User talk:Cyberpower678/Archive 45
 | This is an archive of past discussions with User:Cyberpower678. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
| Archive 40 | ← | Archive 43 | Archive 44 | Archive 45 | Archive 46 | Archive 47 | → | Archive 50 |
IABot dead link fix
Is it possible to have InternetArchiveBot fix dead links like this? --bender235 (talk) 18:00, 13 February 2017 (UTC)
- I'm not sure. I need more details. archive.org is not recognized as an archive. Just web.archive.org.—CYBERPOWER (Be my Valentine) 18:02, 13 February 2017 (UTC)
- Every item at IA (non-Wayback, the audio, texts or video archives) has a work page URL in this format: (https://archive.org/details/manualofconcholo111tryo) using the work ID (manualofconcholo111tryo). A work is composed of multiple files and it's possible to link to a specific file within a work (http://ia700307.us.archive.org/4/items/manualofconcholo111tryo/manualofconcholo111tryo.pdf) but in doing so you also get the machine ID where it's currently hosted (ia700307). If the work is moved internally to a different machine the link will break. That's why it's never a good idea to link to specific files at IA rather the main work URL which stays the same no matter what machine it's hosted on. In cases where a link uses a machine ID URL instead of a generic work URL, and the link is dead, it would be a good idea to convert the link to the generic work URL (which can be created easily using the work ID). -- GreenC 19:42, 13 February 2017 (UTC)
- Still a little lost. Can you elaborate, with examples? IABot isn't designed to run with work URLs but with web snapshots. If it isn't a snapshot, then IABot can't handle it and is treated as a standard URL.—CYBERPOWER (Be my Valentine) 19:47, 13 February 2017 (UTC)
- No problem understood, it's totally different then dealing with wayback. I'll probably add it to WaybackMedic. It's something I've wanted to do anyway for a while. -- GreenC 19:54, 13 February 2017 (UTC)
- That sounds good, thanks. I thought about doing it manually, but it seems to be too much work. --bender235 (talk) 22:17, 13 February 2017 (UTC)
- @Bender235: - The code is done: [1][2]. It only converts if the link tests dead. In the future I may target articles containing machine ID links but for now it just fixes them incidentally as it comes across while doing other work. In a sample of 10,000 articles it found 16 articles... so if there are 5 million articles there will be roughly 8000. -- GreenC 22:29, 16 February 2017 (UTC)
- Still, potentially 8,000 dead links less. Good work. --bender235 (talk) 23:00, 16 February 2017 (UTC)
- bender235, stats showing > 50% dead rate (ie. of all such links > 50% are dead). I'm thinking it would be OK to convert all of these (dead or not) under the assumption sooner than later they are going to be dead anyway. IA infrastructure is dynamic and linking to a specific machine ID is guaranteed to fail as machines get cycled out of production for hardware failure, hardware and software upgrades, network reconfiguration, etc.. machine ID links are good for short term only. -- GreenC 14:34, 17 February 2017 (UTC)
- Started a BRFA since it's not specific to Wayback and will edit a lot of articles. Wikipedia:Bots/Requests_for_approval/GreenC_bot_4. -- GreenC 01:18, 18 February 2017 (UTC)
- bender235, stats showing > 50% dead rate (ie. of all such links > 50% are dead). I'm thinking it would be OK to convert all of these (dead or not) under the assumption sooner than later they are going to be dead anyway. IA infrastructure is dynamic and linking to a specific machine ID is guaranteed to fail as machines get cycled out of production for hardware failure, hardware and software upgrades, network reconfiguration, etc.. machine ID links are good for short term only. -- GreenC 14:34, 17 February 2017 (UTC)
- Still, potentially 8,000 dead links less. Good work. --bender235 (talk) 23:00, 16 February 2017 (UTC)
- @Bender235: - The code is done: [1][2]. It only converts if the link tests dead. In the future I may target articles containing machine ID links but for now it just fixes them incidentally as it comes across while doing other work. In a sample of 10,000 articles it found 16 articles... so if there are 5 million articles there will be roughly 8000. -- GreenC 22:29, 16 February 2017 (UTC)
- That sounds good, thanks. I thought about doing it manually, but it seems to be too much work. --bender235 (talk) 22:17, 13 February 2017 (UTC)
- No problem understood, it's totally different then dealing with wayback. I'll probably add it to WaybackMedic. It's something I've wanted to do anyway for a while. -- GreenC 19:54, 13 February 2017 (UTC)
- Still a little lost. Can you elaborate, with examples? IABot isn't designed to run with work URLs but with web snapshots. If it isn't a snapshot, then IABot can't handle it and is treated as a standard URL.—CYBERPOWER (Be my Valentine) 19:47, 13 February 2017 (UTC)
- Every item at IA (non-Wayback, the audio, texts or video archives) has a work page URL in this format: (https://archive.org/details/manualofconcholo111tryo) using the work ID (manualofconcholo111tryo). A work is composed of multiple files and it's possible to link to a specific file within a work (http://ia700307.us.archive.org/4/items/manualofconcholo111tryo/manualofconcholo111tryo.pdf) but in doing so you also get the machine ID where it's currently hosted (ia700307). If the work is moved internally to a different machine the link will break. That's why it's never a good idea to link to specific files at IA rather the main work URL which stays the same no matter what machine it's hosted on. In cases where a link uses a machine ID URL instead of a generic work URL, and the link is dead, it would be a good idea to convert the link to the generic work URL (which can be created easily using the work ID). -- GreenC 19:42, 13 February 2017 (UTC)
On your article about Blackberry (RIM), reference number 79 has a bad url. It should be http://www.theglobeandmail.com/report-on-business/patently-absurd/article18153839/ (24.80.208.185 (talk) 22:36, 17 February 2017 (UTC))
- It's not my article, and I'm not sure what you want me to do.—CYBERPOWER (Chat) 22:41, 17 February 2017 (UTC)
Bot resetting RPP's "Rolling Archive" subpage (again)
The bot continually resets the weekly past requests. I don't know whether the bot can be simply fixed to properly archive the protection requests. I wonder whether those at WP:VPT can handle it. --George Ho (talk) 01:45, 19 February 2017 (UTC)
Actions
Greetings cyberpower678, I am a member of Wikipedia-pt and I have a doubt. Accessing here in the part that has Admin actions: Delete, what would this delete be?--Agent010 (talk) 10:19, 17 February 2017 (UTC)
- I would not be able to tell you that as I don't know myself.—CYBERPOWER (Around) 13:59, 17 February 2017 (UTC)
- @Agent010: It simply fetches the number of entries in your delete log where you're the performer. --QEDK (愛) 19:07, 19 February 2017 (UTC)
- @QEDK: Okay, but I'm not Eliminator, and this option is for those who delete pages (Eliminator)--Agent010 (talk) 19:15, 19 February 2017 (UTC)
- @Agent010: The logs don't necessarily work that way, I moved a protected page, so now I have one entry in my protected log but it doesn't mean I have the rights to protect (which I don't btw). Just check your deletion log, you should have the answer. --QEDK (愛) 20:31, 19 February 2017 (UTC)
- @QEDK: Okay, but I'm not Eliminator, and this option is for those who delete pages (Eliminator)--Agent010 (talk) 19:15, 19 February 2017 (UTC)
- @Agent010: It simply fetches the number of entries in your delete log where you're the performer. --QEDK (愛) 19:07, 19 February 2017 (UTC)
Cyberbot I: RFPP clerking not working?
