User talk:Coren/Archives/2012/January

Poke

Hi Coren, happy new year! I'm just checking in about the CSBot test – do you want to get it started soon? No rush if you're busy, of course. Just lemme know :) Maryana (WMF) (talk) 17:47, 3 January 2012 (UTC)

I wasn't so much "busy" as "resting". Same level of on-wiki activity, much more fun!  :-) I'll be all yours by the end of the week. — Coren ^(talk)

Ok, cool :) Enjoy the break! Maryana (WMF) (talk) 20:18, 4 January 2012 (UTC)

Praise to the Man (album)

Hello and happy new year! I haven't gotten the information from the site you mentioned. I wasn't copying anything otherwise. Therefore, I just need to get the article completed.Suite1408 (talk) 21:59, 3 January 2012 (UTC)

Question about Deletion of Edgar_J_Brown wiki page

Hi Coren

I'm the person who posted Edgar J. Browns wiki page and was wondering why it was deleted (http://en.wikipedia.org/wiki/Edgar_J_Brown). A lot of the content was taken from his bio page on edgarbrown.com. Since the page was not accepted how can we go about getting an accomplished artist such as Edgar Brown into wikipedia.

Thanks

James — Preceding unsigned comment added by 74.105.37.179 (talk) 05:00, 5 January 2012 (UTC)

Please read the guide to requesting and formalizing permission to use copyrighted works on Wikipedia. Note that, in addition to copyright requirements, the article must still comply with notability guidelines, advertising prohibition and avoid conflicts of interest. Alternately, you can simply write the biography in your own words (not by modifying text you have taken elsewhere), taking care to cite your sources. — Coren ^(talk) 05:02, 5 January 2012 (UTC)

Request for Interview Regarding Wikipedia Bots

Greetings Coren-

My name is Randall Livingstone, and I am a graduate student at the University of Oregon, currently collecting data for my dissertation on Wikipedia editors who create and use bots and assisted editing tools, as well as editors involved in the initial and/or ongoing creation of bot policies on Wikipedia. As a member of BAG and the bot community, I would very much like to interview you for the project at a time and in a method that is most convenient for you (Gchat, another IM client, Skype, email, telephone, etc.). I am completely flexible and can work with your schedule. The interview will take approximately 30-45 minutes.

My dissertation project has been approved both by the Institutional Review Board (IRB) at the University of Oregon, and by the Research Committee at the Wikimedia Foundation. You can find more information on the project on my meta page.

Please let me know if you have any questions, and I look forward to hearing from you to set up a time to chat. Thank you very much.

Randall Livingstone, School of Journalism & Communication, University of Oregon

UOJComm (talk) 00:30, 6 January 2012 (UTC)

I'd be glad to contribute what I can. Given my often sporadic schedule, email is almost certainly the best option. — Coren ^(talk) 01:28, 6 January 2012 (UTC)

Great! Thank you very much for the response and your willingness to be interviewed. Sometime early next week I will email you a set of interview questions (I like to do some background so I don't send you questions that are obviously already answered on your user page).

At your convenience, please review the Informed Consent Statement posted on my User page and let me know if you have any questions. Specifically, let me know if it's ok for me to use your Wikipedia username in my dissertation.

Thanks again...Randall UOJComm (talk) 19:02, 6 January 2012 (UTC)

Do feel free to use my username. I look forward to hearing from you. — Coren ^(talk) 19:59, 6 January 2012 (UTC)

Hello Coren. This is just a friendly reminder to email me back your responses to the interview questions for my research study on bots and bot operators. If you have any questions or concerns, or would like me to send you the questions again, please don't hesitate to contact me. Thank you again for your participation, and I look forward to hearing from you. UOJComm (talk) 23:37, 14 February 2012 (UTC)

Unblock request from User talk:PurpleHeartEditor

Hi Coren!

