Jackaranga (talk | contribs) |
→RfA candidate's attitude to NFC: new section |
||
Line 247: | Line 247: | ||
:::How would the bot know that Boeing is what is required here? And bots don't delete (normally). Humans do. That image requires a human to assess it. Bots are limited in what they can realistically achieve. [[User:Carcharoth|Carcharoth]] 12:35, 18 October 2007 (UTC) |
:::How would the bot know that Boeing is what is required here? And bots don't delete (normally). Humans do. That image requires a human to assess it. Bots are limited in what they can realistically achieve. [[User:Carcharoth|Carcharoth]] 12:35, 18 October 2007 (UTC) |
||
::::Too true, that's one of the reasons I disagree with the new proposal, I don't think the wikimedia resolution meant that people had to go as far as making each element of the rationale machine-readable. Perhaps it will be for the best though in the end. [[User:Jackaranga|Jackaranga]] 16:59, 18 October 2007 (UTC) |
::::Too true, that's one of the reasons I disagree with the new proposal, I don't think the wikimedia resolution meant that people had to go as far as making each element of the rationale machine-readable. Perhaps it will be for the best though in the end. [[User:Jackaranga|Jackaranga]] 16:59, 18 October 2007 (UTC) |
||
== RfA candidate's attitude to NFC == |
|||
I'm concerned that the candidate seems to be pushing the line that the differential between WP's policy and what is strictly required by (US) law is undesirable, and that s/he will work to stop deletions on that basis. I'm not making a judgement on the nomination as whole, but this does need to be cleared up before the keys are handed out. |
|||
[[Wikipedia:Requests_for_adminship/Remember_the_dot#Neutral]] |
|||
[[User:Remember_the_dot#My_views_on_non-free_content]] |
|||
[[User:Tony1|<font color="black">'''Tony'''</font >]] [[User_talk:Tony1|<font color="darkgreen">(talk)</font >]] 03:12, 21 October 2007 (UTC) |
Revision as of 03:12, 21 October 2007
Fair use (inactive) | ||||
|
Archives |
Bot recognition of sources
I see that OsamaKBOT adds no source tags. What is the normal way to make sources machine-readable so the same bot won't come back and tag it again? Is this an acceptable way to add a source so that a bot can recognise it? If not, then OsamaKBOT needs to instruct people on how to do this. I've notified the bot operator. Carcharoth 18:57, 23 September 2007 (UTC)
- that bot needs shut down and indef blocked. There is zero method for a bot to ID sources, Ive thought about it but the human factor makes checking for sources via bot impossible. βcommand 19:00, 23 September 2007 (UTC)
- Ah. Thanks. What about reviewing the tags it added? Carcharoth 19:29, 23 September 2007 (UTC)
- Alas I no longer have access to the tools to mass revert the bot, but that is what should be done, for all edits still on top. βcommand 19:32, 23 September 2007 (UTC)
- Ah. Thanks. What about reviewing the tags it added? Carcharoth 19:29, 23 September 2007 (UTC)
- It's possible for a bot to check images for sourcing in a very limited way. OrphanBot will add no-source tags to newly-uploaded images, but only if the image description page is empty, or the image description page contains only image copyright tags that do not specify the source. --Carnildo 20:41, 23 September 2007 (UTC)
- I've asked on the user's talk page for the user to stop doing this and reverse the tags. They've done this before and apparently stopped and removed all the tags. Both times, going through new and legacy logo images in alphabetical order, proposing deletion to a bunch of perfectly good corporate logos in articles about the organization. Does anyone have any relationship with Osama to ask if they'll work with the rest of us instead to fix these? It would be a lot easier to do this the easy way than to have to go through AN/I, have the bot blocked, etc. Wikidemo 21:12, 23 September 2007 (UTC)
- I can't understand what's the problem is. the most tagging was right. and I have a BRFA to BAG. Images without source must be tagging and deleting or fixing per Wikipedia's policy (not mine!). All one know all bots have some bugs and problems my bot has some and all these has fixed (known ones). I welcome with any post about my bot here.--OsamaK 21:37, 25 September 2007 (UTC)
- The problem is you've now tagged for deletion several hundred perfectly good corporate logos of some of the biggest and most important companies in the world, in longstanding articles where they belong. We're trying to get these images in line with policy, not delete them. You've completely ignored requests to stop, and you're violating Wikipedia by running the bot. I don't buy the argument that it's okay to violate policy in order to enforce policy. I don't know what good it's supposed to do to discuss the issue on your talk page. You've ignored past posts to your talk page on the subject. Is the bot still blocked? If not, please cut it out so it doesn't become an AN/I issue. Wikidemo 23:47, 25 September 2007 (UTC)
- I can't understand what's the problem is. the most tagging was right. and I have a BRFA to BAG. Images without source must be tagging and deleting or fixing per Wikipedia's policy (not mine!). All one know all bots have some bugs and problems my bot has some and all these has fixed (known ones). I welcome with any post about my bot here.--OsamaK 21:37, 25 September 2007 (UTC)
- I've asked on the user's talk page for the user to stop doing this and reverse the tags. They've done this before and apparently stopped and removed all the tags. Both times, going through new and legacy logo images in alphabetical order, proposing deletion to a bunch of perfectly good corporate logos in articles about the organization. Does anyone have any relationship with Osama to ask if they'll work with the rest of us instead to fix these? It would be a lot easier to do this the easy way than to have to go through AN/I, have the bot blocked, etc. Wikidemo 21:12, 23 September 2007 (UTC)
- It's possible for a bot to check images for sourcing in a very limited way. OrphanBot will add no-source tags to newly-uploaded images, but only if the image description page is empty, or the image description page contains only image copyright tags that do not specify the source. --Carnildo 20:41, 23 September 2007 (UTC)
- OsamaK: I'm not sure if you're aware, but the BRFA for that bot task was denied: Wikipedia:Bots/Requests for approval/OsamaKBOT 5. --Dynaflow babble 00:03, 26 September 2007 (UTC)
- Wikidemo, I know deleting is the last choice. So my bot notify the uploaders (All uploaders) if no one have fixed the image it will be deleted after 7 days from tagging date (there is no another choice because there are many "logos of some of the biggest and most important companies in the world" without source). This is not enough reason for stopping my bot. Dynaflow, yeah it does not work from some days ago.--OsamaK 02:28, 26 September 2007 (UTC)
- You've been running an unauthorized bot for a couple weeks now, utterly ignoring everyone's warnings on the subject, and archiving all your talk pages that show those warnings. It's not up to me to argue the merits of image use and justify what's wrong every time someone decides to run a rogue bot. There are productive ways to deal with unsourced images and if you followed the rules and paid attention to the discussion you might be aware of some. It's your burden to get the bot approved. Not here on this page and not by me, but by the proper Wikipedia procedure. Three days ago that authorization request was specifically denied. Are you saying that you will stop running the bot or not? In answering that question, please keep in mind that if you do run the bot you are openly defying Wikipedia policy. In case that is not clear enough, consider this your final warning - stop running unapproved bots. Enough of this nonsense. How can we remove the bot's deletion tags? Wikidemo 05:53, 26 September 2007 (UTC)
- Agreed with Wikidemo (and as he can confirm, that's a rare enough occurrence that it should get your attention). Stop running the bot, or you'll be helped to stop running it. If you'd like to try and fix it so that BAG will approve it, go for it, but until and unless that happens, do not use it. Seraphimblade Talk to me 05:57, 26 September 2007 (UTC)
- lol, has someone seen bot contributions? last edit with this bot was on 00:12, 17 September 2007 :). Seraphimblade, thanks for your helping. I want someone tells me what's the bot problem(s) is/are. then I'll fix them. --OsamaK 11:43, 26 September 2007 (UTC)
- The bot's problem is that the current image description page format doesn't provide any way for a bot to tell whether a source is provided or not. So no bot will be able to tags images for lack of source without human assistance. That's why the bot wasn't approved. — Carl (CBM · talk) 13:08, 26 September 2007 (UTC)
- My bot was smart in the testing (but very few problems have fixed). no? tagging easy and good using bot.--OsamaK 13:26, 26 September 2007 (UTC)
- The bot's problem is that the current image description page format doesn't provide any way for a bot to tell whether a source is provided or not. So no bot will be able to tags images for lack of source without human assistance. That's why the bot wasn't approved. — Carl (CBM · talk) 13:08, 26 September 2007 (UTC)
- lol, has someone seen bot contributions? last edit with this bot was on 00:12, 17 September 2007 :). Seraphimblade, thanks for your helping. I want someone tells me what's the bot problem(s) is/are. then I'll fix them. --OsamaK 11:43, 26 September 2007 (UTC)
- Agreed with Wikidemo (and as he can confirm, that's a rare enough occurrence that it should get your attention). Stop running the bot, or you'll be helped to stop running it. If you'd like to try and fix it so that BAG will approve it, go for it, but until and unless that happens, do not use it. Seraphimblade Talk to me 05:57, 26 September 2007 (UTC)
- You've been running an unauthorized bot for a couple weeks now, utterly ignoring everyone's warnings on the subject, and archiving all your talk pages that show those warnings. It's not up to me to argue the merits of image use and justify what's wrong every time someone decides to run a rogue bot. There are productive ways to deal with unsourced images and if you followed the rules and paid attention to the discussion you might be aware of some. It's your burden to get the bot approved. Not here on this page and not by me, but by the proper Wikipedia procedure. Three days ago that authorization request was specifically denied. Are you saying that you will stop running the bot or not? In answering that question, please keep in mind that if you do run the bot you are openly defying Wikipedia policy. In case that is not clear enough, consider this your final warning - stop running unapproved bots. Enough of this nonsense. How can we remove the bot's deletion tags? Wikidemo 05:53, 26 September 2007 (UTC)
