Talk:Donogh O'Brien, 4th Earl of Thomond

Attribution

This article contains a substantial amount of word-for-word copy-pasted text from Pollard (1895) "O'Brien, Donough (d.1624)" in the Dictionary of National Biography, volume XLI, pages 312–314, which is in the public domain. According to Wikipedia's Plagiarism Guideline (WP:PLAG), such use must be acknowledged (WP:Attribution is not a guideline and serves a different purpose). The section "Where to place attribution" of the Plagiarism guideline prescribes to put "an attribution template in a footnote at the end of the sentences or paragraph" containing the borrowed text. This might be done by using the template "DNB" with the |inline=1 and wrapping it in REF tags. However, this results in treating the attribution like a citation and makes it appear among the list of citations displayed by {{Reflist}}, which is somewhat confusing. I therfore created a group called Attrib to separate the attributions from the citations. There might be better ways. Best regards, Johannes Schade (talk) 20:10, 7 September 2022 (UTC)

I have reverted your recent change.

• WP:Attribution is a false lead. It is an historical page as the box at the top states "This page was proposed as a merger of several core Wikipedia policy pages, but a community poll resulted in no consensus to adopt it."
• to meet the requirements of the guidline WP:PLAGIARISM attribution should either be placed inside inline citations (if they are long citations), or the Attribution should be placed at the end of the bullet pointed list long citiations in the References section. What is not needed is a seperate section containing a list of attributed sources.
• personally I would remove the inline prescript from the inline citations in this case.
• I am going to invite User:DavidBrooks to comment here as he and I have worked extensivly yon Wikipedia:WikiProject Encyclopaedia Britannica and he tends to use more inline attribution than I do and may have a different perspective.

— PBS (talk) 20:21, 9 September 2022 (UTC)

Dear PBS. Regarding your 4 bulleted points above: —(1) we agree; —(2) Why do you say "either-or"? I would say the article, as it stands now, has both: inline attributions and a "general" (not linked to a particular place in the text) attribution. I feel the use of both is justified in the 3rd and last bullet point in the 5th paragraph of the "Where to place..."-section of WP:PLAG, which reads "In a way unambiguously indicating exactly what has been copied verbatim, provide an inline citation and/or add your own note in the reference section of the article." This statement talks about an "inline citation" and keeps me wondering whether it means "inline attribution". In fact WP:PLAG avoids that term. I wonder why. So I marked the borrowings with source-specific inline attribution templates wrapped in REF-/REF tags. One might call them "long citations" as they do not refer to the "source list". Regarding your 2nd sentence, this disputed "Attribution"-section does not contain "a list of attributed sources" but an inline attribution used at the ends of certain sentences and paragraphs. I found it unacceptible to have attributions in the middle of a section headed "Citations". This is wrong: one must not put a pear in a box labelled "Apples". So I placed the inline attribution in its own section. What else could I have done? I feel {MOS:LAYOUT should have foreseen a place for attributions, but it does not and does not even mention them. —(3) The documentation for Template:DNB explains that the inline=1 switch automatically changes the prescript to "One or more ...". Removing the prescript would display nothing (?) Is that what you want? Would that be achieved by the |; or do you mean not to mark the borrowings at all? —(4) please do. Best reagrds, Johannes Schade (talk) 17:25, 10 September 2022 (UTC)

You posted this after my posting your page, but to repost it for clarity

As for WP:PLAGIARISM see the start of the two relevant paragraphs:

As for WP:PLAGIARISM see the start of the two relevant paragraphs: "but there is a paragraph or a few sentences copied from compatibly licensed or public-domain text which is not placed within quotations, then putting an attribution template in a footnote at the end of the sentences or paragraph is sufficient. If a significant proportion of the text is copied or closely paraphrased from a compatibly-licensed or public domain souce, attribution is generally provided either through the use of an appropriate attribution template, or a general attribution template ... placed in a "References section" near the bottom of the page. In such cases consider adding the attribution statements at the end of the Reference section directly under a line consisting of Attribution: My emphasis. — PBS (talk) 20:57, 9 September 2022 (UTC)

I think you are misunderstanding what inline=y does to the attribution templates. It simply changes the wording of the prescript from:

• This article incorporates text from this source, which is in the public domain.
•  This article incorporates text from a publication in the public domain:

to

• This article incorporates text from this source, which is in the public domain.
•  One or more of the preceding sentences incorporates text from a publication in the public domain:

which is more appropriate for an inline citation.