There are currently 56 requests on WP:RFPP to archive, Cyberbot I has not clerked the page since yesterday though the task is enabled. Can you have a look? Regards SoWhy 14:14, 22 February 2017 (UTC)
Book report stuck?
On Book talk:U.S. Roads quality articles, the book report hasn't been updated since July 31, 2016, and it's missing most of the assessments. Thoughts? Imzadi 1979 → 21:59, 25 February 2017 (UTC)
- That's likely because the book isn't made in the proper format. See Help:Books/for experts#Saving books on what the proper format it. In particular, you use headers for chapters, and the non-standard [[: also likely interferes with the bot. Headbomb {talk / contribs / physics / books} 23:45, 25 February 2017 (UTC)
Adminstats stuck
So, my adminstats seems stuck on December 8. Is this something I did, or is this something on the bot's end? Ks0stm (T•C•G•E) 17:19, 6 February 2017 (UTC)
- Dug out of your talk archive, since the bot automatically archived it. Ks0stm (T•C•G•E) 12:17, 10 February 2017 (UTC)
- Sorry for the late response. I have no idea why it's stuck. It seems to crash after a few users. I'll have to investigate. Can you open a phab ticket if you can and assign me to it?
- Dug this out one more time. I have filed the task and assigned it to you. T158897. Ks0stm (T•C•G•E) 19:28, 23 February 2017 (UTC)
- This will help me to remember to look into it —CYBERPOWER (Chat) 09:16, 26 February 2017 (UTC)
- Dug this out one more time. I have filed the task and assigned it to you. T158897. Ks0stm (T•C•G•E) 19:28, 23 February 2017 (UTC)
- Sorry for the late response. I have no idea why it's stuck. It seems to crash after a few users. I'll have to investigate. Can you open a phab ticket if you can and assign me to it?
Cyberbot RFPP archiving
On occasion, Cyberbot strips away huge chunks of the "rolling" RFPP archiving when adding entries from the main page. I assume since that it's called a rolling archive, this shouldn't be happening. [3] — Train2104 (t • c) 23:21, 25 February 2017 (UTC)
Cyberpower, may I take this to WP:VPT? Seems that the bot needs more than just a simple fix. --George Ho (talk) 02:49, 26 February 2017 (UTC)
- I doubt you'll get anything helpful there.—CYBERPOWER (Chat) 09:18, 26 February 2017 (UTC)
Why does this bot do this?
I don't understand this. Why did the bot mark this "|deadurl=yes"? The url is working fine. https://en.wikipedia.org/w/index.php?title=Steve_Israel&type=revision&diff=767481922&oldid=767478114 --Nbauman (talk) 05:27, 26 February 2017 (UTC)
- You completely modified the URL. You can't pass this off as a bot bug.—CYBERPOWER (Chat) 09:19, 26 February 2017 (UTC)
Webcite API loops
Try this: http://www.webcitation.org/query?id=5ZprY4O4g&returnxml=true (from a browser it might hang but curl or wget) -- GreenC 16:45, 22 February 2017 (UTC)
- I fix this by piping it through head -c 1024 though I don't know what the fix might be in PHP curl library. -- GreenC 18:30, 23 February 2017 (UTC)
There is another potential WebCite API bug: this works, this doesn't. The second should return the API data but returns the page data. Seems like returnxml=true has to be the first argument with url= queries. -- GreenC 18:30, 23 February 2017 (UTC)
- Can you clarify?—CYBERPOWER (Chat) 09:15, 26 February 2017 (UTC)
- In API.php line #1604-1605 it looks like "returnxml=true" is being added to the end of the URL which is a source of problem per the example given above. "returnxml=true" should be the first argument of the URL when using the query "url=" to retrieve API data . This is confirmed in the WebCite documentation docs page 6, subsection "XML responses", first example.. "returnxml=true" comes first before the "url" argument. Normally the placement of arguments within a URL shouldn't matter, but for WebCite API using the "url=" query it is significant. -- GreenC 17:47, 26 February 2017 (UTC)
- Their web servers are rather sensitive I would say.—CYBERPOWER (Chat) 02:32, 27 February 2017 (UTC)
- Might be a hack-technique to deal with encoding ie. anything beyond the "url=" is the URL to look up regardless of encoding. They are also sensitive about encoding .. for example the API won't always find a URL if it was saved non-encoded and the API request is encoded. I do two lookups to make sure. But the encoded version can't be fully encoded either ie. not / and : .. it's a complicated mess. Archive.is has similar encoding issues. -- GreenC 15:44, 27 February 2017 (UTC)
- Their web servers are rather sensitive I would say.—CYBERPOWER (Chat) 02:32, 27 February 2017 (UTC)
- In API.php line #1604-1605 it looks like "returnxml=true" is being added to the end of the URL which is a source of problem per the example given above. "returnxml=true" should be the first argument of the URL when using the query "url=" to retrieve API data . This is confirmed in the WebCite documentation docs page 6, subsection "XML responses", first example.. "returnxml=true" comes first before the "url" argument. Normally the placement of arguments within a URL shouldn't matter, but for WebCite API using the "url=" query it is significant. -- GreenC 17:47, 26 February 2017 (UTC)
Administrators' newsletter – March 2017
News and updates for administrators from the past month (February 2017).
