On the contrary, I don't think anyone was suggesting it was easy ‹ quite
the opposite. There are simple situations, at times, where the publisher
is clearly sited in one physical location, but also many times when they
aren't. This is far from simple, which was the situation being identified.
In the end, given the issues that Karen brings up, probably the best
things we can record for any one publisher is the publisher
identification, all of the places named on the piece, and the year.
Hopefully somewhere down the road we will be able to parse that into
something that is meaningful. But yes, publisher information is a world of
hurt. We have certainly experienced this as a metadata aggregator. It
isn't easy, by any means.
Roy

On 8/14/14, 8/14/14 € 5:40 PM, "Karen Coyle" <[log in to unmask]> wrote:

>On 8/14/14, 2:12 PM, Stuart Yeates wrote:
>> I'd like to echo this.
>>
>> If we have a pointer fully-populated authority control record for
>> publisher, adding location of publication is redundant at best and
>> misleading at worst.
>
>If only it were that easy :-(. There is no one-to-one between a
>publisher and a place. There is also no one-to-one between a "publisher"
>(depending on how you define that) and what you find on the title page.
>
>There's been a lot of talk about having authority control for
>publishers, and at some point it may be possible, but with mergers,
>break-ups, buy-outs, etc., keeping up with that will be a costly mess.
>We don't want to head into that morass until it's been given a very good
>analysis.
>
>As for the question that Lars poses:
>
>>
>>>
>>> If we -- however -- talk about place names in publication statements,
>>> URIs (or geographic coordinates) may or may not be useful. For a C17
>>> publication with the (fictive) publication statement "published by
>>> John Nicholson in the High Street at the sign of the lantern and
>>> printed by Jeremiah McTavish in the Lower Mile" it is clear that the
>>> two places mentioned are the locations where the publication
>>> (preparation of manuscript etc.) and the printing took place. We can
>>> replace those two with URIs or geographic coordinates.
>>>
>>> Publication statements by modern scientific publishers sometimes
>>> mention a whole range of place names, as in "Heidelberg, Berlin, New
>>> York, Tokyo, München: Very-Big-Scientific-Publisher, 2012". I'm still
>>> not sure if this means that
>>>
>>> 1) the book was published in all those places (at the same time?), or
>>> if it means that
>>> 2) it was published by the Very-Big-Scientific-Publisher and that the
>>> publisher has offices in all those places.
>
>A better approach may be: for the purposes we can imagine today that
>would take advantage of URIs (linking, searching, other?), does the
>difference between your 1 & 2 matter? What is the expected value of
>retrieving information based on place of publication? If not precise,
>what are the consequences? And does the answer to this question change
>when we contemplate using URIs rather than when we develop cataloging
>rules (which already instruct us to transcribe the places from the title
>page, AFAIK)?
>
>kc
>
>>>
>>> My gut feeling is that it really means 2) and that we really cannot
>>> say that the book was published in http://dbpedia.org/data/Berlin,
>>> http://dbpedia.org/data/Tokyo or any of the other places. Please feel
>>> free to prove me wrong.
>>>
>>> Best,
>>>
>>> Lars
>>>
>>> *** Lesen. Hören. Wissen. Deutsche Nationalbibliothek ***
>>>
>
>-- 
>Karen Coyle
>[log in to unmask] http://kcoyle.net
>m: +1-510-435-8234
>skype: kcoylenet/+1-510-984-3600