User talk:Harej/Archive10

Hi Harej, your bot is removing manually added RfCs. [1] I'm going to restore it, but past experience indicates that it will keep on removing it. Could you fix it, please? SlimVirgin ^{talk|contribs} 02:53, 4 August 2009 (UTC)[reply]

Hi Harej, your bot is removing manually added RfCs. A B. I'm going to restore it, but past experience indicates that it will keep on removing it. Could you fix it, please? GabrielVelasquez (talk) 12:23, 15 August 2009 (UTC)[reply]

Do you see the "add a discussion" link? That is where manually-added RFCs go. —harej (talk) 17:48, 15 August 2009 (UTC)[reply]

No I don't see any "add a discussion" link, but I will guess you mean "new section" tab. but I don't see why that is necessary, your bot shouldn't be interfering that way. GabrielVelasquez (talk) 21:35, 18 August 2009 (UTC)[reply]

This is the instructions on the RFC page that your bot is contrdicting:

Add the tag {{rfctag|sci}}. See Template:Wider attention for a list of abbreviations. To add one discussion into two different categories, use {{rfctag|xxx|yyy}}, where "xxx" is the first category and "yyy" is the second category. To add an entry manually, click the "add a discussion" link.
On the top of the list, note the name of the page and a short summary of the discussion.
For more information, see Wikipedia:Requests for comment.
Report problems to Wikipedia talk:Requests for comment.

GabrielVelasquez (talk) 21:47, 18 August 2009 (UTC)[reply]

You currently appear to be engaged in an edit war. Note that the three-revert rule prohibits making more than three reversions on a single page within a 24-hour period. Additionally, users who perform several reversions in content disputes may be blocked for edit warring even if they do not technically violate the three-revert rule. When in dispute with another editor you should first try to discuss controversial changes to work towards wording and content that gains a consensus among editors. Should that prove unsuccessful, you are encouraged to seek dispute resolution, and in some cases it may be appropriate to request page protection. Please stop the disruption, otherwise you may be blocked from editing. GabrielVelasquez (talk) 21:51, 18 August 2009 (UTC)[reply]

The bot doesn't have a conscience, it's not capable of being belligerent, and by definition, incapable of edit warring. It only does what its instructions tell it to do. And its instructions tell it to removing RFC tags that have been around for one month after the timestamped date, because RFCs have to end sometime. I have added a statement relisting it, giving new life to the RFC. Problem solved. You could have looked at the edit summary where the bot noted "Removing expired rfctag" and asked me calmly to see what the big deal was, but instead you opted to engage in an edit war against an automated process, which is foolish. @harej 22:12, 18 August 2009 (UTC)[reply]

It may appear fooling is you don't see the point. I place a new RFC in a lower section and your Bot still interfered. So I made a new copy of everything in the discussion and placed a new RFC then deleted the old discussion and your bot still interfered. And all this is recorded and the point made. You bot is interfering where it ought not be. I intend to follow up on this, and the diffs are evidence that his bot has too large a scope. GabrielVelasquez (talk) 18:09, 20 August 2009 (UTC)[reply]

An expired RFC is defined by thirty days past the original timestamp. The original timestamp is 9 July 2009. Thirty days after the fact is 9 August 2009. It is well after that point. That is why it is expired. You can get past that by re-listing it, by providing a new "initial" timestamp. @harej 14:39, 21 August 2009 (UTC)[reply]

Hi there. Thanks for deleting this userspace page. But I'm confused -- I don't think I actually got around to inserting {{db-userreq}} onto the page. How did you know to delete the page?

My watchlist is pretty packed, so if you respond, could you do so on my talkpage. Thanks. Andrew Gradman ^talk/_WP:Hornbook 21:27, 15 August 2009 (UTC)[reply]

I came across it somehow. —harej (talk) 00:30, 16 August 2009 (UTC)[reply]

_creepy.

I'm not questioning your actions; I'm just surprised that someone "came across" such an obscure talkspace page. It makes me worried about my privacy. If you can give me a better idea of how you came across this page, that would be comforting. I am a relatively new editor and there are a lot of Wikipedia features that I'm still ignorant of. Thanks. Andrew Gradman ^talk/_WP:Hornbook 05:58, 17 August 2009 (UTC)[reply]

• OK, no problem. Thanks for the fast followup. Andrew Gradman ^talk/_WP:Hornbook 06:03, 17 August 2009 (UTC)[reply]

• edit

As an administrator who has been known for writing on occasion, for some several years now, and who is far from alone in this regard, I call "Phooey!". See what I wrote at the bottom of that very RFA discussion, indeed. We're still editors. Uncle G (talk) 10:53, 16 August 2009 (UTC)[reply]

Oh, Wikipedia uncle who isn't Ed. I actually wrote more articles as an admin than not an admin, but all my desire to work on articles is gone. And there is no doubting that admins have a lot of side activity. —harej (talk) 16:28, 16 August 2009 (UTC)[reply]

Would it be appropriate to relist the proposed move? I feel the outcome was "no consensus"-ish. --Cybercobra (talk) 07:00, 17 August 2009 (UTC)[reply]

Thanks for your participation in my recent RfA. I will do my very best not to betray the confidence you have shown me. If you ever have any questions or suggestions about my conduct as an administrator or as an editor please don't hesitate to contact me. Once again, thanks. ·Maunus·ƛ· 13:17, 19 August 2009 (UTC)[reply]

Stop deleting band profiles. —Preceding unsigned comment added by 65.169.195.238 (talk) 17:56, 19 August 2009 (UTC)[reply]

I'm afraid I don't know what you're talking about. @harej 19:03, 19 August 2009 (UTC)[reply]

Re your edit summary: if that's the case, why in the world doesn't it say so in the instructions at WP:RFC? Rivertorch (talk) 05:02, 20 August 2009 (UTC)[reply]

It states that in Template:RFC list footer but not in the directions on Wikipedia:Requests for comment. This has since been corrected. @harej 05:06, 20 August 2009 (UTC)[reply]

Thanks for fixing that. I've added one more category at Talk:Human. Rivertorch (talk) 07:22, 20 August 2009 (UTC)[reply]

Hi Harej, of the four editors who commented on the proposed move in Talk:Olimpik-Shuvalan#Move.3F, two were in favor of moving Olimpik-Shuvalan to Olimpik-Şüvəlan, and another two neither opposed nor supported the move. Why did you close the move request and declare no consensus to have been reached? Thanks, Atemperman (talk) 18:38, 20 August 2009 (UTC)[reply]

I suppose I interpreted it as two support and two non-support, but if you (or someone else, if you're an involved party) disagree with that, you can re-open it or close it differently. @harej 14:27, 21 August 2009 (UTC)[reply]

I tried :\ --NE2 03:04, 23 August 2009 (UTC)[reply]

I just don't see why this is so damn controversial. @harej 03:26, 23 August 2009 (UTC)[reply]

I think what makes it hard are that there are so many quotes on pages that would need to be dealt with (userbox wars II, I think) if precedent was set that all quotes are verboten... and people don't want to deal with the consequences of setting a precedent one way or the other. Good close by the way, as I opined at the DRV. ++Lar: t/c

You call it a good close because it's in line with your personal opinions, just as you called the original close bad. This really shouldn't result in a purge of all quotes, because no one has ever considered one-sentence quotes to be worth taking to deletion. What caused the ruckus for this is how it is such an extensive quote. @harej 17:48, 23 August 2009 (UTC)[reply]

I hope not... merely because it's a different outcome! Ack. Why would you think that? The original close wasn't bad, and I praised it: [2]... what I criticised was how you responded to comments about it, which you admitted you could have handled better. ("I didn't want to be flippant, I just ended up that way"). Both closes were good in my view. They both matched the apparent consensus of the issue at the time. See the DrV for more I guess. I really wish Erik9 would invest as much effort into getting policy adjusted as he does in argument. ++Lar: t/c 19:56, 23 August 2009 (UTC)[reply]

I would "think that" because I'm cynical like that. :P Thank you, in any case.@harej 19:58, 23 August 2009 (UTC)[reply]

As I'm sure you expected, I have listed your closure of Wikipedia:Miscellany for deletion/User:Jack Merridew/Blood and Roses at Wikipedia:Deletion review/Log/2009 August 23. Erik9 (talk) 03:12, 23 August 2009 (UTC)[reply]

I posted on the deletion review, generally in concurrence. @harej 03:26, 23 August 2009 (UTC)[reply]

Why did you remove a link to it? I posted what I found out on the RFC talk page. Can you explain there why it was removed, I really like it a lot. Ikip (talk) 08:58, 24 August 2009 (UTC) thanks for letting me know that the bot is back :). Ikip (talk) 13:52, 26 August 2009 (UTC)[reply]

I thought you said you were going to get it going? And now I see you marked it as dead. The mouthbreathers sucked away your motivation too? → ROUX ₪ 17:45, 25 August 2009 (UTC)[reply]

I can't do it without my team. @harej 17:46, 25 August 2009 (UTC)[reply]

Now vandalising his user page. HalfShadow 05:40, 28 August 2009 (UTC)[reply]

Duly noted. Talk page protected appropriately. @harej 05:44, 28 August 2009 (UTC)[reply]

I know you are not a big fan of Userboxes, but I thought you would like this one: User:Gu1dry/Userbox/WikiProject Hat Fetishism

『 ɠu¹ɖяy 』 ^{¤ • ¢} 18:38, 29 August 2009 (UTC)[reply]

Hah, thank you. @harej 18:41, 29 August 2009 (UTC)[reply]

I've pretty much finished the date replacement logic and tested that part of the code based on the test cases defined in User:Full-date unlinking bot/Test environment. One thing I was considering was adding counts to the edit summary, so that instead of "...BRreg, AMreg..." the summary could read "...BRreg(×15), AMreg(×2)...", What do you think?

