I feel the vandalism levels of wikipedia getting updated every 30 minutes isn't frequent enough. I think every 5 minutes would be better. Thoughts? — RosscoolguyCVU | 20:29, 22 January 2013 (UTC)[reply]
Please see the comments at our BRFA on this issue. Especially this line from BAG member Hellknowz: "I would say anything less than 20-30 minutes and 2000+ edits is not a large enough sample size. So much can change in a few minutes. I definitely don't think you can judge a site-wide vandalism from just the last few minutes." Vacation921:26, 22 January 2013 (UTC)[reply]
That doesn't even make sense... Anyway, Sigma raised concerns about the edit rate if we edited every five minutes. If it looked for vandalism in the last 30 minutes but edited every five there would be duplicate counts, and it wouldn't be accurate at all. I don't know how that would work. The consensus at our BRFA was for 30 minute edit intervals, and that's how it will stay unless there is an extreme shift in views. Vacation901:11, 23 January 2013 (UTC)[reply]
It seems that the server VoxelBot is currently running on (bots-4 on Wikimedia Labs) is down. We're moving it to another server temporarily. Thanks for notifying us. Vacation901:35, 25 January 2013 (UTC)[reply]
The bot cannot directly replace characters in links, as it would break the link. The first stage, running now, is going through a lot of pages in the Geography of Romania category and moving the page to the correct title, then correcting all mentions of it. The second stage simply corrects the characters outside of links. Sadly, it is not possible to combine the two tasks easily, and it would be very risky. By the end, all those pages should be corrected. Vacation912:21, 14 February 2013 (UTC)[reply]
I'm sorry to hear that. However, please realize no bot can be 100% effective. We only operated on certain subcategories so there would be no false positives. The rest are small in number and can be manually corrected. Vacation911:50, 18 March 2013 (UTC)[reply]
Hi. I just reverted a couple of cases where VoxelBot had created redirect loops: [1], [2]. Then I realized there were several more similar cases; see Double Redirects. So I thought at this point I should pause and ask if you could please take a look. Thanks. – Wdchk (talk) 03:52, 18 February 2013 (UTC)[reply]
Hmm, yes that is true. That task has stopped for now however. Looks like it wasn't too bad. Feel free to fix them if you want. Vacation900:45, 20 February 2013 (UTC)[reply]
Hi. When I saw this article that I created "recently" had been tagged, I was confused as I had no memory of it. When I checked back, I had created it in 2006! I realise it may be difficult to amend the message wording to take dates into account, so how about just removing the word "recently"?Deb (talk) 08:37, 13 March 2013 (UTC)[reply]
On the surface, the number of reverts in the last period of time may seem as a good indication of how much vandalism is flying about. However a simple thought experiment challenges this idea:
Let us assume for the sake of example that at level one, five antivandals patrol, and vice versa (i.e. more patrollers when critical.)
Also assume for the sake of example that the level of vandalism remains constant.
When nobody is patrolling, the report will say level five, because of the lack of reverts. Now someone starts patrolling and they find a fair amount of vandalism, so the report level increases. As the report level increases, more AVs join and suddenly every edit is being scrutinised and all vandalism is found. So the report goes to level one, but the AVs get bored of patrolling, and start to tail off. So vandalism gets missed, and the report level decreases.
The vandalism has remained constant, the reports are merely reporting how many AVs are patrolling. This is the inverse of what should be reported - when there are fewer AVs about, it should encourage more to come on.
I wonder if a better value would be to take all the users who reverted more than five times in the last time period (to remove casual cleanups) and average how many reverts they've made. If the AVs are reverting large numbers, then the level should increase, because there aren't enough to catch all; if fewer reverts per person, decrease as there are enough AVs to clean up.
930913(Congratulate) 21:11, 16 March 2013 (UTC)[reply]
Interesting theory. I see your point, but there are three issues:
This is the way the template has been for years, and I don't want to just change it. People rely on the template and how it is updated.
When people are manually updating the template, for example if VoxelBot is down, they can't do this, as they use Huggle
You're getting into saying whether additional antivandals are needed, and straying away from the current levels of vandalism.