- Wikidemo, I know deleting is the last choice. So my bot notify the uploaders (All uploaders) if no one have fixed the image it will be deleted after 7 days from tagging date (there is no another choice because there are many "logos of some of the biggest and most important companies in the world" without source). This is not enough reason for stopping my bot. Dynaflow, yeah it does not work from some days ago.--OsamaK 02:28, 26 September 2007 (UTC)
- OsamaK: I'm not sure if you're aware, but the BRFA for that bot task was denied: Wikipedia:Bots/Requests for approval/OsamaKBOT 5. --Dynaflow babble 00:03, 26 September 2007 (UTC)
- (←)Can you describe the algorithm the bot uses to tell whether there is a source? Note that no URL is required at the moment, nor is the actual word "source". — Carl (CBM · talk) 13:32, 26 September 2007 (UTC)
- If the human operator doesn't understand that the copyright holder (aka source) of a corporate logo is that corporation, it seems unlikely that the algorithm is of any value. Stan 14:35, 26 September 2007 (UTC)
- Just a comment -- copyright holder != source. The source is the place (often a website or printed material) from which the uploader obtained the image. The copyright holder is the owner of the copyright in the image. The latter is usually the entity identified by the logo, but the former can be quite different. -- But|seriously|folks 17:54, 26 September 2007 (UTC)
- In the case of logos, the only pertinent copyright information for sourcing purposes is whose logo it is, i.e. source as origin, not as URL. Same thing for album and book covers, and other 2D cover art. We can decide for the future if we want to tell people to describe a website. I think it's more pertinent where the image actually comes from. Wikidemo 18:09, 26 September 2007 (UTC)
- CBM, It finds (any URL, "source", and other keywords) -I have written this many :)- that's not very important, let me see what's the problem is :-\. Stan Shebs, sorry, I did not understand your comment. --OsamaK 16:19, 26 September 2007 (UTC)
- But no keywords are actually required, so you shouldn't be tagging images for deletion just because they don't match the list of keywords you compiled. It would be acceptable to use your bot to make a list of possibly bad images, and then check them by hand. But a completely automatic system to detect sources will be too error prone. In any case, you must know by now that you need to get your bot approved before it can restart. To get approval, you will probably need to make sure that its edits are approved by a human before being made. — Carl (CBM · talk) 16:23, 26 September 2007 (UTC)
- Actually, you've got a great idea there. Maybe something like Durin did with articles that use a high number of nonfree images, where a list is auto-generated and then people check through it? I think that would be very helpful, and I bet BAG would approve that in a second. Seraphimblade Talk to me 16:41, 26 September 2007 (UTC)
- But no keywords are actually required, so you shouldn't be tagging images for deletion just because they don't match the list of keywords you compiled. It would be acceptable to use your bot to make a list of possibly bad images, and then check them by hand. But a completely automatic system to detect sources will be too error prone. In any case, you must know by now that you need to get your bot approved before it can restart. To get approval, you will probably need to make sure that its edits are approved by a human before being made. — Carl (CBM · talk) 16:23, 26 September 2007 (UTC)
- CBM, It finds (any URL, "source", and other keywords) -I have written this many :)- that's not very important, let me see what's the problem is :-\. Stan Shebs, sorry, I did not understand your comment. --OsamaK 16:19, 26 September 2007 (UTC)
- In the case of logos, the only pertinent copyright information for sourcing purposes is whose logo it is, i.e. source as origin, not as URL. Same thing for album and book covers, and other 2D cover art. We can decide for the future if we want to tell people to describe a website. I think it's more pertinent where the image actually comes from. Wikidemo 18:09, 26 September 2007 (UTC)
- We should require a templated statement of source in a predictable format, and once that's phased in it will no longer be an issue for new images. Carl, where is our proposal? Until we can discuss and implement the proposal we ought to table any consideration of automatic tagging or deletion of images for lack of an explicit source statement, and ask BAG not to approve any such bots without consulting us. It's an almost identical issue to an image lacking a written use rationale or identification of the article it's used in. In most cases it's in there one way or another and we can fix the images rather than marking them for deletion. Wikidemo 17:21, 26 September 2007 (UTC)
- We can require it, but how many people will actually use it properly? At a guess, based on OrphanBot's encounters with image copyright tags, any "templated statement of source in a predictable format" will be 75% incorrect. You should probably look at User:Carnildo/Things OrphanBot found on image description pages. --Carnildo 19:08, 26 September 2007 (UTC)
- Part of the solution might be to get the upload wizard to add the template automatically, and prompt the uploader for a source. — Carl (CBM · talk) 19:13, 26 September 2007 (UTC)
- Hi there! Check list by hand is very slow way, my bot did well with automatic tagging. All bots has "error prone" my bot has a very few errors as the testing. Wikidemo, it's listed for 7 days there Category:Images with unknown source. It would be good to start a new WikiProject to finding images' sources in these days! I'll help :)--OsamaK 19:36, 26 September 2007 (UTC)
- any news? I think people have no problem with my bot, no? (I hope so :))--OsamaK 10:08, 30 September 2007 (UTC)
- Yes, people have problems with your bot; that's why it wasn't approved. The current policy doesn't require anything that would allow automatic detection of whether there is a source provided. The proposal announced below would change that for new images. In the meantime, the only thing you will be able to do with the bot is make lists for people to check by hand. — Carl (CBM · talk) 12:29, 30 September 2007 (UTC)
- We are in a loop :). The current policy doesn't see anything about check using bot (there are some bots checking at the moment). CBM, list is enable at User:OsamaK/Images bot...--OsamaK 14:09, 30 September 2007 (UTC)
- Yes, people have problems with your bot; that's why it wasn't approved. The current policy doesn't require anything that would allow automatic detection of whether there is a source provided. The proposal announced below would change that for new images. In the meantime, the only thing you will be able to do with the bot is make lists for people to check by hand. — Carl (CBM · talk) 12:29, 30 September 2007 (UTC)
- any news? I think people have no problem with my bot, no? (I hope so :))--OsamaK 10:08, 30 September 2007 (UTC)
- Hi there! Check list by hand is very slow way, my bot did well with automatic tagging. All bots has "error prone" my bot has a very few errors as the testing. Wikidemo, it's listed for 7 days there Category:Images with unknown source. It would be good to start a new WikiProject to finding images' sources in these days! I'll help :)--OsamaK 19:36, 26 September 2007 (UTC)
- Part of the solution might be to get the upload wizard to add the template automatically, and prompt the uploader for a source. — Carl (CBM · talk) 19:13, 26 September 2007 (UTC)
- We can require it, but how many people will actually use it properly? At a guess, based on OrphanBot's encounters with image copyright tags, any "templated statement of source in a predictable format" will be 75% incorrect. You should probably look at User:Carnildo/Things OrphanBot found on image description pages. --Carnildo 19:08, 26 September 2007 (UTC)
We are in a loop. Here it is: Creating a list of images that might or might not have a proper source listed = GOOD. Editing thousands of image pages (e.g., templating) in a way that may lead to images whose uploaders have followed all the rules, and which are thus legitimately resident on Wikipedia, being deleted (and doing all this without the approval of the Community) = REALLY FUCKING BAD. Your bot does both of these things, but the good does not cancel out the bad. This is why your bot would not have been allowed to exceed its 50-edit trial run, had you not decided to keep it running anyway, approval or no approval. --Dynaflow babble 15:10, 30 September 2007 (UTC)
- Hi Dynaflow, I am sorry for this FUCKING :), anyway we do not talk about it now. I want to know what should I do also to get approval?--OsamaK 20:46, 30 September 2007 (UTC)
- See Wikipedia talk:Bots/Requests for approval/OsamaKBOT 5, please.--OsamaK 21:02, 30 September 2007 (UTC)
- If your bot is only compiling a list of images which lack sources your bot can read, and perhaps leaving a nice, polite note on each uploader's page informing them, in the clearest and most precise language possible, that their image might not be properly sourced and how to fix it, and apologizing if they really do have proper sourcing that your bot just couldn't read, I would really have no objection. (This would involve putting together, or having someone else put together for you, a modified, OsamaKBOT-specific version of {{Image source}}.) Unless you are writing OsamaKBOT an artificial intelligence algorithm worthy of a DARPA project, however, I don't think there's any way I or most of the others here would support the bot -- or any bot -- going back to nominating things for deletion per the criteria on which you're focusing. --Dynaflow babble 21:09, 30 September 2007 (UTC) [EDIT:] In regards to the exchange on the bot's BRFA Talk page, I agree completely with Betacommand's evaluation (now there's something I never thought I'd say in a fair-use image deletion debate). --Dynaflow babble 21:25, 30 September 2007 (UTC)
- mmm, your English is very hard in the last comment :). as what I understand you want change bot's message? You can help me with change it. look there.--OsamaK 21:36, 30 September 2007 (UTC)