Looking at some of the other bot code, I have a few comments:

1. I'm new to PHP programming, but from what I read here, it looks like the global statement needs to be present in each function that references a global variable, not just the one that defines it. Can you take a look? For example, in the unlinker function, does $contents need to be defined as global (or passed in as a parameter)?
2. Do you know if the $objwiki->whatlinkshere call also returns redirected references? For example, if called for "January 1", does it include references to "January 01", "Jan 1", "Jan 01", "1 January", etc? That will be needed to cover all of the pages.
3. Also, looking at the code for the whatlinkshere function (here), it appears that the results might be limited to 500 pages. I think many of the month-day pages may have 10,000+ links. I suggest checking with Chris G to determine how to best handle this.
4. In the checktoprocess function, the logic that checks for a previously processed pages might get false results. For example, if pages 12345 and 12346 have already been processed, page 345 would appear already processed, since the string "345," is contained within the string "12345,12346,"). I'd suggest changing strpos($check, $pageid . ",") to strpos($check, "," . $pageid . ","). You would also need to seed the pageid tracking file with a leading comma.
5. With an estimated 650,000+ pages to process, and assuming a get, update, and put cycle time of something on the order of one second, I would expect that you would likely get some edit conflicts, maybe even dozens or hundreds. It might be worth adding retry logic (get again, reprocess, and put again) or at least modify the log message to record the pageid and/or page name. It also might make sense to only record the page in the tracking file after a successful edit.
6. It might be more efficient to compare the updated content to the original and only apply the edit if there is an actual change. If there is no change, no edit need be applied (less work for the Wikipedia servers), but the page is still recorded as having been processed.
7. With a maximum throttle of one edit every 10 seconds (or 5-6 edits per minute), the bot will need something on the order of 75 to 100 days of uninterrupted operation to complete. It might be worth adding logic to the main routine to to allow a restart at some point other than the January 1 what-links-here list and/or possibly limiting the number of dates processed per run, for greater control.
8. I recommend moving the get-contents call to a point after the "already processed" test, to reduce server load.
9. I'm not sure I understand the manual override logic. Will that eventually be linked to a big red STOP button than anyone can access to stop the bot?
10. Before submitting it for BAG approval or attempting any trial runs, I think it would be wise to test as much as the code as possible in a read-only mode. This could be done by: (a) temporarily disabling the apply-edit logic (possibly replacing it with logic that writes changes to a local file), and/or (b) temporarily replacing the what-links-here logic with code that retrieves a predefined list of pages. That list could include known "interesting" test cases such as the one we put together. Running the test multiple times would verify proper operation of the page tracking logic.

-- Tcncv (talk) 20:01, 30 August 2009 (UTC)[reply]

To state the multiplier for each one sounds like a fantastic idea. As for your enumerated statements,

1. I'm not very good with PHP and variable scope, so someone more competent in PHP will have to take a look.
2. Based on experience, I can tell you that whatlinkshere() also returns what links to the redirects.
3. Damn, I didn't know that.
4. I added a comma to the text file that will be used.
5. Chris G added some logic to check for an edit conflict, but I don't really know if it does anything other than detect that an edit conflict happened. Furthermore, said logic only works on edit conflicts and not other API errors.
6. An edit wouldn't actually show up unless the page was actually changed, but this should not be too difficult to implement. After all, it is less work for the API to do.
7. The bot technically does not need to operate uninterrupted. Since WhatLinksHere gets updated as links are removed, then the bot would eventually pick up where to start again. This would not take too much time compared to the long scale of the bot operation. The advantage of not having a reboot point is that if, say, January 15, 1996 is added after the bot processes January in its initial run, then the bot will be able to remove it.
8. What should actually be done is moving the get-pageid operation before the check-if-already-processed operation, since that relies on the page id. Then, after that, the get-page-contents operation can be done.
9. User:Full-date unlinking bot/manual override is the manual override page. If the string Joe Biden appears on that page, the bot will stop running. It should be more prominently linked to (perhaps even in the edit summary), but I've locked the page for admins only considering the ramifications of letting anyone at any time stop the bot.
10. The part of the code that needs the most testing is the unlinker() function, and such testing has been going on. The other things really should be tested in a live environment, in my opinion, so we can see exactly how it runs when we're ready. We could limit this to 20 pages or so, and rollback whenever something goes wrong.

@harej 20:37, 30 August 2009 (UTC)[reply]

I've added match counts to the edit summary as shown in this test. -- Tcncv (talk) 02:56, 31 August 2009 (UTC)[reply]

Looks great! @harej 02:58, 31 August 2009 (UTC)[reply]

Not sure what happened here. Maybe the multi-move wasn't in quite the right format. Kanguole 09:59, 31 August 2009 (UTC)[reply]

There was nothing wrong, and I am not sure why the bot behaved in such a way. @harej 10:00, 31 August 2009 (UTC)[reply]

Ditto here? [3] [4] - 92.3.229.79 (talk) 13:10, 31 August 2009 (UTC)[reply]

Just a note of caution. The bot code in User:Full-date unlinking bot/code contains some unicode characters - namely the "§~§" placeholder used in several places plus several characters in the $part_list_word = ... expression. While testing this on my windows PHP implementation (which does not appear to directly support unicode), I had to encode both the script and the page source as UTF-8, which encodes non-ASCII characters as multi-byte sequences. I'm not sure what support there is for unicode on the toolserver's PHP implementation, but I would be caution when copying the script and setting up the bot. If the bot processes dates containing en-dashes, em-dashes, and the "×" (multiply) symbol, then all is well. -- Tcncv (talk) 16:23, 31 August 2009 (UTC)[reply]

As far as I know, the Toolserver is highly capable of UTF-8 encoding. @harej 17:52, 31 August 2009 (UTC)[reply]

New York City Meetup Next: Sunday September 13th, Columbia University area Last: 07/25/2009 This box: view • talk • edit

In the afternoon, we will hold a session dedicated to meta:Wikimedia New York City activities, review the recent Wiki-Conference New York, plan for the next stages of projects like Wikipedia Takes Manhattan and Wikipedia at the Library, and hold salon-style group discussions on Wikipedia and the other Wikimedia projects (see the May meeting's minutes).

In the evening, we'll share dinner and chat at a local restaurant, and generally enjoy ourselves and kick back.

You can add or remove your name from the New York City Meetups invite list at Wikipedia:Meetup/NYC/Invite list.

