[open-science] Climate Change

Tue Jun 15 19:14:21 UTC 2010

Thanks for this John --
In recent months, I have been speaking (UCB, UCDavis, CalTech) on the theme
"Data as Evidence" -- my intention is to broaden our concerns -- as you do
in your comment...  In using this terminology I am attempting to co opt the
force of argument for "evidence-based" policy...

Respecting terminology, I have been (in the biodiversity community) using
the rubric of "free, open and *effective* " access and use of data. I think
"*effective*" is perhaps better than "*useful*" because it is less
value-laden (?) and implies support for the ability to take and use, to test
and evaluate and validate -- these are actions *assumed to be necessary* as
a part of scientific practice but far too rarely actually practiced.

( Of course *observational* data (as opposed to *experimental *data) pose
problems in that they represent discrete, unique events -- that are by
definition not replicable?  This is where we must rely on both the scope  of
sampling geographically and chronologically ("longitudinally") to evaluate
anomalies.

The import of my focus on "evidence" is that *data are evidential* both in
the *logic of science* and in the *political process of policy formation and
decision making* -- though, I believe that different standards of proof
apply scientifically (where the notion of probability and levels of
confidence apply) and politically -- where I believe something more like the
Anglo-American legal standard of *reasonable doubt* has force (particularly
where acceptance of scientific projections has huge implications for vested
economic power and wealth -- as in the case of climate change).

In my view the special burden that we assume when we move from the *realm of
scientific discourse *to the *realm of political discourse* is created by
problems of scientific literacy in the body politic *and* by the difficulty
many scientists have in formulating their scientific hypotheses in ways that
are readily understandable. (Scientists are, in one sense, rewarded for
their ability to conduct esoteric discourse...)

I believe that the "problem set" we face is something like this:

-- Clear and rigorous expression of the logic of scientific investigation
(including logically supportable alternative hypotheses)
-- Rigorous definition of the evidence that is deemed sufficient to support
one or another hypothesis.
-- Definition of the specific forms of data that provide this  evidence
[At the American Museum of Natural History, a paper was published a few
years ago that in my opinion is a good indicator of how this should be done
SEE: RS Voss and L Emmons,  “Mammalian diversity in Neotropical lowland
rainforests : a preliminary assessment.”  Bulletin of the AMNH ; no. 230
[New York] : American Museum of Natural History, 1996. -- It's freely
available at teh AMNH Library DSpace site  --
http://digitallibrary.amnh.org/dspace/handle/2246/1671 ]
-- Rigorous definition of the methodology -- including apparatus,
calibrations, competence -- necessary to obtain data
-- Rigorous documentation of the actual data collection (I believe that this
can be expressed using work flow applications like Kepler...?)
-- Documentation of the data management procedures -- including capture,
storage, transformations -- this is broadly the issue of lineage and
provenance (many statistical packages including automated "scripting" that
captures this information)
-- publicly intelligible representation of the data (and of all the
preceding documentation)
-- clear argumentation on the basis of this evidence in favor of one or
alternative policies with appropriate risk assessments

Many of these elements are in place (or under development) but they are not
(to my knowledge) coherently expressed as required practice...?

I believe that, at the very least, for certain *canonical *data sets
[Climate Change???] we have an absolute obligation to satisfy these
requirements...

Tom

Tom Moritz
1968 1/2 South Shenandoah Street,
Los Angeles, California 90034-1208  USA
+1 310 963 0199 (cell)
tommoritz (Skype)
http://www.linkedin.com/in/tmoritz

“Πάντα ῥεῖ καὶ οὐδὲν μένει” (Everything flows, nothing stands still.)
--Heraclitus

Please consider the environment before printing this email.