- I've created a draft of a possible substitute for OsamaKBOT's current message here: User talk:OsamaK/Images. I think a lot of the confusion is happening because you may not be completely understanding what we are saying, due to our differing levels of English fluency. Discussions here about copyright law and en.Wikipedia's copyright policies are very hard to follow, even for many native speakers. However, the issue here is much simpler to understand. Your bot, as it was before it was stopped, was causing things to be nominated for deletion that didn't need to be deleted and shouldn't have been deleted. That made people upset. Most of us do not think that a bot would be able to tell the difference between a good source and a bad source, and depending on an administrator to save a "good" image with a misidentified source at the last minute before it is deleted is unrealistic and unacceptable. If the bot is to do anything, it should not put templates on image pages that cause those images to be deleted. I cannot make it much clearer than that. --Dynaflow babble 22:18, 30 September 2007 (UTC)
- mmm, your English is very hard in the last comment :). as what I understand you want change bot's message? You can help me with change it. look there.--OsamaK 21:36, 30 September 2007 (UTC)
- If your bot is only compiling a list of images which lack sources your bot can read, and perhaps leaving a nice, polite note on each uploader's page informing them, in the clearest and most precise language possible, that their image might not be properly sourced and how to fix it, and apologizing if they really do have proper sourcing that your bot just couldn't read, I would really have no objection. (This would involve putting together, or having someone else put together for you, a modified, OsamaKBOT-specific version of {{Image source}}.) Unless you are writing OsamaKBOT an artificial intelligence algorithm worthy of a DARPA project, however, I don't think there's any way I or most of the others here would support the bot -- or any bot -- going back to nominating things for deletion per the criteria on which you're focusing. --Dynaflow babble 21:09, 30 September 2007 (UTC) [EDIT:] In regards to the exchange on the bot's BRFA Talk page, I agree completely with Betacommand's evaluation (now there's something I never thought I'd say in a fair-use image deletion debate). --Dynaflow babble 21:25, 30 September 2007 (UTC)
- See Wikipedia talk:Bots/Requests for approval/OsamaKBOT 5, please.--OsamaK 21:02, 30 September 2007 (UTC)
- I don't think we should approve a bot to do this, nor given the communications problems and unrepentant flaunting of bot policy would I favor of allowing this user to run such a bot even if approved.Wikidemo 23:07, 30 September 2007 (UTC)
- Dynaflow, I understand most discussion here :). Thanks for creating User talk:OsamaK/Images. I'll use it. Dynaflow, can you give me some examples for these images who have tagged but it should not deleted? (There are some examples ,but most them have fixed in last tagging, So give me some mistakes from last tagging.) Wikidemo, /me can't understand why.. ;).--OsamaK 11:42, 1 October 2007 (UTC)
- How is your bot dealing with non-hyperlinked sourcing statements? How would the bot handle a block of text explaining the origins and copyright holder for, say, an obscure, scanned pamphlet, as opposed to a simple statement that says, "Source: www.thesourceplace.com?" What criteria, exactly, is OsamaKBOT using to send things to speedydeletion? Our concern is that your bot will not be able to recognize allowable, but rarer/more-idiosyncratic forms of source citations, and that that will lead to sourced images being unnecessarily deleted. --Dynaflow babble 14:56, 1 October 2007 (UTC)
- Do not worry about how my bot finds unsourced images. Worry about bot's mistakes (I say that for 5th time ;)). anyway my bot finds any image's dicrapsion has (URLs, "source", "from" or other..) if there skipes if there is no list then taqqing with {{subst:nsd}}.. That's it! --OsamaK 19:45, 1 October 2007 (UTC)
- OsamaK, please read what people have written above. Find the alternative that has been suggested and tell us what it is and why it is preferable to tagging with {{subst:nsd}}. Then you might have a very small chance of getting your bot approved. Carcharoth 23:11, 1 October 2007 (UTC)
- The bot's demonstrated errors are only one of three problems. Second, even if the bot were foolproof it steps on the toes of our new proposal on use rationales, a much more serious effort to deal with noncompliant images. We can't have everybody running their own independent bot giving out deletion tags that conflict with ours. Third and finally, even assuming good faith, based on recent history we cannot rely on this editor to run a bot properly, within approved guidelines, communicating quickly and responsively to people who point out errors or otherwise question what he is doing. His ongoing approach to dealing with our questions only drives this third point home. I think we've been gently (and not so gently) suggesting that he give it up. I'm not sure what else we need to do. As far as I'm concerned the bot's rejection means this matter is closed, and I'm watching the bot approval pages to make sure it isn't opened again. Wikidemo 23:36, 1 October 2007 (UTC)
- Carcharoth, What's the "very small chance of getting your bot approved."?--OsamaK 09:37, 2 October 2007 (UTC)
- To be honest, I largely agree with Wikidemo. Given the history, it is very unlikely that your bot will ever be approved without drastic changes. I apologise if I raised your hopes. One thing though, I'm puzzled by this edit. I think you've inserted some text into Wikidemo's reply by mistake. Carcharoth 11:24, 2 October 2007 (UTC)
- Oh! sorry for this edit :). People here talk without examples for wrong tagging. My bot can be fixed if I understand what the problem is (with examples ;)).--OsamaK 14:33, 2 October 2007 (UTC)
- Then..--OsamaK 20:09, 6 October 2007 (UTC)
- Hi there! I'll ask BAG again as no one opposes this bot. ok?--OsamaK 12:09, 8 October 2007 (UTC)
- Then..--OsamaK 20:09, 6 October 2007 (UTC)
- Oh! sorry for this edit :). People here talk without examples for wrong tagging. My bot can be fixed if I understand what the problem is (with examples ;)).--OsamaK 14:33, 2 October 2007 (UTC)
- To be honest, I largely agree with Wikidemo. Given the history, it is very unlikely that your bot will ever be approved without drastic changes. I apologise if I raised your hopes. One thing though, I'm puzzled by this edit. I think you've inserted some text into Wikidemo's reply by mistake. Carcharoth 11:24, 2 October 2007 (UTC)
- Carcharoth, What's the "very small chance of getting your bot approved."?--OsamaK 09:37, 2 October 2007 (UTC)
- The bot's demonstrated errors are only one of three problems. Second, even if the bot were foolproof it steps on the toes of our new proposal on use rationales, a much more serious effort to deal with noncompliant images. We can't have everybody running their own independent bot giving out deletion tags that conflict with ours. Third and finally, even assuming good faith, based on recent history we cannot rely on this editor to run a bot properly, within approved guidelines, communicating quickly and responsively to people who point out errors or otherwise question what he is doing. His ongoing approach to dealing with our questions only drives this third point home. I think we've been gently (and not so gently) suggesting that he give it up. I'm not sure what else we need to do. As far as I'm concerned the bot's rejection means this matter is closed, and I'm watching the bot approval pages to make sure it isn't opened again. Wikidemo 23:36, 1 October 2007 (UTC)
- OsamaK, please read what people have written above. Find the alternative that has been suggested and tell us what it is and why it is preferable to tagging with {{subst:nsd}}. Then you might have a very small chance of getting your bot approved. Carcharoth 23:11, 1 October 2007 (UTC)
- Do not worry about how my bot finds unsourced images. Worry about bot's mistakes (I say that for 5th time ;)). anyway my bot finds any image's dicrapsion has (URLs, "source", "from" or other..) if there skipes if there is no list then taqqing with {{subst:nsd}}.. That's it! --OsamaK 19:45, 1 October 2007 (UTC)
- How is your bot dealing with non-hyperlinked sourcing statements? How would the bot handle a block of text explaining the origins and copyright holder for, say, an obscure, scanned pamphlet, as opposed to a simple statement that says, "Source: www.thesourceplace.com?" What criteria, exactly, is OsamaKBOT using to send things to speedydeletion? Our concern is that your bot will not be able to recognize allowable, but rarer/more-idiosyncratic forms of source citations, and that that will lead to sourced images being unnecessarily deleted. --Dynaflow babble 14:56, 1 October 2007 (UTC)
- Dynaflow, I understand most discussion here :). Thanks for creating User talk:OsamaK/Images. I'll use it. Dynaflow, can you give me some examples for these images who have tagged but it should not deleted? (There are some examples ,but most them have fixed in last tagging, So give me some mistakes from last tagging.) Wikidemo, /me can't understand why.. ;).--OsamaK 11:42, 1 October 2007 (UTC)
- I'll not wait MR.Wikidemo for long time, if you have a really problem let me know!--OsamaK 10:15, 11 October 2007 (UTC)
- I don't know what you're talking about. I have let you know in very clear terms that your bot is a bad idea and that I oppose it. Wikidemo 10:41, 11 October 2007 (UTC)
- Do you have another good idea for tagging all these images which are with no source? it's bad, why? I can't understand! ST47 says your because your English is bad (that's funny reason!).--OsamaK 04:34, 12 October 2007 (UTC)
- The point is, from what I can tell, that it's better not to tag images at all than to tag images with a fundamentally flawed approach. 17Drew 04:59, 12 October 2007 (UTC)
- Why not? there is very large log for images need tagging (then.. maybe deleted). that's Wikipedia's policy--OsamaK 05:13, 12 October 2007 (UTC)
- The point is, from what I can tell, that it's better not to tag images at all than to tag images with a fundamentally flawed approach. 17Drew 04:59, 12 October 2007 (UTC)
- Do you have another good idea for tagging all these images which are with no source? it's bad, why? I can't understand! ST47 says your because your English is bad (that's funny reason!).--OsamaK 04:34, 12 October 2007 (UTC)
- I don't know what you're talking about. I have let you know in very clear terms that your bot is a bad idea and that I oppose it. Wikidemo 10:41, 11 October 2007 (UTC)
STBotI
OsamaK has a point there. User:STBotI does seem to be tagging images for lack of sources. I've dropped a note off on the bot operator's talk page and asked them to clarify here what is going on. It is entirely possible that he is using AWB and human oversight, as opposed to unsupervised automated bot editing. Carcharoth 10:47, 12 October 2007 (UTC)
