Submitting an authority that is a new edition of an old source

[ptgolden]

@atomrab, I have a question for you. Since we're winding down development work, I wanted to submit some periods from books I have laying around the house. One is Lewis Mumford's Technics & Civilization, which was published in 1934, but which I have a 2010 publication of. How should I fill in this authority?

Adding the WorldCat link sets the year published as 2010. While that's technically correct, it's obviously not what we want to have in our data, because the periods were asserted in 1934. I could add a WorldCat link to the 1934 first edition as the authority, but I don't have access to that edition, and the page numbers would be incorrect.

My hunch is to set the year published to 1934, and then write out a full text citation indicating the edition that I have. What are your thoughts?

[ptgolden]

It actually looks like this edition preserves the original page numbering, but I'm still curious what you would say when that is not the case

[atomrab]

I ran into this in a big way with the Priestley chart, which was published in many, many versions, some with significant differences (see my editorial notes at http://n2t.net/ark:/99152/p05hrsf). My own preference is to get back to the original publication wherever possible, especially when we're dealing with reprints rather than revised editions. I would much rather have the older version represented, for obvious reasons, and if the pagination is the same I think you can use the 1934 WorldCat reference with an editorial note that you consulted it in an identical reprint of 2010. It's a little sneaky, but I don't think problematic.

But if the pagination is different, I think we'd be introducing potential error and eroding our commitment to transparency in sourcing. A user might try to track down the reference in the early edition and then not find it. So if there were no way to get to the 1934 version, I would say you'd have to use the 2010 as is, with a note that it was first published in 1934, and then we could try to get the earlier version and add it (or update the existing collection, if it was only a matter of a pagination change). @rybesh, what do you think?

[rybesh]

I guess I would enter the earliest publication details as the main bibliographic info for the authority, but if the page numbers refer to a later edition, then note that in the editorial note for the authority and put a citation to the later edition in the note too.

[atomrab]

I'm comfortable with that, as long as we're dealing essentially with reprints rather than editions with significant updates. But I do think that for later editions with substantial revisions, it's better to have both the older and newer versions, as we did with the Norton anthology. We don't want to attribute a definition that was substantially changed for the 2010 edition to the 1934 edition; this will just muddy the waters for the network of influence we're trying to document.

[rybesh]

Yes, I agree.