture and (b) achieves something. Andrewa 06:39, 9 Dec 2003 (UTC)

[edit] Spell checking a Wikipedia article

What are good techniques that allow quick spell checking of a Wikipedia article? Bevo 19:19, 9 Dec 2003 (UTC)

Unfortunately, for now, the only way that I know of is to use an external spell-checker. You can do this by pasting the displayed text into a word processor such as Microsoft Word or OpenOffice.org and using their spell-checking capabilities, or using some command line program such as aspell or ispell. Dori | Talk 19:44, Dec 9, 2003 (UTC)
Using a Mac, with Safari. The spell checker is available to any editable text, including the text box for editing WP articles. Even as you type. I do wish they would make this panel support Undo though - maybe next version ;-) GRAHAMUK 23:27, 9 Dec 2003 (UTC)
I hadn't actually noticed that before; that's kind of neat! Unfortunately it doesn't automatically detect the language, so I have to manually switch dictionaries on en/fr/etc. And it doesn't have an Esperanto dictionary built-in, I'll have to try and track one down... :D --Brion 00:34, 10 Dec 2003 (UTC)
Dec 2003 (UTC)

[edit] Referrers

Wikipedia:Referrers is interesting but quite old - it hasn't been updated since September 2002. I think it would be worthwhile to have an updated version.

Tualha 06:44, 10 Dec 2003 (UTC)

[edit] Prevalence of pages about places

I wonder why it is that the majority of times I use "Random page" I get a page about a locality, and an American locality at that. Is there some special programming designed to give such results, or is this just an amazing coincidence?

Cheers JackofOz 07:44, 10 Dec 2003 (UTC)

We just have a lot of them. See User:Rambot. --Brion 07:51, 10 Dec 2003 (UTC)

It's because Wikipedia is full of junk. See my The Wikipedia Quality Survey Adam 08:26, 10 Dec 2003 (UTC)

Don't blame JackofOz for the terrible title of this section; mea culpa. JackofOz, you should see a link near the top of the page called "Post a question now" - it creates a new section header. Tualha 13:56, 10 Dec 2003 (UTC)

It's because a lot of stuff was robotically imported from US Census data. There are complaints about this from time to time, which are generally answered by people who note that their first introduction to Wikipedia was searching for their home town or birthplace on Google, which often turns up one of these very common Wikipedia articles. Tempshill 01:49, 11 Dec 2003 (UTC)

I don't see a problem with it myself. Yes, on average, pages are less edited, less "personal", than they were before Rambot did its stuff, and the article count can be considered inflated. But the other articles are no less useful or worthwhile. We just have a whole lot of other information, which is very useful itself, and as Tempshill pointed out, bring in plenty of new people. Tualha 04:40, 11 Dec 2003 (UTC)

So could there be a way to make "random page" produce more interesting results - e.g. maybe it would only select a page above a certain size, so the page is likely to be more "interesting" than a short page. But then I guess it would not be "random page" anymore ! Gandalf61 07:31, Dec 11, 2003 (UTC)

[edit] Slooooow

So, what was the technical problem last night, and presumably now too, since things are working but very slow? Tualha 14:44, 12 Dec 2003 (UTC)

Last night we had the database server fall over, choke, and die with a fishbone stuck in its throat. (Or the equivalent; it really shouldn't have died under the test load it was given.) It's been rebooted and the DB's memory usage adjusted in the hope that this will help.
I can't speak to the specific slowdown you're referring to, but we have had intermittent problems with the web servers which we haven't yet gotten to the bottom of. --Brion 03:53, 13 Dec 2003 (UTC)
Thanks. Tualha 05:35, 13 Dec 2003 (UTC)

[edit] Adding an email/Changing a password

How does one go about adding an email to an account without one? My old account doesn't have an email address, and I forgot the password years ago.

Help!

You probably need to convince one of the developers that you are the same person. I would imagine that that would be a difficult thing, so the only solution is to probably open a new account. Someone correct me if I am wrong on this. Dori | Talk 17:20, Dec 9, 2003 (UTC)


[edit] Can you link to a subheading?

Does the software allow links to a subheading (or other specific part) of a wiki page? I know this is possible with HTML, but haven't found any instructions for how to do it here.

Thalia/Karen 05:38, Dec 13, 2003 (UTC)

Hi Karen, yes you can do this using the standard HTML method. Use the URL for the page as normal, add a hash (#), and then the full name of the section replacing any spaces with underscores. For example Wikipedia:Village_pump#Can_you_link_to_a_subheading? Chris Jefferies 08:25, 13 Dec 2003 (UTC)
Note that it's generally considered a bad idea to jump from a different article into the middle of another. Within the same article, it's generally considered okay. Daniel Quinlan 08:30, Dec 13, 2003 (UTC)
Daniel, I do want to jump from one article into the middle of another, but I think it makes sense in this case. There's already a link to the top of the second article a bit earlier in the one I'm working on; I want to add a second link to a specific subheading that doesn't appear on the first screen of the (linked-to) article. When I'm finished working on it, I'm going to ask for reviews, since this my first substantial writing job here. So if someone thinks it's a bad idea.. :-) Thalia/Karen 19:14, Dec 13, 2003 (UTC)
You don't need to replace spaces with underscores. Wikipedia:Village pump#Can you link to a subheading? works just as well.—Eloquence 13:12, Dec 13, 2003 (UTC)
Thanks for your help, folks. Thalia/Karen 19:14, Dec 13, 2003 (UTC)
Retrieved from "http://en.wikipedia.org../../../v/i/l/Wikipedia%7EVillage_pump_December_2003_archive_1_66c2.html"