Administrator changes
Amortias • Deckiller • BU Rob13
Ronnotel • Islander • Chamal N • Isomorphic • Keeper76 • Lord Voldemort • Shereth • Bdesham • Pjacobi
Guideline and policy news
- A recent RfC has redefined how articles on schools are evaluated at AfD. Specifically, secondary schools are not presumed to be notable simply because they exist.
- AfDs that receive little participation should now be closed like an expired proposed deletion, following a deletion process RfC.
- Defender, HakanIST, Matiia and Sjoerddebruin are our newest stewards, following the 2017 steward elections.
- The 2017 appointees for the Ombudsman commission are Góngora, Krd, Lankiveil, Richwales and Vogone. They will serve for approximately 1 year.
Technical news
- A recent query shows that only 16% of administrators on the English Wikipedia have enabled two-factor authentication. If you haven't already enabled it please consider doing so.
- Cookie blocks should be deployed to the English Wikipedia soon. This will extend the current autoblock system by setting a cookie for each block, which will then autoblock the user after they switch accounts under a new IP.
- A bot will now automatically place a protection template on protected pages when admins forget to do so.
Cyberbot I - AdminStats
Hey, C678, I just found an error in the way the bot determines userrights when it updates the individual stats page for each admin. I noticed that mine is now labeled "Crat", most likely because I became a global renamer at the end of January, and since it sees that I have done some renames, it assumes that I'm a crat now. Not that it really matters to me, but I wouldn't want anyone to get the wrong idea from seeing that heading on my stats. Cheers —DoRD (talk) 16:55, 2 March 2017 (UTC)
- Actually that's the template itself. It can be fixed on wiki.—CYBERPOWER (Message) 17:32, 2 March 2017 (UTC)
- Well, right you are. I guess I didn't drill down far enough to see that. Thanks —DoRD (talk) 17:41, 2 March 2017 (UTC)
Taking the RPP archiving issue to a noticeboard
I wonder whether Wikipedia:Bot owners' noticeboard will do something about your bot's archiving protection requests. Thoughts? --George Ho (talk) 01:12, 5 March 2017 (UTC)
- They will most likely defer you back to me. I'm still looking into the cause.—CYBERPOWER (Chat) 07:48, 5 March 2017 (UTC)
A barnstar for you
|
The da Vinci Barnstar | |
| Adding one more to the pile of thanks you've undoubtedly received for creating InternetArchiveBot, which is simply awesome! --Middle 8 (t • c | privacy • COI) 10:27, 9 March 2017 (UTC) |
Cyberbot thinks I'm no longer an admin?
Hi there. I was surprised to see this edit today to my Adminstats subpage. I'm pretty sure my mop is still where I left it. Regards SoWhy 08:21, 9 March 2017 (UTC)
- Me too. --Redrose64 🌹 (talk) 11:04, 9 March 2017 (UTC)
- Sorry. It got a little tipsy last night and couldn't read right.—CYBERPOWER (Message) 11:06, 9 March 2017 (UTC)
df
In this edit the date format was changed from dmy to iso - that will cause problems in Good and Featured content which requires consistent date formatting. -- GreenC 18:31, 6 March 2017 (UTC)
- The tag that instructs bots and tools on formatting is missing. Without it the bot generally has a hard time determining what format to use.—CYBERPOWER (Message) 18:47, 6 March 2017 (UTC)
- A function that determines the existing format is fairly easy for 99% of cases. If it still can't determine default to iso. -- GreenC 19:01, 6 March 2017 (UTC)
- It tends to already do that. If it can't, it defaults to ISO.—CYBERPOWER (Message) 19:27, 6 March 2017 (UTC)
- Here.—CYBERPOWER (Message) 19:31, 6 March 2017 (UTC)
- What do you think happened? Regex looks OK. Don't know what data is contained in "$default" for this case but that's probably why. -- GreenC 20:39, 6 March 2017 (UTC)
- That cite template is missing the access date parameter. Since most cite templates it touches don't have archive data attached to it, it never relies on that to get timestamp formatting data. So it uses the access date parameter instead. With it missing it can't get a format.—CYBERPOWER (Message) 21:38, 6 March 2017 (UTC)
- It really should respect the existing format of
|archive-date=. In 4 manual runs of IABot it changed the date 2 times. In 1 million runs it can happen 100s of thousands of times. Can "$default" be the full template and the name of the field name in question. If the field exists it parses it out and returns the date format. If it doesn't exist checks access-date and date then fall back to iso. -- GreenC 14:04, 7 March 2017 (UTC)- Please open a phab ticket, as usual. I think I know how to address this.—CYBERPOWER (Chat) 23:48, 8 March 2017 (UTC)
- T160174. -- GreenC 15:15, 10 March 2017 (UTC)
- I'm sorry, but the ticket is locked. I can't do anything with it.—CYBERPOWER (Chat) 18:50, 10 March 2017 (UTC)
- I don't understand. There are no lock or unlock options and it looks the same as every other. It says "Open" and "Public". Should I post for help? --GreenC 20:45, 10 March 2017 (UTC)
- I added "Assigned: Cyberpower 678" .. does it help? -- GreenC 20:49, 10 March 2017 (UTC)
- The problem may be using "InternetArchiveBot (v1.3)" for the tag which doesn't automatically assign to Cyberpower678. -- GreenC 20:54, 10 March 2017 (UTC)
- I'm sorry, but the ticket is locked. I can't do anything with it.—CYBERPOWER (Chat) 18:50, 10 March 2017 (UTC)
- T160174. -- GreenC 15:15, 10 March 2017 (UTC)
- Please open a phab ticket, as usual. I think I know how to address this.—CYBERPOWER (Chat) 23:48, 8 March 2017 (UTC)
- It really should respect the existing format of
- That cite template is missing the access date parameter. Since most cite templates it touches don't have archive data attached to it, it never relies on that to get timestamp formatting data. So it uses the access date parameter instead. With it missing it can't get a format.—CYBERPOWER (Message) 21:38, 6 March 2017 (UTC)
- What do you think happened? Regex looks OK. Don't know what data is contained in "$default" for this case but that's probably why. -- GreenC 20:39, 6 March 2017 (UTC)
- A function that determines the existing format is fairly easy for 99% of cases. If it still can't determine default to iso. -- GreenC 19:01, 6 March 2017 (UTC)
User:InternetArchiveBot screw up
As a side note, this really shouldn't have been done. I've been waiting for 10+minutes to get a confirmation email, and I still haven't received one. Most people will probably just go "this is a hassle, fuck it" and move on without reporting bugs. Talk page protection should be removed per WP:BOTCOMM. Headbomb {talk / contribs / physics / books} 13:59, 11 March 2017 (UTC)