To keep up-to-date on local events, you can also join our mailing list.
This has been an automated delivery by BrownBot (talk) 02:30, 1 September 2009 (UTC)[reply]

Thank you very much for the nomination, harej! I will definitely accept. Your trust means a lot to me, especially considering that you don't seem to nominate people all that often. I'll take a minute to answer the questions, and I'll let you know when it's transcluded. I hope I won't let you down :) Jafeluv (talk) 09:59, 1 September 2009 (UTC)[reply]

Transcluded now. Hope I did everything right... Jafeluv (talk) 15:59, 1 September 2009 (UTC)[reply]

I'm still around, but no longer really do much. But I heard that a West Australian contributor died (horribly) and I had to add my condolences :( Tbsdy lives (formerly Ta bu shi da yu) ^talk 12:45, 2 September 2009 (UTC)[reply]

I see that User:MessedRobot has been inserting "discuss=" into the relevant parameter of {{mergefrom}} and {{mergeto}}. I do not say that the bot is wrong; just that I had not realised that I should have used the |discuss= format, because it's not mentioned at Help:Merging, except for {{Mergefrom-multiple}}. I was going by Help:Merging, rather than Template:Mergefrom/doc, because the latter deals primarily with {{merge}}, mentioning {{mergefrom}} and {{mergeto}} merely as other forms, and not at all in the Syntax section. Unless the documentation shows the correct practice, this bot will have a never-ending struggle. ;) Whose attention should be drawn? --Redrose64 (talk) 14:47, 2 September 2009 (UTC)[reply]

I don't blame you! What you did was current practice at that time. The issue is that I am merging {{mergeto}} with {{mergeto-multiple}}; and {{mergefrom}} with {{mergefrom-multiple}}. In order for there to be a seamless merge, I have to update all the template uses to make sure that they all use the same syntax. In the meantime, I am working to update the documentation. @harej 18:34, 2 September 2009 (UTC)[reply]

I noticed you deleted 2 merge templates, and redirected another 9. Was all of this done after proper discussion? Mind you, I agree. :) Debresser (talk) 16:58, 6 September 2009 (UTC)[reply]

The three I deleted were completely redundant, unused, and no one would miss them. The ones which I redirected were out of a desire to consolidate the system; there is no reason to have that many merge templates. All we really need are {{merge}} (for when the destination isn't known), {{mergeto}} (when the destination is the other page), and {{mergefrom}} (when the destination is that page).

I talked about it on Wikipedia talk:Proposed mergers#Too many templates prior to doing it; bear in mind that I was not trying to seek a grand consensus. I would have, though, if it was more than a mostly-transparent change that is covered by template redirects (the cases that were not covered by redirects were thoroughly managed by my bot). @harej 18:22, 6 September 2009 (UTC)[reply]

I didn't know about that discussion, but, as I said, I agree. Please have a look at Wikipedia:List of monthly maintenance categories with templates, and you'll understand my interest in the matter. Debresser (talk) 20:26, 6 September 2009 (UTC)[reply]

You've been doing some great maintenance work on that page. Luckily, there is less to document now that there are fewer templates. @harej 20:40, 6 September 2009 (UTC)[reply]

The obsolete {{Merge-multiple}}, {{Mergefrom-multiple}}, {{Mergeto-multiple}} and {{Mergetomultiple-with}} templates are all shown at Wikipedia:Merging#List of merge templates - is that an oversight, or a deliberate retention? --Redrose64 (talk) 21:09, 15 September 2009 (UTC)[reply]

It was forgotten, much like Poland. Alas, I have fixed it. @harej 22:21, 15 September 2009 (UTC)[reply]

Specifically this edit. I'm sure you'll agree that the bot isn't meant to do what it's actually gone and done. I've reverted for now as the best way to restore the rest of the content. I also notice that the move request has yet to be closed and that there did appear to be a {{movereq}} template. I'd appreciate you having a look to ensure everything is working correctly. Dpmuk (talk) 12:31, 3 September 2009 (UTC)[reply]

In this edit[5] RFC bot removed 31 kilobytes of talk discussion... did it move it somewhere else? The message given was "Removing unpaired moveheader", but it's a little hard to understand what this relates to, as it doesn't link to a policy or FAQ. Could you let me know what it was in relation to, and possibly consider modifying the bot so that it leaves a linked summary of what the thinking behind the executed rule/action. —Sladen (talk) 12:32, 3 September 2009 (UTC)[reply]

This is the exact same issue I mention in the above section. Dpmuk (talk) 15:25, 3 September 2009 (UTC)[reply]

Someone added a malformed movereq to Category talk:Survivors (aircraft) which caused all of this. I've gone and removed the offending movereq's, and reverted the list at WP:RM to a slightly older version, so the immediate issue is resolved, at least.
— V = I * R (talk to Ω) 16:58, 3 September 2009 (UTC)[reply]

...or not
— V = I * R (talk to Ω) 17:03, 3 September 2009 (UTC)[reply]

Hi Harej. I've blocked User:RFC bot for the above issues; Feel free to unblock when it's fixed. I couldn't find an emergency shutoff switch, so I didn't have another option; my apologies for the inconvenience. Apparently, the bot is also posting blank moveheaders on inappropriate pages (as here). There's some discussion at ANI, as well. Thanks, UltraExactZZ ^Claims ~ Evidence 17:40, 3 September 2009 (UTC)[reply]

• Emeregency shutoff switches are usually just clever links to the block interface or to some fully protected configuration page with a "change this to off to stop the bot" section, btw. Protonk (talk) 17:49, 3 September 2009 (UTC)[reply]

Yes, but I've seen bots that checked simple on/off switches, and wouldn't edit if a certain subpage was changed from "on" to "off", for example - but, yes, mostly it's just a block button. I was trying to go for a more polite version of "I'm in ur pages, blockin ur bot"... ^_^ UltraExactZZ ^Claims ~ Evidence 18:00, 3 September 2009 (UTC)[reply]

In this edit it replaced the contents of Talk:Durham School (Durham, England) with that of Talk:London King's Cross railway station. Kanguole 23:48, 3 September 2009 (UTC)[reply]

That's bizarre. @harej 23:49, 3 September 2009 (UTC)[reply]

It did the same thing at Talk:List of Australian ambassadors to the United Nations. Rrius (talk) 00:06, 4 September 2009 (UTC)[reply]

No it didn't. @harej 00:08, 4 September 2009 (UTC)[reply]

{{move}} says "A bot will list this discussion on Wikipedia:Requested moves within 30 minutes of this tag being placed". Giving a definite timescale is pretty misleading because RFC bot quite often takes a lot longer than 30 minutes — this one took 2½ hours. Adding some sort of qualification in the wording would help, like: "A bot will list this discussion on Wikipedia:Requested moves, typically within 30 minutes of this tag being placed" Dunno. I think it would be a useful clue that things might conceivably take longer than 30 minutes, so there's no need to start thinking, "Hey, what's up with RFC bot, maybe it's skipped the template I added, and I'm gonna have to go and list it manually!?" 85.94.186.91 (talk) 01:24, 4 September 2009 (UTC)[reply]

It usually is done within 30 minutes, though. Maybe not as of late because I've been taking the bot off of crontab to do repairs, but I think qualifying it is a good idea. @harej 01:26, 4 September 2009 (UTC)[reply]

How's this: "Listed in 30 minutes or your move request is free." :D @harej 01:27, 4 September 2009 (UTC)[reply]

Love it ;) 85.94.186.91 (talk) 01:28, 4 September 2009 (UTC)[reply]

...I do need some clarification =) Are you asking me how I would respond to queries about my decisions? –xeno^talk 13:02, 4 September 2009 (UTC)[reply]

That was a literal question, but I suppose you could take it as "how would you deal with your ordained authority being questioned". @harej 13:15, 4 September 2009 (UTC)[reply]

Your bot seems to be having fun edit warring with itself. Just a heads-up for you. --Kurt Shaped Box (talk) 17:34, 4 September 2009 (UTC)[reply]

On Talk:Nanday Parakeet too. Malfunction - or something on the pages that's confusing the bot? --Kurt Shaped Box (talk) 17:38, 4 September 2009 (UTC)[reply]

Italic text

The following pages have been affected by rfc:

• Talk:M108 Howitzer
• Talk:M3 Howitzer

The bot is copying and pasting contents from unrelated articles

username 1 (talk) 17:59, 4 September 2009 (UTC)[reply]

Similar to another complaint filed above, Talk:London King's Cross railway station was pasted to both.

See the ANI thread here. As you're offline at present, I thought it best to block the bot to prevent any further incidents of text being dumped from elsewhere onto unrelated talk pages. --Kurt Shaped Box (talk) 18:49, 4 September 2009 (UTC)[reply]

James - I was passing by and noticed the problem reports and decided to took a look. Looking at the User:RFC bot/requestedmoves.php, it looks like the problem could occur if the $nom loop processes more than one talk page. The suspect statements are:

			while ($talkpage == "") {
				$talkpage = $objwiki->getpage($talkname);
			}

For the first talk page processed, $talkpage is undefined, so the while loop will execute $objwiki->getpage at least once to retrieve the talk page text. However, for any subsequent talk pages, $talkpage is already non-blank, so $objwiki->getpage is skipped, leaving the prior page's content. I think if you change the code to:

			do {
				$talkpage = $objwiki->getpage($talkname);
			} while ($talkpage == "");

getpage will be called at least once and the problem resolved. An alternative would be to blank set $talkpage to an empty string before the loop. I hope this helps. -- Tcncv (talk) 22:50, 4 September 2009 (UTC)[reply]

This is why I publish my source code. Thank you, Tcncv. @harej 23:16, 4 September 2009 (UTC)[reply]

You're welcome. It works both ways. The published code gives me (and others) something to look at to see how things are dome. -- Tcncv (talk) 00:57, 5 September 2009 (UTC)[reply]

