Wikidata:Properties for deletion/P10589

Property:P10589

MangaDex title ID (P10589): (delete | history | links | entity usage | logs | discussion)

The website MangaDex is a website for scanlation (Q557923) and therefore gives people access to copyright protected works for free without holding a license to publish it or consent of the copyright holder. The website infringes copyrights and I don't see any reason why wikidata should link items to such a website.

See also: Wikidata:Project chat/Archive/2022/11#Mangadex?

—Christian140 (talk) 07:58, 7 November 2022 (UTC)

•  Weak oppose Linking to a website doesn’t mean endorsing its practices. The question in this case would be whether the benefits of having this site linked outweigh the concerns about legal issues. --2A02:8108:50BF:C694:E0F2:7F6B:7EAD:26F9 10:12, 7 November 2022 (UTC)

 Delete in the past Japanese publisher sent cease and desist letters to aggregators for Scanlation. Having this property might essentially makes us an aggregator for Scanlation and thus opens up the possibility of legal threads against Wikimedia. I think it's ideal if your community can self regulate in this regard and delete the property without needing to interact with Wikimedia legal. ChristianKl ❪✉❫ 15:12, 7 November 2022 (UTC)

(Japanese publishers having sent cease and desist letters for scanlation sounds … interesting given that it’s not their – arguably monetary – rights infringed upon, but those of the author, and in Japan itself it wasn’t possible until a few years ago to take legal actions on behalf of a third party against copyright violations. Just as an aside. --2A02:8108:50BF:C694:E0F2:7F6B:7EAD:26F9 15:53, 7 November 2022 (UTC))

Scanlation websites are seldomly located in Japan, so the details of Japanese law don't matter here. ChristianKl ❪✉❫ 00:15, 10 November 2022 (UTC)

Admittedly yes, therefore an aside (basically saying that they are taking advantage of another country’s legal provisions where this would not be possible in their own country – indeed not of interest here). --2A02:8108:50BF:C694:1432:F47C:55CC:B105 19:37, 12 November 2022 (UTC)

 Delete I created this property honestly not knowing it was a scanlation website or what scanlation was. Linking to a website that distributes copyrighted material is basically assisting in that distribution which is illegal. Lectrician1 (talk) 20:18, 7 November 2022 (UTC)

 Delete As for scanlation sites, Japanese and U.S. publishers declared in a joint statement in 2010 that they are illegal. By making them available for free, they are infringing on the financial benefits that copyright holders rightfully deserve. Afaz (talk) 04:01, 9 November 2022 (UTC)

While this is undoubtedly true, there are no financial benefits for copyright holders anyway if nobody publishes their work commercially in a country. If there is no “official” translation that is sold in, e.g., the US, anyone who wants to read it (in English) there has to resort to “unofficial” translations, which have no choice but to infringe on copyright. I don’t want to endorse copyright violations, in no way, but the “financial” point of view doesn’t get us anywhere here. That said, what was the point of creating links to the specific site discussed here in the first place? What benefits were seen in linking it? --2A02:8108:50BF:C694:E83C:EBFF:49AF:23CC 10:18, 9 November 2022 (UTC)

As far as the laws are concerned people can import Japanese comics whether or not they are translated. The Berne convention exists to give mutual recognition of copyright and not require products to be marketed in a country to be protected in that country. ChristianKl ❪✉❫ 00:13, 10 November 2022 (UTC)

That’s not the point. There are probably many people in the world who want to read Japanese comics, but cannot read Japanese. That’s the reason why scanlation (and regular translation) exists in the first place. Of course it would be better if they paid the original authors, but the author doesn’t get any money regardless of whether someone abroad reads their comic in scanlation form (without paying) or doesn’t read it at all. Hence the “financial” point of view doesn’t get us anywhere here. There’s more to copyright than remuneration (and the Berne convention presumably exists regardless of financial considerations). --2A02:8108:50BF:C694:E83C:EBFF:49AF:23CC 12:45, 10 November 2022 (UTC)

Copyright is driven by actual laws. You might disagree with those laws but they exist. Scanlation clearly creates deriviative works of copyrighted works. In the US context where Wikimedia has it's legal home, that's forbidden by copyright law unless you have permission or can argue for fair use. Courts have made many rules on copyright and have developed a concept of financial interests in the process. You might not like it or disagree with it, but that's still the law of the land. ChristianKl ❪✉❫ 12:20, 12 November 2022 (UTC)