- ??? Huh? What confirmation email are you referring to in regards to reporting a bug. There's an instruction page on how to report the bug on Phabricator, and what to include for effective bug reporting. Besides that talk page still links to this talk page which is freely editable.—CYBERPOWER (Chat) 14:20, 11 March 2017 (UTC)
- The first link on the talk page directs me to [4], which required me to create to confirm my email before being able to report bugs. Still waiting on it. Now I get a different page, and it's telling me to go to phabricator (it didn't do that the first time). Anyway, I'm done jumping through hoops for this one. It's an old edit, so maybe it's fixed now, but I though I'd point it out. Headbomb {talk / contribs / physics / books} 14:51, 11 March 2017 (UTC)
- Being discussed on IRC.—CYBERPOWER (Chat) 15:45, 11 March 2017 (UTC)
Alt archive bug
There appears to be a fairly serious bug that should receive attention soon. It's reported for T160175, T159975 and T159833 .. those are examples, it likely applies to any archive service outside the core (wayback, webcite, archive.is and a few others). The bug is urgent because there is no way to recover from it automatically (without great difficulty) and it's even somewhat difficult to detect. But it causes a lot of damage by effectively deleting archive URLs. -- GreenC 15:14, 10 March 2017 (UTC)
- User:Green_Cardamom/links for archive links that are currently in Wikipedia or could be due to Memento API injections. This is based on a search of Wikipedia how links appear, and Memento documentation. -- GreenC 17:34, 11 March 2017 (UTC)
Wikipedia:Contact us
Can I contest the cascading protection on this page? It certainly is visible enough but not that frequently trafficked in comparison to the main page. And it forces admins to edit all the templates that are currently used on it when an edit request occurs. Jo-Jo Eumerus (talk, contributions) 13:08, 12 March 2017 (UTC)
InternetArchiveBot errors
Hey, there. InternetArchiveBot incorrectly flagged two URLs on the Brian McClendon page as dead links. They aren't -- they're links to the US Patent Office database. I've fixed the errors. I tried to follow the instructions on the bot's talk page to report false positives, but all the options link me out to https://tools.wmflabs.org which requires a login which I don't have, and I have no interest in setting up yet another account on yet another site.
I don't know who owns this bot, but the last option on the bot's talk page is to leave a message here, so here I am. (Why would anybody let loose a bot and then not make it simple/easy to report problems with it? Why???) Thanks Bookgrrl holler/lookee here 17:44, 12 March 2017 (UTC)
- The bot is mine and you don't create an account, you login with your Wikipedia account. It's a simple 3 second process.—CYBERPOWER (Chat) 17:49, 12 March 2017 (UTC)
- @Bookgrrl: ^. Click the login button on the tool and you'll see how easy it easy to get started.—CYBERPOWER (Chat) 18:55, 12 March 2017 (UTC)
AFD warning
Hello, I removed an AFD notice for an article discussion that had already been resolved, but received a bot warning. What do I need to do to avoid this in the future? Ies (talk) 18:47, 12 March 2017 (UTC)
- The discussion is still open. You handled it incorrectly.—CYBERPOWER (Chat) 18:49, 12 March 2017 (UTC)
- I introduced the initial AFD, and felt the reasons for my introduction had been addressed and a consensus established (it was a peripheral article and I did not anticipate anyone further discussion). What additional steps need to be taken? Is there a minimum time that an AFD notice needs to remain in an article or a minimum number of votes? Ies (talk) 18:52, 12 March 2017 (UTC)
- The AfD needs to be formally closed by an experienced editor. You can withdraw it, but given there's another delete vote there, I doubt it would be appropriate. Once the discussion is formally closed, the tag can be removed.—CYBERPOWER (Chat) 18:54, 12 March 2017 (UTC)
- That explanation seems a bit arbitrary -- ie. at what point is someone experienced enough to close an AfD? 20k edits? 5 years? Also, the asymmetry of being allowed to introduce AfDs at earlier experience levels but not being allowed to remove them unless experienced is difficult to understand. No big deal, will keep in mind. Ies (talk) 19:06, 12 March 2017 (UTC)
- Typically, when you figure out how to close an AfD, then you become experienced at closing AFDs. Read Wikipedia:CLOSEAFD to get a start.--v/r - TP 22:08, 12 March 2017 (UTC)
- Thank you, very helpful. Ies (talk) 04:20, 13 March 2017 (UTC)
- Typically, when you figure out how to close an AfD, then you become experienced at closing AFDs. Read Wikipedia:CLOSEAFD to get a start.--v/r - TP 22:08, 12 March 2017 (UTC)
- That explanation seems a bit arbitrary -- ie. at what point is someone experienced enough to close an AfD? 20k edits? 5 years? Also, the asymmetry of being allowed to introduce AfDs at earlier experience levels but not being allowed to remove them unless experienced is difficult to understand. No big deal, will keep in mind. Ies (talk) 19:06, 12 March 2017 (UTC)
- The AfD needs to be formally closed by an experienced editor. You can withdraw it, but given there's another delete vote there, I doubt it would be appropriate. Once the discussion is formally closed, the tag can be removed.—CYBERPOWER (Chat) 18:54, 12 March 2017 (UTC)
- I introduced the initial AFD, and felt the reasons for my introduction had been addressed and a consensus established (it was a peripheral article and I did not anticipate anyone further discussion). What additional steps need to be taken? Is there a minimum time that an AFD notice needs to remain in an article or a minimum number of votes? Ies (talk) 18:52, 12 March 2017 (UTC)
Rename request
Hi, I would like to change my username as it is my real name and I'd like to change it from 'Sidharthmohandas' to 'TMM1980'. — Preceding unsigned comment added by Sidharthmohandas (talk • contribs) 16:48, 13 March 2017 (UTC)
Done I have started the rename process. It should finish in a few minutes. You can then login in under the new name.—CYBERPOWER (Chat) 16:51, 13 March 2017 (UTC)
Bot
When will the InternetArchiveBot start again? Or will it not be used anymore. Regards,--BabbaQ (talk) 21:36, 13 March 2017 (UTC)
- And, will the bot be available on Swedish Wikipedia at any time? Or has that idea been scrapped :).--BabbaQ (talk) 21:39, 13 March 2017 (UTC)