I wanted to have a look at this block, but I can't find on-wiki documentation for the CU (except the CheckUser logs :)). Can you please have a look at his request and see if things might check out? (it looks like he's using shared IPs at the moment, at first glance). -- Luk ^talk 09:44, 6 January 2012 (UTC)

Thanks! I've been out of the loop for too long :( -- Luk ^talk 15:22, 6 January 2012 (UTC)

Recent Edits

Hello, on WikiPedia I have recently merged and edited several articles into two new articles:
http://en.wikipedia.org/wiki/Leicaflex
http://en.wikipedia.org/wiki/Leica_R8
and these almost immediately attracted the attention of your CorenSearchBot even though all the text already existed in the previous articles. This morning the warnings have disappeared; do I need to do anything else?DesmondW (talk) 16:34, 6 January 2012 (UTC)

No, human reviewers saw what you did and that it was okay. All good.  :-) — Coren ^(talk) 16:49, 6 January 2012 (UTC)

Gary Chapman (footballer)

Another incorrect tag - the article is on an English former professional footballer, the website is on an American "internationally respected marriage and family life expert" - not even close! GiantSnowman 20:00, 7 January 2012 (UTC)

Site may cause problems for Bot.

There is a website at Wikipedia Topics which appears to be causing your Bot some problems. I was looking at the Wikipedia article "English words with uncommon properties" and noted that it had been flagged with a very prominent notice because of issues detected by the bot. I decided to look at the article it had been compared with and discovered that it is, for the most part, a series of links back to the Wikipedia article. They do this kind of thing with a lot of Wikipedia articles. This might result in a number of false flags. It seems to have been done by Swarthmore College Computer Society — Preceding unsigned comment added by Kuitan (talk • contribs) 14:52, 8 January 2012 (UTC)

Wikipedia:Arbitration/Requests/Case/Civility enforcement/Evidence/Blocks

In the course of an ongoing case, the Arbitration Committee has decided to collect all relevant information regarding Malleus Fatuorum's block log and, as such, has created a table of all blocks, which can be found here. Since you either blocked or unblocked Malleus Fatuorum, you are welcome to comment, if you wish. Salvio ^{Let's talk about it!} 13:55, 9 January 2012 (UTC)

Maybe the sensitivity is set a little too high

I recently observed this bot flagging an (admittedly short) article as a "substantial copy" of a completely unrelated web page, apparently just because they both contained the word "Galion". (Other than that, they had no other words in common except ones search engines routinely ignore, e.g. "of", and no phrases in common at all.) I don't think it's reasonable for the bot to conclude one page is a copy of another based on a single word that they have in common. There should have to be multiple words at the very least -- multiple phrases, really. Otherwise it's going to cry wolf MUCH too often. --Jonadab, 2012 January

CorenSearchBot's reliability goes down sharply as the size of the pages gets smaller – and the algorithm isn't particularly stable on very short pages (which causes the occasional really odd false positive like that which you have seen). In the past, the people at WP:SCV have consistently requested that I should not raise the minimal size of comparisons: while the rate of false positives is rather higher, they do catch a number of cut-and-copy they otherwise could not.

On the other hand, if there ever is consensus to raise the minimal size of pages up for comparison, the fix is trivial and takes moments to deploy. If you feel the balance of useful notices to false positive has tilted, discuss it with the people at WP:SCV and WP:CP and I'm sure you can get agreement about what is the "best" useful small size of articles to check and I'll tweak accordingly. — Coren ^(talk) 17:26, 10 January 2012 (UTC)

Xebulon SPI

Did you check against Winterbliss and Oxi42 or just Xebulon?--The Devil's Advocate (talk) 23:57, 11 January 2012 (UTC)

BTW, the Oxi account was a few months old and blocked before Winterbliss registered so that is why I brought up the possible match. I figured a suspicious account that edited at the same time as the Oxi account would possibly be a stronger match.--The Devil's Advocate (talk) 00:04, 12 January 2012 (UTC)

It did justify a check, which is why I did. — Coren ^(talk) 00:15, 12 January 2012 (UTC)

What accounts did you check it against? This sockmaster, from looking over various SPIs has had some success at evading CU inquiries in the past with much stronger conclusions temporarily exonerating the accounts.--The Devil's Advocate (talk) 00:39, 12 January 2012 (UTC)

Ahlfors theory

Hello Coren,

User:Pym1507 received the following message from the bot:

This is an automated message from CorenSearchBot. I have performed a web search with the contents of Ahlfors theory, and it appears to be a substantial copy of http://www.ala.org/alsc/awardsgrants/bookmedia/newberymedal/aboutnewbery/aboutnewbery. It is possible that the bot is confused and found similarity where none actually exists. If that is the case, you can remove the tag from the article. The article will be reviewed to determine if there are any copyright issues. If substantial content is duplicated and it is not public domain or available under a compatible license, it will be deleted. For legal reasons, we cannot accept copyrighted text or images borrowed from other web sites or printed material. You may use such publications as a source of information, but not as a source of sentences. See our copyright policy for further details. (If you own the copyright to the previously published content and wish to donate it, see Wikipedia:Donating copyrighted materials for the procedure.) CorenSearchBot (talk) 22:00, 11 January 2012 (UTC)

This is quite ridiculous, since the webpage has nothing in common with the Alfors theory article (the latter is a mathematical article, the former... well, I do not even understand what is it about So I guess there is some problem with the bot. Could you please check this (and perhaps leave a message at User talk:Pym1507 that the bot message is void)?