Thanks for fixing up the bot, by the way. I wasn't really sure as to whether I should block the bot or not, considering that the last major talkpage goof had occurred several hours previously - but considering that you were offline/in bed at the time, I thought it best to err on the side of caution. --Kurt Shaped Box (talk) 17:25, 6 September 2009 (UTC)[reply]

Nominated for deletion Wikipedia:Miscellany for deletion/Wikipedia:Last Resort Solution. I think this is fair, as it has been around for a long time with no work on it and no consensus. SpinningSpark 17:47, 8 September 2009 (UTC)[reply]

Is there a way of getting the bot to list rfc's to a user page for me? It would be handy if it could, especially if it could do so without adding the description of the dispute, although I'd like the timestamp to remain, if possible. It would save a lot of time when writing up the discussion piece in the signpost. Thoughts? Hiding T 09:39, 9 September 2009 (UTC)[reply]

Wikipedia:Coordination/rfc is a navbox with all the links to RFCs. However, to get them with the date would require the bot to be upgraded. I will think about it. @harej 10:21, 9 September 2009 (UTC)[reply]

This is how I currently write them up: Wikipedia:Wikipedia_Signpost/2009-09-07/Discussion_report#Requests_for_comment. Even if the bot could just add, for example, [[Wikipedia talk:User categories#Guidleline status|Wikipedia talk:User categories]]{{dot}} to a user sub-page or wherever, I could better cut and paste and extrapolate dates myself. Would that level be possible with the current coding? Hiding T 11:39, 9 September 2009 (UTC)[reply]

Hi, this may sound a bit cheeky, but is it possible in your closing edit summaries to say what the result was (e.g. closed as Blank, closed as Delete, closed as Keep). I tend to remember what my !vote was, but I don't always remember to watch the page (especially if it's a multi-list), and this would be useful at a glance without having to always look at the outcome directly!

Thanks! -- PhantomSteve (Contact Me, My Contribs) 22:59, 10 September 2009 (UTC)[reply]

I usually do. I haven't this round for some reason. I will do it from now on. @harej 23:06, 10 September 2009 (UTC)[reply]

Thank you -- PhantomSteve (Contact Me, My Contribs) 23:19, 10 September 2009 (UTC)[reply]

Wikipedia:Bot_owners'_noticeboard#Another_bot_removing_entire_pages_of_content --Admrboltz (talk) 18:57, 12 September 2009 (UTC)[reply]

Thanks for taking care of the issue. --69.225.12.99 (talk) 06:44, 13 September 2009 (UTC)[reply]

when I read that you are an SSD enthusiast.

That's because in my current state I thought that it said, SDS enthusiast.

So . . . . . . . . . . . . . ................. never mind. Carptrash (talk) 00:58, 13 September 2009 (UTC)[reply]

I guess you could say I am also an SDS enthusiast, though I've never affiliated myself with Students for a Democratic Society. @harej 01:00, 13 September 2009 (UTC)[reply]

Also consider that in the same sentence I mention that I am a social democrat. @harej 01:23, 13 September 2009 (UTC)[reply]

Okay. Yes. I'm considering. Probably me too. Life is good. Carptrash (talk) 14:36, 13 September 2009 (UTC)[reply]

thank you for caring re: my loss. DS (talk) 12:45, 13 September 2009 (UTC)[reply]

John Chrysostom's feast day is today. Hear hear! Keegan (talk) 15:33, 13 September 2009 (UTC)[reply]

John Chrysostom is all, therefore his feast day is every day. Regardless, this particular feast day of John Chrysostom is more significant, as it is named after him. Thank you, Brother Keegan. @harej 15:37, 13 September 2009 (UTC)[reply]

Hi! The bot doesn't seem to have updated RM in over two days. Is there something wrong? Jafeluv (talk) 17:57, 13 September 2009 (UTC)[reply]

Since I reworked the innards of the bot, it now gets stuck whenever it can't find which section the move request belongs to. In other words, people doing it wrong. I'm working on fixing the causes right now; fixing the code itself will come when I have more time. @harej 21:34, 13 September 2009 (UTC)[reply]

Thanks for removing the stick from the gears! Glad to see RFC bot updating the WP:Requested moves/current again. -kotra (talk) 18:05, 14 September 2009 (UTC)[reply]

Is it worth pointing out in your closing statement for the above MfD that there is an RFC Wikipedia_talk:Non-free_content#Proposed_clarification about the issues arising in that MfD? If you don't think so, no worries, but if so I think it's only polite for you to add it yourself, as the MfD is closed! -- PhantomSteve (Contact Me, My Contribs) 20:33, 14 September 2009 (UTC)[reply]

I think that, since the MfD is an archive at this point, it would be improper to modify the MfD itself. I'll make a note on its talk page, though. @harej 20:35, 14 September 2009 (UTC)[reply]

Thank you, I just wasn't sure if there was a policy at all! -- PhantomSteve (Contact Me, My Contribs) 20:55, 14 September 2009 (UTC)[reply]

Hello! Please see Template talk:Merge#Template:Mergeto and Template:Mergefrom have different default Discuss pages. Thanks! —David Levy 17:28, 15 September 2009 (UTC)[reply]

The backlog notice on this page (when visible) states "Please remove this notice when the backlog is cleared." However this is not the case as the notice is automatically updated by your RfC bot. As such I propose a change to the {{adminbacklog}} template to add a 'bot' parameter which would change the notice to "This notice is automatically updated by RFC bot (talk) and will no longer be displayed when the backlog is cleared". An example of this in use is here and the updated template is here as the actual template is fully protected. Essentially the change is to replace:

#if:{{{1|}}} | ''This notice will hide when this category has fewer than {{{2|{{{1}}}}}} {{plural:{{{2|{{{1|}}}}}}|item|items}}.'' <small>({{purge|recount}})</small> | ''Please remove this notice when the backlog is cleared.'' }}

with this

#if:{{{1|}}} | ''This notice will hide when this category has fewer than {{{2|{{{1}}}}}} {{plural:{{{2|{{{1|}}}}}}|item|items}}.'' <small>({{purge|recount}})</small> | {{#if:{{{bot|}}}|''This notice is automatically updated by {{User0|{{{bot}}}}} and will no longer be displayed when the backlog is cleared''|''Please remove this notice when the backlog is cleared.''}} }}

There seems little point on adding an {{editprotected}} tag as such a change only makes sense if you also update the bot to use the bot tag when adding {{adminbacklog}}. What are your thoughts? Dpmuk (talk) 23:20, 15 September 2009 (UTC)[reply]

The deed has been done. Expect to see the updated tag in action in ten minutes. @harej 00:20, 16 September 2009 (UTC)[reply]

Cheers, I know it's only a minor thing but it had confused me a bit until I started to dig a bit deeper on the WP:RM page. I've updated the templates documentation. Dpmuk (talk) 10:40, 16 September 2009 (UTC)[reply]

Dear harej, nobody answers what I have written here, Please move that article to the correct name. sicaspi 10:19, 16 September 2009 (UTC) —Preceding unsigned comment added by Sicaspi (talk • contribs)

I saw your edit changing from DMCA to DMC. That was the right thing to do. But we also have images and categories in that same line. Could you please change it, so that only main article namespace and articletalk namespace are sorted there. The rest should sort under "Items to be merged". I'd propose {{ #switch:{{NAMESPACE}} ||Talk={{DMC|Articles to be merged|from|{{{date|}}}|All articles to be merged}} |#default={{DMC|||Items to be merged}} }} . Thank you. Debresser (talk) 06:14, 21 September 2009 (UTC)[reply]

I'd do the same with Template:Mergefrom and Template:Mergeto. Debresser (talk) 06:26, 21 September 2009 (UTC)[reply]

I did the same with Template:Merging, which is not editprotected. It is working perfectly. Debresser (talk) 06:45, 21 September 2009 (UTC)[reply]

I will be sure to do so as soon as I am not on a cell phone. Or if you don't feel like waiting you can do {{editprotected}} on the talk page of the templates (well really it's one talk page shared by the templates. @harej 13:11, 21 September 2009 (UTC)[reply]

If you don't mind, can you please post what the updated code would be on a sandbox page or something? I don't want to risk breaking any more things. @harej 22:58, 21 September 2009 (UTC)[reply]

See Template:Merge/sandbox. Could you please delete that page after copying the templates? Debresser (talk) 10:52, 22 September 2009 (UTC)[reply]

Thanks. You could really have deleted it. I created it only to have a place where to write you. Debresser (talk) 11:30, 22 September 2009 (UTC)[reply]