- I've checked STBotI's edit history, and I think it's necessary but problematic. Whether manual or automated it is suffering from the same errors as OsamaKBOT. For example, see Image:LeRay Toffing.jpg, tagged as "no source" despite having the statement "Author Unknown, Harvard Photo Archive, circa 1915" - a source! However, we don't have the other two concerns I raised with OsamaKBOT, an unwise purpose or a person who has proven unreliable in his running of bots. Regarding purpose, STBotI is filtering images as soon as they are uploaded for lack of use rationale, lack of source, and other problems, and generally notifies people of the error the same day they make the upload. Because it's so fast there is very little disruption and no chance of stepping on our new proposal to clean up old images - at most there is a very mild WP:BITE issue. In fact, I would go so far to say that SBotI is doing a great job and a wonderful, invaluable service. If it weren't operating we would have another 100+ bad images per day added to our stockpile. I really, really don't want to encourage anyone to run bots without authorization. If that's the case, let's ask ST47 to go ahead and ask for approval, and be ready to endorse the bot so people won't feel like they have to run bots on the sly (if not, sorry for any implied accusation).
- The errors in bot recognition of sources point to the need for an explicit "source=" field so it's as plain as day to human and bot whether someone has added a source or not. That's part of our new proposal, so going forward for new images it won't be a problem once the proposal gets implemented.
- Speaking of bad images, could someone with administrative privileges please summarily delete this one? We don't really need yet another copyvio picture of the pope with amateurish swastikas photoshopped on his robe :( Let us know what CSD criterion this fits. Wikidemo 14:39, 12 October 2007 (UTC)
- Yeah I don't know about the CSD criteria, but I deleted it anyways. Somehow I don't think this one is going to end up at WP:DRV... Calliopejen1 14:45, 12 October 2007 (UTC)
- As I already complained at User talk:ST47 this bot has made mistakes with my own images and with the sourcing of other's images, and it seems to have a rather narrow conception of what a source looks like. Dragons flight 17:37, 12 October 2007 (UTC)
- As I explained on my talk page the bot checks for watch words "my", "self", "from", "source", or any link, or tags, such as ".*self" and ".*user.*". This should be more than sufficient to catch any valid source, however I can revise these rules. --uǝʌǝsʎʇɹnoɟʇs 18:12, 12 October 2007 (UTC)
- But that doesn't even come close to catching all possible source notes like "Photograph of Greek statue of Hercules at the NY Museum of Art", "Scan of CD cover art for Raging Librarians", "Cover of Discover Magazine - March 15, 1997", etc. To say nothing of the fact that you missed my {attribution}, citing a Wikipedian as copyright holder, as unsourced. The variety of things that can be used as source descriptions defies your bot's simple logic. And since this task is not approved for your bot, I would ask that would stop it and discuss it at the bot approvals page. Dragons flight 18:25, 12 October 2007 (UTC)
- My bot has less problems it doesn't check free licenses, covers or screenshots. -2% has wrong tag as the test!- it works like ST47's bot (But I'm not in BAG like him!).. anyway 2% is very low for bot which is in test time.--OsamaK 22:30, 12 October 2007 (UTC)
- ST47, I ran across this approval for the bot to check for copyright tag and to guess about the presence of a fur (based on 25+ words in the description, I assume with special allowance for the generally approved templates). But I see no approval for tagging images for lack of source. If the bot has been approved could you point to that? If not, I would support approval. Re. Dragons Flight, I see the problem but now that we know how the bot works, why not simply include a statement "source=" rather than make the bot play guessing games? As long as the bot only goes through new images soon after upload, that's not too much to ask. In fact, maybe we should make it a requirement to include the magic word "source", as a better alternative to hand labor and errors in finding whether there's a source listed or not. Wikidemo 23:39, 12 October 2007 (UTC)
- Bots should behave in a way that never (or rarely ever) surprises normal users that are following our normal expectations. Telling a user that their image has "no source" even when the written description would satisy a human as to what the source is, would be surprising. The logic described so far is, in my opinion, too rudimentary to adequately do the intended job. If a "source" field were part of the upload form, I could understand having a check complain if that were left blank. However, at present "sourcing" is basically free-form, and I do think it is too much to ask for everyone to manually format their descriptions in specific ways to meet the limited capabilities of a bot. One of the reasons we have an approval process is to have trial edits. If we went through 50-100 edits of just the "no source" function, then I do believe the error rate would be unreasonably high at present just looking at my experience and some of the other complaints it has drawn. Dragons flight 00:31, 13 October 2007 (UTC)
- I agree about the need for bot tests and approval. However, we are going in the direction of requiring sourcing and other information to conform to some stricter format guidelines. Overall that will take a lot less effort, not more, than typing out everything freeform. For the moment, I think that having each of the 2,000+ image uploaders per day type "source=" or "==Source==" seems a lot faster a and better use of people's time than having volunteers read through each of those 2,000 images looking for the source, or having errors from the bots. Moreover, putting this and all data fields in a consistent format means they'll be a lot more useful down the road. Wikidemo 00:38, 13 October 2007 (UTC)
- You forget, the least amount of effort is to have no requirments and check nothing. ;-) However, if we are want the upload description to have parts, it really ought to be done be adding parts to the upload form. That said, some months back, I explicitly suggested adding a "source" field to the upload form and the developers rejected the idea as being too Wikipedia specific to warrant including in Mediawiki. (Which is a whole seperate rant since a project our size ought to have software tailored to our needs, rather than allowing ourselves to be constrained by the needs of small third party Mediawiki users.) Dragons flight 00:57, 13 October 2007 (UTC)
- I agree about the need for bot tests and approval. However, we are going in the direction of requiring sourcing and other information to conform to some stricter format guidelines. Overall that will take a lot less effort, not more, than typing out everything freeform. For the moment, I think that having each of the 2,000+ image uploaders per day type "source=" or "==Source==" seems a lot faster a and better use of people's time than having volunteers read through each of those 2,000 images looking for the source, or having errors from the bots. Moreover, putting this and all data fields in a consistent format means they'll be a lot more useful down the road. Wikidemo 00:38, 13 October 2007 (UTC)
- Bots should behave in a way that never (or rarely ever) surprises normal users that are following our normal expectations. Telling a user that their image has "no source" even when the written description would satisy a human as to what the source is, would be surprising. The logic described so far is, in my opinion, too rudimentary to adequately do the intended job. If a "source" field were part of the upload form, I could understand having a check complain if that were left blank. However, at present "sourcing" is basically free-form, and I do think it is too much to ask for everyone to manually format their descriptions in specific ways to meet the limited capabilities of a bot. One of the reasons we have an approval process is to have trial edits. If we went through 50-100 edits of just the "no source" function, then I do believe the error rate would be unreasonably high at present just looking at my experience and some of the other complaints it has drawn. Dragons flight 00:31, 13 October 2007 (UTC)
- But that doesn't even come close to catching all possible source notes like "Photograph of Greek statue of Hercules at the NY Museum of Art", "Scan of CD cover art for Raging Librarians", "Cover of Discover Magazine - March 15, 1997", etc. To say nothing of the fact that you missed my {attribution}, citing a Wikipedian as copyright holder, as unsourced. The variety of things that can be used as source descriptions defies your bot's simple logic. And since this task is not approved for your bot, I would ask that would stop it and discuss it at the bot approvals page. Dragons flight 18:25, 12 October 2007 (UTC)
I had a problem with the bot myself; images tagged with a source and a license still got tagged for no source and or copyright. I talked it over with the botowner on IRC and on my talk page, but I am still just slightly annoyed by it. User:Zscout370 (Return Fire) 00:44, 13 October 2007 (UTC)
- Agreed, bots, Image tagging, and deletion have collectively raised everybody's blood pressure a lot. Hopefully we're all coming to a group understanding about a better and more efficient way to do it, and getting some more accurate and less intrusive tools and procedures. If they won't automatically add it, we can still use templates and maybe some kind of wizard like thing to get everybody on board. Wikidemo 02:28, 13 October 2007 (UTC)
- Another example of the bot malfunctioning: [1] - someone was complaining about this at WP:MCQ. Calliopejen1 20:28, 13 October 2007 (UTC)
- Thanks for informing me. I've corrected this. --uǝʌǝsʎʇɹnoɟʇs 20:36, 13 October 2007 (UTC)
- Heh, are you kidding O ST47? your bot makes more mistakes. It should blocked! Stop it please (2% -maybe more?- is very high as what did you say at the IRC)--OsamaK 20:53, 13 October 2007 (UTC)
- Thanks for informing me. I've corrected this. --uǝʌǝsʎʇɹnoɟʇs 20:36, 13 October 2007 (UTC)
- Another example of the bot malfunctioning: [1] - someone was complaining about this at WP:MCQ. Calliopejen1 20:28, 13 October 2007 (UTC)
Copyright status
- Why does wikipedia allow the non-attribution of copyrighted works ?