Thank you very much, Sasha (talk) 00:06, 12 January 2012 (UTC)

PS I would very much appreciate a response, since the edit of Pym1507 removing the copyright tag was subsequently also categorized as vandalism. Sasha (talk) 01:22, 13 January 2012 (UTC)

Well, the original match was clearly a false positive (and a surprisingly confusing one at that); those unavoidably occur from time to time (as you can see on this very page) but are generally not a significant problem since every tagged page is reviewed by a human before action and would have been quickly noted as such. I think that the original editor, wanting to remove the template, erroneously blanked the whole page instead which – out of context – does look like vandalism and was reverted as such by someone who was not as careful as they should. You obviously did the correct thing by reverting back (although I think that your own qualification of that revert as vandalism itself was a little rash).

That said, while the compound errors are regrettable, I'm not sure how you think I can help? — Coren ^(talk) 01:36, 13 January 2012 (UTC)

Hello Coren,

what happened is the following:

• Pym copied the page twice from his sandbox by mistake
• the bot interpreted it as copyright vio and placed a template on his page
• Pym erased the tag and also the repeating tag
• another automatic tool classified this as vandalism an put another template on Pym's page
• I reverted the edit, and also overreacted a bit (but later apologised on Download's page)

While I understand that the bot has false positives (as any bot does), I do think it should have a feature that places a correction after the template once a tag is identified as false positive (or the owner should do it manually). Otherwise, a) it is rather discouraging to new editors who do not understand what is going on, and b) other editors who see a tag removed and see a warning template at the user's talk page are tempted to place another warning (this is what happened this time). Vice versa, if they see the previous warning was void, they understand they should probably not place another warning.

Best regards, Sasha (talk) 01:56, 13 January 2012 (UTC)

Well, part of the problem is that the bot (obviously) can't know whether a match is a false positive; it's humans that will note this when reviewing the tagged pages. I suppose it would be best to always note this on the editor's talk page when it occurs – but in this case it wouldn't have helped: nobody had gotten around to evaluate that particular tagging in the first place (I expect you noticed it directly through NPP). — Coren ^(talk) 02:15, 13 January 2012 (UTC)

Just for future reference: a) if I see a mis-tagged page, should I manually place a note at the editor's page that the template is void? b) how to make sure that the bot will not tag the same page with the same tag again?

Thanks a lot, Sasha (talk) 02:23, 13 January 2012 (UTC)

Well, you certainly can though if you do so you probably also want to note it at WP:SCV too. That last bit isn't necessary, but it makes sure nobody else goes over that same tagging again.

As for preventing a tag being put again, the bot only tags pages that were just created and that have no deletion or warning templates. Unless the page was deleted, then recreated with the same contents sans the templates, this shouldn't happen. There is no way to prevent the bot from tagging a new recreation with the same content, however, since it doesn't have admin permission it cannot see the deleted revisions. — Coren ^(talk) 02:27, 13 January 2012 (UTC)

thanks for the explanation! I have just left a note at WP:SCV. Sasha (talk) 03:29, 13 January 2012 (UTC)

another error

See CDHRAP for another rather amusing error. i admit being rather curious what CDHRAP and the Acrobat(trade secrets)(TM) web page have in common... CDHRAP have pdf files for download, but that's not part of the Wikipedia article, and CDHRAP don't even seem to provide the ubiquitous false claim that the pdf files "require" using Acrobat(TM)(trade secret) non-free software! Boud (talk) 23:29, 12 January 2012 (UTC)

That... okay, that one has got me thoroughly confused. There's obviously a bug because both pages are large enough that there is plenty of content to (not) match. Looking into it now. — Coren ^(talk) 01:44, 13 January 2012 (UTC)

I've added verbose instrumentation for the next couple of days, so that I can know why the next really odd match occurs. It's not going to help with past matches, but at least I'll have some data to figure out what happened with those. Thanks for bringing that egregious case to my attention. — Coren ^(talk) 02:12, 13 January 2012 (UTC)

Can i place a link on the page i've created acknowledging the source??

