Wikipedia talk:GLAM/ARKive

Attribution template

Hi. In case it's useful, I've created an ARKive attribution template at {{ARKive attribute}}, which could be added after the references section to acknowledge the inclusion of ARKive materials into the article under CC-BY-SA-3.0. Mike Peel (talk) 22:44, 13 July 2011 (UTC)

Thank you. Is it usual to put something like that on an article, as well as the banner on the talk page? I'm happy with either, or both, but am unclear as to which is the norm. If we are going to use it, I;ll add a link to the ARKive article, like that in {{ARKive}}. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 11:42, 20 July 2011 (UTC)

I think so, I've certainly noticed it on articles written using information copied from NASA. SmartSE (talk) 18:13, 20 July 2011 (UTC)

OK, Thank you. I've mentioned it in the project guidelines and added it to the improved articles where it wasn't already present. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 11:39, 22 July 2011 (UTC)

An instance of the template has been removed from Hyacinth Macaw, with the edit summary "…unneeded on the article page. We don't have attribution like this elsewhere". Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 13:07, 28 July 2011 (UTC)

Apparently the person who said that is not familiar with Category:Attribution templates. We have a lot of attribution of the same sort. But the template should go at the very bottom of the article, IMHO. — Carl (CBM · talk) 13:31, 29 July 2011 (UTC)

Maybe I am missing sth...

I am failing to find donated text. Does it inlude entire website? Bulwersator (talk) 15:14, 18 July 2011 (UTC)

No; they have donated 200 articles, the first 80 or so are those listed on the project page. More will follow. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 12:46, 19 July 2011 (UTC)

comments/questions

The animals they cover are pretty notable. For some like Hawkbill, I would be unlikely to want to cut and paste their prose. Our article is already an FA and longer. Even if they had good stuff, would need to integrate it properly and probably just get the base citations and write it in without an Arkive attribution. At the most, it is another competing article on the Hawbill, which I might choose to read and learn from, but no priority for me.

There are a few though, where we have almost nothing (e.g. Common Box Turtle) where just grabbing their prose would make an article pretty quick. could see this being value add. After all we have let it languish for years. Yeah, we could build it from scratch, but have not. So why not take the content. I know Citizendium to Wiki project built some content this way.

Few things I'm not clear on:

1. How to attribute to Arkive? I mean their text itself is referenced, so what we take from them is the prose. Is a citation appropriate for that, or some edit history remark? And what are the exact nicities around it? I guess a good citation would make sure we spec the date and webpage and all that.
2. Is this "approved"? HAve copyright issues been thought through? Is this more than a one person initiative? Imagine running one of these by Mottenen.
3. how physically to cut and paste the text (especially the cites). Or doees that need to be done all manual or any tricks?
4. Anything else to watch out for?

TCO (reviews needed) 02:31, 19 July 2011 (UTC)

I've taken the liberty of numbering your questions:

1. Please see the example edit given, for African Elephant, and the notes on the project page.
2. No, this is not simply a one-person initiative; I have been appointed by Wikimedia UK and ARKive, jointly, as explained on the project page and the Wikimedia UK page to which it links.
3. as [1]
4. The five pillars apply here, as elsewhere.

Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 12:51, 19 July 2011 (UTC)

screw it, will just do one!

I'm going to do common box turtle. You can look it over and see what I missed. My main concern is copying the cites...I can manage the section arrangement of extra text. I'll be your first customer. — Preceding unsigned comment added by TCO (talk • contribs)

Thank you. Please be sure to update the project page when you're done. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 12:51, 19 July 2011 (UTC)

Is Arkive a reliable source for the wiki?

• In the references section Archive says; "Authentication. This information is awaiting authentication by a species expert, and will be updated as soon as possible. If you are able to help please contact: arkive@wildscreen.org.uk". With a notice like this, I think that we can not simply merge Archive into the Wikipedia, as it appears to me that Archive (or at least the part of Archive for the Hyacinth Macaw) does not satisfy the Wiki's reliable source guidelines. Snowman (talk) 07:46, 19 July 2011 (UTC)

disagree. We are not citing them as a source. Their content is reffed itself. For box turtle, pretty decently (I know the field). All we are using is their prose. It's no different than getting info from one article of wiki to another.TCO (reviews needed) 07:51, 19 July 2011 (UTC)

Agree with TCO. also, you made the same comment at Wikipedia talk:WikiProject Birds, where I replied: "the text in the ARKive article on the Hyacinth Macaw is referenced to multiple reliable sources; per the African Elephant example, we can lift the text, with references, directly from ARKive. Accordingly, I have done just that, by way of example". Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 12:53, 19 July 2011 (UTC)

The problem then is that according to referencing, if we attribute fact 'x' in an article, which should have laid eyes on the material we are attirbuting it from. Hence the ARKive webpages are ok in the interim but past that we need to be looking at the sources they point to. Casliber (talk · contribs) 03:07, 28 July 2011 (UTC)

That's not always the case. For example, if we just copied an entire Simple English Wikipedia article here, references and all, without changing anything, there is no reason we need to read all the references inside it first. We do need to attribute the copyright, but we are not using Simple English Wikipedia as a reference in that case. — Carl (CBM · talk) 13:28, 29 July 2011 (UTC)