— PBS (talk) 20:58, 10 September 2022 (UTC)

You (User:Johannes Schade) write above 'I found it unacceptible to have attributions in the middle of a section headed "Citations".' I am not sure why you find it unacceptable, as they are simply citations with additional annotation. The attribution either has to go in the footnot section or in the references section and you do not seem to object to attribution appearing in the references section. If you do not like the section header "Citations" then why not remove it and include all the inline citations, which are after all footnotes, in the "Notes" section or at the start of the References section. However given the number of inline citations why not simply just have the attribution just at the bottom of the bullet pointed list in the References section? -- PBS (talk) 23:56, 10 September 2022 (UTC)

@PBS: I believe your expansion of the |inline=y text was incorrect: it becomes One or more of the preceding sentences incorporate... (from memory). The wording "One or more" is intentionally imprecise; any alternative phrase by phrase would be far too intrusive IMO. Otherwise I don't feel a strong need to intervene in this particular discussion, as there are bigger fish to fry (still plenty of articles that don't acknowledge EB1911 text in any way whatever, for example). David Brooks (talk) 16:29, 11 September 2022 (UTC)

@David Brooks sorry for the confusion the template I was using was changed between my posting here and you reading it so I have fixed it using "subst:". So please take another look :-). -- PBS (talk) 10:24, 12 September 2022 (UTC)

Dear PBS, dear David Brooks. I feel this is as far as we can go for the moment. As apprentice editor I will accept from you two very experienced (Grandmaster and Veteran III) editors that inline-attributions can appear among the citations. There are many examples for this e.g. Thomas Sopwith (geologist). I have not found an example that uses {{Sfn}}. Most examples are rated Stub or Start; I have not found a GA or FA. List of "footnotes" typically appear in a section called Notes of References. With many thanks, Johannes Schade (talk) 19:26, 13 September 2022 (UTC)

See Andreas Aagesen (the first biography article in the EB1911), or for a more developed one Alexander I of Russia, both use {{sfn}}. You will not see inline-attributions with a {{sfn}}, but you will see it in inline long citations such as in Demographics of Angola. I know that there are inline-attributions before inline long citations, followed bu {{sfn}} that link to the inline long citation (because I have edited such articles), but I do not have an article to hand at that moment to which I can provide a link. -- PBS (talk) 22:45, 13 September 2022 (UTC)

Dear PBS, thanks for your further explanations and examples. I had understood attributions as completely separate from citations, mainly because they are prescribed in different guidelines: WP:PLAG and WP:CITE. So I thought that a verbatim loan or close paraphrasing would be marked by an inline attribution (e.g. {{DNB}} with |inline=) in addition to a citation (e.g. {{Sfn}}). The citation would tell where the information comes from, whereas the inline attribution would acknowledge the loan to fend off accusations of plagiarism. All the examples show that this is not how it is done in practice. In practice only one inline template is used that performs both functions. It obviously makes sense to do so. I have therefore added The locations and links that I usually include with Sfn to the {{DNB|inline=1...}} in the article. I link to the djvu page of the Wikisource text so that the reader can work out the location of the loan. I give the location in terms of page, column and line. This is quite laborious to work out and I am not sure I have counted the lines correctly in all cases. I find it is more difficult to do this in Wikisource than in Internet Archive. The prescript is repeated each time and is a bit long. Perhaps I should use |prescript= to replace it with a shorter one. There does not seem to be a link between the inline attribution and the attribution at the end of the reference section. As the inline attribution appears among the short attributions created with Sfn, the reader might expect such a link. I looked with much interest at the examples you provided, but I think I should probably discuss them in their talk pages. With many thanks and best regards, Johannes Schade (talk) 13:01, 15 September 2022 (UTC)