Would you be willing to do the same thing for another three related editprotected templates? Tweak the categorisation, that is. If so, I can give you the details. Debresser (talk) 11:31, 22 September 2009 (UTC)[reply]

Yes. Please give me the details. @harej 14:38, 22 September 2009 (UTC)[reply]

I propose changing

{{{category|<includeonly>[[Category:Categories for merging|{{PAGENAME}}]] [[Category:CfD {{{year4|!!!!}}}-{{{month2|!!}}}]]</includeonly>}}}

to

{{#ifeq:{{NAMESPACE}}|Category|[[Category:Categories for merging|{{PAGENAME}}]] [[Category:CfD {{{year4|!!!!}}}-{{{month2|!!}}}]]}}

in Template:Cfm, Template:Cfd, and Template:Cfr. This will have the same effect of sharpening the categorisation. Debresser (talk) 20:01, 22 September 2009 (UTC)[reply]

 Done @harej 21:12, 22 September 2009 (UTC)[reply]

Oops. In template:Cfr that should have been "for renaming" and in template:Cfd "for deletion". I had thought of that, but forgot to write you about it. Debresser (talk) 21:51, 22 September 2009 (UTC)[reply]

And please also change part of the code back to [[Category:CfD {{<includeonly>subst:</includeonly>CURRENTYEAR}}-{{<includeonly>subst:</includeonly>CURRENTMONTH}}]]}}}. That was from another template. Another mistake of mine. I should have copied only the first part here for you. I'm really sorry about the mistakes. Debresser (talk) 21:56, 22 September 2009 (UTC)[reply]

For the meantime, I have rolled back my edits. Can you please prepare another sandbox in the same manner as you did for the merge templates? @harej 22:00, 22 September 2009 (UTC)[reply]

They are on Template:Merge/sandbox again. Debresser (talk) 22:19, 22 September 2009 (UTC)[reply]

I've also solved an old issue very elegantly, with onlyinclude tags. I have tested it (on a real category) and it works perfect. If you care, you could delete Template:Mergetest. Debresser (talk) 22:27, 22 September 2009 (UTC)[reply]

That solution I was so proud of, had a side effect of breaking bots, as pointed out by User:Kbdank71. He has made an edit to fix this. The nicer way to fix this is on Template:Merge/sandbox again. Would you care to copy them again? Debresser (talk) 00:21, 23 September 2009 (UTC)[reply]

 Done hopefully for the last time. :) @harej 00:53, 23 September 2009 (UTC)[reply]

Thanks. Should be. These last edits were minor, more a matter of standardisation. This is not the first time I have seen an edit to a template had to be undone or changed because bots are used to working in specific ways. Debresser (talk) 00:56, 23 September 2009 (UTC)[reply]

Lately I have been watching the project namespaces (especially WP:DR) more avidly. Because of this I have been seeing your name frequently. As a result of this, I absentmindedly visited your userpage. I decided I just had to compliment you on the humerous and satirical tags on your userpage. They are spot on. Keep up the good work. —Matheuler 02:05, 22 September 2009 (UTC)[reply]

Thank you. Though I am not sure how my name keeps appearing on WP:DR. Also, User:Gigs created the wikibreak template on the bottom of my user page, and I don't remember who created the warning box on the top of mine. @harej 02:08, 22 September 2009 (UTC)[reply]

Hi James, a few editors have asked me what stage the triple-date unlinking bot is at. I heard that a trial run of 50 had been approved / was happening. Tony (talk) 13:01, 22 September 2009 (UTC)[reply]

Some issues with the code. First, it wasn't able to connect to the database at all. Now the process seems to get hung up at a certain point and I am not sure why. I've been adding echoes throughout the code to see where I could track it, to no avail. @harej 14:31, 22 September 2009 (UTC)[reply]

I hope you didn't get discouraged by all the comments when you first put up the code. It must have changed quite a bit if there is now a database. This database, is it MySQL running on your computer? --Apoc2400 (talk) 15:48, 23 September 2009 (UTC)[reply]

Not discouraged at all. The script runs, or at least is supposed to run, on the Toolserver. The MySQL connection works now, but there are other things that need to be addressed. @harej 19:36, 23 September 2009 (UTC)[reply]

AWB has been experiencing problems connected with the recent MediaWiki changes. Could the bot's problems have the same source? Colonies Chris (talk) 20:08, 23 September 2009 (UTC)[reply]

No, unless English Wikipedia's copy of MediaWiki had its database structure changed. The source is most likely problems from within the code (the bot's code) itself. @harej 22:29, 23 September 2009 (UTC)[reply]

Thanks for your hard work. Hopefully these problems will be sorted out soon. Dabomb87 (talk) 02:14, 24 September 2009 (UTC)[reply]

I tried to move Portal:Agropedia to Portal:Agriculture and Agronomy per the MfD, forgetting that all the subpages would need to be moved. If you could clean up my mess I'd appreciate it. :) Gigs (talk) 23:21, 23 September 2009 (UTC)[reply]

 Done @harej 23:33, 23 September 2009 (UTC)[reply]

You missed the main page it seems. Gigs (talk) 23:58, 23 September 2009 (UTC)[reply]

No, I'm confused. Why are all the transclusions on the main page broken though? Gigs (talk) 23:59, 23 September 2009 (UTC)[reply]

Purge thine cache. @harej 00:02, 24 September 2009 (UTC)[reply]

Thanks for your help! Gigs (talk) 13:29, 24 September 2009 (UTC)[reply]

Hi, please see this edit: [6]. Note that inserting the [[ ]] marks into an existing wikilink breaks the link. TJRC (talk) 01:15, 24 September 2009 (UTC)[reply]

I am indeed aware of this. The workaround I've been using is to remove the link in the header. I've been experimenting with a more scalable (i.e. automatic) solution on WP:CORD, and I'll probably bring it over for RFC. @harej 01:20, 24 September 2009 (UTC)[reply]

		The Admin's Barnstar
		For your work at MfD, including cleaning up the messes I make. Thanks! Gigs (talk) 13:31, 24 September 2009 (UTC)[reply]

Yeehaw! But what does "party like it's 2004 refer to? @harej 16:45, 24 September 2009 (UTC)[reply]

Barnstars are somewhat retro these days. Gigs (talk) 18:24, 24 September 2009 (UTC)[reply]

Not really. In fact, someone's RFA criteria calls for having barnstars. @harej 19:05, 24 September 2009 (UTC)[reply]

I had to rollback this erroneous edit: http://en.wikipedia.org/w/index.php?title=Talk:Zenwalk_Linux&diff=311282768&oldid=310866238 The bot blanked the page giving edit summary "Removing unpaired moveheader". --Chealer (talk) 02:54, 25 September 2009 (UTC)[reply]

That happened quite a while ago. I believe at this point the underlying cause of the bug has been fixed. @harej 02:55, 25 September 2009 (UTC)[reply]

Hi, Harej! I wonder why you closed the request for move as 'no consensus'. All the users who have been involved in the discussion supported the move: me (Eleassar) as a proposer, and also Marschner (I fully agree with Eleassar's argumentation and support a move.) and Viator slovenicus (I have to admit that Slovenian Carinthia does seem the best of the proposed options.) The consensus for the move is clear even though none of the proposed names is optimal. --Eleassar ^{my talk} 10:24, 26 September 2009 (UTC)[reply]

I'm sorry, I misread these things sometimes. After re-reading it, moving the page makes more sense. @harej 13:50, 26 September 2009 (UTC)[reply]

As the puppet government of WT:RM is clearly unable to react to the evident threat, I have taken the appropriate action and listed you for deletion at Wikipedia:Editors for deletion. Feel free to present your defence on that page (or just remove the section if you don't feel like participating). Regards, Jafeluv (talk) 11:16, 28 September 2009 (UTC)[reply]

Hi Harej! I'd just like to know how there was a lack of consensus for that move. It seemed 2 against 1 was more than in favor of a move when the dissenting opinion was fraught with errors in argument (count "von" Wisborg, incorrectly stating that the move was to say the royal family's surname is Monpezat, etc). The user was trying to speak for me (and was wrong) in saying that I based the argument on it being their surname. The fact is the queen made the title count(ess) of Monpezat. Thanks! 142.68.80.29 (talk) 16:16, 28 September 2009 (UTC)[reply]

The one objection appeared to be a significant objection, enough to put the request into doubt. Really, there wasn't enough participation, so I have relisted it. @harej 16:51, 28 September 2009 (UTC)[reply]

Hi Harej! Thanks for the color you gave me for the background! That color looks good! RuneScape Adventure 01:33, 30 September 2009 (UTC)[reply]

