Wikipedia talk:WikiProject Dictionary of National Biography/Archive 1

Copyright concern
The current project text makes the claim that "The content of the 1900 works are in the public domain in the US." The discussion I started at Wikipedia talk:Copyright problems/Archive_10 makes the point that, as far as I can ascertain, copyright in the US for DNB is the same as for the UK, which is 70 years after the author's death. By what rationale have we decided that it is all in the PD in the US? --Tagishsimon (talk) 13:50, 9 September 2010 (UTC)


 * It seems a reasonable claim, given that (a) the DNB is being posted at Wikisource, (b) only things PD in the USA are allowed on Wikisource, and (c) there is a considerable body of actual expertise at Wikisource on the legalities. Here (c) is the key point, of course: the Wikisource approach is noticeably less vague than that of the average Wikipedian. Charles Matthews (talk) 14:16, 9 September 2010 (UTC)


 * The DNB does have the advantage that all its authors are a) identified with their articles, and b) notable enough to be traceable, meaning that we can clearly identify the PD-70 cases and identify the (relatively small) fraction who died post-1939. The DNB was published by 1900; most authors were respectable academics, which for the period means late-middle-age or older. Surviving to the 1940s is pretty rare among this cohort, not more than 20% on a quick sampling of wikisource author pages.
 * Given this, it might be productive to work on importing the "safer" articles first - the ones definitely PD in non-US jurisdictions - and then look at the others when we come to it. Shimgray | talk | 09:24, 10 September 2010 (UTC)


 * To make some rough numerical estimates - 27,000 DNB names, say 25% of which (the big names) are covered sufficiently well to require no content addition, 20% of the remainder is approximately 4,000 articles that may pose copyright problems against 16,000 which don't. I think we'll have more than enough to keep ourselves busy for a few years, by which time a lot more of the dubious ones will have come into the public domain - by the time the 1940s came around the remaining contributors were very elderly, and I'd estimate half of those were dead by 1945 - so come January 2016, we'll have another couple of thousand to trickle in. No reason not to take the long view - we've been around ten years, I think we can rely on another five or ten! Shimgray | talk | 09:32, 10 September 2010 (UTC)


 * AFAIK, wherever a work was first published (with some obscure exceptions e.g. Afghanistan), if it was first published before 1923, it is in the public domain in the United States . . In other words, your hypothesis that "copyright in the US for DNB is the same as for the UK" is not correct, and the articles are good enough for Wikipedia and Wikisource, both of which go by copyright in the US and only copyright in the US. The "Museum Copyright Group"'s table refers to the UK status of the DNB, and, as discussed above, some articles may still be copyright in the UK. So, ideally, we'd work on the US and UK-free passages first, but it isn't a question of legality per se, but a question of making them easier to reuse, given the UK-centric nature of the DNB. (The URAA adjustment applied to works published between 1923 and 1977.) - Jarry1250 [Humorous? Discuss.] 16:33, 10 September 2010 (UTC)


 * My concern arises from a reading of the Berne Convention article and the Uruguay Round Agreements Act article. I note your Cornell link. But I note sentences such as "Article 18 of the Berne Convention specified that the treaty covered all works that were still copyrighted in their source country and that had not entered the public domain in the country where copyright was claimed due to the expiration of a previously granted copyright there." Date of death of author plus 70 years keeps some DNB articles in copyright in the UK, and Berne Article 18 + Uruguay seems to me to place these under copyright in the US. If not, what is the point of Berne & Uruguay? --Tagishsimon (talk) 20:12, 10 September 2010 (UTC)


 * Hmm, I fear we are getting somewhat outside my comfort zone. While I have a think - and try to find a source other that Hirtle - I think one might look at section 2 or article 18 of the Berne Convention, "If, however, through the expiry of the term of protection which was previously granted, a work has fallen into the public domain of the country where protection is claimed, that work shall not be protected anew." Maybe we could show that the DNB had been registered for copyright in the US and that's its term had therefore legitimately lapsed? Just a thought. (Incidentally, just to check, these articles are all pre-1909 also, right? That makes them even better, because of that funny 9th circuit decision we won't have worry about.) - Jarry1250 [Who? Discuss.] 21:04, 10 September 2010 (UTC)
 * Oh, here it is, I think. Title V, URAA: "Any work in which copyright is restored under this section shall subsist for the remainder of the term of copyright that the work would have otherwise been granted in the United States if the work never entered the public domain in the United States." (emphasis mine) Now, even the 1909 act only gave 56 years from publication; the CTEA only extended works that were still copyright; the 1976 act of life + 50 only applied to works still copyright in 1976, and to those still copyright it only gave 75 years from publication. In other words, had the DNB been an American work, there is no feasible way it could be still copyright, and hence, under the URAA's own description, it cannot be copyright in the US today. Well, IANAL, but that's my take, anyway. - Jarry1250 [Who? Discuss.] 21:24, 10 September 2010 (UTC)
 * I'll go with that. Thanks for your patience in pursuing the question. --Tagishsimon (talk) 21:28, 10 September 2010 (UTC)


