[open-science] the early-career guide to doing open science?

Fri Mar 16 17:15:06 UTC 2012

On Fri, Mar 16, 2012 at 1:01 PM, Puneet Kishor <punk.kish at gmail.com> wrote:

>
> On Mar 16, 2012, at 11:44 AM, Carl Boettiger wrote:
>
> > A related question would be: why do many people use github when they
> could
> > just host there own git?
>
>
> Well, that is a bit of an apples-oranges comparison. Github is for code
> that people (mostly) are interested in forking, improving, modifying... it
> is constantly in motion. An archive is, well, an archive.
>

True.  Maybe using arxiv instead of putting on your personal website would
be a better comparison -- there's value in a central archive.

>
>
> Also, not picking on figshare per se, but I'd rather not put my research
> data on the servers of an enterprise backed by a private company with only
> a couple of years of history, and governed by the laws of a different
> country.

True again, but there's no reason for the data to be in just one place, if
you're trying to share.  I wouldn't rely on github alone for the future of
my code, but it's easy to maintain copies there, on my servers, my laptop,
and still use figshare to facilitate getting the data to others.  This goes
back to Peter's comments -- no doubt we need a variety of solutions for
these different challenges.

> Besides, before dumping my data in any repo, I'd look at its longevity, or
> at least, purported promise of longevity. Statements such as "Company
> reserves the right, at its sole discretion, to ... discontinue the Service
> (including without limitation, the availability of any feature, database,
> or content) at any time by posting a notice on the Site, on or through the
> Service, or by sending an email to the email address associated with your
> account" scare the crap out of me.
>
> So, for mostly unscientific reasons, I believe that I should be personally
> responsible for my own data. But, for practical reasons, I realize that I
> can't do it alone. I generally trust my university (I am fortunate to work
> in a mostly fairly enlightened university when it comes to tech and data
> use and deployment), but there too, I know that my needs are not shared by
> everyone else. Hence, I would tend to seek a small group of like minded
> researchers on my campus and, hopefully with the backing of my institution,
> develop a focused repo. After all, my university has 30+ libraries -- one
> mongo library couldn't serve the needs of humanities and medical sciences
> and law and library sciences and geosciences alike. Well, the same logic
> applies to IRs.
>
>
>
>
>
> --
> Puneet Kishor http://punkish.org
> science http://earth-base.org
> advocacy http://creativecommons.org
>
>
>

-- 
Carl Boettiger
UC Davis
http://www.carlboettiger.info/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-science/attachments/20120316/3782286f/attachment-0001.html>