[open-science] github/R stack for the nomadic researcher - permanence?

[Matthew Todd]

All,

This point may have been made already in the cascade of mails (no longer
grouped?) on this subject, but we're dealing with this issue of data for
our open projects right now.

All our primary data on the schisto and drug discovery projects are shared
through Labtrove's electronic lab notebook. While we post new experiments
every day (i.e. the content changes) *we are required by law to keep the
data essentially forever*. This is because the data involve experimental
science, and as part of safety regulations we are required to keep lab
books for 70 years after the completion of the relevant procedure (i.e.
forever). I would urge any open scientist to look into this - their
university/institution/country probably has similar requirements. Ours come
from New South Wales state law. I can forward the relevant clause if anyone
desperately wants to see it.

Now I'm in the middle of talking with Australian data guys (at NeCTAR - the
cloud data service specifically designed for things like this) about
permanent backup of our data. If for some reason NeCTAR goes under, or we
need to pay NeCTAR and cannot, then it would be my responsibility to ensure
my University backs up the data. There aren't clear routes for this, so I
have no clear answers. It's important because if I don't keep the data I
could be liable years down the track.

If I tell my University adminstrators that I am using a commercial service
to back up my data (data which have arisen from government-funded research)
they will tell me that that is not good enough, and will ask me what
happens if the company I am using goes bust in 30 years' time. That the
University has no viable alternative won't be a productive argument - I
will be told to continue using paper lab books until there is a better
alternative. Hence for me it's important that the data I generate are
stored in some "official" system, meaning a composite of national and
library archives.

I flag this up just because I've had to look fairly far into this because
of the nature of the open, paperless data we're generating. I know a
commercial service will not be looked upon kindly. Clearly (I hasten to
add) I don't necessarily agree with this. Clearly commercial services (such
as Mark's wonderful Figshare) are useful for sharing nimbly and
effectively, and other things we're using for our open science projects
(like G+ and Twitter) are not archived but are still great and useful in a
way that an archive is not. But *permanence of the repository use to house
the raw data* (i.e. the electronic lab notebook needed for open science)
and *contingency plans for the repository's failure* need to be in place
for us to have a general solution that could gain traction. My feeling is
that data arising from University-generated research needs to be managed by
the University through support of something like a national service. That's
slowly happening in Australia right now, and I hope elsewhere.

...unless other services are able to guarantee their existence in 70 years'
time, in which case: problem solved.

Cheers,

Mat


-- 
MATTHEW TODD | Senior Lecturer and Honours Coordinator
School of Chemistry | Faculty of Science

THE UNIVERSITY OF SYDNEY
Rm 519, F11 | The University of Sydney | NSW | 2006
T +61 2 9351 2180  | F +61 2 9351 3329  | M +61 415 274104
E matthew.todd at sydney.edu.au  | W
http://sydney.edu.au/science/chemistry/research/todd.html

CRICOS 00026A
This email plus any attachments to it are confidential. Any unauthorised
use is strictly prohibited. If you receive this email in error, please
delete it and any attachments.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-science/attachments/20120412/8b1ba54b/attachment.html>