 * To be honest, I think the best argument in favour of "all pre-1923 anywhere is PD-US", to a first approximation at least, is that I've never seen anyone seriously challenge the presumption until now! It's pretty widely relied upon, including by significant chunks of the "classic" publishing industry, who presumably have lawyers looking for this sort of problem.
 * Even so, though, I'd be happier if we focused on PD-70, which is pretty much universal. Our potential reusers are international; we should probably avoid, where possible, knowingly introducing material that we're aware has copyright implications for downstream users. (On the other hand, there's a reasonable argument that that bird has long since flown - the Wikipedia corpus is heavily "contaminated" with PD-US-only text already, and so our reusers already have this problem...) Shimgray | talk | 21:30, 10 September 2010 (UTC)
 * Quite, but it's always nice to be thorough :) 1909-1923 has been challenged though, on the Ninth Circuit, but pre-1909 never has. And I definitely agree about the focus on PD-70, as I said above, "ideally, we'd work on the US and UK-free passages first, but it isn't a question of legality per se, but a question of making them easier to reuse, given the UK-centric nature of the DNB." So yes, you have my 100% support on that point. - Jarry1250 [Who? Discuss.] 21:35, 10 September 2010 (UTC)...

Re DYKs
I should have thought the preparation of a new article with the aim of DYK nomination (per the comment by User:Victuallers) would be similar to what is already on the page. Mostly I would expect it to go as:


 * 1) search through articles that could be created at DNB No WP for something promising and not already on WP;
 * 2) edit according to the recipe;
 * 3) finish off with some further research in other sources, look for some images, do a final copy edit.

Charles Matthews (talk) 15:45, 9 September 2010 (UTC)

Template:DNB Cite
It seems to me that icon displayed by the template Cite DNB is a little obstrusive and perhaps non standard? Example: Is there a concensus or am I alone? --Senra (Talk) 22:37, 9 September 2010 (UTC)


 * It does seem a little unorthodox, but on the other hand I think it's a nice compromise between the conventional plain citation (which doesn't make it clear it's one of our resources) and the little floating "see Wikisource" box, which doesn't fit in with the normal citations. There is something to be said for flagging up our own resources, I think. Shimgray | talk | 09:53, 10 September 2010 (UTC)
 * I think it might conflict with MOS:ICON. P. S. Burton  (talk)  20:11, 14 September 2010 (UTC)
 * I'd be happier if was removed. --Tagishsimon (talk) 20:42, 14 September 2010 (UTC)
 * Bear in mind that it does have a purpose (across interwiki links to reference articles to Wikisource), and the icon is more important in other templates of the same family. Those who would prefer to read the Catholic Encyclopedia at Wikisource without extraneous messages can rely on this icon (and there are at least four other sites you might be routed to, about which the same cannot be said). Similarly with the 1911 Britannica. I don't think it does contradict the idea of MOS:ICON: it gives navigational information. In other words uniformity for citation to Wikisource is a plus, and it's not just us concerned. Charles Matthews (talk) 21:30, 14 September 2010 (UTC)
 * I don't buy it. In DNB Cite it's merely an inline branding & promotion exercise. The wstitle link is what gives the user access to a document at wikisource without extraneous messages. The icon itself has no function (i.e. is not clickable). It serves no encyclopaedic purpose. And as mere cross-site promotion, it's a typographical mess when preceded with a bullet point, and in fact the globe will always look a mess whilst it is sized and positioned so as to be positioned lower on the line than the bottom of the text that follows it. Charles, do you know of any discussion or guidelines elsewhere on this issue, noting your point about uniformity and "not just us"? --Tagishsimon (talk) 21:54, 14 September 2010 (UTC)