- Even the license templates say "To the uploader: please add a detailed fair use rationale for each use, as described on Wikipedia:Image description page, as well as the source of the work and copyright information."
- Take a look through Category:Film poster images, you will see almost no image there has a phrase such as "Copyrighted by Universal".
- This is the only required feature by law, nowhere does it say in US law that you need a fair-use rationale. Stop arguing over the rationales missing the title of the article, this is a minor problem. If Jimbo is likely to land in jail, it's not going to be for a rationale missing an article name. There is absolutely 0 requirement by law for this. However not attributing the author of a Copyrighted work is illegal, even under Fair Use claim. Almost all other sites apart from blogs, even though they are much less strict than wikipedia, have a little line under the photo saying "Copyright XXX". And yet here there is systematically no attribution. And if you look through the backlog of pages tagged for missing an article name in the rationale, you will see that most in fact have no copyright information either. Sometimes there is a line like "Copyrighted by the film/music company", that is not sufficient and totally redundant anyway, as there is already a great big template above saying it is copyrighted. We need the actual name of the company, group, organisation.
- When the template says the source of the work and copyright information, do people really think it means write some bogus line like "Copyrighted by whomever made this image" ? Of course not, if it were that simple it would say it in the template already.
- Instead of spending time arguing with bot owners about tagging of pages with non-standard rationales, try to fix the real problem.
- How can I tag pages such as Image:Forbidden quest.jpg ? (Using an image I uploaded so as not to offend anyone in particular). Should I tag as a copyright infringement ? Or as missing rationale ? I can't see anything in Twinkle that really applies to this case. Maybe I should use the IfD process ? (please don't say db-self, I am asking as an example :P) Jackaranga 11:10, 8 October 2007 (UTC)
- For the example you gave I'd say just put "Copyright by CJ Entertainment" (asuming the distributor created the poster) somewhere on the image page. But yeah we can defenently get a lot better at stating who the copyright holder is, a lot of people (om either side of the non-free issue) seems to think that source = URL, for example some people would tag a logo that says "Microsofts logo, copyrighted by Microsoft" as having no source because there was no URL on the page... In generaly people seem to add the bare minimum of information on uploads. I think' I did a fairly good job on this image Image:Straxus robot mode.png I uploaded a few years back (although in retrospect it doesn't explicitly say that Marvel hold the copyright (though it's implied by saying it was reprinted with theyr permission). --Sherool (talk) 14:11, 8 October 2007 (UTC)
- Copyright notices are not a legal requirement, either for the copyright holder or any fair use re-user. We do ask for sourcing information, and we're working on affixing copyright ownership information to older images ona remedial basis. For the film poster example, it's clear in nearly all cases who owns the copyright - the studio. Nothing bogus there and no problem worth getting worked up over; going forward we can be more explicit about exactly what information we expect from the image uploader. Wikidemo 14:50, 9 October 2007 (UTC)
- Well nobody seems to know what deletion template I should use for Fair use images that fail criterion 10a) because they are lacking specific copyright attribution, (neither here or on administrator noticeboard), so I guess I will have to use the IfD process. Jackaranga 21:09, 14 October 2007 (UTC)
- I don't think you should be deleting these kinds of images. The source is obvious, reproduceable, and makes no difference. One gets movie posters from the distributor, directly or indirectly. Copyrights are owned by them or someone in the distribution chain or, in a few rare cases, the artist who made the poster - either the copyright notice is on the image, or if it isn't there's no way anyone here will ever find out.Wikidemo 21:47, 14 October 2007 (UTC)
- Well nobody seems to know what deletion template I should use for Fair use images that fail criterion 10a) because they are lacking specific copyright attribution, (neither here or on administrator noticeboard), so I guess I will have to use the IfD process. Jackaranga 21:09, 14 October 2007 (UTC)
- Copyright notices are not a legal requirement, either for the copyright holder or any fair use re-user. We do ask for sourcing information, and we're working on affixing copyright ownership information to older images ona remedial basis. For the film poster example, it's clear in nearly all cases who owns the copyright - the studio. Nothing bogus there and no problem worth getting worked up over; going forward we can be more explicit about exactly what information we expect from the image uploader. Wikidemo 14:50, 9 October 2007 (UTC)
- I would prefer to attempt to fix them, but there are so many that are missing it, and nobody seems to want to help (understandably) , I can never do all on my own, so it's better if I apply WP:NFC#Enforcement, it will at least get a few. Well I tagged a few yesterday anyway, I will wait and see if they get deleted before doing more. Jackaranga 08:26, 15 October 2007 (UTC)
Album covers in band articles, yet again
I have for some time been gradually trying to ensure that fair-use album cover images are not used in band articles. I last brought it here in August; here is the relevant archive. There was another relevant discussion here more recently.
My stance is (briefly) that using album covers in band articles is not good fair use (it contradicts the boilerplate text in the fair use license) and in many cases may also discourage people from uploading free images, where that is possible.
While wary of the danger of instruction creep, it might be worth having a specific clause in the policy that makes it clear that this sort of thing is not valid fair use. Wiggle-room could be left for truly exceptional cases, but I am concerned that each of these (what seem to me like) consensuses exists in isolation and there is not really a central point we can refer to on this, thus leading to the argument having to be refought for every single band article. Alternatively, it may be that I am wrong and that there is a project-wide consensus that this is, or may sometimes be, a valid fair use, although it is hard for me to see that. The same issues apply to book covers used in authors' articles, and maybe elsewhere too.