Can i place a link on the page i've created acknowledging the source??-Ankit Agrawal 16:24, 13 January 2012 (UTC)

That wouldn't be sufficient; you need explicit permission to use text copyrighted by someone else under a licence suitable for Wikipedia. Please read the guide to requesting and formalizing permission to use copyrighted works on Wikipedia. Note that, in addition to copyright requirements, the article must still comply with notability guidelines, advertising prohibition and avoid conflicts of interest. — Coren ^(talk) 16:51, 13 January 2012 (UTC)

MediaWiki API, & committing to Subversion

Hi! Thanks for writing this bot.

I'd welcome your thoughts on ways the mw:API documentation should be improved. We're trying to make it more useful for developers like you.

And would you be interested in using the main Wikimedia Subversion repository to store the bot's code, to make it easier for others to improve it? You can request commit access pretty easily to commit your bot to SVN.

Best, Sumanah (talk) 01:58, 13 January 2012 (UTC)

Isn't a bot out of scope for that repo? If not, I should clean up the code a bit first so that it doesn't contain secrets anymore (and move those to an external file); I can sure push it there if useful.

As far as the docs go, they've been adequate enough I suppose, although I'm known to be able to understand IBM docs so I might not be the best judge.  :-) Lemme read it again top to bottom and figure out if there are points worth raising. — Coren ^(talk) 17:54, 13 January 2012 (UTC)

Scratching my head here ...

I just started a new article at Rakni's Mound and immediately got templated for suspected copyvio - of Rakni's Mound! All that previously existed was 3 lines at Romerike#Raknehaugen to which Raknehaugen previously redirected. I searched for Rakni's Mound and clicked on the redlink to start the new article. Whu hoppen? All I can think of is I triggered something by previewing the new article a couple of times. Yngvadottir (talk) 17:52, 15 January 2012 (UTC)

A similar situation with 2012 Clemson Tigers football team. Wikipedia can't copyvio itself. -- P.T. Aufrette (talk) 00:26, 16 January 2012 (UTC)

Same with Buddleja davidii 'Grey Dawn'. -- P.T. Aufrette (talk) 00:58, 16 January 2012 (UTC)

New Zealand 10 cent coin

You should exclude domains like uk.ask.com which host mirrors of Wikipedia's own content. In this case, the newly-created New Zealand 10 cent coin article obviously resembles the already-existing New Zealand 20 cent coin article. -- P.T. Aufrette (talk) 00:22, 16 January 2012 (UTC)

skip all articles with a quote mark in the title

I noticed the discussion at User talk:Yngvadottir. The bug should not be that hard to fix, but until it is, the bot should simply skip any article with a quote mark in its title. -- P.T. Aufrette (talk) 01:10, 16 January 2012 (UTC)

blank (unspecified) target page

The bot seems rather buggy: in some cases it reports a copyvio from blank (no page specified). For instance, [1] [2] [3] [4]. You really should add some simple sanity checks to avoid this and the quote-bug. -- P.T. Aufrette (talk) 01:30, 16 January 2012 (UTC)

iCarly - Season 6

Don't mind this Mr. Coren. I was just moving these pages back since I have provided enough evidence on behalf of the Season 2 production split of this show. I was moving the information from the Season 4 page over to here, and then change the name of the page to Season 5. - Jabrona - 01:03, 17 January 2012 (UTC)

Wilmette Institute

Hello Coren Bot,

I am the author of both the Wilmette Institute pages on Wikipedia and Bahaikipedia, as you can see by the same username on both sites. It's a bit of two for the price of one? The articles will look substantially different by the time I'm done. Thanks,I'm Nonpartisan 18:20, 19 January 2012 (UTC) — Preceding unsigned comment added by I'm nonpartisan (talk • contribs)

BOT function and copyright issues

Hello! Please do not take this as an angry message. Below is the content of your message to me, my instant reply, and my subsequent thoughts.

"Conus nocturnus

Hello, Shellnut, and thank you for your contributions!

An article you worked on Conus nocturnus, appears to be directly copied from http://en.goldenmap.com/Conus_spurius. Please take a minute to make sure that the text is freely licensed and properly attributed as a reference, otherwise the article may be deleted.