DYK - not original content

You currently recommend to nominate articles expanded using the info from ARKive for DYK, but I don't think you should as WP:WIADYK states "Try to select articles that are original to Wikipedia (not inclusions of free data-sources)..." therefore meaning they should be excluded. It's unfortunate, but I know that at DYK we have rejected articles made from free sources before, so it only seems fair that these shouldn't be eligible. I think it is a reasonable rule too, as it so much easier to create articles from sources like this. SmartSE (talk) 18:08, 20 July 2011 (UTC)

My reading of that is that it refers to articles created by cut'n'pasting a full article from another source, not an article which has been expanded by weaving donated text and references into an existing article; or indeed including donated text and original material such as in the currently-nominated Terrapene carolina. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 09:00, 21 July 2011 (UTC)

Brachypelma albopilosum

I was unable to find the Curly haired Tarantula (what redirects to Brachypelma albopilosum). Bugboy52.4 ¦ =-= 16:23, 21 July 2011 (UTC)

If you look at Brachypelma albopilosum, you will see it says "Brachypelma albopilosum is a species of tarantula known commonly as the Honduran curlyhair or simply curlyhair". The ARKive article confirms that "Curlyhair tarantula" is another name for Brachypelma albopilosum. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 16:41, 21 July 2011 (UTC)

Peer review request: African elephant

I've asked for peer review of African elephant. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 22:14, 21 July 2011 (UTC)

authentication?

On rock python, the Arkive article has an authentication. Not looking to make work, but think noting that on article talk page, maybe? I do think it is a useful reference if we ever take the article to FA, to try to get outside peer review (really would not bother an academic until then).

Authenticated (28/04/09) by Dr. Luca Luiselli, Senior Researcher in Ecology, Institute Demetra, Rome, Italy. http://www.intecol.net/pages/002_personal.php?id=lucamlu

TCO (reviews needed) 00:23, 22 July 2011 (UTC)

Perhaps we should add a parameter to {{ARKive attribute}}? Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 11:26, 22 July 2011 (UTC)

Possible plagiarism at ARKive: Galapagos penguin

I was just starting to import text from the ARKive article on the Galápagos Penguin, when I noticed that the ARKive article was very similar, though largely better-written than the Wikipedia article. Examining the sources used, it is clear that at least some parts (like the main part of the "Description" section) on the ARKive page are closely copied from the sources cited. These sources were those formerly given in the external links section of the Wikipedia article, and it looks very much like the article was written using the Wikipedia article. (I don't think it's the other way around, given the nature of the Wikipedia article; the Wikipedia article's usage of these sources doesn't include any outright plagiarism or clear close paraphrasing, as far as I've seen yet.) Here's the ARKive text:

This diminutive penguin has a black head and upperparts, with a narrow white line extending from the throat around the head to the corner of the eye. The underparts are white with two black bands extending across the breast. The upper part of the bill and the tip of the lower part of the bill are black, the rest of the bill and a bare patch around the eye and bill are pinkish yellow. Although the sexes are generally similar in appearance, males are larger than females.

The second sentence is cited to a now non-existent BirdLife International page. Here's the text from the International Penguin Working Group, the main source:

They have a black head and upperparts, with a thin white line running from the throat, up around the head to meet the corner of the eye. The underparts are white, but are bordered by a black line which extends down to the blackish legs. The upper bill and tip of the lower bill are black, with the remainder of the lower bill and surrounding skin around to the eye being pinkish yellow. The females are smaller than the males, but have similar plumage.

And the Wikipedia article:

They have a black head with a white border running from behind the eye, around the black ear-coverts and chin, to join on the throat. They have blackish-grey upperparts and whitish underparts, with two black bands across the breast, the lower band extending down the flanks to the thigh. Juveniles differ in having a wholly dark head, greyer on side and chin, and no breast-band. The female penguins are smaller than the males, but are otherwise quite similar.

The ARKive article also is poor writing in other ways, as with the citation to the Penguin Taxon Advisory Group (archive), which refers to a genus (Spheniscus) as a family, and from which the Galápagos Penguin was said to be the "third smallest penguin in the world [and] the smallest member of the Spheniscidae family", though all penguins are Spheniscidae. Furthermore, some sources are definitely not reliable, like this class paper also used in the Wikipedia article.

Not great having to point this out, and hope this isn't found to be a problem with other ARKive entries. I do remember, though, seeing other ARKive articles that may have drawn on the sources present in the Wikipedia articles. —innotata 18:32, 25 July 2011 (UTC)

I find that it varies article by article and you pretty much have to consider it the equivalent of cut and pasting from another part of Wikipedia (i.e. no assurance). I pretty much concentrate on using them for places where we have not much of an article and they have more. I think it can get you to a C-B category article...but to go to GA-FA, you will need to do the thing over yourself. Of course the same applies with any wiki article one takes over. Need to look at old content with skepticism if you have not checked it yourself.TCO (reviews needed) 18:47, 25 July 2011 (UTC)

Definitely agree with this assessment of the quality; I think the ARKive donation should be valuable, though it's going to be less so and harder to use now we need to be as wary as I've found. The articles reviewed by species experts are mostly very good, and the writing is usually high quality. —innotata 23:27, 25 July 2011 (UTC)

Hug.