RfC talkpage for Wikipedia:Requests for comment/Policies‎ has a first topic where the title is screwed up due to links in the original header. This confuses the bot. Can you take them out, and also the color tags? That makes the title serve as proper link back to the page. THe bot will not let me fix it,and keeps reverting to the bad version which will not backlink to the WP:Red link talkpage. Thanks! SBHarris 07:52, 2 October 2009 (UTC)[reply]

It appears to have been fixed. @harej 11:22, 2 October 2009 (UTC)[reply]

You just can't beat that old English humor. ESCapade (talk) 14:09, 2 October 2009 (UTC)[reply]

{{Tb|Wikipedia:Miscellany_for_deletion/User:ESCapade/Zenfolio}}

{{rfctag|proj}}

I will have to test my bot's new edit summary feature somehow. @harej 06:05, 3 October 2009 (UTC)[reply]

RFCbot seems unhappy about edit summarying correctly on RFC/Style. Fifelfoo (talk) 06:39, 3 October 2009 (UTC)[reply]

I am trying to implement a new edit summary feature. Except it didn't work, and I was too tired last night to fix it, so I implemented a hack until I could really fix it. @harej 15:43, 3 October 2009 (UTC)[reply]

I'm a bit scared by this edit...how does the bot know what my signature looks like? Xenon54 / talk / 22:38, 3 October 2009 (UTC)[reply]

There is a government conspiracy to deprive you of cheese My bot ascertained the signature from the entry on Wikipedia:Good article nominations. @harej 22:54, 3 October 2009 (UTC)[reply]

Harej, I hope you're in a position to ask for further approval, having got so far already. Perhaps you might consider asking for approval for a sequence of trials, with sets of edits separated by pause periods. For example: 1,000 edits, 2,000 edits, 5,000, 10,000, 20,000, and 50,000 edits.

Each set could be followed by a mandatory pause of x days (e.g. 2 days) before continuing to the next block. This would allow ordinary editors time to identify errors in the bot coding and complain.

Or BAG could limit the delinking rate (x per hour).

Just some thoughts. Tony (talk) 10:26, 4 October 2009 (UTC)[reply]

Wikis Take Manhattan Next: Saturday October 10 This box: view • talk • edit

WHAT Wikis Take Manhattan is a scavenger hunt and free content photography contest aimed at illustrating Wikipedia and StreetsWiki articles covering sites and street features in Manhattan and across the five boroughs of New York City.

LAST YEAR'S EVENT

• Wikipedia:Wikipedia Takes Manhattan/Fall 2008 (a description of the results, and the uploading party)
• Commons:Wikis Take Manhattan (our cool team galleries)
• Streetfilms: Wikis Take Manhattan (our awesome video)

WINNINGS? The first prize winning team members will get Eye-Fi Share cards, which automatically upload photos from your camera to your computer and to sites like Flickr. And there will also be cool prizes for other top scorers.

WHEN The hunt will take place Saturday, October 10th from 1:00pm to 6:30pm, followed by prizes and celebration.

WHO All Wikipedians and non-Wikipedians are invited to participate in team of up to three (no special knowledge is required at all, just a digital camera and a love of the city). Bring a friend (or two)!

REGISTER The proper place to register your team is here. It's also perfectly possible to register on the day of when you get there, but it will be slightly easier for us if you register beforehand.

WHERE Participants can begin the hunt from either of two locations: one at Columbia University (at the sundial on college walk) and one at The Open Planning Project's fantastic new event space nestled between Chinatown and SoHo. Everyone will end at The Open Planning Project:

• Wikis Take Manhattan page at The Open Planning Project

148 Lafayette Street

between Grand & Howard Streets

FOR UPDATES

Please watchlist Wikipedia:Wikipedia Takes Manhattan. This will have a posting if the event is delayed due to weather or other exigency.

Thanks,

Pharos

You can add or remove your name from the New York City Meetups invite list at Wikipedia:Meetup/NYC/Invite list.
This has been an automated delivery by BrownBot (talk) 21:03, 4 October 2009 (UTC)[reply]

Please take a look at User talk:Tony1#Date link request. You may wish to comment as my questions are directly related to User:Full-date unlinking bot. Cheers! Vassyana (talk) 10:59, 8 October 2009 (UTC)[reply]

James, thanks for your response. I have added [a note there after yours. Tony (talk) 03:18, 9 October 2009 (UTC)[reply]

I saw that you ran the next 200 edits for the Full-date unlinking bot. I spot checked a couple dozen edits and they all look good. So far I see examples of seven of the nine edit patterns. The YMD ([[yyyy]] [[month dd]]) and ISO2 ([[yyyy]]-[[mm-dd]]) patterns have not yet been encountered, as they are rare, but based on earlier controlled tests, I'm fully confident that they will work fine.

I did notice that the edit summary codes for List1 and List 2 are missing the "×", but that is a very very minor item. Probably not worth the risk of changing already-tested code. -- Tom (tcncv)^T/_C 01:34, 10 October 2009 (UTC)[reply]

Here's YMD. As for the ×, I have added them in for List1 and List2. @harej 01:38, 10 October 2009 (UTC)[reply]

Good. By the way, did you find a solution to the whatlinkshere 500 link limit? If not, the solution might be to make a private copy of that routine and bump up the limit parameters. My sampling of what-links-here for month-day pages shows 5000+ linked pages per date (example: [7]). I t might be worth running a stand alone test (using a fragment of your code) to confirm that your version of whatlinkshere gets (roughly) the expected number of articles in the PHP call. -- Tom (tcncv)^T/_C 02:09, 10 October 2009 (UTC)[reply]

I haven't even been thinking about that. The most obvious solution is to use a custom query that will increase the limit, but what should I set the limit as? (I don't want it too high, for the risk of pissing off the wrong people). @harej 02:11, 10 October 2009 (UTC)[reply]

In my URL link above, I set the limit to 999999 and got between 5000 and 6000 articles for each date I tried. The results came back fairly quickly. From my database experience, I don't think overstating the max causes any real extra work for the server. The server will stop retrieving data once everything is retrieved. Setting the limit to 9999 (or 10000 if you prefer) would produce the same result with the same amount of server load. It might be worth asking the database folks though.

By the way, I don't think the loop that looks for bare month links is needed. All of the dates that we are unlinking will have a month-day combination link. -- Tom (tcncv)^T/_C 02:41, 10 October 2009 (UTC)[reply]

As for the page blanking, I'm looking into that. I grapped a copy of the Common year starting on Thursday and ran it through my stand-alone copy of the unlinker code, and also ended up with an empty result. I'm looking into that now. -- Tom (tcncv)^T/_C 03:04, 10 October 2009 (UTC)[reply]

You mean it doesn't have to do with the API mucking up and accidentally loading an empty page? @harej 03:04, 10 October 2009 (UTC)[reply]

That is correct. Apparently, a couple of regular expressions had some ambiguities that made them nondeterministic, in that there were multiple path through some of the subexpressions that matched the same string. In cases of a successful match, this was not a problem. But in cases of a partial match, the regular expression processor seems to backtrack extensively trying every possible combination of the partially matching subexpressions, eventually reaching some limit and giving up. That's when it returns an empty result. (Example: The expression "Ax*B?x*C" will match "AxxBxxC" or "AxxxxC", but will not match "AxxxxZ". Since the x's can match either the left or right "x*" subexpression, the regular expression processor will try every combination as it attempts to complete that match. If there are ten or more x's the process becomes too complex and will fail. However, if the expression were coded as "Ax*(Bx*)?C", the problem is eliminated. The second "x*" is only considered if the optional B is present, so no ambiguity exists.)

I've got two of the three problem cases fixed and am working on the third. I'll release the changes once I have everything fixedand tested. I'm also going to review all of the expressions to check for other potential problems. -- Tom (tcncv)^T/_C 15:28, 10 October 2009 (UTC)[reply]

I have updated the bot code to fix the regex ambiguity problems. I also discovered that cutting&pasting the code between the Wikipedia edit window and my local editor was replacing the graphic non-breaking space with a regular space, which introduced additional ambiguities, so I decided to eliminate that case. It was extremely rare anyway. The updated regexes have been successfully tested against the three pages that previously had problems and the test environment page. I also added (but am not in a position to test) [few lines of code] to detect and report page blanking cases before they are saved. Please take a look.