Neither do I disagree with copyright laws (where did I claim that?) nor do I deny that scanlation violates them. All I’m saying is that it does not hurt the authors financially and that the claim by Afaz that they infringe on “the financial benefits that copyright holders rightfully deserve” is therefore misleading. Of course they have “financial interests” – they are selling their works in Japan, after all –, but that’s different from “financial benefits”. (Easy example: A greengrocer has a financial interest in getting vegetables sold, but no financial benefit if nobody buys them – a reason for which might be that all the people who would like to buy them live in another city. Does this make stealing the vegetables from the greengrocer and giving them away for free in that other city legal? Obviously not. Does the greengrocer have a financial damage? No, he doesn’t receive money for the vegetables anyway.) But let’s stop this pointless discussion here – both of us agree that scanlation is a copyright violation, while we seem to disagree on why it is (or maybe not; the deriviative work argument is independent of financial aspects, and I’m not sure the US context is actually necessary for it, but anyway). The reason why it derailed was probably my justification for the continuing widespread existence of scanlation despite its illegality – which is unnecessary for the point I wanted to make, I think (now). --2A02:8108:50BF:C694:29E6:BE9C:1625:78C8 20:07, 12 November 2022 (UTC)

The reason why I’m so nit-picky about this is that it often gets mixed up. If remuneration were the problem, scanlators could solve it by taking money from their “customers” and using it to pay the original authors – but in the absence of permission to do so this would still be a copyright violation. That there is more to copyright than remuneration can also be seen in the advent of Creative Commons licences, where copyright holders waive their right to remuneration without (necessarily) waiving other rights they deserve, such as proper attribution (a misconception many have: “It’s free, so I can use it any way I want”). --2A02:8108:50BF:C694:29E6:BE9C:1625:78C8 20:33, 12 November 2022 (UTC)

•  Question I see questions of copyright here but if this was actually an issue of concern with wiki projects merely linking, then wouldn't this be a major issue with wiki projects linking to the Internet Archive (Internet Archive ID (P724) and the works there that are still under copyright? If there isn't an issue with that I dont see the issue here. -Jeanjung212 (talk) 20:51, 18 November 2022 (UTC)

@Jeanjung212: Can you link any item there where the site infringes the copyright? On the first look, all the content looks like public domain and creative commons as well as previews. Also, not that copyright is not the problem. There are also links to Netflix. Copyright infringement is the problem. --Christian140 (talk) 07:58, 19 November 2022 (UTC)

This website (linked from Tetris (Q71910)), for example, doesn’t seem to be public domain, so technically (ianal) the Internet Archive is infringing on the creator’s copyright by making a copy of it available. (It’s just that no one bothers to sue the Internet Archive, I think.) --Data Consolidation Officer (talk) 21:11, 21 November 2022 (UTC)

Hmmm. But on all websites where every user can upload content, copyright infringement happens. On wikipedia and commons, too. Just, eventually it gets deleted. However, for mangadex, copyright infringement is the core of the website. For internet archive, they have this site: Rights – Internet Archive Help Center. So, you could report content you think that infringes copyright. But here, I am actually not sure if it is copyright infringement. A lot of old software is made available for free and you can download them from many serious websites. --Christian140 (talk) 08:01, 22 November 2022 (UTC)

Internet Archive, or archiving in general, might even be covered by Fair Use (I simply don’t know). And given the large number of pages archived there, reporting copyright violations would be a Sisyphean task. As I stated below, I don’t think there’s a legal issue with mere linking, but P10589 is very dispensable anyway. --Data Consolidation Officer (talk) 20:23, 22 November 2022 (UTC)

The Internet Archive is also recognised as a library by the US government. Thibaut (talk) 10:39, 13 February 2023 (UTC)

•  Comment Linking doesn’t mean endorsing, afaik, so the site’s copyright violations alone wouldn’t be valid grounds for deletion of this property. Having a look at the property proposal discussion, however, it seems that the property was created without thorough discussion, basically because “I think properties for it would be useful”. Wikidata should, imho, be extremely restrictive with respect to which external databases it chooses to systematically link, given the considerable effort of maintaining such link collections, avoiding inconsistencies and so on. That’s why I’d tend to vote for deletion at the moment, unless someone provides a good reason why having external identifier links to the site in question is essential. --Data Consolidation Officer (talk) 21:03, 21 November 2022 (UTC)

  It's 100% endorsing. You're exposing the copyrighted works to a wider audience by linking to them. You're clearly assisting in their distribution. Lectrician1 (talk) 21:30, 21 November 2022 (UTC)

  Sorry, I should have made clearer that I was specifically talking about legal issues. There can of course be ethical issues with linking (depending on intention), but afaik (and ianal, so please correct me if I’m wrong) courts in various contries have established that website operators cannot be held liable for criminal violations by other sites they merely link, so Wikimedia Foundation could not be (successfully) sued for those links or something like that. Anything else is a question of whether we, as a community, want those links, but as I said, I don’t really see any reason anyway why we should. --Data Consolidation Officer (talk) 20:12, 22 November 2022 (UTC)