It's entirely possible that this bot made a mistake, so please feel free to remove this notice and the tag it placed on Conus nocturnus if necessary. CorenSearchBot (talk) 22:55, 16 January 2012 (UTC)

It is not a copy. It is an entirely new article with new and different content. ..."

There was a second species article also flagged by the BOT while I was working adding species articles to remove red-lined species names. I immediately removed the BOT flags and fixed up the articles to finish them.

Now, a day later in reflection I have looked to the material referenced by the BOT. That other website has taken verbatim from Wikipedia language which is used in "our articles." The content was originally written by Wikipedia members/editors who are on the WikiProject Gastropods team. That language is not copyrighted and when another website copies it and uses it on their site we are getting further distribution of Wikipedia's materials - their republication of Wikipedia articles does not make it subject to copyright protection. Therefore our re-use of our own standard layout and formating, even actual copying of a sentence or two that are reused in many many Wikipedia articles is not a copyright violation. I recognize that a BOT is merely a computer program and is designed to protect against inadvertent or intentional copyright violations, and that another website using Wikipedia's articles verbatium is indistinguishable from new content. Can anything be done, or is this just going to be frustrating while writing new species articles? Shellnut (talk) 02:00, 18 January 2012 (UTC)

Actually, as sites that copy Wikipedia contents are found, they can be added to a specific whitelist of sites we know have contents it is generally safe to copy from. I have done so for that site, so it should no longer consider matches from there at all. — Coren ^(talk) 02:55, 18 January 2012 (UTC)

Thank you! Shellnut (talk) 01:46, 21 January 2012 (UTC)

Hi. I recently updated the entry on me (created by Dr. James K. Woosley) to reflect new publications (as Dr. Woosley was too busy to do so) and to create links to entries on those books as requested by fans. These entries were based on the promo blurbs created for each book, and which I and my coauthors (such as Travis S. Taylor) use on our websites as a quick means of updating them, but are IN NO WISE copyright-protected, as this would prevent newspapers, etc. from running with the blurbs as is. I was suddenly bombarded by a slew of messages saying that I was in copyright violation, that I was not unbiased (when I even made a point of ensuring that Dr. Woosley would be editing behind me afterward to ensure that, as well as fix glitches - I have a hard time with your editing system), etc. These came so fast and furious that I was unable to answer them before another came in and overrode my attempt to post an answer. By the time I was done answering, I found that at least half of the entries I worked so hard to enter had been deleted without even bothering to check for a reply. Please check the user talk for steph-osborn and respond. Steph-osborn (talk) 22:01, 25 January 2012 (UTC)

I miss CorenSearchBot!

Looks like he's been down for a couple days. D: — madman 02:34, 23 January 2012 (UTC)

Gah! So it has – I'm used to my Nagios install warning me if the bot goes down, but that's no longer the case since it moved to the labs cluster so I never noticed. Need to fix that. — Coren ^(talk) 03:04, 23 January 2012 (UTC)

Thanks for the quick response! :D One more thing: CorenSearchBot seems to routinely flag articles with special characters as copyright violations of themselves. (e.g. the first article will have the character escaped and the second article will have the article unescaped). Is this possibly something you could look at? If not, I could possibly look at the source code linked to on CSB's User page and suggest a patch. Thanks! — madman 15:22, 23 January 2012 (UTC)

That version needs to be updated, there were a few changes since. But you're correct that this has been a little niggling bug I need to fix. I'll have a look this evening when I get a minute; I know what is going on (the title of an article isn't escaped exactly the same way depending on how it was found) I just need to trace the data paths to make sure all of them have the same postprocessing. — Coren ^(talk) 15:30, 23 January 2012 (UTC)

Potential Gremlins

I just had this message which claimed a page I created, O'Rahilly was a copyvio of another Wikipedia page of the same name. It then directs me to the exact same page. How can this be? The page didn't exist until this evening, and I see no other page of that name, so I'm guessing there's a glitch that needs fixing. :) Paul MacDermott (talk) 20:11, 25 January 2012 (UTC)

There is, per above; it'll be fixed as soon as possible. :) — madman 04:16, 26 January 2012 (UTC)

Hi, the bot tagged a page split to Thurston (name) on the grounds that it matched http://en.wikipediamindmap.com/2000/07/thurston-usa-county_5185.html – please exclude that domain from the bot's search, as it is clearly derivative from Wikipedia. – Fayenatic (talk) 22:00, 25 January 2012 (UTC)