The template's for user talk pages. And ho hum. This is serious and unpleasant; things need to be done yet. —innotata 01:55, 26 July 2011 (UTC)

I thought it would be a teensy one, like a "done" icon. TCO (reviews needed) 02:12, 26 July 2011 (UTC)

I've raised this with ARKive, and await their response. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 09:23, 26 July 2011 (UTC)

ARKive have provided this response, and asked me to post it here on thier behalf:

The paragraph marked "ARKive text" has a similar structure and the same factual content as the source, but the wording is different so it's different text, and should therefore not be considered plagiarism. ARKive have clearly referenced their source.

ARKive always try to make their texts as different as possible from the original source but apologise if on this occasion it is similar to the original text. When talking about characteristics of a species, it is often hard to re-word without compromising accuracy. ARKive feel that re-wording where possible and citing their sources is sufficient.

ARKive do not ever use Wikipedia as a source. The authors of the ARKive texts are all qualified zoologists / biologists and experienced researchers. Further to this, ARKive get species experts to authenticate the accuracy of their texts. While this particular text is still awaiting formal authentication, it has been written by a biology graduate, and so is to the best of their knowledge accurate.

This particular text was written in March 2004 and has not been updated since.

Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 10:34, 26 July 2011 (UTC)

Some, at least, of the text quoted above appeared in the Wikipedia article on 31 March 2004. It was a new article that day, begun by User:Smallweed. At that stage it included no references or external links.

What exactly this means for the sourcing of the Wikipedia page and the ARKive material, I don't presume to say, but it seems possible that both are derived from www.penguins.cl (the link to this site was added to our page, by a second editor, in April 2004). Andrew Dalby 13:02, 26 July 2011 (UTC)

Surprised by some of the message from ARKive, though I was imagining they might say something similar. This is plagiarism, of one sort: as defined by the OED and Merriam-Webster, it can be copying ideas, or copying text. Rewording and citing the source can avoid plagiarism, but doesn't mean that this close an adaptation is not plagiarised. In the section above quoted, the order of the sentences, and the text of every sentence, is copied in its basic form from penguins.cl (which has been up at least since 2002); I managed to write up what was overall a very different structure and phrasing from that in all three texts, by adding bits of the ARKive text to the Wikipedia texts, before I noticed the copying, similarly to what I've managed to with a good deal of descriptions of species characteristics. And this sort of plagiarism is clearly stated to not be allowed on Wikipedia, so it can't be added to the article (WP:PLAG).

The article is indeed not well researched: one source is a student paper from a sophomore seminar, and the calling a genus a family after penguintag.org makes it seem the author did not research relevant general topics at all. At least they're probably right about copying sources from Wikipedia's: the Wikipedia article (of the current title, at any rate) only got all the same sources in the external links, with information presumably from them, well after March 2004. This article, and more so ARKive's response (text can only be changed so much from sources, rewording and citing is sufficient, this is accurate to the best of their knowledge) make it very much seem their standard of desired quality is below Wikipedia's, and that the donated texts are mostly not worth working on. —innotata 16:06, 26 July 2011 (UTC)

What this seems to come down to is that ARKive's definition of an "acceptably close paraphrase" is not always the same as Wikipedia's, at least for older articles. Calling it a plagiarism allegation makes it sound worse than it is: copying of ideas is a necessary part of what both Wikipedia and ARKive are doing, so it's irrelevant that the dictionary definition of plagiarism mentions that. ARKive are clearly not in the business of copying-and-pasting others' work: the material is paraphrased. As others have said, we may have to exercise a bit of caution with older articles that haven't been reviewed. Not all ARKive fact-files are at the same level of quality: they are open about that and give indicators of quality in a similar sense to how Wikipedia does. MartinPoulter (talk) 18:54, 26 July 2011 (UTC)

The copying of text and too closely paraphrasing (not exactly the right word) is the plagiarism; and of course copying ideas is not plagiarism. This not only fits into the definition of plagiarism we use, but that I've assumed and been told I'd be held to. This sort of plagiarism is unnecessary, and like the choice and reading of sources for this article, lazy writing. I don't think ARKive is clear about quality, especially looking at the response above, where they backed up what was definitely a very poor job. It's going to be rather difficult adding ARKive texts having to check for plagiarism as well as the other issues. —innotata 19:49, 26 July 2011 (UTC)

Sigh. I guess my problem is this - the British Museum and Smithsonian etc. are learning institutions which (I guess) function themselves as secondary sources and receptacles of knowledge that fit more easily with the wikipedia structure. The link between experts and the writing of Hoxne Hoard was a great example of cooperation. Here we have an entity, ARKive which functions as a PR and fundraising vehicle, and is in some ways in competition with wikipedia. To me, the connection on the 'knowledge chain' is somewhat suboptimal as it doesn't make it much easier for us to access secondary sources or images, just text, the benefits of which are patchy. In return, there is a push for acknowledgement on the article page (see Hyacinth_Macaw#References) in excess of what other source have had (e.g. no mention on the Hoxne Hoard). I don't think it is appropriate for the tag to appear in article space - other GLAM collabs have had it in the talk space.