  There's a difference between having a simple link to MangaDex and having a system on Wikidata that tells Wikidata users for every manga, the exact page where they can download a copyright violating copy of that manga. Having a link to every single manga, is like torrent websites that link to individual content and torrent websites do face legal problems. ChristianKl ❪✉❫ 11:12, 27 November 2022 (UTC)

  Which all the more raises the question why Wikidata would want such a system in the first place; a question the answer to which I still don’t see. Given that it took nine months (the property was created in early April) until someone noticed that there are copyright violations linked, I wouldn’t consider any claim about copyright infringement endorsement intentions plausible (in contrast to torrent sites; and indeed those links will have been created in good faith in most cases), but let’s the lawyers fight that out (or not). --Data Consolidation Officer (talk) 19:36, 27 November 2022 (UTC)

   Delete, as indicated, on the grounds that there has been no good reason given why having external identifier links to the site in question is essential. --Data Consolidation Officer (talk) 19:36, 27 November 2022 (UTC)

• As the property proposer, I have no objection to deletion based on the arguments provided. Tol (talk | contribs) @ 01:01, 30 November 2022 (UTC)
•  Comment I won't comment on the actual scanlated content MangaDex works, but they are a gold mine of information, as they maintain links to many of the other manga databases on the internet. Maybe deletion can be waited on until my bot is able to copy as many of the external linkings as possible. In that case, there is another issue brewing, as whenever my bot pulls information from MangaDex it makes a reference and puts the full URL into the reference URL property, although I theorize it would be trivial to clean those up (SPARQL query for stated in MangaDex would bring them all up). RPI2026F1 (talk) 01:49, 9 December 2022 (UTC)

  Alternatively, would the problem not solve itself if the link was simply removed, rather than deleting the entire property? RPI2026F1 (talk) 01:51, 9 December 2022 (UTC)

   Comment Looks like the property is going to be removed. It seems reasonable however that the actual removal can be put on hold for a period of up to 3 months (or less) to allow for links to other sites to be extracted from this identifier. Infrastruktur (talk) 20:54, 9 December 2022 (UTC)

  Yes, I'm trying to get a whole bunch of properties created so I can extract maximal information from the source. I can see about 8 or 10 new properties being partially populated on top of what is already being extracted. RPI2026F1 (talk) 16:17, 21 December 2022 (UTC)

•  Delete per the above copyright and legal concerns. ミラP@Miraclepine 19:47, 9 December 2022 (UTC)
•  Keep Wikidata is neither a judge nor a police officer. Besides problematic links, the database contains other useful data as well (date of publication, artist, genres, alternative titles, even links to official shops). --Jklamo (talk) 18:30, 21 December 2022 (UTC)
•  Keep per Jklamo. We shouldn't censor the identifier of this useful database unless we have clear evidence of law. Laftp0 (talk) 13:47, 22 December 2022 (UTC)
•  Delete per the above copyright and legal concerns, we already have better manga/anime databases properties like ANN, MAL and Anilist that don't host illegal content, we don't need some random scanlation website. --Thibaut (talk) 14:30, 20 January 2023 (UTC)

 Weak oppose Linking isn't endorsement, and it's very useful for getting links to other manga services. Although, ultimately if it is deleted we can still use its API to grab links as long as one of AniList/MyAnimeList/Kitsu are linked, so it wouldn't be the end of the world. Ultimately, my opinion would be based on the opinion of the Wikidata team as to whether this kind of site should be linked to. FWIW, from what I can tell MangaDex does respect the wishes of copyright holders if they do request a takedown, although whether that redeems the site is debatable. Nicereddy (talk) 02:09, 21 January 2023 (UTC)

 Keep, Mangadex is just a platform (non-commercial and ad-free), which like other platforms like YouTube or Facebook could be used for publishing anything, but no evidence provided by nominator that this website opposes copyright holders in any way (other than "it is free, therefore it is illegal"). The rules are pretty restrictive there, cases when obtaining a license is required are mentioned. Lockal (talk) 04:51, 13 February 2023 (UTC)

The “rules” are not that “restrictive”:

Any scanlated release is allowed to be uploaded regardless of the existence of official translations […]

And even if there’s no official translation or it’s out of print, translating something that is copyright-protected and uploading it to the web is still illegal per the Berne convention (see above).

The difference with Facebook or YouTube is that they disallow illegal content and respond to DMCA requests. Thibaut (talk) 09:53, 13 February 2023 (UTC)

A single proof that Mangadex hosts illegal content and does not respond to DMCA requests? Lockal (talk) 16:04, 12 March 2023 (UTC)

The website is literally designed to host illegal content, like Christian140 said above: it's at its core.

