[open-bibliography] (Final?) discussion of the openbiblio principles

[Peter Murray-Rust]

On Mon, Jan 10, 2011 at 2:39 PM, Dr. Micah Altman
<Micah_Altman at harvard.edu>wrote:

> Hello,
>
> Thanks for the thoughtfule posting


> I've been mostly a listener in this ongoing discussion and I had read
> closely developing drafts of the principles, but have hesitated to
> edit or suggest changes because I wasn't sure I understood the primary
> argument / models driving them. It does seem now that there
> may be different views underlying the concepts of bibliographic data
> and the main arguments underlying the draft principles.
>

There are differences and we should welcome them.  The differences require
us to look for what is common to all of us (and "us" means any members of
this list and beyond). We have agreed that we have something important to
take to the community.

>
> Some gray areas at the boundaries of the definition of "bibliographic
> data" is to be expected. However, the principles might be made more
> compelling if the main arguments backing these principles were made
> clear -- probably mostly by reference to previously published
> statements on the value of open access, and some text to put
> bibliographic data in the overall context of open access.
> Making these arguments explicit may also help to reduce future
> ambiguity in interpreting the principle statements.
>

By "Open Access" I assume we mean the general approach of the BBB
declarations - that was certainly in my mind as soon as we started.  I'd be
happy to say that the principles (i.e. the motivation and general
abstraction) are heavily informed by Open Source, Open Access and Panton
Open Data. That in itself creates a strong pre-amble

>
> My  current take on the  four core principles is that the most
> compelling cases for them are for scholarly publications, and
> specifically to the sharing of  quasi-factual bibliographic
> information (over which multiple coders are unlikely to significantly
> disagree -- identifying and locating information, but not necessarily
> subject classification and abstracting).  If this data were available
> broadly, the research impact across fields would be tremendous; and
> the incentives to produce such information in future would  not  be
> substantially reduced. I think one can also make a very compelling
> argument about the need to have open  taxonomic/ontological
> information (including controlled vocabularies, etc.) that are meant
> to be shared across a community -- simply in terms of network effects.
>

No secret that this is my motivation!


> As I get farther from these cases, I'm finding it somewhat harder to
> apply these unconditionally...
>
>
Again a major strength is that we have active participation from a wide
range of interests. It's really valuable to have the cross-fertilization
between A&H and sciences (excuse the simplicity), between monographs, theses
and articles, and between "library" and other sectors. A lot of the early
discussion highlighted the very different backgrounds and we felt that
having a single approach was valuable to the community as a whole. (I would
hate to have "POBD for libraries").


> Alternatively, one could (e.g. in the preamble) reframe these
> principles as defining "open bibliographic data" rather than as
> unconditional recommendations for all possible bibdata. And then list
> some the compelling advantages of open bib data for research and other
> applications... This might be a good rhetorical move -- establishing
> the definition of OBD would be very useful (and likely not to generate
> much resistance) for those who already the see the value of open
> publication, and any criticism directed at whether or not an open
> model was appropriate to a particular set of pulications and/or types
> of information would then be separated from the definition of what it
> means to be open.
>

I have some sympathy with this - we create a generally acceptable term "Open
Bibliographic Data"  - rather like we created "Open Data" (which is anything
but simple, but seemed so back in 2006). But it shouldn't by itself exclude
anything.

>
> I'm probably missing important parts of past conversations, and I
> apologize if this rehashes. At the same time, my guess is that similar
> questions will  come up when presenting these principles to new
> audiences... Is there a consensus among the main authors about what
> arguments are compelling for these principles, and could this be made
> somewhat more explicit?
>

I hacked a few ideas in a previous post. We could start off like OA by
asserting the universal and self-evidential truth of OB. That's a political
rather than an operational statement. It might be valuable.

Remember also that OB is not just a warm fuzzy approach - it's going to have
serious impact in some sectors. It's effectively saying "some of what you
have been treating as under your political and financial control actually
belongs to the public good". If we define this too precisely we may draw
lines that we would redraw in the future. What we are saying is that
"bibliographic data" should be Open. I believe that most people and
organizations will go along with that just as most people want democracy. We
do believe there are a set of inalieanble rights - e.g. author names,
titles, etc. We also know there are others which will cause dispute. We
thought we would start with those things which it would be very difficult to
challenge.

My inclination is to publish as-is, add pre-ambles and additional
categorization and see how far we get.

P.


> best,
>
> Micah
>
>
>
>
> On Sun, Jan 9, 2011 at 2:59 PM, Peter Murray-Rust <pm286 at cam.ac.uk> wrote:
> > It is not surprising that we have slightly different views at this stage
> > about what bibliographic data are. The same problems will occur in
> > Scientific (Panton) data. But that should not prevent us going ahead with
> > the principles.
> >
> > We are not sure exactly where the line will be drawn but we are
> reasonably
> > clear that there is a concept of Bibliographic Data, just as we agreed
> there
> > was a concept of Scientific Data. Some of that will be problematic but we
> > shouldn't work too hard to try to draw lines. We are clear that there is
> > Core data, secondary data and content. Let's leave it as that at present.
> >
> > It's possible that some of our anticipated problems will not materialise
> and
> > vice versa. Using the general terms will help concentrate discussion and
> > greater adoption of the Principles.
> >
> > Let's go with the wording we have got and - if necessary - refine it
> later
> > in the light of experience.
> >
> > P.
> >
> >
> >
> > --
> > Peter Murray-Rust
> > Reader in Molecular Informatics
> > Unilever Centre, Dep. Of Chemistry
> > University of Cambridge
> > CB2 1EW, UK
> > +44-1223-763069
> >
> > _______________________________________________
> > open-bibliography mailing list
> > open-bibliography at lists.okfn.org
> > http://lists.okfn.org/mailman/listinfo/open-bibliography
> >
> >
>
> _______________________________________________
> open-bibliography mailing list
> open-bibliography at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/open-bibliography
>



-- 
Peter Murray-Rust
Reader in Molecular Informatics
Unilever Centre, Dep. Of Chemistry
University of Cambridge
CB2 1EW, UK
+44-1223-763069
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-bibliography/attachments/20110110/9574831c/attachment-0001.html>