You are far more detailed in your citations than is necessary. Do not link to the djvu pages. They are there so if someone doubts the veracity of the machine readable transcript (as accessed through {{DNB}} etc}}, they can varify the transcript from the photocopy (and just like on Wikipedia edit the text and fix OCR errors etc). If you spot an OCR error in the text on Wikisource and edit, you will usually find embeded comments for "end of column" etc. It is possible to transcribe the Wikisource machine readable text in columns with similar hyphenation and paragraph indentation etc, as in the original. This was not done because a decision was made aeons ago to format such Victorian formatting in a more modern (html) way, partly for ease of reading for the modern reader, but mainly to allow the physical display (and the html skin used) to determine how the text should be displayed. Projects on Wikisource, such as DNB and EB1911, have style guides for formatting and templates to aid in that formatting (DNB MOS EB1911 MOS).

In a citation all you need to do is provide the page number to DNB and similar sources, because as it is machine readable the reader can search for the specific text, or read the page. If you click on the page number in the wikisource you will see that it highlights just that page in the text in the whole article (so highlighting the specific citation in the linked Wikipedia citation). So you do not need to provide a link in every short citation any more than you would do to an prited book. The long citation links to the specific article and the reader can look up the page in the DNB article on Wiki page if they wish. If you update the page using the online edition of any of the OUP sources such as the Oxford English Dictionary or the Oxford Dictionary of National Biographies, the OUP do not include page numbers, even if the text is just copied from the old DNB, so including the page number as well as the article name is enough to meet the verification policy without linking to a page that includes the column in the printed original.

Similarly there is usually no need to include a quote from Wikisoure in a citation because either it is obvious in the source, or if it is not the Wikipedia article often includes a SYN (pun intended).

As the examples I gave you above there is no need to include attribution to the source if the short citations are backed up with a long citation in the references section (as is the case here).

References

— PBS (talk) 19:58, 16 September 2022 (UTC)

Arbitrary break

Dear PBS, thanks for your remarks. I have changed my inline attributions again. Your edit on the article made me realise that I can use {{Sfn}} with an attribution postscript instead of <ref>{{DNB|inline=1|wstitle=...}}</ref> . This has the advantage that the inline citations-attributions link through to the full descriptions of the source list, just like ordinary citations do. In the article information coming from the DNB sometimes is summarised in the editors own words , sometimes it is borrowed verbatim. In both cases the citations link to the same full description in the source list. The general attribution declaration at the end of the of references section is given in addition to that. In the sfns I used "Borrowed text attributed" to indicate that the text is borrowed and must be attrivuted. This wording is perhaps too cryptical. I wanted it short, but you can probably find better. —Regarding the citations. You are of course right: precise location beyond the page and footquotes are not required, but I find them helpful. I find that when the the information in the article is summarised or entirely reworded, I often have difficulties to find the corresponding passage in the source within reasonable time. Pages are long in works like the DNB. I find it also very useful when one can directly click through to the source, It can be tiring to find the book online, then the page, especially when doing this repeatedly when reviewing an article e.g. for GA. I have recently have had remarks from a reviewer who thanked me to have made these checks easier, but admittedly others said as you say or even delete the needless details. I have used this type of "detailed citing" on about 100 articles of which 2 passed GA and 1 A-class. —I refer to the DjVu page because the reader cannot find locations per column and line on the "main page" in Wikisource. In some respects the DNB is easier to cite at Internet Archive, but Wikisources has its advantages as well. —I understand what you say about the advantages of using the {{DNB}} but should each source (millions of them) have its own template? With many thanks for talking to me. Best regards, Johannes Schade (talk) 19:58, 17 September 2022 (UTC)

I think you need to consider is the fact that this is viewed on many types of screens and more importantly slow or metered internet connections. The current size of the article is 35,242 bytes without any of the inline citations it is approximately 23,000 bytes. You really do not need to add "Borrowed text attributed" or the line number to the inline citations. I suspect that far fewer people are going to view each inline citation and want the line number than are going to read the article on a slow or metered line. The first ("Borr...") is superfluous to requirement WP:PLAGIARISM and the second (line number) is superfluous to WP:CITE. Further it clutters up the text in edit mode making it more difficult to read and edit; and it clutters up the reflist for example:

• 34 Pollard 1895, p. 313, right column, near bottom. Borrowed text attributed
• 35 Pollard 1895, p. 313, right column, penultimate line. Borrowed text attributed
• 36 Pollard 1895, p. 314, left column, line 5. Borrowed text attributed
• 37 Pollard 1895, p. 314, left column, line 9. Borrowed text attributed

which to meet the requirements of WP:CITE should appear as

• 34 Pollard 1895, p. 313
• 35 Pollard 1895, p. 314

nothing more or less.