Here is the most recent manifestation of this lack of clarity. Your comments would be welcome, there or (preferably) here. --John 20:51, 11 October 2007 (UTC)
- I believe the NFC policy permits use of non-free album covers in band articles if the album in question is significantly discussed in the band article. The best case for this would be an article about a band with only one album, where there is no separate article about the album. -- But|seriously|folks 20:56, 11 October 2007 (UTC)
- I could see that as being a viable exception to what I believe to be the correct interpretation of our current policy. The boilerplate text I mentioned above currently reads, in part, "solely to illustrate the audio recording in question". It seems clear that this allows its use on the article about the album, less clear that it allows its use on band articles. I would fall back on Minimal number of uses. As few non-free content uses as possible are included in each article and in Wikipedia as a whole. Multiple items are not used if one will suffice; one is used only if necessary. At present we have "An album cover as part of a discography" as an example of unacceptable fair use; we would only have to refine that example slightly to make it clear that there are also problems with using fair use images willy-nilly to decorate articles like this. --John 21:08, 11 October 2007 (UTC)
Album covers are a significant part of the brand image created by most professional bands and, obviously, a central element of their marketing. They thus provide valuable information—irreplacable by text alone or by free images—for articles on virtually all professional popular music acts. As for how many such images should be called upon to serve this significant purpose, that clearly varies from band to band, from article to article. To have a sweeping policy clause that their appearance in band articles is not valid fair use is (a) not supported legally and (b) does not best serve Wikipedia's mission. We can promote the use of free images and the judicious use of fair-use images without restricting the latter in a way that unnecesarily undermines the quality of the encyclopedia.—DCGeist 21:03, 11 October 2007 (UTC)
- The key word, as DCGeist says, is 'judicious'. Ultimately each instance needs to be judged on its own merits. If an album is discussed significantly in a band article, there may well be fair use grounds for including the album cover - though in most cases 'significant discussion' of an album should be in its own article, where use of the album cover is perfectly reasonable. I expect John's main concern - and I'd agree with it - is where album covers are used in band infoboxes or distributed throughout the aricle to pretty it up without any real justification in terms of discussion of the relevant albums. Don't get me wrong, I'd prefer to see more liberal use of such images to enhance the appearance of articles, however fair use has been abused and will continue to be abused if we don't encourage clear justification for each instance of a cover's use. Cheers, Ian Rose 01:30, 12 October 2007 (UTC)
- I thought of a one hit wonder from my youth and there was an article on John Fred and his only significant album, Judy in Disguise (With Glasses). The use of album cover in this article on the band seems acceptable. It is only use once. There is no need to move it to the stub on the album. The same could apply to a book cover in an article about the author with only one major book. -- SWTPC6800 01:58, 12 October 2007 (UTC)
- Maybe a suggestion could be that by default fair use images like album and book covers should only appear once on Wikipedia, and that any exceptions would be argued on a case-by-case basis. The fair use rationale on the image should really take care of this and state categorically why its use on each article is essential, to comply with "As few non-free content uses as possible are included in each article". At the moment there is still an expectation among some that by default every album cover is acceptable on every band article. I have been gradually challenging this and I think it is starting to shift to a more encyclopedic and discriminating approach more in line with our stance on limiting fair use. I would not be averse to one or two fair use images on band articles, where they are particularly "iconic" or where the cover image itself is discussed in the text; but that leads us to some difficult case-by-case arguments. We need a coherent discussion to balance the needs of our articles to conform to our own core values with the value on any one article of any one image. I was going to make a hypothetical suggestion that we could include Image:TheClashLondonCallingalbumcover.jpg on The Clash article; then I looked at the file and saw that it
iswas (I sorted it out!) used 16 times! And this is the problem. As a beginning to solving this, I'd like to adjust the wording of the policy to make it clear that such use is not automatically permitted, and then try to thrash out some ideas of what good practice would look like. --John 03:08, 12 October 2007 (UTC)
- Maybe a suggestion could be that by default fair use images like album and book covers should only appear once on Wikipedia, and that any exceptions would be argued on a case-by-case basis. The fair use rationale on the image should really take care of this and state categorically why its use on each article is essential, to comply with "As few non-free content uses as possible are included in each article". At the moment there is still an expectation among some that by default every album cover is acceptable on every band article. I have been gradually challenging this and I think it is starting to shift to a more encyclopedic and discriminating approach more in line with our stance on limiting fair use. I would not be averse to one or two fair use images on band articles, where they are particularly "iconic" or where the cover image itself is discussed in the text; but that leads us to some difficult case-by-case arguments. We need a coherent discussion to balance the needs of our articles to conform to our own core values with the value on any one article of any one image. I was going to make a hypothetical suggestion that we could include Image:TheClashLondonCallingalbumcover.jpg on The Clash article; then I looked at the file and saw that it
- Just to be clear: right now, our policy (Wikipedia:Fair_use#Acceptable_images) states that if there is critical commentary on an item--say, an album of popular music--an illustration of that item's cover art for the purpose of identification of the item is acceptable fair use. The policy does not restrict where that critical commentary appears--it might be in an article specifically on that item or in an article covering that item and several others (e.g., because they were all produced by the same musical act or because they are all important examples of the same musical genre); it might appear in the primary running text or it might appear in the caption. And the policy requires neither that the cover art itself be iconic nor that the critical commentary directly address the cover art.
- Now, you've said you'd "like to adjust the wording of the policy"--obviously to make it more restrictive. Exactly what new sort of language do you have in mind? And why do you believe it is necessary to change the language in order to "thrash out some ideas of what good practice would look like"? Can't we establish a best-practice guideline without changing policy language?—DCGeist 03:35, 12 October 2007 (UTC)
- Either way would work, really. We need to make it clearer that album cover images in articles are not an automatic entitlement; we need to better and more clearly and transparently balance "Cover art: Cover art from various items, for identification only in the context of critical commentary of that item (not for identification without critical commentary)." (from Wikipedia:Fair_use#Acceptable_images) with "Minimal number of uses. As few non-free content uses as possible are included in each article and in Wikipedia as a whole. Multiple items are not used if one will suffice; one is used only if necessary." As to wording, a few extra words alongside the existing mention of discographies as an unsuitable venue for album covers. I'll try to draft up something suitable once I have thought about it some more. --John 04:00, 12 October 2007 (UTC)
- This doesn't have so much to do with the policy as it does with the issue itself. Using a cover is good to identify the work, but what is usually most helpful is a sample of a song or two from the album, in the context of discussion of how the album sounds. That's what I did with, for example, the Gwen Stefani article. 17Drew 09:03, 12 October 2007 (UTC)
- And, usually, a 20-30 second clip will work. It also has to be in OGG. User:Zscout370 (Return Fire) 09:09, 12 October 2007 (UTC)
- This doesn't have so much to do with the policy as it does with the issue itself. Using a cover is good to identify the work, but what is usually most helpful is a sample of a song or two from the album, in the context of discussion of how the album sounds. That's what I did with, for example, the Gwen Stefani article. 17Drew 09:03, 12 October 2007 (UTC)
- You are all overlooking the most important fact, which is that a large majority of album covers and film posters, have no copyright information, (even though it is the only actual information required by law, rationales being simply a wikipedia policy). The template tags and this page clearly say there should be copyright information. There is so much copyright infringement on wikipedia anyway, and a lack of will to delete 1000s upon 1000s of images at a time that there really is not much that can be done. If we take into consideration the lack of copyright info, I think we would have to delete more than half the images on wikipedia. Just look through Category:Album covers and Category:Film poster images, most don't have a line such as "© xxx Record Company". I think it is a case of don't ask don't tell unfortunately. Jackaranga 09:48, 12 October 2007 (UTC)
- Properly identifying the copyright holders of copyrighted material is indeed very important, but the participants in this discussion have not been "overlooking" that fact--it's simply not pertinent to the specific topic being debated in this thread.—DCGeist 09:52, 12 October 2007 (UTC)
- That's true sorry. Jackaranga 09:56, 12 October 2007 (UTC)
- You also raised this in another thread on this page, and were told that you were overstating the case. Maybe you should reply up there before repeating this down here? Carcharoth 10:53, 12 October 2007 (UTC)
- To be specific: "Copyright notices are not a legal requirement, either for the copyright holder or any fair use re-user." - in other words, we need to say who owns the image (what we call "source"), but we don't need to explicitly put a copyright notice on the image, though in fact we do. We have a generic "copyright" tag, and then a separate "source" bit. Carcharoth 10:55, 12 October 2007 (UTC)
- You also raised this in another thread on this page, and were told that you were overstating the case. Maybe you should reply up there before repeating this down here? Carcharoth 10:53, 12 October 2007 (UTC)
- That's true sorry. Jackaranga 09:56, 12 October 2007 (UTC)
- I'm sorry, I hope that in time you may forgive me. Jackaranga 11:41, 12 October 2007 (UTC)
- Only time will tell. ;) My contribution to this (and John is already aware of my position) is that, in accordance with the various laws applicable and the Wikipedia policy, it's actually ok to post all album covers (yes, all) on the artist's page in addition to the album article. It's not, however, appropriate to do this as we prefer (not, do not require, but prefer) context based inclusions only on the Wikiguidelines. In many cases, every album cover may be appropriate to add. In some, many may be. In some, yeah, none would be suitable to add. It's based on commonsense as to the quality that the image (or sound file, since someone tangented to that) provides to the page. There is no copyright or fair use law in the world that would prohibit use of every album cover by an artist being listed on the artist page and on the album page. We do choose not to do it because it's inappropriate, but it's certainly not illegal, and definitely within Wikipolicy, despite extracts from the policy that lead one to believe it's a questionable practice. --lincalinca 12:50, 12 October 2007 (UTC)
Suggested amendment
"The use of non-free media in lists, galleries, discographies, and navigational and user-interface elements usually fails the test for significance (criterion #8), and if it fails this test such use is unacceptable.