The mere fact that their domain reseller and/or Cloudflare had to kick them away because of the number of DMCA requests they were getting is a strong indicator ([1][2]). One of these requests was from VIZ Media, which is owned by two major Japanese publishing companies (Shueisha and Shogakukan).

Now, please enlighten me how a website hosting full manga releases translated in multiple languages without the copyright holders' permission doesn't infringe Japanese copyright law and therefore the Berne Convention? Thibaut (talk) 18:17, 12 March 2023 (UTC)

My point was that there is a way to legally host a scanlate website (or any other types of derivative works). It is not difficult to receive a permission from copyright holder to publish your own translation under well defined conditions: non-commercial (optionally providing additional details to help copyright holder to verify that translator are not seeking profit with translation) and only on specific website. Actually, I did it multiple times (not for manga, but it does not matter). Consider that all mindful translators received a permission: it is called "presumption of innocence".

The links you provided mentions that some time ago a fan group that has been coloring the Boruto manga used official scanlation, which resulted in DCMA takedown. Such types of uploads are not allowed on MangaDex:

Scans of physical official releases or rips of digital official releases/webcomics from official sources, such as original releases (raws) or officially translated releases, are not allowed to be uploaded.

Lockal (talk) 10:58, 13 March 2023 (UTC)

 Delete - per legal concerns. Under Japanese copyright law, it is a violation of copyright law to link to a site that is known to copyvio. Links that may violate laws should not be kept. The server for this site may not necessarily be located in Japan, but it should be sensitive to the law. The server for this site is not necessarily located in Japan, but I think it should be as sensitive to the law as possible. Syunsyunminmin (talk) 10:15, 19 February 2023 (UTC);edit  – The preceding unsigned comment was added by Syunsyunminmin (talk • contribs).

 Delete - Link to a site that is a violation of copyright law. --Fralambert (talk) 19:22, 12 March 2023 (UTC)

 Keep As far as I am aware:

1. MangaDex acts in full compliance with U.S. law under the DMCA act

2. MangaDex has never faced charges for hosting what they do. (they have been subpoenaed once, but that's very much not the same thing) Binarycat32 (talk) 00:52, 31 January 2024 (UTC)

@Binarycat32: One of their staff literally says that "Mangadex doesn’t adhere to DMCA requests".

Speaking of DMCA, see also above. Thibaut (talk) 06:06, 31 January 2024 (UTC)

•  Delete the copyright problem alone is enough to delete (in itself and because this make this website less likely to be perennial). In addition to that, I see that this property is use only on ~2800 items and ~25000 references, plus in most cases there is other identifiers and others references. It's maybe a "gold mine of information" but it's clearly not the only one, deleting these data would mean only a negligible lack of information in the end. Cheers, VIGNERON (talk) 15:30, 1 February 2024 (UTC)

  "in most cases there is other identifiers"

  a lot of those identifiers have been imported from mangadex. as far as i'm aware, mangadex is the only site other than wikidata that maintains links to other manga sites in this way. Binarycat32 (talk) 22:05, 2 February 2024 (UTC)

  It is possible to correlate titles with their MangaDex IDs but it would require maintaining a database of MD ids and other IDs linked to MD and then regularly updating this database both with new titles and if existing titles change their IDs. RPI2026F1 (talk) 02:39, 3 February 2024 (UTC)

  There's already a bot that does all of that, besides changing IDs, which happens approximately never. Binarycat32 (talk) 00:09, 5 February 2024 (UTC)

  Yea, that's me, I wrote the bot that does that RPI2026F1 (talk) 01:26, 5 February 2024 (UTC)

  Yeah I wondered as much, but I figured someone wouldn't write a bot then act like it didn't exist. Binarycat32 (talk) 18:30, 5 February 2024 (UTC)

  The current problem with said bot is that it only adds stuff from given properties. Basically, it won't try to approximate a MangaDex ID from just the MAL ID, but it will add a MAL ID if there is a MangaDex ID because MangaDex lists a MAL ID for that entry. RPI2026F1 (talk) 13:19, 6 February 2024 (UTC)

  finding the mangadex id is fairly easy with animanga-db-matcher[3]. however, this tool does not work well (or at all) for several other sites that often have their identifiers listed on MangaDex. Binarycat32 (talk) 18:51, 8 February 2024 (UTC)

  Yea that would be my bad, I wrote that tool as well but I had a very limited understanding of React and I found it easier to work on the auto-import bot than the finder. The reason it's not as effective is because I designed it for items that had no identifiers whatsoever, and at that point title searching was the only way to find potential IDs. I could make a future update that looks up other identifiers as well. RPI2026F1 (talk) 21:03, 8 February 2024 (UTC)