 * Since the place for the comments really is Template talk:DNB Cite, and there is an indication there of what is happening in parallel for EB1911 and CE, I'm going to suggest that some of this discussion is adjourned. As for being mean to Wikisource ... it really is a well-kept secret, you know. It would be better for others to argue on the "brand" issue. I gave a talk earlier in the year about "Reference Commons", namely a way of thinking about WS that looks on it as more analogous to Wikimedia Commons (with reference material instead of media files). The fact that this is not the normal way of looking at it is obviously (to me) related to the business you are talking about. Charles Matthews (talk) 22:08, 15 September 2010 (UTC)


 * I think this is really a minor issue, and something we can sort out later, lets instead concentrate on the more vital parts of this project. Cheers P. S. Burton  (talk)  22:15, 15 September 2010 (UTC)


 * There's more discussion at Template talk:1911 which may throw some light on things. P. S. Burton, you're welcome to determine whether or not this is a priority for you, but suggesting others should amend their priorities to suit your world view is a bit much; which is to say I didn't find your contribution very useful, all things considered. Accretion of cruft, especially typographically rotten cruft, is a matter of great concern, not least when it touches on (for which read "buggers up the aesthetics") of hundreds and thousands of articles. --Tagishsimon (talk) 22:27, 15 September 2010 (UTC)
 * I agree with Charles Matthews comments at 21:30, 14 September 2010. Tagishsimon you make statements like "Accretion of cruft ..." as if it is a matter of fact rather than you opinion. Likewise "matter of great concern" great concern to whom? So far the number who have expressed an opinion either way is about half a dozen editors. "hundreds and thousands of articles" where did you get that number from because AFAICT the total is closer to 15,000 (How do I come to that figure? I had a look at the numbers in the categories attached to the templates). By far the biggest is undefined and in my opinion that there are less than 2,000 links to article titles, and over 8,000 usages of the template with no article (in other words Wikipedia is saying -- somewhere in the 29 volumes of the EB there is some text copied into this article), is in my opinion (to put it mildly) less than helpful. In this specific case the DNB Cite template is used on just under 2,600 articles but just over 2,000 of those it is called from inside DNB or DNBfirst, so the total of just over 500 articles call DNB Cite. I think that the icons help both readers to help them navigate, and for editors to fix problems by giving them a visual clue. -- PBS (talk) 21:26, 19 September 2010 (UTC)

Use of the new DNB
Every article in the old DNB has been updated for the new one--in addition to the new articles added. Since every patron of a public library in the UK, and almost any user of a college library elsewhere, has remote access to the ODNB, I do not think it good practice to add material from the old dnb without checking it in the new. The virtue of using the old is that it provides a basic text, so that does not have to be written from scratch, but just modernized. But not to take readily available modern sources into account is not good work. Personally, I'd find it easier to write from scratch ignoring the old dnb, but many people like to start with a framework.  DGG ( talk ) 02:44, 10 September 2010 (UTC)


 * Well, I can't concede the point that only UK editors and college students should be allowed to do the work. That really amounts to closing down the project as open to all participants. Charles Matthews (talk) 07:11, 10 September 2010 (UTC)


 * It is not the best practise to import DNB without checking ODNB. But it is far far far quicker and more expedient. So whereas we advise that ODNB should ideally be checked, our first priority is the DNB insert whether or not ODNB is checked. Doubtless eventually we will run an ODNB lookup project to check that all articles have been updated from ODNB. In about 2015, I'd reckon. See you then. --Tagishsimon (talk) 16:51, 11 September 2010 (UTC)

Moving ahead
Two things. Firstly, I'd propose that we choose a single volume (e.g. vol. 28) that has complete articles on WS, and see how it goes to "process" it:


 * check the listing here thoroughly;
 * in parallel, because it very much makes sense, make sure the 'wikipedia' field is filled for all the biographies on WS that exist here;
 * look at what the stock of vol. 28 articles that need creation actually is.

At that point, we could consider the issue that prompted all this, namely bot importation of text. It might make sense to import a volume's worth of DNB text into our own project space, with some basic topping-and-tailing, so that we have an area of collaboration with one list of several hundred articles (I suppose) to work on. Articles could be moved out to mainspace whenever anyone felt they were ready.