Ultimately I commend them for their drive and wish them success in promoting an environmental agenda, but am wondering about the extent of the benefit for the PR that is being sought. Casliber (talk · contribs) 03:23, 28 July 2011 (UTC)

PS: They've spelt this wrong, it is Boletus satanas. Casliber (talk · contribs) 03:33, 28 July 2011 (UTC)

• ARKive is *potentially* a competitor to Wikipedia, and to me that means it's a good thing that there is a collaboration project.
• I agree that a special template for article space is a step too far. (withdrawn MartinPoulter (talk) 09:16, 30 July 2011 (UTC)) However, the case for external links to ARKive media seems very strong: they give further information, it's quality content that readers may well want to see once they've read the article, it's run by a charity, and it can't be copied into WP because of copyright.
• I don't accept that the point of ARKive is "PR and fund raising". Their mission is raising awareness of threatened species and they are part of a charity organisation dedicated to that topic. It's like saying that the point of Wikipedia is "PR and fund raising".
• Yes this is different from other GLAM projects. Again I take that as a good thing, in that it shows we can work with a variety of partner organisations, including ones that have not released free content before. If you don't find it interesting, fine: no one's forced to participate in it. The fact is that it speeds up the improvement of a range of articles on a scholarly topic. MartinPoulter (talk) 08:02, 29 July 2011 (UTC)

To innotata, I take your point about article quality, but in ARKive's defence, the problematic article above was an old one that wasn't marked as expert-reviewed. MartinPoulter (talk) 08:02, 29 July 2011 (UTC)

To comment on the article space template: the difference here is that the material arkive has released is under a CC-BY-SA license, and hence needs to be attributed to them (particularly due to the GFDL/CC-BY-SA dual licensing thing). My understanding was that the appropriate way to do this was to use a template at the bottom of the article (following the lead from a number of other attribution templates in Category:Attribution templates). It's very different from a Hoxne Hoard-style collaboration, in that Hoxne was about writing new content with curatorial input in standard Wikimedia style, rather than reusing pre-existing content. Also, take a look at {{1911}}, and e.g. Angle#References, where a similar approach is taken even for a PD work. Mike Peel (talk) 08:47, 29 July 2011 (UTC)

Hmm, the 1911 content is a good analogue actually - I just wish this collaboration had been made five years ago. Of course their aim is good, I am talking more about how they have set up their website - it is written by designers and then vetted by experts, rather than being a source written by experts. I am still having a nose around so will comment more later. Casliber (talk · contribs) 11:08, 29 July 2011 (UTC)

ARKive's text is written by biologists, with BSc or MSc qualifications in relevant fields. The checking by experts is an additional step, and its absence does not reduce the quality of their (cited) text. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 13:40, 29 July 2011 (UTC)

Mike, you make a very good point. There is ample precedent for crediting shared text in article space and we should keep the template. Casliber: "written by designers"??? where do you get that idea? MartinPoulter (talk) 09:16, 30 July 2011 (UTC)

WP:SAYWHEREYOUGOTIT when writing in-line citations

I have looked at several of the articles where ARKive content has been transcribed to the Wiki. I think that this project needs to be clear about sourcing the material correctly. In many instances contributors to this project have been writing the in-line sources as if the contributor had read the original articles and were sourcing the original articles, when they were actually sourcing from ARKive. Please make the necessary amendments across many articles. The guidelines on sourcing are all laid out in MOS and and WP:SAYWHEREYOUGOTIT is highly relevant. Snowman (talk) 08:43, 29 July 2011 (UTC)

The difference here is that the person that originally wrote the text for arkive will have looked at the references and verified them - and the contributor making the changes to the article pages is simply inserting the material from the arkive source (complete with the original references). That situation doesn't appear to be covered in the citing sources guideline. Mike Peel (talk) 08:50, 29 July 2011 (UTC)

WP:SAYWHEREYOUGOTIT is explicit about this. ARKive is no different to any other source where authors have used information from a number of places having analysed the relevant information. The person who is writing the Wiki should say where he or she sourced information, and not only where the source got the information from. Snowman (talk) 09:02, 29 July 2011 (UTC)

It's explicit about the situation where you're paraphrasing content from another source, which is different from this situation. An an illustrative example: if you were merging two Wikipedia articles together, would you modify all of the references in the material you copy over into another article referencing the original Wikipedia article? E.g. saying "Smith, John. Name of Book I Haven't Seen, Cambridge University Press, 2009, p. 1, cited in Wikipedia article Blah"? That's a more analogous situation here. Mike Peel (talk) 09:15, 29 July 2011 (UTC)

I have some difficulty in interpreting "Blah" in your comment, so I would be grateful if you could explain your use of this word. I think that merging two Wikipedia articles is special case, and where the page histories are preserved. Snowman (talk) 09:27, 29 July 2011 (UTC)

Update: I have left messages at relevant talk pages (Wikipedia talk:Citing sources, Wikipedia talk:WikiProject Mammals, Help talk:Merging) for users who may wish to advance the discussion here. Snowman (talk) 09:37, 29 July 2011 (UTC)

By Blah, I just meant linking to the appropriate Wikipedia article in the reference (using "Blah" simply as a random name) - it might make more sense to say link to the other Wikipedia article. Apologies for the confusion it caused. Mike Peel (talk) 10:12, 29 July 2011 (UTC)

I am with Snowman here in that we would reference the webpage unless we'd seen the original source - which is ok as alot of us have university access and can check them. Casliber (talk · contribs) 11:10, 29 July 2011 (UTC)

It would seem to me that the majority of the project's work to date is flawed by in-line referencing that is below Wikipedia standards. Some examples such referencing are seen on : Giant Armadillo, South Asian River Dolphin, Wisent, and Bornean orangutan. Snowman (talk) 11:23, 29 July 2011 (UTC)

