Ed Jones wrote:
<snip>But if ONIX data comes from publishers, they should know better than anyone else in the world where they are located. All they have to do is look out the window or if that doesn't work, ask the person in the next desk what town they are in!
For modern publications, I have never seen a convincing argument for spending time and effort on the place of publication except for obscure publishers (where the full address might be even more helpful). AACR2 tried to make this element optional back in 1978, but by the time the code was implemented in 1981 most libraries (including LC) had decided they would treat it as mandatory. RDA tried to make it optional again but this effort has been successfully beaten back.
This is an example of a possible area of *cooperation,* i.e. when *everybody* involved can change. Instead of just giving up and saying it can't be done or it's not worth the effort, we need to look at the entire system and see which communities can provide specific information the most easily. Getting publishers to add a place of publication should be the easiest thing they can do--certainly much easier than expecting them to buy into this obscure system of FRBR/RDA works/expressions/manifestation/items.
Do we need place of publication? That is a matter of debate, but in my opinion, if we want systems to somehow automatically know what copyright laws are in effect for a resource, place of publication becomes vital because the laws are different all over the place. For example, it is such a pain for me in Italy to not see many items in full text from Google Books that I should. I would love having better and more reliable information in these areas. While I hope that the copyright laws unify someday, I am not holding my breath.