Wikipedia:Bots/Requests for approval/BetacommandBot Task 7

The following discussion is an archived debate. Please do not modify it. Subsequent comments should be made in a new section. The result of the discussion was Approved.

BetacommandBot

• (BRFA
• Approved BRFAs
• talk
• contribs
• count
• SUL
• logs
• page moves
• block log
• rights log
• flag)

Operator: User:Betacommand

Automatic or Manually Assisted: auto unsupervised

Programming Language(s): m:pywikipedia

Function Summary: Moving images to the Wikipedia Commons.

Edit period(s) (e.g. Continuous, daily, one time run): Continuous

Edit rate requested: maxlag = 5

Already has a bot flag (Y/N): Y

Since Commons is a repository for free images that can be used across wiki and cross-project, wikipedia should move all its Free images to commons. But a major problem with that is that moving such images is time consuming and semi difficult in that you have to maintain the image history. Thus I have created a bot to do this. But, because there are problems with any current system (categories,templates) and quality control and improper labeling, I have set up a new method to make it simpler to transfer images to commons. Only approved users may add {{commons ok}} (users who add the image and are not approved will be ignored) after the image is tagged it will be automatically transwikied to commons.

Table filled in by Mets501. Function details taken from User:Betacommand/Commons. Betacommand, please fill in the rest of the fields.

Discussion

Per User:Betacommand/Commons moving and tagging image to commons. β^_command 06:10, 12 August 2007 (UTC)

Please use the proper template. Also, could you please point out the approval on commons, or bot flag, or some consensus? Matt/TheFearow (Talk) (Contribs) (Bot) 06:34, 12 August 2007 (UTC)

How will your bot:

• Tell if the images license is undoubtedly accepted at Commons
• Properly categorize the image on commons(so it is not orphaned there)
• handle image pages that transclude Wikipedia templates no available on commons

This sounds like a good idea, but I would be interested in the details of percisely what it does. Until(1 == 2) 15:23, 12 August 2007 (UTC

The request for approval needs to be here, not in your userspace, so we can archive it. — madman bum and angel 16:38, 12 August 2007 (UTC)

A few questions from — xaosflux ^Talk

1. Other then AGF of someone tagging, will compatible licsense checking be done?
2. Will all versions of files be copied?
3. Is this a 'manual' type of transwiki, or will it be using export/import?

Also, why do you refuse to use the generally accepted form for requesting bot approval? —METS501 (talk) 22:42, 12 August 2007 (UTC)

Second request in the last few days that hasn't - maybe we need a new rule, requests without the form will be removed? It gets frustrating having to figure it out, the form is much more convenient and takes about 3 seconds to fill out. Matt/TheFearow (Talk) (Contribs) (Bot) 23:00, 12 August 2007 (UTC)

Only the current version of a file will me transwikied, and there is no checking done by the bot, as users who are trusted with the tagging know what they are doing. Im not sure what templates commons uses that we dont. and categorization is done automatically by licensing. β^_command 23:04, 12 August 2007 (UTC)

as for filling out that dumb form, this is my tenth BRFA its all the same and that form is a pain its bloated with HTML comments that are annoying and Ive not use that in the past see those request. β^_command 23:04, 12 August 2007 (UTC)

In the case of the form, just ignore the comments. You just have to put in a couple values then after that you can directly edit the discussion section. It makes it a lot easier. Just fill in the two remaining values on the form, and we can continue. Matt/TheFearow (Talk) (Contribs) (Bot) 23:08, 12 August 2007 (UTC)

how about we not worry about the format of this request, and stop giving me a hard time for doing the exact same thing ive been doing? is called Wikipedia:IAR if BAG doesnt know my style then we have a problem. Its the same method as always. So any issues about the task? β^_command 23:13, 12 August 2007 (UTC)

The forms are helpful, but we can certainly review idea without them, but they do prove useful to new/inexperienced readers of these pages trying to figure out what is going on....now back to the real business:

(may have got missed with all the form talk but) What transwiki method will you be using to maintain the edit history of the description page? — xaosflux ^Talk 23:41, 12 August 2007 (UTC)

I dont keep the image description history, what I do keep is the File history which is more important than the description page because the file is whats being copied. see commons:Image:Robert burns.JPG β^_command 23:48, 12 August 2007 (UTC)

Ok, can you please point me to consensus on commons, or a similair thing, that a bot is wanted for the task? Matt/TheFearow (Talk) (Contribs) (Bot) 23:50, 12 August 2007 (UTC)