Snowman, instead of thinking of it as flawed, think of it as a stepping stone. I've used web refs that were verging on okay - some zoo sites etc - until we found some peer-reviewed material. We'll get there just is going to take a long time. Casliber (talk · contribs) 11:29, 29 July 2011 (UTC)

Sure; it is an intermediate stage, which enriches the Wikipedia. However, I would like to prompt this project to be more helpful and reduce confusion in providing in-line references that indicate the actual source of the new text. I guess that the ARKive project editors have not been checking the references used to make the ARKive website, going on the speed in which the additions to the Wiki are being done, but I might be wrong. Some of the ARKive contributors have been writing in-line refs that indicate that a source from ARKive, and I think that this is more consistent with the guidelines; see Hyacinth Macaw on 29 July 2011, where the new in-line references end with "via ARKive". Snowman (talk) 11:53, 29 July 2011 (UTC)

This seems to involve what most style guides refer to as a secondary source— a publication that you did not read yourself, but is cited in a publication you did read. You can't directly cite a secondary source as if you read it, but you can mention it in-text. See Reference List: Other Print Sources at Purdue OWL. This unsigned edit was made by Gadget850 (talk) at 12:30, 29 July 2011.

So just add "via ARKive.org" to each citation that's copied from ARKive and not personally checked by the author: that's a sensible proposal that brings the project into compliance with policy and highlights the sources that might need further checking. It sounds like we should all be happy with that(?). MartinPoulter (talk) 09:50, 30 July 2011 (UTC)

Difference between citing a source and re-using free content

There is a difference between citing a source and re-using freely copyrighted content from another place. If we have another freely copyrighted source that we incorporate directly into Wikipedia, so that the text of that free source becomes the text of our article, that is not a use of the free text a reference. For example, if the Simple English Wikipedia had a sentence with a reference on it, and we copied that sentence into our article, we could also simply copy the reference as well. We would not say, "according to Simple English Wikipedia", because we are not referencing them, we are simply re-using their free content. This is the point of free content: we are able to re-use free text from other sources as part of our articles.

It is important, though, to indicate when we have re-used other free content. For example, at the bottom of Median, it says "This article incorporates material from Median of a distribution on PlanetMath, which is licensed under the Creative Commons Attribution/Share-Alike License.". That text is not a reference. It is a copyright acknowledgment that indicates that some text from PlanetMath was used in our article. We would not use PlanetMath as a reference, but we can and do share free content with it. — Carl (CBM · talk) 13:19, 29 July 2011 (UTC)

Is PlanetMath checked against the references given by PlanetMath before text from PlanetMat is added to the Wiki. How are the in-line references done on the Wiki for the added text that is from PlanetMath? Snowman (talk) 13:59, 29 July 2011 (UTC)

I agree with CBM; it has long been the practice on Wikipedia to import articles intact from other sources that have appropriate license terms. Where it gets tricky isn't the portions of the imported text that has inline cittions; the tricky part is if the imported source is reliable, and as such, is allowed to state the author's conclusions without inline citations. Later, when we make further improvements to the article, it becomes difficult to distinguish novel statements copied from the original outside source from novel statements by Wikipedia editors. Jc3s5h (talk) 14:06, 29 July 2011 (UTC)

Are you aware that this is not compatible with WP:SAYWHEREYOUGOTIT? Surely, if the additions to the Wiki from PlanetMath were added with references ending "via PlanetMath" then some of this confusion would be reduced, and the additions would be consistent with the Wiki's guidelines on citations. Is PlanetMath text copied to well developed Wiki articles, or is PlanetMath mainly used to create articles or enhance Stub articles? Snowman (talk) 14:23, 29 July 2011 (UTC)

When we import text from another place, like PlanetMath or another language WIkipedia, the author of the original text becomes an author of our article, and SAYWHEREYOUGOTIT applies to them. But SAYWHEREYOUGOTIT is about material that is used as a reference for our articles, not about free material that is incorporated to become a part of our articles. — Carl (CBM · talk) 15:38, 29 July 2011 (UTC)

Where do the guidelines say WP:SAYWHEREYOUGOTIT does not apply to text under CC licence transcribed to the Wiki from external websites? Snowman (talk) 16:45, 30 July 2011 (UTC)

Because that text is not used as a reference, and the policy you are looking at is about text that is used as a reference. SAYWHEREYOUGOTIT says that you should not cite source B if you are using source A as a reference. But if source A cites B, and we re-use the text from A in our article, then our article also using B as a reference, and is not using A as a reference. That's why SAYWHEREYOUGOTIT is irrelevant. I keep bringing up the easiest example: if we copy an article directly from Simple English Wikipedia, we would not try to claim that we are using them as a reference, we are just re-using their text. — Carl (CBM · talk) 18:56, 30 July 2011 (UTC)

Text copied from an external website is not used as a reference, however WP:SAYWHEREYOUGOTIT says that it should be clear where all text added to the wiki came from. All references added to the wiki should have been read by the Wiki editor unless it is made clear in the citation that the editor read a specified block of CC text, but not the sources for the block of CC text. WP:SAYWHEREYOUGOTIT says says: "Don't cite a source unless you've seen it for yourself". It could not be clearer; "Don't cite a source unless you've seen it for yourself" means that you can not copy creative commons text from an external website and quote only their references. Snowman (talk) 19:16, 30 July 2011 (UTC)

But we can indeed copy creative commons text verbatim from other websites without double-checking every reference the text uses before we make the copy. We do this all the time, in fact. It's one the benefits of free content that we can use free content from other sites without having to re-do all the work ourselves. The assumption is that the person who wrote the text originally read the reference. Remember we have AGF as a policy - do you have some reason to suspect that here that the person who wrote the article did not read the source? — Carl (CBM · talk) 19:42, 30 July 2011 (UTC)

I am perfectly happy to accept that the authors read the references they quote. Surly, you can see that coying text with references and not saying where it is from is not consistent with WP:SAYWHEREYOUGOTIT. Snowman (talk) 20:58, 30 July 2011 (UTC)

What?

