Jump to content

User talk:Shirik/Archives/2010/April

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


Stop killing me!

Why are you trying to kill my page?
⎠o　～°　⎛
　　　　　　 ⎬
⎞o。　　　⎝ —Preceding unsigned comment added by Being.someone.human (talk • contribs) 07:17, 1 April 2010 (UTC)

Your initial page was obviously misinformation. On Wikipedia, these types of pages are speedily deleted. Your current page has some useful information about the character, I will give you that, however it still doesn't meet notability guidelines. All articles must be about a subject that is notable, that is, worthy of inclusion in an encyclopedia. Unless you can assert this, this page will be deleted. --(ʞןɐʇ) ʞıɹıɥs 11:34, 1 April 2010 (UTC)

New good/bad revisions data set

I've generated a data set (for the first 2000 revisions of the Rocket article) identifying good/bad revisions. See: User:Dc987/Dataset_for_the_Rocket_article. Most of it is auto-generated, but some questionable edits were human verified. I guess it is pretty accurate. I'd like to see it cleaned up even more. Would you be willing to run your favorite vandalism identification bot against it?--Dc987 (talk) 00:20, 2 April 2010 (UTC)

I'm sure we can rig something together. How did you evaluate this dataset? If it's from a bot, I should point out User:Tim1357 and I are currently testing out the initial prototype of the CollabRC collaboration cloud and are very pleased with the results right now. I'd be more than happy to see more bots participating in the cloud if you're willing to write a small IRC wrapper (see User:Shirik/CollabRC/Commands for a brief specification, or feel free to ask for more details). We're really excited at how well this is going. I'll get back to you when I find a way to analyze this. --(ʞןɐʇ) ʞıɹıɥs 00:51, 2 April 2010 (UTC)
Er... Guess I'm just making an obvious statement here, but the only way of analyzing this data set automatically (that I can see) would be running it against your (or some other) bot. And manually evaluating false positives/negatives. Some of them are going to be your bot false positives, some - mistakes in the data set. Knowing them could help improving both your bot and the data set. --Dc987 (talk) 02:20, 2 April 2010 (UTC)
Sorry, I meant to ask how you collected the data set in the first place. And yeah, we'll run it against one of the bots in the collaboration cloud as well, I just need to figure out the easiest way to feed it in. --(ʞןɐʇ) ʞıɹıɥs 02:22, 2 April 2010 (UTC)
In a few words it's based on the user reputation (based on the reverts analysis), plus extra verification by the CRM114 engine plus human verification. The current source code (that creates datasets like this) is available here: User_talk:Dc987test, a brief description of the approach is also available on the Wikipedia:Bot_requests page. Not sure how to feed it into your bot, the easiest way to read/write the dataset is to execute/pprint it in python. I guess writing some hack that feeds it into the bot should be possible. --Dc987 (talk) 06:11, 2 April 2010 (UTC)

Filter 306

When you get a chance, would you be able to investigate this false positive report? I can't figure out what's causing it to behave that way, whether it's intentional, or how it could be fixed. Thank you. —Soap— 10:22, 2 April 2010 (UTC)

 Done as well as bugzilla:23027 as discussed privately. --Shirik (Questions or Comments?) 23:41, 2 April 2010 (UTC)

Filter 287

It seems to be working pretty effectively, and no false positives yet -- and honestly, I think false positives are pretty unlikely. (What user within the filter parameter would legitimately add the term in question to an article that didn't already mention it?) I was about to add a ! (removed lines) condition, but X! just disabled the sucker. I'm talking w/X! now. NawlinWiki (talk) 20:38, 2 April 2010 (UTC)

Sorry, you caught me just as my flight was landing so I couldn't respond. Anyway, the edit filter in question effectively denied all new users from writing any content about Harland Sanders (including the page itself). This would be overkill. Of course, adding the "removed lines" condition would alleviate this problem, but there are still cases that this would not cover (such as a new paragraph in the Col. Sanders article). Yes, newer users contributing to these articles would be rare, but not unheard of. It will happen. In any case, such a dramatic change to a filter is, in my opinion, unacceptable to leave at disallow without testing for a little time. I'll admit it was effective in stopping those edits, but it also would have been effective at disallowing effective edits, which we don't want to do. Honestly, at this point, I think we're better off using a log-only filter. He's not editing fast enough to rule out rolling back upon detection, and trying to encode a natural language processor is a bad idea. So you were right to simplify it, I just have trouble accepting that condition as a disallow filter as there will be legitimate edits that hit it. Hope that helps point out my concerns, and sorry for the longwinded response, Shirik (Questions or Comments?) 23:40, 2 April 2010 (UTC)