Here are other images that if non-free would almost certainly not satisfy the policy.
1. An album cover as part of a discography, as per the above. Album covers in band articles, and book covers in author pages, are not automatically good fair use. Additional fair use of this kind would have to be justified by a special rationale on the image page showing the unique contribution the image makes to every article in which its use is proposed." ...
Suggested amendment (2)
"The use of non-free media in lists, galleries, discographies, and navigational and user-interface elements usually fails the test for significance (criterion #8), and if it fails this test such use is unacceptable.
Here are other images that if non-free would almost certainly not satisfy the policy.
1. An album cover as part of a discography, as per the above. Album covers in band articles, and book covers in author pages, are not automatically good fair use. A separate and specific rationale must be provided on the image page for every particular use of a given fair-use image showing the unique contribution the image makes to every article in which its use is proposed." ...
Discussion
I've drafted something which seems to balance our wish to use these images occasionally under special circumstances with the principle of minimizing the use of nonfree images which the project has adopted. It is not perfect I am sure, but I think it would be better than the current situation. --John 07:13, 14 October 2007 (UTC)
- A book cover should be allowed in an author article to identify a significant book. No special discussion of the cover design or such is required. The cover can be used for identification. If an author wrote 10 books there is no need to have 10 covers. -- SWTPC6800 15:32, 14 October 2007 (UTC)
- Identification of significant works is a fundamental and valuable purpose served by illustration. Significant comment--even if only a sentence or two--about an author's significant books or a band's signficant albums most definitely has a place in author/band articles.
- As for the proposed change, I would trim the additional language in order to focus on the essential point: emphasizing what is already policy—a separate and specific rationale must be provided for every particular use of a given fair-use image.—DCGeist 20:56, 14 October 2007 (UTC)
- Thank you, I have amended my proposal with that in mind. If we went for something like that we would still probably need to thrash out here just what is and isn't good fair use in this area, as such a wide range of views have been expressed. Exemplars of good fair use in articles, anybody? --John 04:18, 16 October 2007 (UTC)
- The revised language looks good to me. Unquestionably, when a Wikipedia editor adds a fair-use image to an article, he or she needs to think about what purpose it will serve there--your proposed revision should help encourage that.—DCGeist 04:41, 16 October 2007 (UTC)
- As for an examplar, I really like the choice of two album covers that appears in Bob Dylan. The first is for Blonde on Blonde, the album most often cited as Dylan's greatest--a clear candidate for identification as particularly significant. The second is for "Love and Theft", the album most often cited as the best of Dylan's most recent...good lord...three decades (I don't like it much, but the critical consensus seems to regard it as his best since Blood on the Tracks). And, more than that, the substantive content of the cover is very similar to that of Blonde on Blonde: Dylan's face. In a very effective way, we can see how the man has changed over forty years...and has not. Despite his fame and prolific output, Dylan's album art has rarely been regarded as particularly notable, so the restrained use here strikes me as appropriate. The one cover image that does strike me as iconic in of itself, that of The Freewheelin' Bob Dylan, in fact does not appear. I think it could well, but you'd want a sourced sentence or at least a sourced phrase on it to explain its appearance.—DCGeist 08:43, 16 October 2007 (UTC)
- The statement that certain uses of album covers are not automatically fair use has nothing to do with the example it's attached to, that they're not permitted in discographies. It also goes without saying, because nearly everything is in that category. That's the default, that a separate rationale must be provided. The limitation on the album cover example is already reinforced by the "cover art" example of acceptable use farther down the page, where it says it is only for identification of the item. Another problem is that your proposal misstates the rule. "Specific" is problematic and ambiguous over the question of whether the rationale has to be different in each case. For album covers used for band articles and book covers in article pages, that's not really true. Each is evaluated on its own merits, but specificity has nothing to do with it. Moreover, "unique contribution" is not the standard for inclusion. The standard is whatever it says in Criterion #8 ("if its presence would significantly increase readers' understanding of the topic, and its omission would be detrimental to that understanding") as affected by the other criteria (e.g. #3, minimal use, #1, replaceability). I wouldn't encourage rewriting the proposal to try to match the rules. Though a noble ambition, I think it's a hopeless and boundless task to call out specific contexts where the rules have to be applied and repeat a statement or interpretation of the rules. Most people know this already and the few who don't will be reminded when they try to add inappropriate album covers to artist articles. Something like this might fit somewhere in the music projects as a guideline for how to write music articles, though. Just be sure to state the policy and guideline correctly. Wikidemo 09:22, 16 October 2007 (UTC)
Proposed change to the WP:NFC#10
Introduction:
- I would like to propose a change. An overwhelming number of images in Category:Film poster images and Category:Album covers do not specify the author or copyright holder of the poster/cover, probably over 90% have this problem.
Policy excerpts
The image or media description page contains the following:
- (a) Attribution of the source of the material, and of the copyright holder if different from the source.
It is important that you list the author of the image (especially if different from the source), which is important both for copyright and for informational purposes. Some copyright licenses require that the original author receive credit for their work. Examples:
- Author: The British Broadcasting Corporation ...
Proposed change (in red)
10 Image description page. The image or media description page contains the following:
- 10 (a) Attribution of the source of the material, and of the copyright holder if different from the source. See: Wikipedia:Citing sources#Images.
- 10 (b) A copyright tag that indicates which Wikipedia policy provision is claimed to permit the use. For a list of image copyright tags see Wikipedia:Image copyright tags/Non-free content.
- 10 (c) The name of each article in which fair use is claimed for the item, and a separate fair use rationale for each use of the item, as explained at Wikipedia:Non-free use rationale guideline. The rationale is presented in clear, plain language, and is relevant to each use.
- to
10 Image description page. The image or media description page contains the following:
- 10 (a) A copyright tag that indicates which Wikipedia policy provision is claimed to permit the use. For a list of image copyright tags see Wikipedia:Image copyright tags/Non-free content.
- And unless dictated by common sense
- 10(b) Attribution of the source of the material, and of the copyright holder if different from the source. See: Wikipedia:Citing sources#Images.
- 10 (c) The name of each article in which fair use is claimed for the item, and a separate fair use rationale for each use of the item, as explained at Wikipedia:Non-free use rationale guideline. The rationale is presented in clear, plain language, and is relevant to each use.
Benefits
- Only added six words
- Human readers still get as much information
- deletion bots are now rendered useless because they can't estimate if something is dictated by common sense, but can still be used for blatant violations such as lack of license tag, a role IMO more appropriate for a bot
- the NFC policy becomes more coherent with the spirit of the wikipedia policies, which are meant to be used by humans, (read WP:BURO)
- I feel WP:BURO is a more general policy, defining the spirit behind other policies, and WP:NFC should make an effort to be in compliance, and should be written in such a way as to leave a place for common sense
- Applying WP:NFC as it is currently would lead to the deletion of nearly all album covers and movie posters, as well as almost all other Fair Use images, this is in contradiction with WP:Ignore all rules, I don't believe this policy should be used thousands of times a day, as it is currently every time someone looks at almost any fair use image. If Ignore all rules were meant to be used all the time like that, other policies would not have been written. Instead of hiding behind this policy all the time, lets change WP:NFC.
Jackaranga 09:26, 15 October 2007 (UTC)
Comments
Comment - Thanks to Jackaranga for his interest in the problem of noncompliant images, but another proposal on the subject has recently gained nearly universal consensus (see heading earlier on this page, and Wikipedia:Non-free content criteria/Proposal ). Jackaranga's proposal conflicts with the earlier one, so I do not think we should approve it. The earlier proposal follows what I believe is a strong consensus that all images need source information and use rationales, but calls for templates that greatly simplify the process of uploading and using images, tagging noncompliant images by bot, and fixing tagged images. Although film posters are not in our initial batch of templates to develop (those are logos, book covers, album covers, and 2D and 3D artwork), it would be a logical next step to propose something for film posters once we get the initial ones in place. The target date is January 1, 2008. In the meanwhile I don't think the deletion issue is as bad as Jackaranga suspects. Those doing the large-scale tagging and deletion efforts have agreed informally to direct their efforts to new images uploaded after January 1, 2007 for the time being, so the "legacy" images uploaded before then are getting a reprieve until we can figure out what to do. The current round of large-scale deletions began sometime in June or July, and since then they have vigilantly challenged new images uploaded without sources and use rationales. So we only have a five month window, from January to May of this year, where there are a lot of noncompliant images in immediate danger of deletion. That's a lot of images, but nowhere near 90%. I don't think we should move to a "common sense" test on whether sources are needed or not. Of course common sense should be applied to all rules, but we need more guidance than that. The simplest approach, which i think has wide consensus, is to require people to say what they know about where they got an image and who owns the copyright, at the time they do the uploading. There's strong consensus for continuing to require use rationales every time an image is used, and film posters are no exception. But we will be making the process easier. Wikidemo 10:02, 15 October 2007 (UTC)
Jurassic Park screenshots
I would appreciate the opinions of other users on whether this screenshot Image:JPvelociraptor.png from the Jurassic Park film is fair use in anything other than the Jurassic Park articles. I am pretty sure it isn't, as otherwise we could use any depiction of anything in the article of that thing which I am pretty sure would send Wikipedia the wrong side of fair use law. Gustav von Humpelschmumpel 11:16, 16 October 2007 (UTC)
- I believe there is no such rule that states a movie screenshot can only be used in an article about the film. In Velociraptor the image is used to describe the Velociraptor in the film so I believe it is OK, as no free replacement of a Velociraptor in Jurassic Park could be found.