- (edit conflict)Don't worry, it's not gone. I'm just hammering out bugs and doing limited testing before fully deploying it again. Feel free however to use this tool in the meantime to fix articles you want fixed. This tool has a direct link in the page history of articles now. Look for "fix dead links". Feel free to spread the word of this new tool. I'm sure there are other users itching for a tool like this.—CYBERPOWER (Chat) 21:42, 13 March 2017 (UTC)
- The svwiki deploy hasn't been scrapped, just delayed. In order to ensure less problems in the long run, I'm working on perfecting the code on this wiki first and making the maintenance tools available first. That way this development process and bot reception goes over smoothly on svwiki. The aim is to run this on every Wikipedia when it's done.—CYBERPOWER (Chat) 21:42, 13 March 2017 (UTC)
- Oh, thank you for notifying me of that new tool. I will have good use of that. But I can not find the Fix dead links link in the page history. Take a look at my latest edits with the tool. Or are my eyesight the problem :) --BabbaQ (talk) 15:56, 14 March 2017 (UTC)
- The tool is really helpful :)BabbaQ (talk) 00:09, 15 March 2017 (UTC)
:80
What's the bot policy for ":80" formatting? Here it added the :80 [5] and here it removed it. [6] I'd like Medic to be on the same edit policy so we are not doing many small counter-edits. I don't care either way how it is, though preference is to remove to reduce complexity. -- GreenC 19:43, 15 March 2017 (UTC)
- There's a phab ticktet open on this.—CYBERPOWER (Chat) 19:55, 15 March 2017 (UTC)
- The tool is adding archives to still functional references that are not dead. I guess the tool is not 100% perfect as of yet. Work in progress.--BabbaQ (talk) 21:47, 15 March 2017 (UTC)
- There's always going to be factors that causes the tool to goof. You can use this tool to report those working URLs for me to look into them. List the URLs you found in that text box and submit it.—CYBERPOWER (Chat) 21:55, 15 March 2017 (UTC)
- @BabbaQ:: I just fixed a bug related to false positives. Give it a go now.—CYBERPOWER (Chat) 22:20, 15 March 2017 (UTC)
- There's always going to be factors that causes the tool to goof. You can use this tool to report those working URLs for me to look into them. List the URLs you found in that text box and submit it.—CYBERPOWER (Chat) 21:55, 15 March 2017 (UTC)
- The tool is adding archives to still functional references that are not dead. I guess the tool is not 100% perfect as of yet. Work in progress.--BabbaQ (talk) 21:47, 15 March 2017 (UTC)
Ping. Didn't know if you'd forgotten about this. Ks0stm (T•C•G•E) 21:06, 14 March 2017 (UTC)
- It's still not working for you?—CYBERPOWER (Chat) 21:09, 14 March 2017 (UTC)
- Nope. Still stuck on December 8. Ks0stm (T•C•G•E) 21:16, 14 March 2017 (UTC)
- It just doesn't like you. :p—CYBERPOWER (Chat) 21:17, 14 March 2017 (UTC)
- Heh maybe. XD I did have a thought, though. Would the bot recreate and start updating the page if it were deleted? I've noticed it does something similar when transcluded adminstats are deleted, but I don't know if it would work when it's stuck like that. Ks0stm (T•C•G•E) 21:23, 14 March 2017 (UTC)
- I'm doing a run of it on my terminal right now to see what the problem is.—CYBERPOWER (Chat) 21:29, 14 March 2017 (UTC)
- Well, apparently no luck getting it to recreate the template page. Were you able to figure out what the problem was? Ks0stm (T•C•G•E) 20:12, 15 March 2017 (UTC)
- It's not acknowledging the template being transcluded on your page.—CYBERPOWER (Chat) 21:01, 15 March 2017 (UTC)
- Well, apparently no luck getting it to recreate the template page. Were you able to figure out what the problem was? Ks0stm (T•C•G•E) 20:12, 15 March 2017 (UTC)
- I'm doing a run of it on my terminal right now to see what the problem is.—CYBERPOWER (Chat) 21:29, 14 March 2017 (UTC)
- Heh maybe. XD I did have a thought, though. Would the bot recreate and start updating the page if it were deleted? I've noticed it does something similar when transcluded adminstats are deleted, but I don't know if it would work when it's stuck like that. Ks0stm (T•C•G•E) 21:23, 14 March 2017 (UTC)
- It just doesn't like you. :p—CYBERPOWER (Chat) 21:17, 14 March 2017 (UTC)
- Nope. Still stuck on December 8. Ks0stm (T•C•G•E) 21:16, 14 March 2017 (UTC)
- @Ks0stm: I'm not one for cryptic section titles, and this is a good example. If I had realised what it was about, I could have helped - or at least directed you to this thread. --Redrose64 🌹 (talk) 09:03, 16 March 2017 (UTC)
List of archives on Wikipedia
Wikipedia:List of archives on Wikipedia. All but the ones in the "other" section have been fully incorporated in Medic and the pages containing the links processed. Only took a few weeks of work ugh :/ If you see anything else please feel free to add or edit the page as a central place as new info comes to light. IABot doesn't need to fully support them all, but at a minimum it should know the domains and not delete them. Also, I found about 13,000 links in |archiveurl= that are not web archives .. editors seem to dump anything there. -- GreenC 17:10, 16 March 2017 (UTC)
- Yea, I'll get to it when I can. There's lots of work to be done to the bot. My latest update to the code forces validation of URLs in the archive field. It won't delete them, but it will ignore them, and overwrite them if it needs to have an archive attached to it. However it will have the unfortunate side effect of wiping out archives not recognized by the bot.—CYBERPOWER (Chat) 18:09, 16 March 2017 (UTC)
- What can we do to prevent the bot from deleting archive links? I have simple regex statements written and tested for these. Would it be possible to add a check to skip any links that don't match? -- GreenC 18:31, 16 March 2017 (UTC)
Pandora etc
Hi, I disabled because alt archives are getting misdone per T161162 and possibly others. What would you think about doing a full test cases page with every link variety listed at here in different formats (cites, bare link, webarchive) inside and outside refs. They are hard to spot when things go wrong due to relatively rarity. -- GreenC 22:03, 22 March 2017 (UTC)
- I think that would be a fantastic idea and the tool makes it easy to target the bot code on any page you want. When I implemented the archive, I tested this on a single URL, which seemed to work no matter how I formatted it. I didn't think that others wouldn't be able to handle the same kind of reformatting. :/. Back to the drawing board.—CYBERPOWER (Chat) 22:06, 22 March 2017 (UTC)
- @Green Cardamom: If you can do me the favor, and have your bot revert the edits that modified pandora links. It shouldn't be too many. I only started the bot a half an hour ago.—CYBERPOWER (Chat) 22:08, 22 March 2017 (UTC)
It made 1124 edits of which 4 contained "pandora.nla.gov.au".