There is also a problem in my opinion with some of you citations quotations. See the the first sentence in the first section ("Adding information to Wikipedia") in WP:EDIT "Wikipedia summarizes accepted knowledge. As a rule, the more accepted knowledge it can encapsulate, the better it is."

Text	Number	Citation	Quotation	Comment
On his father's death in 1581 he succeeded as 4th Earl of Thomond	18	McGurk 2004, p. 360, right column, line 41	"O'Brien succeeded his father as fourth earl of Thomond in 1581 ..."	You have summarizes (per WP:EDIT) the knowledge, anyone interested can check the page there is no need to spoon-feed them and quote the line from the reference.
Henry (1588–1639), succeeded his father as the 5th Earl of Thomond and married Mary Brereton in 1608 but died without male issue in 1639	22	O'Donoghue 1860, p. 258	"Henry, earl of Thomond, the fifth of that title, dying without male issue, was succeeded by Sir Barnabas, his brother."	Apart from the quote being unnecessary in the citation, it does not support the fact that "Henry" was the son of "Donogh and Elizabeth" so another source is needed eg Pollard 1895, p. 314,. Also the quote does not support the fact "and married Mary Brereton in 1608". However the second half of the sentence in the text "and married Mary Brereton in 1608 but died without male issue in 1639" is superfluous. It is the sort of thing included in genealogies but is unnecessary in a Wikipedia biography where the subject of the line in this biography "Henry" has their own Wikipedia biography -- besides which 1639 appears twice in the same line.

If on consideration you either can not see the validity of the points I have been raising in the whole of this thread and still choose to ignore them and add such detailed in-line citations, then I guess we will have to agree to disagree. -- PBS (talk) 11:58, 18 September 2022 (UTC)

Dear PBS. Thank you for the amazing patience you had with me. You practically taught me all I know about the subject. I am thankful for that. I have not ignored your explanations in the whole of this thread but have come quite a way from my initial misunderstanding (I thought that verbatim borrowings even from PD sources were unwelcome, except in quotions or indirect speech ("in-text attributions"), then my first misunderstood approach to attribution (inline attributions separate from the citations using {{DNB}} displayed in a separate section called "Attributions"), to my latest approach using the "ordinary" citations (i.e. {{Sfn}}) with an attribution postscript {{PD-notice}}, which I discovered only today and is much better than what I had invented.

A crucial question probably is in which cases WP:PLAG prescribes, tolerates, or forbids the use of inline attributions in addition to the general end-of-referenc-section attribution notice for unquoted borrowed text passages. I would expect it allows it in the present case. Just like WP:V seems to accept general references only for early-stage articles and expects in-line citations for more advanced ones, so does WP:PLAG. Under "A practice preferred by some" WP:PLAG says "specifically mention the section requiring attribution" and "In a way unambiguously indicating exactly what has been copied verbatim". The next question then would be which form the in-line attribution should take. WP:PLAG mentions {{Citation-attribution}}, which probably remains enigmatic for most readers. {{DNB|inline=1...}} seems to implement {{Citation-attribution}}. I find WP:PLAG difficult to read. It seems always to say "this but also that" due to the history of the finding a consensus between you and some other editors such as Franamax and SlimVirgin (both deceased). It sometimes also uses terminology that is perhaps not common knowledge. I had difficulties to grasp that "inline" and "in-text" attribution is not the same and that "in-test attribution" means by indirect speech. The articles mentioned as examples do not seem to be particularly well-chosen. Do you think there would be a chance for straightening WP:PLAG out and make it a bit more prescriptive?

