Wikipedia talk:WikiProject Missing encyclopedic articles/DNB

Many of the volumes of the DNB not found at archive.org can be found at google books. They can be hard to locate, though, because google books thinks they're all different editions of the same book. I can add links to them as I happen to find them if that would be useful. Loren Rosen 00:12, 6 September 2007 (UTC)

Interesting. I'm a UK user, and Google shows the DNBs, but with limited or no preview. I understand there are geographical differences in the availability of full view of google books; looks like the UK is bitten on this one. Which is a shame. It would be absolutely great if you could add links to the google book instances of DNB. --Tagishsimon (talk) 08:38, 6 September 2007 (UTC)

I wasn't aware that google was checking your location for "full view" books. I suppose it's possible. But also, there are truly different editions, and google has several of them with varying rights. So as an experiment, see if you can access vol. 26 (in the original 1891 version) from http://books.google.com/books?id=iycJAAAAIAAJ

I'd be interested to know about the geographical variation for full view books. I've contemplated adding links in references to google books that are out of copyright, but only if truly everyone can make use of them. Loren Rosen 15:24, 6 September 2007 (UTC)

No, I cannot see the text of that one. How much does that suck? Seethes quietly. --Tagishsimon (talk) 08:58, 7 September 2007 (UTC)

Good enough?

How should we decide when to remove an entry from the list? The stated criterion is "when an article has all of the ODNB information in it." More reasonably, I think it should be "When an editor determines that There is nothing more that should be added from this source based on Wikipedia's policies."

However, this was not the criterion used when the list was generated. Of he ODNB "top 3000", we mechanically deleted 3000-198 = 2802 articles based merely on the existance of WP article with a matching name and date: no human check was done for completeness, if I understand the process.

Given this, should we now hold ourselves to a higher standard for the remaining articles? I think the answer is "no." If we have an article that is more than a minimal stub, I think we should remove it from our current list. after all, if the article had existed, even as a minimal stub, before the list was created, the article would not have appeared on the list. So, if it is even a little bit better than minimal, just remove it.

This brings up the need for a new "article improvement" project. We need a second project page (well, maybe ten pages) with all 3000 names. Editors with ODNB access should be invited to check articles on that list and remove them only if the article is "complete" with respect to the ODNB.

I'm raising this issue for two reasons:

• I have been dumping DNB1900 articles into WP and removing the names from the list. These articles need attention from someone with ODNB access.
• I have encountered some existing articles that match list entries, but that do not appear to include all useful DNB information (I don't know about ODNB information.)

Please comment. -Arch dude (talk) 03:11, 19 November 2008 (UTC)

I've removed your poll, since I think we'll get consensus from this discussion. There are only three or four of us hammering away at this thing, with you currently well in the vanguard.

My feeling is that we should from hereon expect that there will be a full DNB dump into wikipedia, or put another way, a check that no more useful content can be added from the DNB before we remove the item. That appears to be the MO of the small group that's chipping away at the work, and it is only for that reason that I suggest this approach. I don't think we should be too fussed about making exceptions for, for instance, a long & well written article that was sourced from other than the DNB. So "containing all DNB data or of length comparable with DNB entry" might be a rule of thumb.

More broadly, because of the mechanical exclusions, I think we need a second project - check each DNB entry against its wikipedia article and check for additional DNB content and then a third project - check each ODNB entry against its wikipedia article & make such changes as appear appropriate. Proejcts 2 & 3 should be run in parallel, since in most cases doing a thorough ODNB check will either cover the DNB check or else prompt the DNB check (DNB archives being available through ODNB).

However P2&3 are of less interest to me right now, since we have so much of P1 to do. P2 & P3 we might set up any time we like - rather like the EB1911 verification project. --Tagishsimon (talk) 03:24, 19 November 2008 (UTC)

Progress and the future

Much has been happening on Wikisource to produce the "raw material" - finished DNB articles - since the discussion above. In the view of a couple of us, it is time for the DNB MEA subproject to raise its profile and develop more of a corporate identity. This page seems the right place to start a discussion of how to go ahead from here. Charles Matthews (talk) 08:45, 27 July 2010 (UTC)

As a participant in the other place, you have my support. Were you thinking of a specific hook? billinghurst sDrewth 10:10, 27 July 2010 (UTC)