-- GreenC 01:01, 23 March 2017 (UTC)
- That's not so bad. Thanks for checking. :-)—CYBERPOWER (Chat) 01:03, 23 March 2017 (UTC)
Tool
Hi, the ref tool seems to not register dead links that are not already tagged with the Dead link tag. I have noticed it several times. Other than that, the tool is great.--BabbaQ (talk) 21:20, 22 March 2017 (UTC)
- You're going to need to give me an example.—CYBERPOWER (Chat) 21:53, 22 March 2017 (UTC)
- By the way, now I have found the tool in the display that you mentioned. Good! I must check my eyesight ;) No, lets just leave it, it is cosmetic at best. The tool is working great overall. It must be complicated to set tools like that up and working.--BabbaQ (talk) 20:41, 23 March 2017 (UTC)
- I've been developing the code behind these tools and IABot for almost 2 years now to put it into perspective. ;-) This is the code that drives it all. Feel free to have a look.—CYBERPOWER (Chat) 21:54, 23 March 2017 (UTC)
- But just in the name of wanting to help you improving the Bot for the future. For example, Idol 2007 (Sweden),[7] the bot did not (search) for the dead link. So I had to add a Dead link tag and then run the bot again. No big problem but better to bring it up :) The same occured at the article Lolo Ferrari (feeling ashamed that I edit such articles:) ) Were I had to tag 3 dead links with the tag and then use the bot as the bot did not register them. --BabbaQ (talk) 01:13, 24 March 2017 (UTC)
- Oh, it looked complicated. I do admire your stamina for this code work.. Two years! :)--BabbaQ (talk) 01:17, 24 March 2017 (UTC)
- The bot may not see it as dead yet. When it encounters a dead link, it doesn't immediately get marked as dead. It checks to make this isn't a temporary problem.—CYBERPOWER (Chat) 01:21, 24 March 2017 (UTC)
- Ok, I see. Well, then I know so I will just leave it and let the bot work in due course.--BabbaQ (talk) 01:22, 24 March 2017 (UTC)
- The bot may not see it as dead yet. When it encounters a dead link, it doesn't immediately get marked as dead. It checks to make this isn't a temporary problem.—CYBERPOWER (Chat) 01:21, 24 March 2017 (UTC)
- Oh, it looked complicated. I do admire your stamina for this code work.. Two years! :)--BabbaQ (talk) 01:17, 24 March 2017 (UTC)
- But just in the name of wanting to help you improving the Bot for the future. For example, Idol 2007 (Sweden),[7] the bot did not (search) for the dead link. So I had to add a Dead link tag and then run the bot again. No big problem but better to bring it up :) The same occured at the article Lolo Ferrari (feeling ashamed that I edit such articles:) ) Were I had to tag 3 dead links with the tag and then use the bot as the bot did not register them. --BabbaQ (talk) 01:13, 24 March 2017 (UTC)
- I've been developing the code behind these tools and IABot for almost 2 years now to put it into perspective. ;-) This is the code that drives it all. Feel free to have a look.—CYBERPOWER (Chat) 21:54, 23 March 2017 (UTC)
- By the way, now I have found the tool in the display that you mentioned. Good! I must check my eyesight ;) No, lets just leave it, it is cosmetic at best. The tool is working great overall. It must be complicated to set tools like that up and working.--BabbaQ (talk) 20:41, 23 March 2017 (UTC)
Error
I issued an AfD on OlOne, which got speedied by HighKing. I then retracted the AfD, since the CSD seemed more appropriate and did not wish to clutter the AfD-list needlessly. I should have closed the AfD, admittedly, byt issueing warnings per bot seems a bit too drastic. Kleuske (talk) 14:19, 27 March 2017 (UTC)
Wrong Warning issued by bot
I removed WP:PROD and issued WP:CSD for Ayesha Jahanzeb. Please fix your bot in issuing warning for such cases and it would be great if you could rectify your warning message on my talk page Bishal revenger (talk) 04:52, 31 March 2017 (UTC)
- You opened an AfD. in failing to close it first, procedure was not followed. All open AfDs, will trigger the bot to replace the missing tag on the respective article. Nothing to rectify here.—CYBERPOWER (Chat) 08:10, 31 March 2017 (UTC)
InternetArchiveBot
What caused this error? Archive date was correct as 11 August 2014. Bot changed it to 13 August. -- Ham105 (talk) 00:22, 31 March 2017 (UTC)
- I have no idea what caused that. It could be how PHP handles dates. I need to investigate. Please open a ticket on Phabricator using these instructions.—CYBERPOWER ( Chat) 08:12, 31 March 2017 (UTC)
Block notice (JOKE)
Hey! I'm on Australia, get me back on Northern Virginia by plane and get me in I-66! Perhaps I'm a a sock of Supreme Genghis Khan and unblock me lol :D [April Fools!] KGirlTrucker81 huh? what I've been doing 22:37, 31 March 2017 (UTC)
- How offensive. I'm a guy. :p—CYBERPOWER (Around) 22:46, 31 March 2017 (UTC)
Administrators' newsletter – April 2017
News and updates for administrators from the past month (March 2017).