Commons BRFA currently underway, commons:Commons:Bots/Requests for flags/BetacommandBot and there was a discussion a month or two ago on their version of AN. β^_command 23:53, 12 August 2007 (UTC)

It's technically not an RfA because as I understand it your bot does not need adminship there, correct? What it is, if so, is just a request for the bot flag. The commons process is a lot less rigid than this one, but it nevertheless is helpful to fully and politely answer questions when they are asked, it helps everyone get nice warm fuzzies. (and I share a lot of Until(1==2)'s concerns about how this would work) ++Lar: t/c 01:03, 13 August 2007 (UTC)

Do both of these requests have to be approved (essentially) at the same time? Or eventually, before the bot runs? As of right now the commons request has not been approved, and there have been some concerns raised that need to be addressed. (that is not the same as saying no bot is wanted for the task, btw) As a fairly new commons 'crat, I'd probably not be the one to turn the flag on myself unless consensus was a lot clearer than it is at present. One comment, I think there is a lot of value in bringing both the file AND image histories along, if at all possible, as the file history has information about the licenses (if there was change) that may have a great deal of bearing on the validity of the transfer. ++Lar: t/c 01:03, 13 August 2007 (UTC)

there is no adminship involved, and it will only run when bot BRFA's are approved. If you want the full image description moved also I can do that too. β^_command 01:07, 13 August 2007 (UTC)

The description itself would be really great, ya.... If you can bring the history of it along too, so much the better. I realise that most other processes don't do that, but I think knowing what has happened in the past to the description is goodness. Better too much info than too little, especially if the image then gets deleted on en and there is a question later (on Commons) about how it might have been licensed... we'd have to go dig up an en: admin. We have lots of those, true, but still... (scanning every version for possible license changes is probably huge overkill :) ) ++Lar: t/c 01:36, 13 August 2007 (UTC)

the discription is already transwikied, but ill add the page hist too. β^_command 01:39, 13 August 2007 (UTC)

They do not have to be approved at the same time, it's whenever they are both finished. If they both are approvals, it goes ahead, otherwise, it doesn't. Matt/TheFearow (Talk) (Contribs) (Bot) 01:48, 13 August 2007 (UTC)

Does commons: have Special:Import active, and would commons: allow a bot to run it? — xaosflux ^Talk 01:48, 13 August 2007 (UTC)

That might be possible for the main import, but the major issue there is setting up an interwiki import which would annoy the devs, and then assigning the bot to the "import" group, which would annoy the stewards, and if it also allows XML upload, then most likely would not be given, so again the devs have to be bothered to create a new group, then the stewards to assign it. It's easier to just copy+paste the file and description history. Matt/TheFearow (Talk) (Contribs) (Bot) 02:08, 13 August 2007 (UTC)

Again, not knowing what is set up on commons: well, they may be more lax on granting bots access to what ever existing group (sysop?) has access to Import: (assuming it even exists on commons:) I've tested it on meta: and it works well there. — xaosflux ^Talk 05:20, 13 August 2007 (UTC)

Maybe. I know the import group has the permissions, so that is most likely. Also, special:import is always there, just often doesnt have the wikis you want configured for it. Matt/TheFearow (Talk) (Contribs) (Bot) 05:50, 13 August 2007 (UTC)

I am not using the import functions. β^_command 01:52, 14 August 2007 (UTC)

About categorization: how will the bot determine an appropriate category for the images? —METS501 (talk) 00:04, 14 August 2007 (UTC)

it will only use the cats that are present in the licensing templates. β^_command 01:52, 14 August 2007 (UTC)

That is unacceptable. All images sent to commons must be placed in a category or an article there. Can you use CommonSense? —METS501 (talk) 04:17, 14 August 2007 (UTC)

it now uses commonsense. β^_command 23:45, 14 August 2007 (UTC)

I'm still not getting around the lack of edit history, e.g. with Image:Robert burns.JPG, the uploader (who is attributed) was not the person who declared the licensing, it was done by someone else. In some cases licensing may legitimatley change. Now while PD is PD, other compatible licenses could make use of this, no? It also appears to fail the note "The file was properly uploaded (preserving GFDL required history of revisions)" Is this something that can be gathered and uploaded outside of using a tranwiki import? — xaosflux ^Talk 04:53, 14 August 2007 (UTC)

Xaoxflux GDFL issue has been fixed and is compliant β^_command 23:45, 14 August 2007 (UTC)