The Mourning Sickness

I am very disappointed that you chose to delete my page, "The Mourning Sickness." I understand the reasoning behind your decision, but yet I still find these policies to be frighteningly hypocritical. It seems any corporate mass-produced product will have a page (representing little more than free marketing), yet when a real band, with real members - that has notably been in existence for over 15 years - is noted in a page, you pull the plug. Seriously, "Filet-O-Fish" and "Hamburger Helper" have more validity than "The Mourning Sickness?" I can see Diderot turning over in his grave. —Preceding unsigned comment added by Rattmahem (talk • contribs) 21:45, 6 April 2010 (UTC)

Please note that I am not an administrator, and thus do not have the power to delete pages. I merely nominated it for deletion, and consensus was reached that the page was not fit for inclusion. Anyway, you state "that has notably been in existence". The article was deleted because this was not shown. Wikipedia relies heavily on the concept of notability. We have guidelines for which bands are considered notable. If you can show that this band meets a condition there, then I'm sure we can get the page restored. However, I was unable to find any such evidence. Do you have any third-party reliable sources that suggest otherwise? Thanks, Shirik (Questions or Comments?) 21:51, 6 April 2010 (UTC)

My edits

hello, i appologise for the page. but for real, the girl i wrote about ruined my life. i thought it was only fair.

regards. —Preceding unsigned comment added by Lexa123 (talk • contribs) 16:44, 7 April 2010 (UTC)

Hi. I'm sorry you feel that way, but Wikipedia is not a place for attack pages, regardless of the reason. We have a very strict policy regarding this. Please feel free to review our welcome page regarding how you might be able to contribute constructively. Thanks, Shirik (Questions or Comments?) 17:09, 7 April 2010 (UTC)

Wikipedia:School and university projects/Dr. Aaij's Freshman comp

Thank you for your assistance! Shirik, you've been a great help. Dr Aaij (talk) 01:40, 9 April 2010 (UTC)

Not a problem. I hope everything went smoothly for you and hope you choose to do something like this again! Take care, Shirik (Questions or Comments?) 07:35, 9 April 2010 (UTC)

Howelseornotso + Loosmark

Hi Shirik, I left you a note at the SPI page Wikipedia:Sockpuppet investigations/Howelseornotso, regarding your statement about a "right to a clean start". Cheers, --Fut.Perf. ☼ 17:19, 9 April 2010 (UTC)

Hi Fut.Perf., thanks for the note. I missed the comment about the arbcom sanction. You are right in that this would be inappropriate and thus I have endorsed the original request as well and moved the case back to its original page. In the future if you could tag it with check code "A" as well (for arbcom ban/sanction evasion) that would be great. I have added it to this case. Thanks, --Shirik (Questions or Comments?) 17:44, 9 April 2010 (UTC)

MacBook Pro GA review

Hello Shrik, thanks for reviewing! I'll try to work on your concerns for next time. Regards, Airplaneman ✈ 21:27, 1 April 2010 (UTC)

Feel free to correct them and renominate at any time. I'm going to be doing a ton of GA reviews this month and will be more than happy to re-review it, even if you renominate in days. --(ʞןɐʇ) ʞıɹıɥs 00:47, 2 April 2010 (UTC)
Cool, thanks :) Airplaneman ✈ 00:50, 2 April 2010 (UTC)
OK, I'm getting to work on this. Is it fine if I just collapse the wikitables (I'll try trimming them as well) and keep their current width? I agree they are quite cumbersome. Airplaneman ✈ 19:37, 11 April 2010 (UTC)

I have renominated. Please drop by and take a look if you wish. Thanks, Airplaneman ✈ 20:09, 18 April 2010 (UTC)

Sorry for the delayed/lack of response. The last few weeks have been quite packed for me and your first reply I didn't even see. I will be sure to drop by. --Shirik (Questions or Comments?) 16:13, 19 April 2010 (UTC)

need your advise

Hello Shirik!