I would recommend running another test of a few hundred or a thousand edits to exercise the changes before any full scale operations. -- Tom (tcncv)^T/_C 18:30, 10 October 2009 (UTC)[reply]

I left some feedback at Wikipedia_talk:Good_article_nominations#comments_by_CBM. I also wrote some PERL code to parse the sections out of GAN directly, so that the "live code" thing isn't necessary. It would have been slower for me to write it in php, which is basically a read-only language for me. I put the code at ga1.pl. — Carl (CBM · talk) 13:48, 10 October 2009 (UTC)[reply]

Please pay attemtion to the edit summary of my last edit to Wikipedia talk:Manual of Style/Archive 103. Has that problem been fixed? Debresser (talk) 20:08, 11 October 2009 (UTC)[reply]

It was fixed a looooooooooong time ago. @harej 20:09, 11 October 2009 (UTC)[reply]

Ok. that's good. Appearently not all affected pages were restored. Debresser (talk) 21:47, 11 October 2009 (UTC)[reply]

Your attention is brought to a motion currently being considered by the Arbitration Committee:
Wikipedia:Arbitration/Requests/Motions#Motion to amend Wikipedia:Requests for arbitration/Date delinking.

At the time this notice was posted the text of the motion read:

“

Wikipedia:Date formatting and linking poll, Wikipedia talk:Full-date unlinking bot#RFC, and Wikipedia:Bots/Requests for approval/Full-date unlinking bot indicate that Full-date unlinking bot (BRFA · contribs · actions log · block log · flag log · user rights) fulfills the requirement for "a Community approved process for the mass delinking" in "1.3 Mass date linking" and the requirement for "[d]ate delinking bots [performing] in a manner approved by the Bot Approvals Group" in "2.1 Date delinking bots". The Committee thanks the participants for their efforts and encourages them to continue with their contructive work and consensus building.

”

This wording may have since changed; please see the above link for the current wording.

On behalf of the Arbitration Committee, Manning (talk) 09:46, 12 October 2009 (UTC)[reply]

Really reads like a court summons... @harej 11:10, 12 October 2009 (UTC)[reply]

But this is a very welcome gesture and proper procedure. Looking forward to your further work, James! Tony (talk) 13:42, 12 October 2009 (UTC)[reply]

Hmmm... I wrote the notice. Sorry if it was too stiff. This was my first attempt - later versions of this notice started with "As a potentially interested party your attention is brought..." which I think softened the tone a bit. Cheers Manning (talk) 22:30, 12 October 2009 (UTC)[reply]

[8] --Rschen7754 (T C) 06:35, 15 October 2009 (UTC)[reply]

Thank you for bringing this to my attention. @harej 14:31, 15 October 2009 (UTC)[reply]

James, it's good to see that User:Tcncv is suggesting another, bigger date-bot run. Are you going to apply for the necessary permission? Cheers. Tony (talk) 14:58, 16 October 2009 (UTC)[reply]

James, I see that Rich Farmborough has offered to help with the bot. Since it's taking a lot longer than anticipated, for whatever reason, do you think a collaborative effort might speed things up? Tony (talk) 11:24, 20 October 2009 (UTC)[reply]

We have enough programmers. Simply, I am awaiting approval to continue testing. @harej 13:22, 20 October 2009 (UTC)[reply]

Thanks, James. I presume a formal application has been made to BAG. Tony (talk) 10:33, 21 October 2009 (UTC)[reply]

I requested a move by cutting and pasting the recommended text at 20:38 UTC on Oct 16 at Talk:United States House of Representatives special elections, 2009. As of 00:41 on Oct 17, it has still not been added to WP:Requested moves/Current. -Rrius (talk) 00:43, 17 October 2009 (UTC)[reply]

I'm looking forward to your move! --Ludvikus (talk) 01:19, 17 October 2009 (UTC)[reply]

What are you talking about? @harej 02:11, 17 October 2009 (UTC)[reply]

[9] [10] --Ludvikus (talk) 04:11, 17 October 2009 (UTC)[reply]

Thank you for participating in WP:Requests for adminship/Kww 3
Sometimes, being turned back at the door isn't such a bad thing

—Kww(talk) 18:13, 20 October 2009 (UTC)[reply]

How is there no consensus? It is 4 against 2, and the first voted oppose because he believes that the other Wikis call him Bardiya and that is what the English Wikipedia should do too. I believe you should re-look at the decision you have made. warrior4321 21:27, 20 October 2009 (UTC)[reply]

I'm somewhat concerned about your addition of this category - I can currently see no explanation of what the difference between this category and the policy category. I think the RfC page really does need a clear description of what the different categories should be used for, but I can't see a clear description of this category even in your proposal. As such I would imagine it's confused many editors. I also think your edit summary when you've made the change to existing RfCs (e.g. here is not informative enough - it doesn't even mention a new category, never mind why it's been changed. It was only because of this unexplained change (linked above) that I even become aware of your proposal. Dpmuk (talk) 09:16, 21 October 2009 (UTC)[reply]

Was this more or less what you were suggesting? --SmokeyJoe (talk) 04:24, 22 October 2009 (UTC)[reply]

To which of my comments are you referring? @harej 20:12, 22 October 2009 (UTC)[reply]

Wikipedia_talk:Sock_puppetry/Archive_6

Let's make it clear that alternate accounts cannot be used for abuse. When this abuse happens, the connection between the two accounts must be posted, with both blocked indefinitely. How's that for a deterrent? —harej (talk) (cool!) 00:00, 22 July 2009 (UTC)

I asked if this would conflict with the privacy policy? I expected an answer, but none came. On my own reading. to my best judgment, under the privacy policy, an abuser cannot expect to be covered. --SmokeyJoe (talk) 20:24, 24 October 2009 (UTC)[reply]

The changes you have made to the policy page seem to be an accurate reflection of what I posted on the talk page then. @harej 20:30, 24 October 2009 (UTC)[reply]

Hi, I saw you nominated moved. Would the same apply to {{notmoved}}? Thanks! Plasticspork (talk) 20:35, 22 October 2009 (UTC)[reply]

The same principles apply. @harej 20:48, 22 October 2009 (UTC)[reply]

I'd like your opinion as to whether we should add code to the bot to handle additional date forms as discussed in User talk:Full-date unlinking bot#False negatives. -- Tom N (tcncv) talk/contrib 19:39, 24 October 2009 (UTC)[reply]

Assuming there are no objections, it will take me a couple of days to update and test the new date expressions. Before I make any edits, can you confirm that the posted code is the latest and greatest? Thanks. -- Tom N (tcncv) talk/contrib 20:47, 24 October 2009 (UTC)[reply]

It is actually later than my own local copy, making it the latest and the greatest of them all. @harej 20:52, 24 October 2009 (UTC)[reply]

Hi Harej, and thanks for all of your good work so far. Do you know when the next bot run will be? Thanks, Dabomb87 (talk) 23:45, 28 October 2009 (UTC)[reply]

Tcncv is making some changes to the bot code, I believe, so I am waiting for those changes to be posted. @harej 00:18, 29 October 2009 (UTC)[reply]

I placed a request on Wikipedia:Requested moves/current, and your bot removed it. ??? Teh Rote (talk) 14:48, 26 October 2009 (UTC)[reply]

I am trying to move my first article and I am having trouble doing so. Could you please help me? Thank you in advance! Elysian1503 (talk) 10:57, 28 October 2009 (UTC)[reply]

Where is it? @harej 20:03, 28 October 2009 (UTC)[reply]

Hi. Have you noticed that zour bot was approved for a third trial? Check Wikipedia:Bots/Requests for approval/Full-date unlinking bot. -- Magioladitis (talk) 14:36, 30 October 2009 (UTC)[reply]

I noticed, and I am waiting to see if Tcncv has any updates to the code. @harej 14:38, 30 October 2009 (UTC)[reply]

I have made changes locally to recognize ordinal days, but have not finished the changes to handle fully piped dateforms. I expect to have them done and unit tested this weekend. -- Tom N (tcncv) talk/contrib 01:50, 31 October 2009 (UTC)[reply]

I've posted an updates to User:Full-date unlinking bot/code that implement changes to support ordinal dates and fully piped dates. Results of unit testing can be seen in this diff. A comparison with earlier test results shows no regression.

Although we are approved for 500 edits, I would suggest starting with 50 – 100 edits and checking the results before completing the full run. There were a lot of changes. -- Tom N (tcncv) talk/contrib 21:52, 31 October 2009 (UTC)[reply]

I scanned the results of the first 84 edits of the current test run and did not see any problems. There were only a few ordinal and one piped link encountered, and the results were as designed. I think we're good-to-go on the remaining 400+ edits of the test. -- Tom N (tcncv) talk/contrib 01:31, 3 November 2009 (UTC)[reply]

The bot has made 117 more edits, bringing its third trial edit count to 201. @harej 02:08, 3 November 2009 (UTC)[reply]

Looks good to me. -- Tom N (tcncv) talk/contrib 01:51, 4 November 2009 (UTC)[reply]

[11] I believe that was a bit premature. One of the opposers was of the opinion that {{All plot}} serves no distinct purpose and nominated it for discussion. The TfD is still open, but overwhelmingly leaning towards a consensus to keep. With that concern out of the way, there is nothing in the arguments against moving {{Plot}} that I could make out. I wanted to wait until the TfD is officially closed though before noting the result and its immediate consequences for the move request at {{Plot}}'s template talk page.