Administrator changes
- TheDJ
- Xnuala • CJ • Oldelpaso • Berean Hunter • Jimbo Wales • Andrew c • Karanacs • Modemac • Scott
Guideline and policy news
- Following a discussion on the backlog of unpatrolled files, consensus was found to create a new user right for autopatrolling file uploads. Implementation progress can be tracked on Phabricator.
- The BLPPROD grandfather clause, which stated that unreferenced biographies of living persons were only eligible for proposed deletion if they were created after March 18, 2010, has been removed following an RfC.
- An RfC has closed with consensus to allow proposed deletion of files. The implementation process is ongoing.
- After an unsuccessful proposal to automatically grant IP block exemption, consensus was found to relax the criteria for granting the user right from needing it to wanting it.
Technical news
- After a recent RfC, moved pages will soon be featured in a queue similar to Special:NewPagesFeed and require patrolling. Moves by administrators, page movers, and autopatrolled editors will be automatically marked as patrolled.
- Cookie blocks have been deployed. This extends the current autoblock system by setting a cookie for each block, which will then autoblock the user if they switch accounts, even under a new IP.
FAQ
Hi, I'm going to presume that the Cyberbot FAQ is skew-wiff for April Fools and will be back to normal tomorrow. If not, is it possible to get an easier to read version of the Cyberbot FAQ? Red Fiona (talk) 16:13, 1 April 2017 (UTC)
- Correct. When April Fools begins, the page flips upside down and then during the day slowly rotates back to its upright position.CyberP(talk) 16:36, 1 April 2017 (UTC)
Hi
Smash!
You've been squished by a whale!
Don't take this too seriously. Someone just wants to let you know you did something really silly.
For the fake new message thing and the flipped user page.
Crunch, crunch!
Here are some chips to go with your fish!
For your good adminship and exemplary use of humor.
AnAwesomeArticleEditor (talk) 18:39, 1 April 2017 (UTC)
ANI
This message is to inform you that you have been identified as a turrible admin and haz been reported to the noticeboard of doom. — Tea // have a ☕️ // leaves // 22:44, 1 April 2017 (UTC)
IAR bot
This could really liven things up! Perhaps it should block every tenth editor to ANI or fully protect the page for random lengths of time! ;) btw, it's nice to see some April Fool's humour that's both harmless and actually amusing, like the flipped userpage! HJ Mitchell | Penny for your thoughts? 22:45, 1 April 2017 (UTC)
- Indeed. I came up with the flipped user page idea a while ago. It's a fun little script I installed into my layout that flips the page upside down at the start of April Fools and slowly rotates back during the day. :-)Cyberpower How can I help? 22:50, 1 April 2017 (UTC)
Dead URLs
Hi Cyberpower. For info this edit marked the URL as dead, but it works fine. Thanks. Lugnuts Precious bodily fluids 19:45, 2 April 2017 (UTC)
- Please follow the instructions on the bot's talk page.—CYBERPOWER (Around) 19:46, 2 April 2017 (UTC)
- It wants me to log into something to report it - I don't have an account for whatever that site is. If it keeps happening, then I'll just request for the bot to be blocked at WP:ANI. Lugnuts Precious bodily fluids 07:12, 3 April 2017 (UTC)
- You login with your Wikipedia Account.—CYBERPOWER (Chat) 07:14, 3 April 2017 (UTC)
- Anyway, now that I have more time. I went ahead and did the report for you. It's a very simple tool. You just list all of the false positives you found, and it either fixes it internally on the spot, or it will escalate to an actual report for me to look at to figure out why the URL is coming back as it should. It's meant to streamline the false positive reporting process. One simply clicks login to attach their account to the IABot Management Interface. It's approved to do so.—CYBERPOWER (Chat) 08:21, 3 April 2017 (UTC)
- You login with your Wikipedia Account.—CYBERPOWER (Chat) 07:14, 3 April 2017 (UTC)
- It wants me to log into something to report it - I don't have an account for whatever that site is. If it keeps happening, then I'll just request for the bot to be blocked at WP:ANI. Lugnuts Precious bodily fluids 07:12, 3 April 2017 (UTC)
bot=<name>
[8] feature or bug? -- GreenC 21:13, 2 April 2017 (UTC)
- Feature.—CYBERPOWER (Around) 21:14, 2 April 2017 (UTC)
- Guess I'm a bot. Seems odd as the purpose of
|bot=is to track work done by bots so it can be checked by tools in the future ie. re-check all dead tags made by IABot. Theoretically the user is responsible but in practice they accept as-is most times. -- GreenC 21:42, 2 April 2017 (UTC)- Well that was the idea. It was supposed to identify the editor that placed it there, be it through automated means or not.—CYBERPOWER (Around) 21:52, 2 April 2017 (UTC)
- Hmm.. I thought it was to identify a bot edit, bots need tracking for a number of reasons. Human editors we don't track in the wikitext (that I know of). BTW it looks like the IA API has been down for a few hours. -- GreenC 22:45, 2 April 2017 (UTC)
- Oh I know. There was a config issue on their server. They got flooded with error emails from my bot.—CYBERPOWER (Around) 06:17, 3 April 2017 (UTC)
- Hey, the API has been down for me since around 5pm yesterday, but IABOt is still running. Did something changed with the API? -- GreenC 13:46, 4 April 2017 (UTC)
- Oh I know. There was a config issue on their server. They got flooded with error emails from my bot.—CYBERPOWER (Around) 06:17, 3 April 2017 (UTC)
- Hmm.. I thought it was to identify a bot edit, bots need tracking for a number of reasons. Human editors we don't track in the wikitext (that I know of). BTW it looks like the IA API has been down for a few hours. -- GreenC 22:45, 2 April 2017 (UTC)