I needed your help on understanding a comment received for my article . My article has been contested for deletion becuase one of the refrences seemed to be violating the copyright policies. I did not really understand this allegation. The refrence was readily and easily available on the internet, aka public domain. The Refrence white paper also does not mention any copyright permissions that should be taken. Also, why has it been nominated for "Strong Delete" I am here to stay and write. I have really worked hard on this article. I am ready to go the extra mile to save the article. Please advise where I am going wrong and what am I missing. Also, how can I retreive my original article?I wish to study the article again. Please help PCJain (talk) 11:43, 13 April 2010 (UTC)

Please note that the internet is not the public domain, and just because something doesn't have a copyright explicitly stated does not mean it is not copyrighted, so it is generally inappropriate to copy content directly. You should write the content yourself, using the reference as only a basis for the information. However, if you don't think it should be deleted, you should add a comment at Wikipedia:Articles for deletion/Mobility-as-a-Service where the discussion is ongoing. The reason is explained there by the nominator. Please note that this is a discussion so just because one person says it should be deleted does not mean it will be, however after 7 days the discussion will be evaluated and consensus will be determined over whether or not it should be retained. You should post your arguments for keeping the content there. --Shirik (Questions or Comments?) 15:28, 13 April 2010 (UTC)

Thank you for the input. How do I see my article? I wish to review it to study it further..122.172.163.201 (talk) 15:53, 13 April 2010 (UTC)

If you go to the article's history you will see all of the old versions. You can click any of the dates there to view that version. --Shirik (Questions or Comments?) 15:55, 13 April 2010 (UTC)

False Positive Help

Thank you for fixing the filter on the Brendan Loy page. Every couple of months the same guy (I believe his name is Swarthington) comes back and edits the page with several dead links and unverified information. —Preceding unsigned comment added by Snickerdoodle22 (talk • contribs) 19:07, 14 April 2010 (UTC)

Thanks for the info. I will be sure to keep an eye on it. And no problem with the filter. --Shirik (Questions or Comments?) 16:14, 19 April 2010 (UTC)

Hello Shirik, may I copy your navigation template User:Shirik/Navigation for my userspace? Airplaneman ✈ 21:14, 18 April 2010 (UTC)

Sure, though I'm not sure I can call it "mine". If you check the history, you'll note it's based off User:Ronhjones/Navigation. But feel free! --Shirik (Questions or Comments?) 16:11, 19 April 2010 (UTC)
OK, thanks! Airplaneman ✈ 01:50, 20 April 2010 (UTC)


SPI

I feel this case has raised compelling questions and warrants further attention. The recent hasty closure seems premature to me. ܥܝܪܐܩ (talk) 17:06, 20 April 2010 (UTC)

You will want to take this up with Amory, the patrolling administrator (which also happened to be a clerk, so it was archived immediately after the decision). My level of involvement was merely to get the case endorsed for further information. Amory has come to the conclusion that, given the information received from checkuser, there isn't enough to go on for a block, and I respect that decision. If you feel otherwise, perhaps inquire at User talk:Amorymeltzer. Thanks, Shirik (Questions or Comments?) 17:09, 20 April 2010 (UTC)
Will do then. I just wanted to seek your opinion since I felt your initial clerk review was objective. Thanks for the quick reply. ܥܝܪܐܩ (talk) 17:22, 20 April 2010 (UTC)
If you're looking for my opinion on the matter, I feel Amory is probably correct. As I pointed out in my endorsement, the evidence I uncovered that persuaded me to endorse the case was merely of note; enough to warrant further checking. This is much like a search warrant where the evidence isn't enough to indicate that someone is guilty, but it is peculiar. Note that common hours of editing could be for a variety of reasons, including being in the same country, as they are. Given the information uncovered at checkuser we've uncovered that one is likely innocent and the other is only on the same ISP (not uncommon for a location like Sweeden). Along with Amory's diff in the closing note, I feel the correct decision was probably made. But, of course, if you have any reason to feel otherwise, please feel free to bring it up. --Shirik (Questions or Comments?) 17:36, 20 April 2010 (UTC)