On a related note: I am under the impression that all discussions on Wikipedia should aim a achieving consensus on how best to proceed in a given situation. This is also means that all input to the discussion should be evaluated on its own merits, not as some kind of vote.

Agreement may of course be noted, but discussions on Wikipedia are not supposed to be straight votes. I really don't want to be a dick about this, but the fact of the matter is that no arguments have been presented against the move by any of the opposers. If you think differently, please point out which arguments in the discussion led you to determine there was a rational consensus against the move? Otherwise, please consider reopening the move request.

Alternatively, I'll have to start a new move request after the TfD has concluded. --87.79.84.238 (talk) 18:38, 30 October 2009 (UTC)[reply]

Let's wait until the TfD is finished. @harej 19:19, 30 October 2009 (UTC)[reply]

Ok. --87.79.84.238 (talk) 20:46, 30 October 2009 (UTC)[reply]

Thanks for uploading File:Led Zeppelin - All My Love.ogg. The image description page currently specifies that the image is non-free and may only be used on Wikipedia under a claim of fair use. However, the image is currently orphaned, meaning that it is not used in any articles on Wikipedia. If the image was previously in an article, please go to the article and see why it was removed. You may add it back if you think that that will be useful. However, please note that images for which a replacement could be created are not acceptable for use on Wikipedia (see our policy for non-free media).

If you have uploaded other unlicensed media, please check whether they're used in any articles or not. You can find a list of "file" pages you have edited by clicking on the "my contributions" link (it is located at the very top of any Wikipedia page when you are logged in), and then selecting "File" from the dropdown box. Note that any non-free images not used in any articles will be deleted after seven days, as described on criteria for speedy deletion. Thank you. ww2censor (talk) 18:15, 2 November 2009 (UTC)[reply]

I've gone and deleted it for you. @harej 20:13, 2 November 2009 (UTC)[reply]

New York City Meetup Next: Sunday November 15th, Columbia University area Last: 09/13/2009 This box: view • talk • edit

In the afternoon, we will hold a session dedicated to meta:Wikimedia New York City activities, review the recent Wikipedia Takes Manhattan, plan for the next stages of projects like Wikipedia at the Library and Wikipedia Loves Landmarks, and hold salon-style group discussions on Wikipedia and the other Wikimedia projects, for example particular problems posed by Wikipedia articles about racist and anti-semitic people and movements (see the September meeting's minutes).

In the evening, we'll share dinner and chat at a local restaurant, and generally enjoy ourselves and kick back.

You can add or remove your name from the New York City Meetups invite list at Wikipedia:Meetup/NYC/Invite list.

To keep up-to-date on local events, you can also join our mailing list.
This has been an automated delivery by BrownBot (talk) 03:26, 3 November 2009 (UTC)[reply]

More strange behaviour from RFC bot: [12]. I won't try to fight it since I know from experience it's very persistent, but please could you fix whatever the bug is and correct the page in a way that the bot won't immediately change it back. Thanks,--Kotniski (talk) 18:43, 3 November 2009 (UTC)[reply]

It's futile to edit an automatically generated list. The weirdness in question though was most likely caused by a combination of factors that don't affect most page move requests but this one because of the title. Since the glitch doesn't seem to be causing problems anywhere else, I closed and processed the request since it was due to be processed anyway. @harej 19:00, 3 November 2009 (UTC)[reply]

Thanks. I'm guessing it was caused by the presence of the word "Talk" in the title - is the bot programmed to remove that word because requests to move talk pages should be converted into requests to move the subject page? (Or something like that?)--Kotniski (talk) 19:02, 3 November 2009 (UTC)[reply]

The bot indeed converts talk page titles to regular titles, though it goes about it in a smarter way than just removing the word "talk". In order for me to make more conclusions about the cause, this kind of thing has to happen more than once so I can see a pattern develop. Still, my money is on the combination of "Wikipedia" and "talk" in the title. @harej 19:09, 3 November 2009 (UTC)[reply]

I saw you just fix up Talk:I'm Your Captain (Closer to Home) and Talk:Echoes (Pink Floyd song) for their GA nominator details. However, neither article is listed at WP:GAN; I think the nominator forgot that step. Should the nominations be deleted off the talk pages or added to the GAN page? Wasted Time R (talk) 04:15, 6 November 2009 (UTC)[reply]

Go ahead and add them to WP:GAN. In my opinion no one should have to do that, hence my efforts to create an automated list. @harej 04:21, 6 November 2009 (UTC)[reply]

OK, I've now done that. Wasted Time R (talk) 05:43, 6 November 2009 (UTC)[reply]

Hi! I've moved the article Salaried Employees' and Civil Servants' Confederation to FTF – Confederation of Professionals in Denmark, since this now is the correct name of the confederation. How do we move the page [[Category:Salaried Employees' and Civil Servants' Confederation]] to [[Category:FTF – Confederation of Professionals in Denmark]]...? (Because I don't know how to do it...) --Wikibruger • Write to Wikibruger... • ♂ 16:06, 6 November 2009 (UTC)[reply]

Categories can be renamed through WP:CFD. @harej 17:05, 6 November 2009 (UTC)[reply]

Thanks a lot - it's done now! --Wikibruger • Write to Wikibruger... • ♂ 21:30, 6 November 2009 (UTC)[reply]

Hey. I've taken care of the comments you left there — plus some questions on a few — so feel free to check it out. The Flash ^{talk} 17:37, 7 November 2009 (UTC)[reply]

As per this edit I can't tell why this bot is changing my edits? Am I doing something wrong? If not why is the bot acting in such a strange manner? Surely there must be some kind of reasonable explanation, and this is one heck of a glitch (if it turns out to be one). Outback the koala (talk) 07:09, 9 November 2009 (UTC)[reply]

Read the directions, then get back to me. @harej 12:39, 9 November 2009 (UTC)[reply]

See WT:GAN. I've blocked it until you can sort it out! Thanks Geometry guy 18:17, 9 November 2009 (UTC)[reply]

Thanks for sorting this out. Good decorum would be to apologize to the editors concerned. Geometry guy 21:54, 9 November 2009 (UTC)[reply]

Hi RFC! I think the bot went berserk again. Keeps removing move requests on the grounds of a regular update? Regards. Cretanforever (talk) 13:35, 11 November 2009 (UTC)[reply]

Can you please be more specific? Where is this happening? @harej 15:11, 11 November 2009 (UTC)[reply]

Sorry, my mistake! I was still trying to add the move request manually. It seems to be working now. Thx and rgds. Cretanforever (talk) 10:02, 12 November 2009 (UTC)[reply]

The bot appears to have stopped working at 17:18, 12 November 2009. Vegaswikian (talk) 09:16, 13 November 2009 (UTC)[reply]

I don't understand -- the bot is still editing, and there does not appear to be anything ostensibly wrong with its edits. @harej 12:20, 13 November 2009 (UTC)[reply]

It stopped updating Wikipedia:Requested moves/current. the history there shows the pause. Vegaswikian (talk) 02:11, 14 November 2009 (UTC)[reply]

Harej, are you an admin who regularly takes part in WP:RM? I'm looking for an uninvolved admin to supervise and close a poll regarding a controversial move. Would you be willing do that in principle? If yes, I'll write up the details for you. Cheers, SlimVirgin ^{talk|contribs} 02:14, 11 November 2009 (UTC)[reply]

Helloooooooooooo ... :) SlimVirgin ^{talk|contribs} 06:03, 12 November 2009 (UTC)[reply]

You look rather stupid if you make edits like this one, when the whole donation banner has been disabled for the past 2 hours. At the same time, you are also ignoring ALL warnings and cautions presented to you in the editnotice which is also a somewhat dumb thing to do. You ought to know better harej. —TheDJ (talk • contribs) 15:47, 11 November 2009 (UTC)[reply]

I think this is a little harsh TheDJ. Obviously rollback wasn't appropriate here and there is the consideration of caching, but on the other hand, the other editor who made exactly the same edit was awarded a barnstar, and I don't see you complaining to him ... — Martin (MSGJ · talk) 12:12, 12 November 2009 (UTC)[reply]

As you participated in the recent Audit Subcommittee election, or in one of two requests for comment that relate to the use of SecurePoll for elections on this project, you are invited to participate in the SecurePoll feedback and workshop. Your comments, suggestions and observations are welcome.

For the Arbitration Committee,
Risker (talk) 08:09, 12 November 2009 (UTC)[reply]