Hm, this is actually more complicated than it appears; the site doesn't actually indicate it's a Wikipedia mirror, it's not licensed properly, and it doesn't include all articles so it's hard to tell if content's from Wikipedia or from another source. So I'm not sure this one can be added to the whitelist, but there are volunteers who watch everything CorenSearchBot tags and will indicate false positives like this when they haven't been indicated already. Cheers! — madman 04:16, 26 January 2012 (UTC)

Mildred Seydell

Your bot tagged it. Not directly copied. Fully referenced. Got it wrong. Happy editing. 7&6=thirteen (☎) 23:31, 24 January 2012 (UTC)

When you get a chance, you can see how it all turned out. 7&6=thirteen (☎) 20:26, 26 January 2012 (UTC)

Discussion continued: → User talk:7&6=thirteen#Mildred Seydell. — madman 21:06, 27 January 2012 (UTC)

DRV

A notification that the Templates for Discussion discussion (oy, repetition) has been taken to a deletion review discussion. The Article Rescue Squadron was notified, and as notifications to previous involved parties isn't normal practise, I and a few ARS members agreed that, in the interests of transparency and fairness, we should let everyone know...hence this talkpage message ;).

If anyone has an issue with me sending these out, do drop me a note on my talkpage. Regards, Ironholds (talk) 10:29, 28 January 2012 (UTC)

Deletion of Davis Carroll

Davis Carroll was my father and I am one of the original co authors of the obituary which is a public posting at legacy boston.com This wikipedia page that I attempted to create is for public information on a artist who has passed away. Is this not appropriate for wikipedia? Thank you

---

— Preceding unsigned comment added by Johndaviscarroll (talk • contribs) 19:09, 28 January 2012 (UTC)

Well, it depends. You should check our notability guidelines for people to see if he's suitable for inclusion on Wikipedia. Unfortunately, we can't really include information about every artist on Wikipedia. If he is suitable for inclusion, then you'll want to check out Wikipedia:Donating copyrighted materials to see how to give us permission to use the obituary. Since you can't give such permission on the boston.com site, the easiest way to do it is probably going to be to e-mail permission to permissions-en at wikimedia.org. Please let me know if you have any further questions. Cheers, — madman 20:07, 28 January 2012 (UTC)

Bot boo-boo

Manfred Roeder (judge) was tagged in error. The site that was cited was itself a copy of the German Wikipedia. This may be moot, since I quickly translated it to fix the problem, but thought I'd mention it in case it comes up again. Marrante (talk) 22:56, 28 January 2012 (UTC)

Hmmm. I'd add that site (de.inforapid.org) to the whitelist since the information in the footer is correct, but I think there's still some value in the bot flagging these articles. You knew enough to place the {{translated page}} template but a less experienced user might not, in which case we'd be in violation of our licensing. — madman 18:17, 29 January 2012 (UTC)

Hello, Coren/Archives/2012. Please check your email; you've got mail!
It may take a few minutes from the time the email is sent for it to show up in your inbox. You can at any time by removing the {{You've got mail}} or {{ygm}} template.

CSBot process question

I have a process question.

While checking into a CCI, I decided to run CSBot, which got a hit. However, the site affirmatively noted the reuse of Wikipedia material, so no copyright issues. However, that doesn't definitely say whether the Wikipedia material was first, it simply confirms that one place isn't the source. My process question – if CSBot finds a hit, does it stop looking? Can I infer from the fact that it found one hit, that it didn't find any others, or should I assume that it returned the "best" hit, and there's still a chance that there is another site which is the ultimate source of both?

I think you can answer in the abstract, but if the specifics matter, I summarized in this post to MRG.--SPhilbrick(Talk) 20:27, 30 January 2012 (UTC)

I don't know that Coren's been active recently, but looking at the source code, it would appear that the bot gets the first three results from Yahoo!, excludes ones that meet certain criteria (in userspace on-wiki, in exclude list, etc.) then flags the first one above the minimum score. So, I would assume that it's the best hit that is returned. — madman 21:09, 30 January 2012 (UTC)

Thanks for the answer. I thought something like that might be the case, which is why I went on to look further, but I was interested to know for future use.--SPhilbrick(Talk) 21:26, 30 January 2012 (UTC)