- However there are several problems:
- The image has no source or copyright holder info (i will tag it for deletion in a sec) (criterion 10a)
- It is used in some articles without a rationale
- A discussion about a dinosaur in a film may not be appropriate for a "scientific" or paleontology article, but I don't know about that and it's debatable, and is also a produce of the popularisation of wikipedia (ie more and more fancruft, and minor details from films and cartoons being spread across articles that many would consider unrelated).
- As to your particular concern though I believe from a Fair Use Point of view the image can be used in articles about the dinosaur as long as they discuss the dinosaur in the film. For another example see Image:Playboy 0603.jpg You may not see the relation at a first glance but it is in fact exactly the same because it's a case of the fair use image being used in an article about the subject depicted in the image. Jackaranga 12:40, 16 October 2007 (UTC)
- I don't believe the use of that magazine cover is fair use as it is not vital to the article to display it and could one could perfectly well explain that that person appearing on the cover of a magazine without using a copyrighted image. If we allow the depiction of a dinosaur in Jurassic Park film to be used in this article then it is going to open the floodgates to thousands of depictions of people, creatures, commercial items, brandnames, places, buildings to be used in articles which I believe will put a big stretch on the claim that Wikipedia is the free content encyclopedia. Gustav von Humpelschmumpel 16:23, 16 October 2007 (UTC)
- Unfortunately it's kind of too late as far as opening the floodgates goes :P. You may be right, I don't know, criterion 8 is very subjective, and one can wonder for example if having a picture of an album cover in an article about the album that does not mention the cover even once, satisfies this criterion. Jackaranga 16:46, 16 October 2007 (UTC)
- There is no such rule which states that a screenshot of a film must only be used in the film's accompanying article, only that a Fair Use image can be used if its presence would increase readers' understanding of the topic. Detailed Fair Use rationales were provided for uses in both Velociraptor and Deinonychus months ago. Please do not remove FU rationales (Gustav), and please do not tag images for deletion which have both detailed Fair Use rationales and which have clear source information ("from Jurassic Park" is the source) (Jackaranga). Firsfron of Ronchester 04:30, 17 October 2007 (UTC)
- Doesn't "if its presence would increase readers' understanding of the topic" mean of Jurassic Park and not any article that contains something in the picture? This is just a license for proliferation of dubious fair use. Gustav von Humpelschmumpel 15:48, 17 October 2007 (UTC)
- No please re-read WP:NFC#10a. "Attribution of the source of the material, and of the copyright holder if different from the source". Is "Jurassic Park" the copyright holder for itself ? No, a film can't be a copyright holder, only a person, an organisation, or a government may be. Jackaranga 09:42, 17 October 2007 (UTC)
- I am familiar with the policy. You tagged the image for deletion with {{di-no source}}, which stated the article had no source information, which is not correct. Now you are stating the image should have been deleted because the copyright holder wasn't provided. However, the licensing tag states "This image is a screenshot from a copyrighted film, and the copyright for it is most likely owned by the studio which produced the film." Tagging an image for deletion with a false tag isn't helpful. Removing detailed fair use rationales isn't helpful. I didn't upload this image, but I think the willful destruction of someone else's work is pretty awful, and telling someone to "re-read" the policy when you yourself didn't even use the correct tag is beyond the pale. Please avoid copyright paranoia. The image clearly falls into the WP:NFC#Acceptable_images category: "Film and television screen shots [used] for critical commentary and discussion of the cinema and television."; the image is accompanied by several paragraphs of critical commentary of the animal's depiction in the film. Finally, I will note that the article is a Featured Article, representing what the Wikipedia community (or a stringent portion of it) believes is the best work on Wikipedia, and there were no reservations about this image when it became a Featured Article. Since you also left a note on my talk page reminding me to re-read the policy, I'm cross-posting this to your talk page as well. Firsfron of Ronchester 16:07, 17 October 2007 (UTC)
- There is no such rule which states that a screenshot of a film must only be used in the film's accompanying article, only that a Fair Use image can be used if its presence would increase readers' understanding of the topic. Detailed Fair Use rationales were provided for uses in both Velociraptor and Deinonychus months ago. Please do not remove FU rationales (Gustav), and please do not tag images for deletion which have both detailed Fair Use rationales and which have clear source information ("from Jurassic Park" is the source) (Jackaranga). Firsfron of Ronchester 04:30, 17 October 2007 (UTC)
- Unfortunately it's kind of too late as far as opening the floodgates goes :P. You may be right, I don't know, criterion 8 is very subjective, and one can wonder for example if having a picture of an album cover in an article about the album that does not mention the cover even once, satisfies this criterion. Jackaranga 16:46, 16 October 2007 (UTC)
- I don't believe the use of that magazine cover is fair use as it is not vital to the article to display it and could one could perfectly well explain that that person appearing on the cover of a magazine without using a copyrighted image. If we allow the depiction of a dinosaur in Jurassic Park film to be used in this article then it is going to open the floodgates to thousands of depictions of people, creatures, commercial items, brandnames, places, buildings to be used in articles which I believe will put a big stretch on the claim that Wikipedia is the free content encyclopedia. Gustav von Humpelschmumpel 16:23, 16 October 2007 (UTC)
I believe what Jackaranga was after was what I've put at Image:JPvelociraptor.png#Source. This was already obvious from the summary, which said the source was the film "Jurassic Park". I would point out to Jackaranga that copyright ownership can change over the years, so the requirement should be to give the copyright details at the time of uploading, not the present details. Carcharoth 16:43, 17 October 2007 (UTC)
- Hey, you are making me say loads of things I didn't ! :p I agree that it's only fair to require the copyright as it was at the time of uploading, never said otherwise, I didn't delete any rationales from that image, I believe its use is valid, and is consistent with other cases. The only problem was it didn't say the name of the copyright holder. I believe {{di-no source}} is the template to use because the user notice that goes with it ({{Di-no source-notice}} says:
- I noticed that the file's description page currently doesn't specify who created the content, so the copyright status is unclear. If you did not create this file yourself, you will need to specify the owner of the copyright.
- Like I said though I asked here and as you can see my question was archived without any answer being given. I will ask again maybe I will get more luck this time. As for the copyright holder being obvious, I think the policy should be changed so that the copyright holder does not have to be stated if it is obvious (see above), however the current main proposal goes against this, as does the current policy. Note it says: and of the copyright holder if different from the source and not if different from that of the source. For example Image:Boeing-Logo.svg could be deleted by a bot for not saying ©Boeing. Jackaranga 21:03, 17 October 2007 (UTC)
- How would the bot know that Boeing is what is required here? And bots don't delete (normally). Humans do. That image requires a human to assess it. Bots are limited in what they can realistically achieve. Carcharoth 12:35, 18 October 2007 (UTC)
- Too true, that's one of the reasons I disagree with the new proposal, I don't think the wikimedia resolution meant that people had to go as far as making each element of the rationale machine-readable. Perhaps it will be for the best though in the end. Jackaranga 16:59, 18 October 2007 (UTC)
- How would the bot know that Boeing is what is required here? And bots don't delete (normally). Humans do. That image requires a human to assess it. Bots are limited in what they can realistically achieve. Carcharoth 12:35, 18 October 2007 (UTC)
- Hey, you are making me say loads of things I didn't ! :p I agree that it's only fair to require the copyright as it was at the time of uploading, never said otherwise, I didn't delete any rationales from that image, I believe its use is valid, and is consistent with other cases. The only problem was it didn't say the name of the copyright holder. I believe {{di-no source}} is the template to use because the user notice that goes with it ({{Di-no source-notice}} says:
RfA candidate's attitude to NFC
I'm concerned that the candidate seems to be pushing the line that the differential between WP's policy and what is strictly required by (US) law is undesirable, and that s/he will work to stop deletions on that basis. I'm not making a judgement on the nomination as whole, but this does need to be cleared up before the keys are handed out.
Wikipedia:Requests_for_adminship/Remember_the_dot#Neutral