The other thing, arising from the previous comments: it would be a good idea, as other projects do, if we tagged in some fashion articles with DNB text added (can be through DNB) and then tried to check them against the ODNB and other sources. In other words have some sort of rating system (not a new idea, see Merging encyclopedias). This is how I'd like to address the issue of reliability of the old DNB (it is a "reliable source" in jargon terms, and contains some mistakes - we should discuss this on a project page, though). Charles Matthews (talk) 10:05, 10 September 2010 (UTC)


 * What do other participants here think of these proposals?Dsp13 (talk) 23:35, 15 September 2010 (UTC)
 * Very much in favour of this experiment. --Tagishsimon (talk) 00:58, 16 September 2010 (UTC)
 * I think I could autogenerate a list of those who are listed in DNB Volume 28 but not tagged with . Would that be useful? Dsp13 (talk) 01:16, 16 September 2010 (UTC)
 * That'd surely be a step to take once we've disambiguated the links in #28. Right now that page is quite scrappy. And if by DNB Poster (a template new to me) you mean to find articles in wikipedia that do not have a pointer to wikisource, I think you'd also have to look for absences of DNB or DNB Cite having wstitle arguments, since these in my experience are used as the pointers. --Tagishsimon (talk) 01:26, 16 September 2010 (UTC)
 * re wstitle, hmm. That makes it more difficult than I'd hoped - maybe it's beyond my competence. (I don't think disambiguation need precede this step, though it could.)Dsp13 (talk) 01:45, 16 September 2010 (UTC)
 * I stand corrected; it's already been dabbed. I assumed it hadn't. JamAKiska has been a busy boy or girl. --Tagishsimon (talk) 01:57, 16 September 2010 (UTC)


 * DNB now has hiddent categories for the numbers of articles. The sub categories are listed in template:DNB/doc. The numbers are currently:
 * 1,840 Category:Articles incorporating DNB text with Wikisource reference
 * 192 Category:Articles incorporating DNB text without Wikisource reference
 * 0 Category:Articles incorporating DNB template with an unnamed parameter
 * Another useful task would be the importing articles into Wikisource that make up the 10% of Wikipedia articles that use DNB text but for which there is no corresponding Wikipedia article. -- PBS (talk) 22:48, 19 September 2010 (UTC)

Bot building
I don't know if Rich has any ideas on how best (from a technical perspective) to proceed, but the above list contains a nice number of redlinks, which seem like the top priority (maybe those PD-70 also?). We could distil that list, thrash out some automated edits to make whilst importing, then transfer from WS into projectspace, ready to be tweaked and moved into mainspace by human editors. Does that seem reasonable, or have I misread the discussion? - Jarry1250 [Who? Discuss.] 09:15, 18 September 2010 (UTC)
 * Yes I had a good look at this, now I am prohibited from editing HTML comments.... I have imported the DNB contributor templates for starters, and am looking at the 5000 articles in DNB No WP.  There's lots of fix-ups that can be done, I will probably get 100 trial imported into workspace, fairly soon (say by Thursday) for comments. Rich Farmbrough, 10:59, 26 October 2010 (UTC).


 * Ah, there is a new tool for that category from Magnus Manske. Trial run for letter T: edit the browser line for any initial letter. The very strong matches to an enWP article are in the right-hand column, in bold. There are a few false positives in the matching, but in general this tool is very good at finding such matches. This means that some whittling away by humans of s:Category:DNB No WP is going on. Just so you don't duplicate efforts. Charles Matthews (talk) 11:25, 26 October 2010 (UTC)
 * I thought that was already done. It is no great matter though, but of course useful to understand why the cat size may be changing, since once something is removed from that cat I shan't be looking at it in the short term. (I may need to use them all to enhance the data set, or may wish to to provide merge material cheaply but that's two different stories.) Rich Farmbrough, 14:54, 26 October 2010 (UTC).


 * I have linked a few, and made my first article move on Wiksource. I press on with the work, but distractions are abundant. Far too early for feedback to be useful (there is so much wrong and missing that I am not short of thing to improve) however for the curious, a set of 10 pages can be found at User:Rich Farmbrough/DNB Rich Farmbrough, 23:24, 26 October 2010 (UTC).


 * OK now have 43 articles in test at Category:DNB drafts will increase this to 100 as time goes by. Those not listed under "A" have clear problems demarking the beginning of the article in the source - that's on the back burner, but seems an easy fix. I am working on linking algorithms right now. What would be useful is lists of anachronistic collocations with their modern equivalent. I do not intend to bring the text completely up to date - part of the style involves tortuous, even by my standards, sentences - these are best left form manual attention, merely to bring the language up-to-date where convenient. Rich Farmbrough, 03:58, 29 October 2010 (UTC).