Additionally, the current method is reasonably accurate, as _most_ vandalism is reverted within thirty minutes, and outliers will vanish in the 30 minute average. Thanks for writing this, but unless clear consensus is shown for this change, we won't implement it at this time. Vacation922:52, 16 March 2013 (UTC)[reply]
Here's a thought: How about basing the information off Cluebot NG's reverts a minute. That pretty much fixes the problem. The only issue is what to do when Cluebot is down :P nerdfighter21:36, 15 April 2013 (UTC)[reply]
Oh :) The only thing that I can think of off the top of my head is the IRC feed, but in the BRFA we already agreed that IRC would be difficult to implement... hmmm... I know VoxelBot keeps a log here, but that's not really relevant. I'll look into it, but our current detection method seems to work well enough. It's a Fox!(What did I break)22:18, 23 April 2013 (UTC)[reply]
Hi, the bot is not archiving FACs correctly. It is describing the articles as lists and adding the wrong closing delegate. [3] and wrong date and time of closure. I have fixed the ones from today and yesterday, but I notice that this has been a problem for some time. The bot saves a lot of time and tedium and it's really appreciated. It would be great to get it fixed. Best wishes, Graham. Graham Colm (talk) 11:00, 24 March 2013 (UTC)[reply]
In response to Graham, describing articles as lists was an error with the substed template, not with the bot. I fixed the template. About adding the wrong delegate, it adds the most recent delegate who edited the page at the time of its daily run. We thought this would be fine as it is run every day and the list usually isn't updated that often and we had to get this going ASAP. The code is available at my GitHub if anybody would like to implement it. To Tony, the bot doesn't include the oldid. That can be implemented by anybody and then making a push request at GitHub. Vacation913:11, 25 March 2013 (UTC)[reply]
Thanks for the detailed explanation, and thanks again for your help. 19:41, 25 March 2013 (UTC)
The promotion dates seem to be consistent with the old algorithm. I think the promotion date has been historically the date that the delegate closed the discussion rather than the date that the bot updates the page. Unless you choose the right date, it makes it less likely that people will choose the proper oldid number.--TonyTheTiger (T/C/BIO/WP:CHICAGO/WP:FOUR) 15:22, 31 March 2013 (UTC)[reply]
Also, the bot is adding the new T:AH entry at the bottom of the template, it should be added above the general information about the talk page within the template at the bottom of the discussion list.--TonyTheTiger (T/C/BIO/WP:CHICAGO/WP:FOUR) 15:30, 31 March 2013 (UTC)[reply]
To echo Tony, the bot failed to also swap the "GA" flag to an "FA" one on the Marie Lloyd talk page. See here. Not a problem (I changed it), but I wouldn't want the problem to persist. --CassiantoTalk19:40, 21 May 2013 (UTC)[reply]
This seems like a really great project that I would be happy to take on when I have the time. That being said, when I have the time. Don't expect it to be done right away, as I'm getting more and more busy. I'll put it in my priority list though. How exactly do you want the requests archived? Vacation903:29, 5 May 2013 (UTC)[reply]
The same way that FAC works, I think, is the way GimmeBot (talk·contribs) used to do it. Just check the logs and the WP:FPORTC page, see which ones were promoted/archived, and update those discussion subpages, add the star to the portal main page, and update the ArticleHistory on the portal talk page. You can see an example by GimmeBot (talk·contribs), with the recently promoted Portal:Society. Thank you very much, — Cirt (talk) 03:48, 5 May 2013 (UTC)[reply]
Reply to VoxelBot about Article notability notification left on my talk page[edit]
On 24 May 2013 at 22:19, VoxelBot left the following listing on my talk page:
== Article notability notification ==
Hello. This message is to inform you that an article that you wrote, Only Time Knows, has been recently tagged with a notability notice. This means that it may not meet Wikipedia's notability guidelines. Please note that articles which do not meet these criteria may be merged, redirected, or deleted. Please consider adding reliable, secondary sources to the article in order to establish the topic's notability.
First off, I want to say there really ought to be a better way to respond.
Second, although the article may not meet the impossibly high notability guideline, it has plenty of secondary sources (five independent, one liner notes, and one by their current publisher), and I have added several more since the notability notice was added. Did you or any uber-editor re-review the article, or only notice that at one point a notability notice was added.
Per the notability guidelines, it does not matter if the band is notable, but only if each album is notable. Since this was the band's first album and self-published, it can be argued that the album is not notable. However, I can find many other albums that fall into that category. Dixie Chicks first album Thank Heavens for Dale Evans is a very similar article to Only Time Knows and has way less references. The album is only notable because of who the Dixie Chicks are, and not for the album. That argument also applies to Bearfoot's first album. While Bearfoot is not as successful as the Dixie Chicks, in the Bluegrass scene, their fourth album was #1 in the Billboard bluegrass charts in 2009 and their fifth album was #7 most played bluegrass album in 2012, which means the band is reasonably notable. So that brings us to the question of when an album article should be allowed. If the Dixie Chicks album Thank Heavens for Dale Evans is not marked for notability failure, that seems to indicate a bias on the part of the uber-editors or a work over-load. In addition, there are all the EPs by The Cranberries, none of which are notable, and all of whose articles are simpler than for Only Time Knows.
If needed, I can find a hundred more albums that are not notable that have articles (not hyperbole), but the fact is that I don't want any album article to be deleted. Album articles are necessary to gather facts and details about the album in one easy to reference place, Wikipedia. If all the article did was to provide track names, that would be one thing, but a good album article lists track times, and who wrote each song, and the performers for the album.
You could put that information into the band article, but doing that for five albums would be clumsy, and doing it for just one album would be odd. Therefore, leave the album article alone, delete the notability notice, and move on.