Snowman, I don't think I understand this question. It seems to me that this is what happens:

1. Editor goes to ARKive and finds a page on, say, Hyacinth Macaw: http://www.arkive.org/hyacinth-macaw/anodorhynchus-hyacinthinus/
2. Editor clicks on the #Find out more or #References links on that page.
3. Editor clicks on one of the links to some non-ARKive website in that section.
4. Editor reads and uses information from that non-ARKive website to write the Wikipedia article.
5. But you want the editor to give a shout-out to ARKive anyway.

Is that right? WhatamIdoing (talk) 16:35, 29 July 2011 (UTC)

Not involved, but no. I think he meant that aside from copying the text and references verbatim from ARKive, we should also indicate that it was lifted off directly from ARKive using sources you haven't yourself touched. e.g.

1. Editor goes to ARKive and finds a page on Hyacinth Macaw: http://www.arkive.org/hyacinth-macaw/anodorhynchus-hyacinthinus/
2. Editor copies usable text with accompanying references verbatim into a Wikipedia article.
3. Editor then indicates that the ref of the text is not something he actually read himself, but is from ARKive.

Don't see the big deal really. It's donated text after all. If we have to go doublecheck everything, we might as well just grab the list of refs, ignore everything else, read it ourselves, and rewrite it our way. Which renders the whole donation thing moot.

Personally I'd treat ARKive simply as another editor. The person copying the text from there is not part of the equation, he simply copies the text and ref. The rest can be handled the same way we handle reffed additions by other users - we haven't read their sources, but unless we find conflicting claims or unless the info is really really dubious, we let it be, don't we? AGF and all that, assume the ARKive people did their jobs.

But a compromise somewhere between the two is good enough I guess. As you can't like, just go to ARKive's talk page and challenge his additions, LOL. Including the original ref used by ARKive and indicate that the ref and the text are both from there, would probably be good enough.-- Obsidi♠n Soul 18:13, 29 July 2011 (UTC)

I like your idea of treating ARKive as analogous to another editor; in which case this project talk page is analogous to that editor's talk page. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 18:45, 29 July 2011 (UTC)

User Jc3s5h has explained it better than I did, and it is in agreement with WP:SAYWHEREYOUGOTIT, which says - Snowman (talk) 19:13, 29 July 2011 (UTC)

"Don't cite a source unless you've seen it for yourself. Where you want to cite John Smith, but you've only read Paul Jones who cites Smith, write it like this (this formatting is just an example; there are several ways this can be written):

Smith, John. Name of Book I Haven't Seen, Cambridge University Press, 2009, p. 1, cited in Paul Jones (ed.). Name of Encyclopedia I Have Seen. Oxford University Press, 2010, p. 2." Snowman (talk) 19:13, 29 July 2011 (UTC)

Eh?! Are you saying that volunteers who are moving the donated text from ARKive are now obligated to reread everything? You won't get any, which perhaps may be your intention.

Let me spell it out: Wikipedia editors who are transferring the text are not the authors. Nor should they be forced to then be responsible for the contents of the donated material. If they had to reread all the references used, then we really shouldn't be asking for donations then. We can just go to some other website, and look at their list of references.

ARKive is already saying where they got it by providing references. You don't need to do it all over again. It would be nice of course, and doing so would be admirably thorough. But it's an unrealistic expectation to have to redo all the things they already did, for a reason which as far as I can tell is simple distrust because they are not part of Wikipedia. At least ARKive's text contains inline citations.

I'm pretty sure we don't act this way every time another contributor adds something to an article, do we? Do we revert all additions because we can't read the offline source provided? Do we zealously look up their refs, order the book, download a copy, go through a paywall, etc. just so we can read it ourselves and confirm that what he added was true? No. We WP:AGF. Otherwise, Wikipedia would not work at all.