Question about Sockpuppetry case

Hello, since I am still new here, I do not know all the rules. Can I delete this case Wikipedia:Sockpuppet investigations/Shmayo from my talk page (I was "acquitted", because I come from a completely different country). —Preceding unsigned comment added by Destudent (talk • contribs) 21:53, 21 April 2010 (UTC)

Of course! Your talk page is yours to use and manage in whatever way you feel appropriate (within reason). In general, you are free to remove whatever comments you want (though archiving is preferred). I apologize that you got dragged into this case so early; please note that my notification to you was merely to let you know of an ongoing case as part of my duties as a clerk, not an expression of suspicion. If you have any questions regarding how to proceed in the future, please feel free to drop me a line. Of course, you can always stick {{helpme}} to attract attention from more than just me, as well. Happy editing! --Shirik (Questions or Comments?) 03:34, 22 April 2010 (UTC)

Copyedit Backlog Elimination Drive

Hi, as a member of the Guild of Copy Editors you're hereby notified of and invited to participate in the WikiProject Guild of Copy Editors/Backlog elimination drives/May 2010. Please help us eliminate the 8,000+ copyedit backlog! Participating editors will receive barnstars and other awards, according to their level of participation. ɳorɑfʈ Talk! 00:20, 22 April 2010 (UTC)

Thanks! Unfortunately I might be out of town for much of May, but I will try to help out where I can! --Shirik (Questions or Comments?) 03:28, 22 April 2010 (UTC)

GAN backlog elimination drive - 1 week to go

First off, on behalf of myself and my co-coordinator Wizardman, I would like to thank you for the efforts that you have made so far in this GAN backlog elimination drive. It has been nothing short of a success, and that is thanks to you. See this Signpost article about what this drive has achieved so far.

We're currently heading into the final week of the drive. At this time, if you have any GANs on review or on hold, you should be finishing off those reviews. Right now, we have more GANs on review or on hold than we do unreviewed. If you're going to start a GA review, please do so now so you can complete it by the end of the month and so that the nominator has a full 7-day window to address any concerns.

See you at the finish!

–MuZemike delivered by MuZebot 16:24, 22 April 2010 (UTC)

Filter 81

NawlinWiki recommended that I talk to you about filter 81's performance. I depend on filter 81, but it has a reputation as being a horribly expensive filter, so people keep disabling it. Unfortunately, I can't figure out why it is so expensive: it should normally just check to see if the article includes the word "chart" and terminate. While the big test is relatively expensive, it should only execute rarely. Is there any way to determine what path the filter normally takes? Is there a better way to optimize the normal case?—Kww(talk) 04:41, 24 April 2010 (UTC)

At first glance, it looks like the repetitive calls to execute a regex test on lcase(added_lines). These should be able to be condensed into a single condition. I think that will go a long way to helping it. Let me get back to you in a few minutes. --Shirik (Questions or Comments?) 04:44, 24 April 2010 (UTC)
Ok, I made a significant change that should help a lot. I won't be able to tell its performance for about 15 minutes, but I'm feeling fairly confident. That was my 500th filter edit, so there must be some sort of luck embedded in it! --Shirik (Questions or Comments?) 04:57, 24 April 2010 (UTC)
Oh, by the way, about the short circuiting of the long check: I can't be sure, but I've long been of the opinion that short circuiting of subexpressions doesn't work quite the way we want it to, that is, it actually evaluates that long string of "ored" conditions first. My recreation of the filter intentionally avoided that possibility. --Shirik (Questions or Comments?) 05:01, 24 April 2010 (UTC)
Ouch. If it was taking the longest path all the time, no wonder people were unhappy with it.—Kww(talk) 05:06, 24 April 2010 (UTC)

USAD

Hey there Shirik. Hope you're doing well. I'm not going to be around until much later today, and one of the articles I have contributed to is currently on the main page. Do you think you (and any talk page stalkers of yours / people on IRC that you can find) could keep an eye on the article? Thanks, NW (Talk) 01:31, 26 April 2010 (UTC)

