R&S Example page/OCLC as a verification source
Jump to navigation
Jump to search
OCLC has recently been added as a Primary Verification source. It is certainly an invaluable one, but as I use it more and more, a 'problem' with their data becomes more noticeable. A publication/edition with only a single binding can have up to six records, seemingly all for the same printing, but often only one or two actually match the book. I find, especially for books after 2000, just as many, if not more, 'stub' entries (basically the same data that Amazon spits out for pre-release) that are quite useless. We use them on our DB until a source can 'correct' them, but they don't seem to disappear from OCLC. While the need to search multiple records to find a correct one is just part of the work, to me that kind of lessens the reliability as a Primary Verification source. So far I've just been mentioning these 'other' records in the notes. Not really saying it's a problem, just noting what exists. ~Bill, --Bluesman 15:04, 22 July 2009 (UTC)
- Perhaps you should only cite the more reliable or complete OCLC record. I don't think it does much good to cite them all. Some of them will be initial entry records, probably based on publisher catalogs, that somehow got overlooked when the actual books arrived and a new record was created. The only time I'd mention a stub record is if that's the one that the OCLC link goes to when it's clicked from the ISFDB record. This is the most frustrating part: when OCLC has two records for the same ISBN, what determines which record is linked to the ISFDB record? I was recently made aware of the British Library Integrated Catalog, which I find to be a better source for UK pubs than OCLC. The best part is that most of the records show the publisher's list price, which is extremely rare for an OCLC record. I suggest taking a look there for UK publications before going to OCLC. (This is from a diehard fan of OCLC.) MHHutchins 15:46, 22 July 2009 (UTC)
- That's what I've been doing, putting the number of the most complete record (though at times I put two numbers if each has data the other doesn't) and then just note that there are other records. I'll check out the UK site! Thanks! ~Bill, --Bluesman 17:32, 22 July 2009 (UTC)
- One give-away is a record with no page count or physical size. Such records seem pretty much always based on advance publisher info, AFAICT, and of lesser value. Note that if you go in via the "First Search" interface, the "Vendor Info" often displays a price, and in every case I have been able to check, this is the publisher's list price. One minor issue, there is no currency symbol and in a few cases it is not clear which currency this is in for a UK published book, but usually it seems to be in US dollars. Note also that the "responsibility" field is good at weeding out "co-authors" who are actually illustrators or authors of forewords or the like. -DES Talk 21:41, 22 July 2009 (UTC)
- I Usually cite multiple OCLC numbers only if I got some data from each. 21:42, 22 July 2009 (UTC)
- It might be worth an additional software check that an OCLC/Worldcat reference is recorded in notes before an OCLC/Worldcat verification is allowed, but basically it's uncontrollable and should be addressed in Help - "don't verify against OCLC/Worldcat without crediting the records involved". I don't believe that's a Primary verification though, where did that idea come from? BLongley 21:58, 22 July 2009 (UTC)
- So why should the standards for OCLC/Worldcat verifications be different than those for any other verification? Should we cite Reginald's reference number before we're allowed to do a Reginald1 verification? Should we have to cite the page and column for a Tuck or Currey verification? Hell, we can't even codify Primary verifications, much less Secondary verifications. I'll accept that standards for each of the verifications should be different. I don't have a problem with the fact that standards for one area don't necessarily have to apply to all areas. Just give me "clear criteria with principled justifications". :-) Oh, BTW, in every submission that Bluesman has verified for OCLC/Worldcat, he's provided the OCLC number. MHHutchins 00:35, 23 July 2009 (UTC)
- I do make a practice of always including an OCLC number, and in some cases a link, when doing an OCLC verification. (If I have failed to do so, it has been an oversight.) The reason that justifies this? it ensures that any future researcher can fairly easily find and view the exact record I was using. If I don't provide a record number, a search would have to be done on name, title, or perhaps ISBN, and multiple records might well be returned, and it might never be certain which one or ones I had used. For a Tuck or Curry reference, is the author and title sufficient to unambiguously identify the entry used? Is it fairly easy to find the reference without a page/column cite? If so, no great need for them, IMO, although they might save a future researcher time and effort. I don't know that I would make it a "RULE" that every OCLC verification must have an OCLC record number in the notes, but I do think it is a "best practice". -DES Talk 00:53, 23 July 2009 (UTC)
- Thank you, DES, for the rationalization, but it wasn't entirely necessary. I wasn't serious about needing a justification for having different standards for each verification source. In fact, it was my position that each require a different set of standards. One thing I should have learned from my time here is that it's very difficult to convey a playful sense of "tongue-in-cheek" with mere words. It's in the tone of voice, the glint in the eye, the mouth on the verge of a smile. None of which can be easily replicated in print. I'll try in the future to avoid attempting such as this. Sometimes a smiley face just doesn't do the job. MHHutchins 01:15, 23 July 2009 (UTC)
- No sometimes it doesn't. I did understand your reference to my comment from the ebook thread to be in jest, but it wasn't quite clear if your were seriously, even if lightly, asking why an OCLC link should be provided for every pub being OCLC-verified. and even if you weren't, Bill had suggested making it a software-enforced rule, even if he dismissed the suggestion a sentence later, so it seemed worth giving my thoughts on why it was a good idea. I apologize if I said something too obvious for words. Sometimes the obvious needs spelling out. -DES Talk 03:39, 23 July 2009 (UTC)
- I agree with DES, etc. that it would be a "good practice" to always include the OCLC accession number so that we'll know which of several possible records was verified or used as a source. Having the number also makes it easy to search for the record as you can just copy/paste it into the search field. When using a book as the source for something I try to always include the page number, and sometimes even the paragraph or line number. --Marc Kupper|talk 05:43, 23 July 2009 (UTC)
- No sometimes it doesn't. I did understand your reference to my comment from the ebook thread to be in jest, but it wasn't quite clear if your were seriously, even if lightly, asking why an OCLC link should be provided for every pub being OCLC-verified. and even if you weren't, Bill had suggested making it a software-enforced rule, even if he dismissed the suggestion a sentence later, so it seemed worth giving my thoughts on why it was a good idea. I apologize if I said something too obvious for words. Sometimes the obvious needs spelling out. -DES Talk 03:39, 23 July 2009 (UTC)
- Thank you, DES, for the rationalization, but it wasn't entirely necessary. I wasn't serious about needing a justification for having different standards for each verification source. In fact, it was my position that each require a different set of standards. One thing I should have learned from my time here is that it's very difficult to convey a playful sense of "tongue-in-cheek" with mere words. It's in the tone of voice, the glint in the eye, the mouth on the verge of a smile. None of which can be easily replicated in print. I'll try in the future to avoid attempting such as this. Sometimes a smiley face just doesn't do the job. MHHutchins 01:15, 23 July 2009 (UTC)
- See first line of this section. So "Blame Bluesman", if it wasn't mentioned elsewhere earlier. BLongley 23:28, 22 July 2009 (UTC)
- Just chiming in to confirm that OCLC is a Secondary verification source. Ahasuerus 23:40, 22 July 2009 (UTC)
- It was a simple momentary lapse in thought. I knew what he meant, and anyone who has been here more than a few weeks should have known what he meant. God, people, lighten up. MHHutchins 00:35, 23 July 2009 (UTC)