However, if you do decide to delete the article, who may I appeal to about the biased way the notability notice is applied, or even appeal about my opinion that every album article is necessary?
There was a discussion in the Notability (music) talk page on merging the merging of non-notable albums (page 14 of the archive). At the end of the discussion, one commenter posted this "Suggestion - Another option would be to delete the album article, add the basic album info in the artist or artist discog page, and provide a reference (such as Allmusic) that has the track listing and other detail. Happy new year! GoingBatty (talk) 06:09, 2 January 2013 (UTC)". This is a poor suggestion as AllMusic is sometime incorrect about song composers, and appears to have stopped listing song composers for album tracks. This means that Wikipedia is the correctable proper source for album information, and AllMusic is only of use as a limited reference source.
Third, I wouldn't be using the bullets at the front of each paragraph, but Wikipedia merged all the paragraphs into one big block, so if there is a better way to separate paragraphs, I apologize for not using it. Mburrell (talk) 01:57, 25 May 2013 (UTC)[reply]
It's possible WMFLabs is acting up, but just looking at VBot's edits this morning it seems like it closed them. Let me know if this happens again. Vacation917:20, 8 June 2013 (UTC)[reply]
{{WikiProject Military history|class = GA |importance = high |African=yes |African-importance = top |Biography=yes}}
However, the bot did not pick up on this and change it to FA when it edited this page recently. I assume it is because the class parameter has spaces around the = symbol, whereas the others don't. I will make this change manually now, but thought I'd report this as something to consider with the bot. Thanks — Amakuru (talk) 14:06, 23 June 2013 (UTC)[reply]
Unless I'm missing something, the bot has not run on the FAC archived/promoted stuff since 7 July, and there are quite a few articles/FACs that have not been updated. I don't know if there is any particular reason, or if you are aware. Sarastro1 (talk) 19:34, 15 July 2013 (UTC)[reply]
On 8 June 2013 this bot added a featured article template to SMS Thüringen but did not delete the existing good article template.[4] The result was that the corresponding images (specifically File:Symbol_support_vote.svg and File:Cscr-featured.svg) were displayed at the same time and in the same place. I have since solved the problem. However, I'd like to suggest to the operators to look into this bug. Thanks, --190.19.66.81 (talk) 05:23, 22 July 2013 (UTC)[reply]
List of Detroit Red Wings draft picks was promoted to Featured List recently, however when VoxelBot updated the article page it did not update the talk page at the same time (it appears that doing both at the same time is the norm). I would do it manually, however the instructions specifically say not to. Thank you for looking into this. Regards, Rejectwater (talk) 11:05, 25 July 2013 (UTC)[reply]
Hi guys, just letting you know the bot generally seems to be working but there are still odd omissions/inconsistencies:
When God Writes Your Love Story and three other articles were promoted on 31 July but haven't been touched by the bot, yet those promoted in August have been closed/updated as expected.
William Hely was promoted on 28 July and the bot closed the FAC page and added the FA star to the article but didn't update article history.
Priyanka Chopra was promoted by Graham Colm on 16 July but the bot reckoned it was me. I recall this used to happen quite a bit but generally not anymore.
Anything you can do to help... If there's something about a particular article or talk page, pls let me know. Thanks/cheers, Ian Rose (talk) 00:59, 6 August 2013 (UTC)[reply]
I have manually closed all of these today, but I'm concerned that they were missed, especially as there's no obvious pattern: they are from six different dates, and from both promoted and archived FACs. Hope you will look into this so we can get to the root cause. Thanks as always. Maralia (talk) 01:18, 21 August 2013 (UTC)[reply]
"S&M" was promoted three days ago but none of the history has been updated and the star hasn't been added. Articles promoted to FA after "S&M" have been updated and the nomination archive, but "S&M"'s hasn't. — ₳aron14:36, 15 August 2013 (UTC)[reply]
@Vacation9: The bot frequently stalls out after adding the FA star but before updating article talk. This has been an ongoing problem. I've put together a list of examples that may help in debugging:
There appears to be a code problem -- the bot is still running, but failing to run. I'll ping Vacation9 on IRC, and in the meantime, I'll look at the code myself. Thanks for letting us know :) It's a Fox!(What did I break)22:43, 5 December 2013 (UTC)[reply]
What I mean is that the bot isn't updating the template every 30 minutes like it's supposed to. It updates around 1 AM - 1:30 AM UTC instead of every half an hour like it's supposed to. Human editors need to update the template manually. Any information on how this is going to be fixed? K6ka (talk | contribs) 18:33, 3 April 2014 (UTC)[reply]
It looks like it was not updating it because the level it had cached was level 4. It only makes changes to the template if the level changes. I removed the level cache and it should be working again. Vacation915:16, 27 April 2014 (UTC)[reply]
hey how could I have a bot on the page List of Turkic dynasties and countries Turkic_ Warrior 22:01, 7 December 2014 (UTC) — Preceding unsigned comment added by Mehmeett21 (talk • contribs)