I think the bot needs to detect licensing changes, and at the very least ensure that the license is the same from the last time the original uploader touched it this can be done easily by comparing changes from the page's history (with content) which is available vie api.php. One query will get you all the data you need for one image. —— Eagle101^{Need help?} 05:19, 14 August 2007 (UTC)

• A small trial should help determine any outstanding GFDL issues that may be remaining, and show affects on various types of image pages. Approved for trial (50 edits or 7 days). Please provide a link to the relevant contributions and/or diffs when the trial is complete.. — xaosflux ^Talk 04:45, 16 August 2007 (UTC)

Post-trial

OK, a couple of things:

• Can you place the {{NowCommons}} template at the top of the page with a line break below it (not two or three)? It helps it to be spotted.
• Instead of just "page history" can you make it "Image description page history"?
• Instead of a table like this:

date/time	username	resolution	size	edit summary
19:47, 2 December 2006	Wdflake	300×300	1,239	(Basic production possibilities frontier)

can it be a table like this:

Date/Time	Username	Resolution	Size (bytes)	Edit summary
19:47, 2 December 2006	Wdflake	300×300	1,239	(Basic production possibilities frontier)

and instead of

oldid	date/time	username	edit summary
91639956	11:01, 18 August 2007	Sopoforic
91639956	19:51, 2 December 2006	Wdflake
91639262	19:47, 2 December 2006	Wdflake	(Basic production possibilities frontier)

can it be like:

Link	Date/Time	Username	Edit summary
Permalink to the revision here	11:01, 18 August 2007	Sopoforic
Permalink to the revision here	19:51, 2 December 2006	Wdflake
Permalink to the revision here	19:47, 2 December 2006	Wdflake	(Basic production possibilities frontier)

• Also, can you remove the empty <nowiki></nowiki>'s from the table?
• Can you capitalize "Source: Self-made"?
• Which style are you using? [1] or [2]? If it's the second, can you add a <hr> between the bot's message and the image description/source?

Thanks. —METS501 (talk) 00:51, 20 August 2007 (UTC)

Also, please add the interwiki prefix "wp:" to the start of all links on the page, to stop redlinks from appearing. Matt/TheFearow (Talk) (Contribs) (Bot) 01:10, 20 August 2007 (UTC)

No, add the prefix w:, that's the interwiki prefix!! :) Cbrown1023 talk 01:18, 20 August 2007 (UTC)

Make a random number generator choose out of w: wp: wikipedia: w:en: meta:wp:en: meta:w:fr:en. Or just use the shortest one there is, which is probbly w:. Matt/TheFearow (Talk) (Contribs) (Bot) 02:15, 20 August 2007 (UTC)

The permalinks would only work for sysops once the image on the English Wikipedia has been deleted, no? — madman bum and angel 04:29, 20 August 2007 (UTC)

Yep, but it's good for double-checking in the case of complaints or doubts it was correct. Matt/TheFearow (Talk) (Contribs) (Bot) 05:53, 20 August 2007 (UTC)

I fixed the commons template placement, as for the table format, Im going to have to bug a pywiki dev to fix that properly as that is a pywikipedia function. β^_command 01:28, 22 August 2007 (UTC)

well because of the changes you asked for a a little specific I went ahead and coded it in manually. β^_command 16:40, 22 August 2007 (UTC)

Since those issues are fixed, Approved for trial (7 days). Please provide a link to the relevant contributions and/or diffs when the trial is complete. Matt/TheFearow (Talk) (Contribs) (Bot) 22:28, 22 August 2007 (UTC)

see Image:Harpers weekly 1897.jpg for the first move after the improvements. β^_command 23:08, 22 August 2007 (UTC)

The perma-links on the commons' page appear to be broken for pages with spaces in the titles. — xaosflux ^Talk 02:44, 23 August 2007 (UTC)

permalink fixed. β^_command 04:27, 23 August 2007 (UTC)

{{BAGAssistanceNeeded}} Is this ready for a decision? It's been waiting for a few weeks now... GrooveDog (talk) (Review) 22:32, 7 September 2007 (UTC)

We're waiting for the commons BRFA, as commons is affected by this more than we are. --ST47Talk·Desk 22:48, 10 September 2007 (UTC)

Lets just approve it here and Ill sort commons out and get that approved. β^_command 15:56, 13 September 2007 (UTC)

OK.  Approved. —METS501 (talk) 18:55, 14 September 2007 (UTC)

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made in a new section.