Interesting to see articles in this sort of intermediate state. My eye lit on expressions like "is now at" (update might be needed), "poetical writer" (poet), "deservedly popular" (popular), "abounding charity" (charity). A common one is the use of "removed" of a person going from place A to place B, where modern English is simply "moved". Rich, the citation template to use now is. I mean to put up a page or two for the project on all the relevant templates and categories, since there has been substantial development recently. Charles Matthews (talk) 07:13, 29 October 2010 (UTC)
 * Thanks I grokked "removed" but put it in a context (s)"he removed to" of course then I found adverbs before he removed. The "peacocky" terms .. abound? and are definitely worth addressing  Do you have an example of "is now at"?  This sort of process is very odd, because sometimes I spend hours wrestling with a punctuation mark, while supposed difficult things (like the bluelinks) take scant minutes.  Rich Farmbrough, 16:53, 29 October 2010 (UTC).


 * "Her portrait was painted by Cornelius Jansen, and is now at Welbeck." This is from [User:Rich Farmbrough/DNB/M/a/Mary Armine this draft]. So it may have been there 120 years ago ... who knows now? If you set up a special page in your userspace, I could add many more examples as I came across them. Charles Matthews (talk) 16:02, 31 October 2010 (UTC)
 * Rich, these are fantastic. I know from experience how it's the punctuation marks that cause merry hell with this sort of processing. A few (I'm sure obvious) issues: 1. death date seems to be repetition of death date. 2. I'm not convinced about the [DNB n] references listed at the end of the article. Given we have no info about where these references should go, I think the Attribution subhead I've seen on some WP bios from DNB is as good as anything. 3. Presumably elementary categories (birth and death yrs) and the DEFAULTSORT template will go in at some point.
 * A couple of questions: 1. what would you like unbotty editors to do with Category:DNB drafts at present? Start moving bios agross to mainspace with a bit of manual tidying? Or is that premature until you've got the process sorted? 2. I have a spreadsheet with ODNB ppl in - no record of whether or not they're in DNB, but 13K links to WP - would that be useful to you? Best, Dsp13 (talk) 19:57, 30 October 2010 (UTC)

Possibly broadening scope
While the British DNB is certainly a valuable source for information about the individuals mentioned there, it covers only Brits. There are a number of other, similar, works of "national biography" for other nations, like and probably others as well. Would this group have any intention of addressing those works as well, particularly when the British DNB is finished? John Carter (talk) 15:56, 11 September 2010 (UTC)
 * American National Biography online
 * African American National Biography
 * Australian Dictionary of Biography
 * Dictionary of Canadian Biography
 * Dictionary of Irish Biography
 * Dictionary of Welsh Biography
 * Dictionary of German National Biography
 * Dictionary of Georgian Biography
 * Biographical Dictionary of Republican China
 * Dictionary of Indian Biography


 * Many of the above may not be in the public domain. The DNB is low hanging fruit because it is PD. This group might move on to any of them, or sibling groups might be established. Or someone might extract the indexes of the Dictionaries and make lists for the missing encyclopaedia articles page. But there's pretty much no way this project will take these on, not least since we have a few years of work ahead of us to complete the DNB. (But thanks for the very useful list;)--Tagishsimon (talk) 16:49, 11 September 2010 (UTC)


 * The German analogue is the old Allgemeine Deutsche Biographie - it's close to the same size as the first edition of the DNB, and it is being posted at de:s:Allgemeine Deutsche Biographie. For French, see fr:s:Biographie universelle ancienne et moderne (older, and less advanced as a project). Also s:A Compendium of Irish Biography has been started. The Catholic Encyclopedia and Jewish Encyclopedia contain many biographies, of course. One further resource that is worth mentioning too, is the "Protestant Encyclopedia", the Schaff-Herzog Encyclopedia of Religious Knowledge which is public domain and fairly easy to get hold of (in the English translation), for example at archive.org (I have a list of scans somewhere, and having this on Wikisource would be good, some day). All these are potentially of interest when working on biographies.


 * I think it is fair to say that if the point is just to fill in a redlink, you don't need a project. Projects score by providing full listings, topical lists, tracking of progress, tools and bots and templates: apparatus and support for getting the most out of a resource. It's about "getting serious" in certain ways. People will go where they want to work, as usual. Charles Matthews (talk) 20:55, 11 September 2010 (UTC)
 * I have picked up some stubs from the ANB update lists, and Dr Blofeld and others have already made Sanapia into a respectable article. Rich Farmbrough, 09:11, 30 October 2010 (UTC).