On Tue, Jun 15, 2010 at 10:40 AM, John Wilbanks <
wilbanks at creativecommons.org> wrote:

> We've spent a lot of time on climate change and open science at Creative
> Commons. I have a personal interest, as my father is a climate change
> researcher and was an author on the most recent IPCC report. He and I
> co-wrote a paper on open innovation in sustainable development earlier this
> year which was OA, and the references for that paper are a good start for
> the non-data side of the problem. It's at
> http://www.mdpi.com/2071-1050/2/4/993/
>
> In most cases in climate change science, impacts, and adaptive responses,
> the hurdles for open science are not intellectual property rights but
> scientific practices related to confidentiality and protecting one's own
> data and models - a different challenge. The current evaluation of iPCC
> being done by the Interacademy Council at the request of the UN is beginning
> to take a look at how such conventional scientific practices can become a
> threat to the perceived integrity of science. IP is a footnote in the
> debate, unlike in OA or in free software or in free culture. Our successes
> in these spaces have sadly conditioned us to look at "free" legal tools as
> our hammers, and see the world as a bunch of nails. It's a great irony
> actually.
>
> In the case of climate change mitigation, of course, the open science
> issues are similar to those in other areas of traditional manufactured
> technology - accentuated by the fact that the main drivers of increases in
> global GHG emissions are now in the larger developing countries, while the
> industrialized countries still control a lot of the intellectual property
> for addressing that problem....
>
> In many ways the "open" debate about data fails to capture the reality of
> these issues. Making data open, even fully compliant with the Science
> Commons protocol, is actually far from enough. I hope that we can make these
> debates nuanced enough that we don't push "open" as the end game, because I
> can comply with the protocol, or with Panton, and still have my data be
> worthless from a scientific perspective. An extreme example would be that I
> publish PDFs of my data under PDDL, and claim the mantle of "open". If we as
> a community push "open" as the goal, and not "useful" as the goal, then we
> enable that outcome.
>
> Open climate science, at least as it regards data, is almost never an
> intellectual property problem. It's a culture problem, it's a technology
> problem (formats, ontologies, standards), and it's a language problem. It's
> a political problem, it's an incentive problem. Getting rid of the IP is no
> more than table stakes. And if we don't deal with the inventions - the
> technologies that both create climate problems and that promise to mitigate
> them in adaptation - then we won't be changing the world the way we want.
> That's a big part of why our science work has shifted to focusing
> significantly on patent licensing and materials transfer...
>
> jtw
>
>
> On Jun 15, 2010, at 5:42 AM, Lance McKee wrote:
>
> Peter,
>
> I call your attention to one activity of the Open Geospatial Consortium
> (OGC): the GEOSS Architecture Implementation Pilot 3 (AIP-3) data sharing
> activity:
> http://sites.google.com/a/aip3.ogcnetwork.net/home/home/aip-3-kickoff/data-sharing-guidelines.
>
> There are many in the OGC (http://www.opengeospatial.org) who share your
> concerns about climate data. OGC runs a consensus process in which
> government and private sector organizations collaborate to develop open
> interfaces and encodings that enable, among other things, sharing of
> geospatial data, including climate data. I think the OGC is likely to play
> an important role in the opening up of climate science.
>
> I invite you to look through a presentation in which I gathered my
> learnings and musings about the importance, feasibility and inevitability of
> persistent and open publishing of scientific geospatial data:
> http://portal.opengeospatial.org/files/?artifact_id=37254 .
>
> Lance McKee
>  Senior Staff Writer
> Open Geospatial Consortium (OGC)
> 508-752-0108
> lmckee at opengeospatial.org
>
> The OGC: International Location Standards
> http://www.opengeospatial.org
>
>
>
>
>
>
> On Jun 15, 2010, at 3:33 AM, Peter Murray-Rust wrote:
>
> I have posted a report on a meeting I went to last night.
> http://wwmm.ch.cam.ac.uk/blogs/murrayrust/?p=2449
> I believe that this is an area in which the OKF's involvement will be
> positive and important.
>
> --
> Peter Murray-Rust
> Reader in Molecular Informatics
> Unilever Centre, Dep. Of Chemistry
> University of Cambridge
> CB2 1EW, UK
> +44-1223-763069
> _______________________________________________
> open-science mailing list
> open-science at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/open-science
>
>
> _______________________________________________
> open-science mailing list
> open-science at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/open-science
>
>
>
> _______________________________________________
> open-science mailing list
> open-science at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/open-science
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-science/attachments/20100615/853dd559/attachment-0001.html>