Will do! Regards, Shirik (Questions or Comments?) 01:32, 26 April 2010 (UTC)

Special:AbuseFilter

The filter that User:NawlinWiki set up to protect me from very real attacks and threats appears to have been first disabled by you, then deleted. This is typical, and has now been going on for over THREE MONTHS, and before I go to the police, would prefer some sort of internal solution. Any ideas, hmmm? Rodhullandemu 22:28, 23 April 2010 (UTC)

As I mentioned in that filter's notes, the filter was doing little to no help beyond that of semiprotection. Instead, I merged that filter into a more powerful filter in a different section because of performance problems (179). I didn't just delete it without any concern to the problem it was addressing. That filter was simply illogical, it makes more sense to merge it into 179.
I have also just expanded it due to the most recent activity. If 179 can't keep up, I would advise semiprotection; as much as I don't like the problems you're going through, it simply doesn't make any sense to use a filter that essentially does the same thing as semiprotection. --Shirik (Questions or Comments?) 22:33, 23 April 2010 (UTC)
You'll notice that I have had to set up a second, unprotected, Talk page, but that, too, has been under attack. It is intolerable for an Admin to have to even semi-protect both pages, because unconfirmed users have just as much right to contact me as registered users. So be it. I have a few things to do, then I will be leaving the project if it will not protect me. I have my sanity and general health to think of as well as this place. Rodhullandemu 22:39, 23 April 2010 (UTC)
I've also added your pages to filter 294. Between 179 and 294, we should be able to shut off attacks against your pages pretty quickly. NawlinWiki (talk) 22:39, 23 April 2010 (UTC)
I understand your frustration, but try to understand, the filter I disabled was disabled because it was semiprotection. Instead, I added capabilities into 179 (and NW into 294) to do more powerful protection instead. 319 had to be deleted because it was causing edits to pass the edit filter without being checked (due to the condition limit) and it had zero purpose; semiprotection would have the same effect. The additions to 179 and 294 aim to provide a better level of protection while still allowing non-autoconfirmed users to edit, which seems to be your goal anyway. The old 319 filter would not have met that goal. --Shirik (Questions or Comments?) 22:42, 23 April 2010 (UTC)
I should also point out that 179 and 294 both auto-report to WP:AIV, whereas 319 could not. --Shirik (Questions or Comments?) 22:44, 23 April 2010 (UTC)
None of which seem to work. Rodhullandemu 14:17, 24 April 2010 (UTC)
I'm honestly doing everything I can to help you, but there is so much I can do (and I'm only human, despite what some people like to think). You're asking me to filter out some very erratic edits while not causing something like this which just occurred. That is an extremely fine line to walk and there are going to be mistakes along the way. I'm certainly open to suggestions, but try to understand it's going to take time to get this perfect (or it may simply not be possible). --Shirik (Questions or Comments?) 18:06, 24 April 2010 (UTC)
Thanks. There are certain tropes in these posts which exhibit commonality, and it occurs to me that a suitable minimal subset of keywords might be more successful as filter terms. Otherwise, it just isn't possible for non-auto confirmed editors to contact me, which (as above) creates problems. Rodhullandemu 18:22, 24 April 2010 (UTC)

Filter 295

This filter is directed at a group of vandals who share accounts (Users JarlaxleArtemis, Johnny the Vandal, and ByAppointmentTo) and who created a whole bunch of sleeper socks in March 2009. When this filter is disabled, they start doing multiple vandal page moves (see the log for 4/18, after the filter was restored). I've reactivated the filter but added a throttle, which should clear up the false positives problem. Thanks, NawlinWiki (talk) 18:02, 30 April 2010 (UTC)

Ok; the throttle sounds like a good idea because pretty much every false positive in there has been a one-shot miss. I didn't think of that just because I didn't know what it was directed at. Thanks, Shirik (Questions or Comments?) 18:03, 30 April 2010 (UTC)
Retrieved from "https://wiki.riteme.site/w/index.php?title=User_talk:Shirik/Archives/2010/April&oldid=1141023801"