Topical topics
I put up WikiProject Dictionary of National Biography/Topical lists as a reference on how to get working lists of DNB people. There is an example currently in the news (beatification in a few days of John Henry Newman), meaning it would be highly desirable to create WikiProject Dictionary of National Biography/Tractarians where Tractarians = Oxford Movement. Perhaps this can be used as a prototype collaboration. In any case thre will be quite a large number of related articles that either need creation or expansion: it's the typical situation that clicking away from the Newman article quickly gets into territory needing much attention. The DNB will have most of the figures. See for example Library of Anglo-Catholic Theology for some idea of the secondary figures that are needed. Charles Matthews (talk) 14:12, 14 September 2010 (UTC)

Well, I've started a prototype. This is a table with columns according to how the WP article stands, and topics could be moved from left to right. This kind of listing might be suitable for quite special subjects, not vaguer ones such as "politicians". Charles Matthews (talk) 21:14, 14 September 2010 (UTC)
 * I've autogenerated a list of people likely to go on this page from the epitome pages. Let me know if/how you think this sort of thing could be useful.Dsp13 (talk) 16:13, 15 September 2010 (UTC)
 * Certainly useful. I don't quite know where to go yet with this idea. User:Charles Matthews/DNB Officers of Arms is an example where the idea is to start with lists of office holders. User:Charles Matthews/DNB Cambridge heads of houses is self-explanatory. These are tentative beginnings (more on User:Charles Matthews). I think that a list on the East India Company would be particularly interesting to have. But then a list of alchemists in the DNB would also be fun (and shorter). The analogy with the Catholic Encyclopedia topic lists would suggest starting with one-line summaries and sorting everything, which is perhaps too heavyweight for the DNB because of the scale. What granularity to look at? "Theatre" is a good candidate for a big listing, because the DNB has what appears to be very thorough coverage from 1660 onwards. Lists by date of birth would be good for historians. Charles Matthews (talk) 21:52, 15 September 2010 (UTC)
 * one question is which other WikiProjects, particularly UK WikiProjects, might be excited about plundering the DNB. There are a bunch of regional UK WikiProjects, the most specific ones looking like those for Cardiff, Coventry and Wikiproject Sheffield. Other fairly specific WikiProjects might be British Royalty, or Music of the United Kingdom or WikiProject Elizabethan theatre. Perhaps Quakers, composers, Peerage. At the other end of the scale, of course, is the colossal WikiProject Biography, which I think at the moment is so large and overwhelmed with BLP issues that it isn't much concerned with historical redlinks. As far as historical eras go, WikiProject Middle Ages might be worth a look. Dsp13 (talk) 23:31, 15 September 2010 (UTC)

Yes, very much a good thought about how to get others interested: a path


 * start specialist list -> work over it by various methods for a bit -> pay attention to creating the WS articles -> present it to a WikiProject as an opportunity for them to expand their area, once it looks reasonably mature.

Charles Matthews (talk) 09:51, 16 September 2010 (UTC)

ODNB check
Pardon if this is too specific, move it to a better page if there is one. I created a new article for a painter, then poked around for an image and found that google cast up an artist with a similar name, medium, and subject matter. What I found is noted at Talk:Charles Vacher, but could someone glance at the current article and see if they mention the second artist? Cheers, cygnis insignis 13:41, 2 October 2010 (UTC)


 * No there is not. -- PBS (talk) 02:01, 3 October 2010 (UTC)

the old dnb
I continue to object to the practice of using material from the old dnb for anything more than the bare facts of someones careers: everyone in the UK has access to the new one, the ODNB, and should use it. And all people in the old dnb were included in the new ODND.  DGG ( talk ) 01:03, 31 October 2010 (UTC)


 * That's just the best driving out the good. It is way faster to create an article from the DNB than the ODNB. Sure, one created well from the ODNB will in many cases be better than the DNB text (not all cases; many have few revisions in the new version). It is not (in the main) that DNB is woefully wrong, more that the ODNB often has more information and a rebalanced focus. But this is besides the point. Simply, there is no should about it. You have your approach and preferences, and the rest of us have ours. You don't get to dictate. --Tagishsimon (talk) 01:16, 31 October 2010 (UTC)