This is not WP:FAR. It's a donation. You know... something freely given? And all I'm seeing here are people looking the gift horse in the mouth. If this is the usual reaction, I'm not surprised that other databases are not exactly jumping to cooperate with Wikipedia.-- Obsidi♠n Soul 20:15, 29 July 2011 (UTC)

P.S. I'm not mad, lol. As I said, I'm uninvolved. Just my honest impression at reading all this, really.-- Obsidi♠n Soul 20:27, 29 July 2011 (UTC)

Obsidian Soul, it is not a "gift horse in the mouth", it is a transaction with quid pro quo benefits. Wikipedia is often the first page listed in google for a search item. This agency is has a large Public Relations element to it, hence they are interested in cross-promotion. I respect their aims and all, I am just uneasy with how this is all working out. Casliber (talk · contribs) 21:06, 29 July 2011 (UTC)

Oooh. Fair point. Heh. Still, the atmosphere here is rather... I dunno. I'd really rather assume good intentions. Some of those articles would really benefit their Wikipedia counterparts. Mostly the more obscure species. I understand perfectly why people are leery of integrating donated text into perfectly good articles, which is the case with some of those dealing with more familiar animals.-- Obsidi♠n Soul 21:24, 29 July 2011 (UTC)

No, Snowman's right. If you go to WebsiteX.com and copy sentences off it, then you must cite WebsiteX.com as your source. WebsiteX.com might have a dozen sources, but those dozen sources are not your sources. What you should do to accommodate licensing and copyrights has nothing at all to do with what you should do for citations. Your citation must name your source, not your source's source.

Besides, this ought to make everything easier for the people doing the transfers: You can name the ARKive page alone as your sole citation for every single sentence. You have the option of passing along ARKive's sources, but the only thing that CITE actually requires you to name as your source is your actual source—and your actual source is never a page that you've never even laid eyes on. WhatamIdoing (talk) 21:09, 29 July 2011 (UTC)

Come to think of it, the only time I've ever copied something wholesale from another site was from TOLweb, a single paragraph from a page which explicitly states that the text is under CC-BY-SA. And rather than cite their source, I did cite TOLweb instead. And yeah it's easier in a way. My concern is that it loses something in the process. What if the ARKive page changes? Knowing your source's source has to be good for something, and it would be that. You'd be able to [theoretically] read it yourself.-- Obsidi♠n Soul 21:24, 29 July 2011 (UTC)

In this instance, I agree that it would be wonderful if editors choose to pass along these sources. CITE is satisfied if they provide only their source, but it would be lovely to have both if they don't mind. WhatamIdoing (talk) 21:41, 29 July 2011 (UTC)

It would also help those who don't quite view ARKive as a reliable source. By retaining the inline cites, you can at least figure out which came from unreliable sources, which are solid, which are problematic, etc. Then you can improve it without having to worry about which sentences were reffed to what source. So yeah... a compromise? We can easily add a (from ARKive) or something like that in between the <ref> tags in addition to the actual ref cited in the original. While we may not have read the original, we did read the ARKive material.-- Obsidi♠n Soul 21:54, 29 July 2011 (UTC)

WhatamIdoing, I am sorry to say you're simply wrong in the post dated 21:09, 29 July 2011. If we copy an article wholesale from Simple English Wikipedia, we are not using them as a reference for our article. In fact they are not a reliable source according to our standards, so we can't use them as a reference. But we can still copy articles from there to here. When we do that we are simply using their text here as part of our article, we are not using it as a reference. This happens all the time with text from many free sources, which is why we have so many attribution templates. — Carl (CBM · talk) 23:34, 29 July 2011 (UTC)

I like Obsidian Soul's suggestion that we think of ARKive as another editor. Here's where the similarity between ARKive factfiles and WP articles is a benefit. However, unlike a real user, ARKive is not automatically credited in page history. That suggests the need for mechanisms to credit that user, and the attribution template and "via ARKive" addition to citations seem to satisfy that need.

Casliber, above, you say that this is not a donation but a transaction: surely you could say about any donation to Wikipedia projects (including a donation of labour) that the donor benefits. In Wikipedia educational assignments, students donate labour and content but they benefit from it in terms of grades. I benefit in various personal ways from editing WP, but it's still true to say my work is donated to it. What's wrong with that? I can't understand why you're trying to make a special case of ARKive. They want people to be more aware of, and informed about, threatened species. We want people to be able to inform themselves about anything at all. There's a very strong overlap of goals there, which is obscured when we incorrectly describe them as "designers" doing "pr". Wikipedia and ARKive will both benefit: that's the point of arranging a collaborative project. MartinPoulter (talk) 09:50, 30 July 2011 (UTC)

WP:SAYWHEREYOUGOTIT

WP:SAYWHEREYOUGOTIT is a sub-section of a page, which opens with the words:

This page documents an English Wikipedia content guideline. It is a generally accepted standard that editors should attempt to follow, though it is best treated with common sense, and occasional exceptions may apply.

(my emphasis). HTH. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 20:51, 30 July 2011 (UTC)

High visibility template

Re: high visibility template, {{ARKive attribute}}, which says:

"This article incorporates material from the ARKive fact-file Pagename xyz, which is licensed under the Creative Commons Attribution-ShareAlike 3.0 Unported License but not under the GFDL."

It is present on the following pages: Snowman (talk) 12:32, 29 July 2011 (UTC)

• Wisent
• South Asian River Dolphin
• Coco de Mer
• Giant Armadillo
• Apollo (butterfly)
• Common box turtle
• Python sebae
• Bornean orangutan
• Erioderma pedicellatum
• Freshwater pearl mussel
• Maxima clam
• Southern giant clam
• Brachypelma albopilosum
• Brachypelma boehmei
• Dracula vampira

I think that this template is too prominent should be removed from all these articles and that the template be deleted. ARKive have a box for the talk page which will suffice. There is also external links to ARKive (under discussion on the WP Bird talk page at Wikipedia_talk:WikiProject_Birds#External_links_to_Arkive), and sometimes end references with "via ARKive". Snowman (talk) 12:32, 29 July 2011 (UTC)

