[Taxacom] Proofs for opinion

[Stephen Thorpe]

No. Possibly Paul can corfirm this, but, somewhere, there almost certainly WAS AT SOME STAGE a strong implication to the effect that a publication must be registered on Zoobank before it is published electronically. The problem of course is that once it is published without registration, you can't go back in time to preregister it, so it cannot be published electronically. But now Art. 8.5 doesn't seem to say that! Hmmm ...

I think that the meaning of "issued" is actually quite clear, and so that a pre-2011 electronic publication cannot be rendered valid by adding registration etc. In fact, nothing before mid-2012 can be, because the archiving statement could not be made on ZooBank before then. There is published discussion on this point, somewhere.

You say [quote]I think that the earliest reasonable date in that case would be the date on which all requirements were fulfilled[unquote]. But that is the problem! How can you tell on what date a missing LSID was added to a PDF if the publisher doesn't come clean and state that date?

We don't have the same problem with dating paper publications as we have for electronic ones, as I'm sure that your colleague Neal can explain to you. Ever been to a library? Journal issues in libraries have accession date stamps (and librarians have no reason to falsify them). The original idea behind archiving electronic publications was to make such archives the electronic equivalent of libraries, but, because Zootaxa output is faster than archiving can keep up, this requirement got watered down to a mere statement of intent to archive. Hmmm...

Cheers,

Stephen

--------------------------------------------
On Fri, 9/1/15, Richard Pyle <deepreef at bishopmuseum.org> wrote:

 Subject: RE: [Taxacom] Proofs for opinion
 To: "'Stephen Thorpe'" <stephen_thorpe at yahoo.co.nz>, "'Paul van Rijckevorsel'" <dipteryx at freeler.nl>, taxacom at mailman.nhm.ku.edu
 Received: Friday, 9 January, 2015, 1:16 PM
 
 > This was my point
 from before. Oddly, I can't now seem to find the bit
 that I
 > thought was in the
 Code requiring Zoobank registration BEFORE publication!
 > Art 8.5. (below) seems OK from this point
 of view:
 
 It's implied
 that registration must happen before a work can be
 considered
 available, because the
 registration number (or other evidence of
 registration) must be included within the work
 itself.  Basically, in order
 to be
 available, an electronic work must ITSELF include:
 1. the "date" of publication (which
 can be only a year)
 2. evidence that
 registration has occurred
 3. all the other
 stuff that is required for both paper and electronic
 works.
 
 Separately, the
 associated ZooBank record must include:
 4.
 the intended archive
 5. ISSN or ISBN
 
 In my mind, the work becomes
 available as soon as all five points are
 fulfilled.  
 
 The real problem that we're dancing around
 is not the meaning of the word
 "published", but rather the meaning
 of the word "issued".
 
 For example, suppose a PDF was put online in
 2011, and fulfilled all five
 points above. 
 It was not available in 2011, because Art. 8.5.1 says that
 it
 must "have been issued after
 2011".  So, what does "issued" mean? 
 Assuming
 the PDF was still downloadable
 (=obtainable) on 1 January 2012, then is it
 considered to have been "issued" on
 that date (as well as every other date
 that
 the work is obtainable)? I can see arguments both ways.
 
 The word "issue" or
 "issued" appears 17 times in the Amended Article
 8. But
 it's not defined in the Glossary
 (closest definition is within the
 definition
 of "publish").
 
 > But 8.5 now opens another can of worms.
 Now it looks like a forgotten
 ZooBank
 > registration can be added in at any stage,
 thereby making a valid
 publication
 > from the date of registration (provided
 that the LSID is added to the PDF
 on the
 > same date)! 
 
 I don't think anyone would argue that
 retroactively adding a ZooBank number
 to an
 electronic publication pushes the date of availability (for
 purposes
 of priority) back to the
 *registration* date.  I think that the earliest
 reasonable date in that case would be the date
 on which all requirements
 were fulfilled. 
 That is, the date on which an electronic work with the
 ZooBank number included was obtainable
 (assuming other criteria met).
 
 > Given that the date of publication cited
 in the work itself can be
 > incomplete or
 incorrect, how can we tell when the Zoobank LSID was
 > subsequently added to a PDF? How can we
 tell the valid publication date??
 
 Indeed!  This is getting to the crux of my
 original long post on this
 thread. Of
 course, we have the exact same problem with paper-based
 publications too, so this problem is not
 peculiar to electronic
 publications.
 
 Aloha,
 Rich