The same source and even the same passage in the source might be sometimes be cited verbatim and sometimes summarised. The citations that also serve as inline attributions must be distinguishable from the ordinary ones. I still feel that it is helpful to give column and line so that the reader has a better chance to find the borrowed passages in the source within reasonable time. I do not think it is fair to cover all in a general attribution alone and let the reader who is worried about possible plagiarism (probably a GA or FA reviewer) err around in the text trying to find the relevant locations. GA criterion 2d and FA criterion 1f require the reviewer to check for plagiarism and will have to discount the ones that have been attributed and point out the remaining ones to the nominator. These GA and FA nominations are often quite long articles and have hundreds of citations as you know well.

My use of footquotes WP:FOOTQUOTE in the ordinary citations is a separate topic, which we should perhaps discuss in a different thread. You are not the only one to criticise me in that regard.

I am not so sure I understand the sentence above the table "There is also a problem ...". Concerning the second row of the table, you rightly point out that the given citation does not entirely cover the given information. Either another citation must be added or the given one must be replaced with a better one as you propose. It is one of the advantages of the footquotes that it makes it much easier to find such shortcomings. The article is rated C, perhaps somewhat generously. I agree also with your warning against an apprach that is too genealogical. I often fall into this trap because many historical biographies are of noblemen for which genealogical sources, always the same, are readily available. With many thanks and best regards, Johannes Schade (talk) 21:14, 18 September 2022 (UTC)

Apart from too many instances of inline attribution "There is also a problem" to do with quoting something that the reader can easily check themselves by looking at the sources. In the second example there is a further problem of {{failed verification}}. I think you misunderstand my offering a second source. As Wikipedia often summarises several sources simultaneously you probably need two or three sources for that sentence whether in their own ref..tag pair or bundled into one). However in the instance as I noted in the table, I would remove the second half of the text in the sentence (rather than adding citations to support it), because it can be found via the link Henry. This is the difference between a biography and a genealogy. Modern biographies in works such as the ODNB usually do not mention children unless they are notable (ie have their own biography in the ODNB), or appear naturally in the text as a notable event (eg who inherits the title), and they do not detail their genealogy in the main subject's biography. In Wikipedia's case because some people are interested in genealogy, it is usually considered acceptable to give a one sentence summary of a non-notable child. This is particularly useful for daughters because thanks to European society before the 20th century they often are not notable enough in history to warrant an separate biographical article, but their marriages into other notable families can help someone understand the politics of powerful dynasties, both at the country and country level in part through family ties.

As it happens I have been working on the DNB articles with broken links thanks some editors replacing dash with ndash within the dab extensions of some Wikisource DNB articles. These are two articles were I cleaned up the sources and work as examples

1. James Howard, 3rd Earl of Suffolk — inline short citations with the attribution before the long citation in the references section.
2. George Birkbeck Hill — attribution before the inline long citations (in this case for EB1911 but it demonstrates the point).

So I would argue that:

• The first example works better because it is referencing multiple pages in the original references, which is why short and long citations are usually used. If someone is really interested in the citations then they will see the attribution in the references section. It helps keep the inline citations short which makes editing the text much easier, and presents the inline citations in a clean way in multiple columns.
• In the second example, the EB1911 is only one page long so it fits in neatly with the other inline long citations and cleans up the Appendix sections
• In the second example if the EB1911 was multiple pages long then as the style already developed was for inline long citations citations, a solution is to keep the first citation to EB1911 long with the attribution prescript, but with specific page number (or numbers), and then use short inline citations with different page numbers linked to the first instance that has a the attribution, but the later inline citations are just short citations {{sfn}} or similar with no attribution (unfortunately I do not have an example to hand).
  • The other alternative is to move all the long citations down into a bullet list in the references section and then add short inline citations linked to the bullet list (like example one above), however that is a change in citation style and someone might object. What is not acceptable is multiple inline long citations to the same book (that is not covered by change in citation style ]] see the section Duplicate citations in WP:CITE—it is a mistake that many editors make when they first start to edit Wikipedia articles and have yet to learn how to format inline citations (because at first ref...tags are complicated and confusing for many).

This will be my last posting to this section, because I think we have said all there is to be said about the citation style in this article and we are drifting off into a more general conversation. I will continue on your talk page. -- PBS (talk) 10:59, 19 September 2022 (UTC)