- Well that was the idea. It was supposed to identify the editor that placed it there, be it through automated means or not.—CYBERPOWER (Around) 21:52, 2 April 2017 (UTC)
- Guess I'm a bot. Seems odd as the purpose of
IABot is leaving a lot of {{dead link}} in recent edits, and the few I checked have snapshots available and the API looks down since yesterday around 5pm. It looks like there is a problem. If I don't hear back soon I will do a shutdown. WaybackMedic will be able to fix all those dead links it's not a problem long term. -- GreenC 15:31, 4 April 2017 (UTC)
Labs terms of use
Hello! Thank you for your edit counter! But please reply in this "bug report" about the publication of the source code: [9]. --Valerio Bozzolan (talk) 00:01, 5 April 2017 (UTC)
Wikispecies
Per discussion at s:Wikispecies:Village Pump#Fixing dead links, please could you operate your Internet Archive bot at Wikispecies? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:54, 6 April 2017 (UTC)
- I'd like a little more discussion than that. Also, please open a ticket on Phabricator with a link to the discussion so I can track it. Use this tool to do it.—CYBERPOWER (Chat) 15:56, 6 April 2017 (UTC)
- Sure: T162378. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:42, 6 April 2017 (UTC)
Cyberbot I offline
Cyberbot I stopped updating User:Cyberpower678/Tally and User:Cyberpower678/RfX Report around the time a lot of unclosed RfAs got deleted. (See Wikipedia talk:Requests for adminship#Graveyard RfAs) I wonder if the sudden removal of these caused it to fall over? Ritchie333 (talk) (cont) 23:30, 6 April 2017 (UTC)
- Seems to be updating just fine looking at the timestamp.—CYBERPOWER (Chat) 07:50, 7 April 2017 (UTC)
- Yes, it seems it came back up again a few hours after I wrote this. Ritchie333 (talk) (cont) 09:24, 7 April 2017 (UTC)
Detected an error
Hello Cyberpower - it appears that your bot has made a minor error mentioned in detail within this discussion. Put simply, at WP:RFA, there is a message on GoldenRing's RFA which appears to be an extract taken from there. I look forward to your reply at your earliest convenience with regards to this matter. Patient Zerotalk 12:14, 7 April 2017 (UTC)
- The elements the bot looks for are missing, namely the close timestamp is gone, so the bot is not going to deliver the right output. The RfA normally gets untranscluded when closed.—CYBERPOWER (Chat) 12:19, 7 April 2017 (UTC)
- They don't get untranscluded when there's a cratchat though. Is it possible to program the bot to just leave that section blank if it can't find the close timestamp? WJBscribe (talk) 12:22, 7 April 2017 (UTC)
- The bot's code is antiquated. No doubt would that be the solution, but it could benefit from a complete rewrite, which is something I need to find the time for. I have a different solution for the time being.—CYBERPOWER (Chat) 12:24, 7 April 2017 (UTC)
- I just edit conflicted with you; I was about to ask something similar. Patient Zerotalk 12:25, 7 April 2017 (UTC)
- Thanks for your help and for solving the error, Cyberpower. Patient Zerotalk 12:25, 7 April 2017 (UTC)
- Well my solution is to simply add the timestamp back in place. That keeps the bot happy.—CYBERPOWER (Chat) 12:27, 7 April 2017 (UTC)
- Anything to keep it happy! Take care. Patient Zerotalk 12:29, 7 April 2017 (UTC)
- Mmm, that just about works, but can we lose the confusing "Voice your opinion on this candidate" link without confusing the bot? WJBscribe (talk) 12:37, 7 April 2017 (UTC)
- Let's find out.—CYBERPOWER (Chat) 12:38, 7 April 2017 (UTC)
- Looks like the answer is yes.—CYBERPOWER (Chat) 12:44, 7 April 2017 (UTC)
- Great job! Thank you once again. Patient Zerotalk 12:57, 7 April 2017 (UTC)
- A great use of my mop, edit a fully protected RfA. :p—CYBERPOWER (Chat) 13:02, 7 April 2017 (UTC)
- Great job! Thank you once again. Patient Zerotalk 12:57, 7 April 2017 (UTC)
- Looks like the answer is yes.—CYBERPOWER (Chat) 12:44, 7 April 2017 (UTC)
- Let's find out.—CYBERPOWER (Chat) 12:38, 7 April 2017 (UTC)
- Mmm, that just about works, but can we lose the confusing "Voice your opinion on this candidate" link without confusing the bot? WJBscribe (talk) 12:37, 7 April 2017 (UTC)
- Anything to keep it happy! Take care. Patient Zerotalk 12:29, 7 April 2017 (UTC)
- Well my solution is to simply add the timestamp back in place. That keeps the bot happy.—CYBERPOWER (Chat) 12:27, 7 April 2017 (UTC)
- Thanks for your help and for solving the error, Cyberpower. Patient Zerotalk 12:25, 7 April 2017 (UTC)
- I just edit conflicted with you; I was about to ask something similar. Patient Zerotalk 12:25, 7 April 2017 (UTC)
- The bot's code is antiquated. No doubt would that be the solution, but it could benefit from a complete rewrite, which is something I need to find the time for. I have a different solution for the time being.—CYBERPOWER (Chat) 12:24, 7 April 2017 (UTC)
- They don't get untranscluded when there's a cratchat though. Is it possible to program the bot to just leave that section blank if it can't find the close timestamp? WJBscribe (talk) 12:22, 7 April 2017 (UTC)
Swedish Wikipedian in exile
Lika several others, I have been blocked permanently on Swedish WP for objecting to the incorrigible bullying done by the elite there, year after year, world without end, amen. Ergo, no use to write to me there. Sorry. Best wishes with your project. --SergeWoodzing (talk) 14:15, 8 April 2017 (UTC)
- @SergeWoodzing: I'm sorry to hear that. This wiki is my primary base of operations. If you are willing you can also help over here. I have a copy of the JSON needing translating. Since I am an admin here, I was able to change the content model of the page to make it more readable. You can find it here. If you have any questions, let me know. :-)—CYBERPOWER (Chat) 14:19, 8 April 2017 (UTC)