 * Not to forget that if the DNB article is from Wikisource, then it can be linked to directly and be freely available with neither a special subscription nor a login. Not all of us have the luxury or the immediate convenience of a UK-based library to backend our access. billinghurst  sDrewth  07:51, 31 October 2010 (UTC)

Well, we cannot do more than skim the modern ODNB for a few facts, can we? Close paraphrase is against our copyright rules. So, David, I don't take too kindly to all this. "Use" could mean quite a number of things. As I have been saying at the Wikisource end of this project, we now have a good matching tool provided with Magnus Manske. There are 5000 articles in s:Category:DNB No WP; and with the aid of the tool, it is quite easy to find some very poor stub articles on WP that can quickly be expanded with old DNB text to something much better. Are you saying that I shouldn't do that if I have a chance? Because, if so, I think I'm going to play a strong card. Don't criticise the editor, criticise the current article. This is one of our community norms, and breaking it can be very deleterious. Please don't dictate how others should edit. Charles Matthews (talk) 15:53, 31 October 2010 (UTC)


 * agree with the DNBers here, but DGG if you'd like the ODNB list I mentioned above then in the bot section thn do get in touch. Dsp13 (talk) 16:20, 31 October 2010 (UTC)

Census
Given the awesome scale of everything DNB, I think we need concrete proposals for finding and logging everything. The tool I have mentioned above is over time going to help with finding the WP articles that should be assessed for DNB text inclusion (because they'll be listed in the WP field of the WS DNB header). Another aspect should be to get ODNB ext lks here into the ODNBweb template, and upgrade the template to include the WS DNB link (this really is the way to address what DGG raises above, in the longer term). Charles Matthews (talk) 16:07, 31 October 2010 (UTC)
 * ODNBweb looks to need to be converted to an inline form. Not sure how you would want to incorporate cite DNB, though maybe we would look to have the first template call the second and we could have both inside the same &lt;ref> tag.  Putting them as separate isn't correct for pretty much the same data. billinghurst  sDrewth  13:04, 1 November 2010 (UTC)

I was thinking of folding DNBfirst into ODNBweb, with some flags to raise if there is no matching DNB article at all, resp. no existing DNB article. Charles Matthews (talk) 17:17, 1 November 2010 (UTC)

The external link finder shows nearly 8900 links to http://www.oxforddnb.com. Before much more is done, I'd like to consider tweaking ODNBweb purely for format: see Template talk:ODNBweb. There is a great deal of the ODNB stuff that we should look at getting under control. That template really ought to admit author information, for example. That in addition to incorporating and expanding what DNBfirst does. Charles Matthews (talk) 09:17, 8 November 2010 (UTC)

So now I've tweaked the template, and checked existing occurrences. There is plenty to do with the External link finder in the direction of placing the ODNB links in templates. (Try moving ahead a few hundred in the search.) Existing usage of ODNBweb tends not to invert the name, but it doesn't matter so much as long as there is a convention. The current pattern is to place DNBfirst afterwards, as in



(note punctuation). That's when (a) there is a DNB article by 1912, and (b) it exists at Wikisource. A few things have come up already. There are some long ODNB ids like 11000999, and they need to have the first part taken off. The template ODNBweb starts to look a bit basic, since it is not possible to put author information into it as it stands. And I have met a case of doi used to give the ODNB link - at Richard Bagot (bishop). I left it there for the moment, though I think probably it is better to use something purpose-made. Charles Matthews (talk) 09:46, 10 November 2010 (UTC)


 * Yes, I'm reluctant to replace templates, if they're kitted out with author, date of access, etc. - see e.g. John Adam (administrator) - with  if that means losing information. Any template fiends to give ODNBweb more bells and whistles?Dsp13 (talk) 20:10, 10 November 2010 (UTC)


 * Its easy enough to do, as we can simply include cite encyclopedia for the ONDB, and cite DNB or DNBfirst in it. But see the next section is that what we want to do? -- PBS (talk) 23:40, 10 November 2010 (UTC)


 * But as a first step I could knock up a version of ODNBweb calling cite encyclopedia in the ODNBweb if it is wanted. -- PBS (talk) 23:40, 10 November 2010 (UTC)