• We need to have something on the article page, because of the copyright license variation (ie not GFDL) and it is fair enough to attribute that site in a references section. It could however be less prominent. So don't delete the template. Graeme Bartlett (talk) 12:45, 29 July 2011 (UTC)

• I do not understand what difference the different copyrights between the Wiki and ARKive should make to the need to have this template. When text is appropriately added to the Wikipedia it is licensed with the Wikipedia's copyright. I see no reason why the Wikipedia should need to give the licence of an external website in this way. This template does not help to indicate which text is from ARKive, so it is useless at pointing to any particular part of the article that is derived from ARKive. The Wikipedia is not required to explain the copyright licences of books or any other source. Is there a copyright problem about cutting and pasting text from ARKive to the Wiki? Snowman (talk) 12:56, 29 July 2011 (UTC)

  I'm sad that we have to add these. It seems that it will be very difficult to get rid of these, and I would hate to have something so tacky on these articles in perpetuity! ErikHaugen (talk | contribs) 13:06, 29 July 2011 (UTC)

• I think that the text added from ARKive should be attributed to ARKive with an in-line ref and that there is no need for this template. To me, this template which has been added to existing Wiki articles (including well developed articles) explaining the copyright of an external website seems to be as bad as spam. Snowman (talk) 13:13, 29 July 2011 (UTC)

The text here seems very similar to Template:Planetmath and the other templates in Category:Attribution templates. I believe these are usually placed at the bottom of the main article, like Template:Planetmath, but there may be some variation in practice. The use of this sort of attribution is well established and is important to acknowledge when we have re-used freely copyrighted material. — Carl (CBM · talk) 13:21, 29 July 2011 (UTC)

See also prior discussion, above. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 13:28, 29 July 2011 (UTC)

I have not seen evidence of any prior negotiation with the WP projects affected about the roll out of this template. Snowman (talk) 13:43, 29 July 2011 (UTC)

Surely "... but not under the GFDL" at the end of the template could be removed. Snowman (talk) 13:50, 29 July 2011 (UTC)

There is no requirement for negotiation. There was, however, consultation, as you are well aware, having replied to it. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 14:03, 29 July 2011 (UTC)

"All Wikipedia content is open to being edited collaboratively". Snowman (talk) 14:26, 29 July 2011 (UTC)

As per above, if Arkive text is considered WP:RS, with its own internal references, use of the text should be referenced to Arkive, not to the unseen references within the Arkive page. This is the same principle as adding references to a book or a site like BirdLife. The reference is to the book or the website, not their primary sources. If an Arkive page isn't RS, it shouldn't be used. Thus, any appropriate use of Arkive text will be referenced to the site. Why do we need a spammy high-visibility template to tell us the text has been used? Jimfbleak - talk to me? 14:53, 29 July 2011 (UTC)

Here are the references listed on ARKive for the Hyacinth Macaw. Not one of these references is the page where information was sourced from. All these references are either home pages of websites or "Not Found". These have been copied to the Wiki, and because they do not contain the information they are not adequate of the Wiki for verification. Surely, it would be better to use ARKive with a url to the information on ARKive in the in-line citation (assuming that ARKive is RS). As noted above, this would make the high-visibility template unnecessary. Snowman (talk) 15:33, 29 July 2011 (UTC)

WP:BIRD has in the past had discussions on photographers adding external links to their own sites marking them as "photographs by XYZ" and these additions were largely in good faith. The privilege of including names is largely unavailable to the authors or editors of text. The idea that one could use a template to indicate source could set an interesting precedent. An expert editor could write something perfectly reliable and with reliable sources on a personal blog and mark it under a suitable creative commons license and then copy-paste it into Wikipedia with a template at the bottom saying "text included from _mysite.com_". Allowing such templates could lead to more unfair practices. Shyamal (talk) 13:26, 30 July 2011 (UTC)

I think that those are good points not raised before: 1. I would not like to see scores of these templates added to an article by users jumping on the band-wagon after writing a short referenced paragraph under CC licence on a website somewhere. 2. By comparison photographers release images under CC licence and their websites are not featured in templates on wiki articles. Snowman (talk) 16:28, 30 July 2011 (UTC)

ARKive is not "a personal blog", so your example does not help us. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 20:56, 30 July 2011 (UTC)

OA support for the project

Hi Andy,

I had a look at your project and think there might be ways to join forces: If you feel that the result of complementing the original Wikipedia entry with ARKive text still does not yield a decent entry, I could try to find some Open Access resources (text, images or media) on the topic. Please post the requests at meta:Wikimedian in Residence on Open Science/Requests for Open Access materials.

I won't comb systematically through your pages, though I may check some of them on occasion, as I am also interested in zoology, animal physiology, taxonomy, biodiversity and conservation issues.

-- Daniel Mietchen, GLAM/OA 22:28, 29 July 2011 (UTC)

Thank you, Daniel, that's a very kind offer, and I would encourage anyone helping with this project to consider taking advantage of it. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 14:30, 30 July 2011 (UTC)

More donated articles

ARKive have just released another batch, of 58 texts. I've added these to the list on the project page. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 14:32, 30 July 2011 (UTC)

It might be useful to know at this juncture if the ARKive project would like to make any changes however simple and small to address or partly address any issues that have been raised above. Snowman (talk) 21:02, 30 July 2011 (UTC)