 * First cut using the example on the talk page:
 * The url is broken but it is also broken for
 * There are two ways to fix it. Either remove the link and rely on the "doi" (my preferred option and that of the ODNB ) or add "101093" internally to the URL string. What do other think? -- PBS (talk) 01:31, 11 November 2010 (UTC)
 * The url is broken but it is also broken for
 * There are two ways to fix it. Either remove the link and rely on the "doi" (my preferred option and that of the ODNB ) or add "101093" internally to the URL string. What do other think? -- PBS (talk) 01:31, 11 November 2010 (UTC)
 * There are two ways to fix it. Either remove the link and rely on the "doi" (my preferred option and that of the ODNB ) or add "101093" internally to the URL string. What do other think? -- PBS (talk) 01:31, 11 November 2010 (UTC)

Today I went in adapted Template:ODNBweb by lifting most of Template:Springer. This gets us some of the way. Now the old two fields are id= and title=; I have just gone through changing that everywhere. Also author names from the ODNB now go in the usual last/first/authorlink fashion. That deals with the issue of including information already given. (It would be easy enough to add print publication info via the old Springer fields).

What comes next? We have three-valued logic on wstitle=; namely present, absent on WS; and none. We have two authors, original on DNB (maybe = none) and in the ODNB. Some ODNB articles are explicitly revisions. From here, I'd like to see conditional statements so that we have maintenance categories for the unfilled author and wstitle fields. Collectively we have the know-how ... Charles Matthews (talk) 13:11, 21 November 2010 (UTC)

Too much of a good thing
See history of Henry Anderton

If there is a DNB or a cite DNB in the References section, I do not think it appropriate to add DNBfirst as well. -- PBS (talk) 20:30, 10 November 2010 (UTC)


 * Well, if we start attributing intelligence to the reader, it's something of a game-changer! There is a clear argument for accompanying the use of ODNBweb everywhere by the choice of looking at a free version. I think you mean that if DNB is linking to Wikisource, then DNBfirst is duplicative. Only so if everyone will realise that (at whatever distance down the page) the DNB represents the "freemium" aspect of the ODNB (well, should be). I wouldn't bet on that.


 * In any case I'm for having the functionality of DNBfirst put in ODNBweb and gradually retiring this template (it was a stopgap when first introduced, see my Talk archives). Then we should ask that it can be deliberately switched off when it seems intrusive. Charles Matthews (talk) 21:52, 10 November 2010 (UTC)


 * I have qualms. First, I agree with PBS that using both DNB and DNBfirst on a page is duplicative - duplicating information for those in the know, and duplicating enough verbose text to be irritating for those that don't care. We don't want to annoy readers & editors on the way to where we want to go. Second, in cases where the DNB & ODNB differs, it's sometimes necessary to cite each separately. Automatically bundling them together gets in the way there.Dsp13 (talk) 22:37, 10 November 2010 (UTC)
 * I presume you are referring to the problem that not all the entries in the ODNB have older entries in the DNB, and it is not just 20th century entries. I came across such a biography William Eyre (leveller) who became notable because of Christopher Hill's work done between the publication of the DNB first edition and the ODNB. If the two templates are combined how do we handle that? -- PBS (talk) 23:27, 10 November 2010 (UTC)


 * The DNB topics are a subset of the ODNB topics (to a first approximation: there is some merging as well as splitting). As said above, what is really required is an ODNB template with the option to add in the WS title of a DNB article.


 * To be clear, cite DNB can perfectly well stand on its own. In fact under general WP principles, we shouldn't at all be using the ODNB to substitute for references to the DNB (at Wikisource); where the old DNB article is flawed, as it is in numerous cases, we should try to annotate it at Wikisource for the general benefit. Putting up two footnotes, one with the ODNB ref and one with the old DNB ref, possibly commented, is a decent solution, when the articles concerned are sufficiently different to warrant it.


 * Talking here about DNBfirst, that template is there to solve the problem of giving a free "backup" for those who can't access the ODNB site. It is a verbose but clear way. The issue in this thread is the verbosity, when duplicative on the page. It really would be easy to create Template:ODNBweb1 which did away with the verbose phrasing but still gave a "free version" link to Wikisource, for use on occasions where the fuller wording is thought obtrusive. Charles Matthews (talk) 07:00, 11 November 2010 (UTC)

Just as a note, if pages with both DNB and DNBfirst are to be cleaned, you can find an automated list here. --Magnus Manske (talk) 16:42, 11 November 2010 (UTC)

A critic
See Wikipedia talk:WikiProject Biography. Charles Matthews (talk) 14:49, 23 November 2010 (UTC)

{Now archived.) Charles Matthews (talk) 10:50, 27 January 2011 (UTC)