[open-science] Thoughts on Having a Meta Open Science Community

Thu Oct 2 15:02:32 UTC 2014

What an important post!
I would like to toss out a few thoughts, which, it says here in the fine
print, offer ideas for a way forward.

Item 1 is a complex issue which some of us call the "subject identity"
problem.  The term "firehose aggregator" is used, and that's the term to
which I would like to respond. There is a new term (well, it's not that
new) "knowledge federation" (see http://www.knowledgefederation.org/ ) in
which the very processes of aggregation -- bringing together -- are
combined with other processes which take subject identity seriously and
organize that which is brought together as a topic map: the promise is that
there will be one and only one location in that map (as with any good geo
map) for each and every individual subject teased out of that which is
aggregated.  It's a profoundly complex job, but we need to start doing
that.  In fact, Google figured that out and bought freebase.com just to get
their hands (so to speak) on a kind of topic map.

A topic map, like a road map, wires topics together with whatever relations
it can tease out of the goo.

It is true that there will always be ambiguities. My favorite is this query
"jack park".  It gets zillions of hits, easily half of which are men of
Korean persuasion, unlike those of us who have Scottish persuasion in us.
So, if I just query "jack park topic map", I own google. It's amazing what
just the right set of keywords will bring up. But, topic maps, well,
knowledge federation platforms will make all that easier (more fine print,
I suppose).

Here's an interesting aside in relation to those of us with Scottish blood
rather than Korean blood: We northern Europeans must pay attention to
something called HH (hereditary hemochromatosis); there is no (known to me)
reason to believe that Asians must do so. That is because the biography of
HH reads more like it got its start during the well-known potato famine;
while it helped a boatload of people survive long enough to breed, there is
no free lunch. A good topic map would allow for the fabrication and
discovery of such wormholes as would make those connections.

Aggregation is a start, but, in some sense, it doesn't go far enough.

Item 2 speaks to using SN (which I disambiguate to Social Network) tools
for doing science. The pathological extreme which strikes me is trying to
do science on FB.  This is a huge issue. On other threads in this forum,
there are conversations about that very topic, and, in fact, there are
several SN platforms emerging strictly for scientists.

But, the key point (for me) in item 2 was the question "how do we get to
hear from those who are silent?".  That's important.  It speaks to
crowd-sourced research, which, in some sense, is what should go on in
science all along; we are all doing science together, even if sequestered
in quiet, private offices. So, if the SN platform we use affords the right
kinds of behaviors, it should be a slam dunk for those who know how to do
social bookmarking and structured conversations to give voice to that which
they uncover while doing research.  The term for that is "stigmergy", which
is best recalled as what ants do when they find food; they lay down trails
which go from foot to home; more ants build that trail stronger, and when
the food is gone, the trail evaporates. Feedback and decay, the two key
elements of complex adaptive systems.

So, overall, I read this post as a conversation in quest to find or
co-create a next-generation sensemaking portal for scientists.

Just a couple of Euros...
Jack

On Wed, Oct 1, 2014 at 10:26 PM, SVAKSHA <svaksha at gmail.com> wrote:

> On Wed, Oct 1, 2014 at 5:32 PM, Bill Mills <mills.wj at gmail.com> wrote:
> > But a big
> > firehose aggregator might help do an even better job of what you
> describe;
> > DJing open science is a big job in itself, and might be made more
> convenient
> > (==realistic) if the author could periodically survey the landscape from
> > just one planet.
>
> This discussion got me thinking as I've always struggled to manually
> parse data the spider pulls for the the 'exact' information I seek at
> any point in time and because the word 'Science' encompasses many
> sub-fields, with many cross-linkages across multiple sub-domains, it
> can be very interesting, but overwhelming too. Here are two scenarios
> for which I could not find a solution:
>
> 1. For example, if a blogger tags a post with "HMM", she could be
> writing about "Heavy meromyosin", "Hidden Markov models" or "Human
> malignant melanoma" which can have a parent category of Biochemistry,
> Statistics (but some may argue that HMM can be used in bioinformatics
> and other "science" research), and "Cancer Research" (say, genetics,
> epidemiology, cell biology, etc..), respectively.  How would a single
> firehose aggregator with a few "specialist" authors filter this
> mountain of information and would'nt the information flood, despite
> the tags for all things 'science' or 'meta open science', quickly
> overwhelm subscribers?
>
> 2. Secondly, there are some researchers and scientists who refuse to
> blog for whatever reason - sometimes its sheer lack of time, lack of
> interest in blogging, whatever...; and yet they have a body of work to
> back them up.  How do you get to hear their voices as opposed to the
> louder and more vocal folks? Unfortunately, most of the SN tools only
> end up promoting the most narcissistic self-promoting personalities,
> not necessarily 'science'. This is something to think about before we
> obsess over the tools - what is the goal and how can we use the tool
> to achieve it while promoting 'science' per se!
>
> /my zero paise worth,
> SVAKSHA ॥  http://about.me/svaksha  ॥
> _______________________________________________
> open-science mailing list
> open-science at lists.okfn.org
> https://lists.okfn.org/mailman/listinfo/open-science
> Unsubscribe: https://lists.okfn.org/mailman/options/open-science
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-science/attachments/20141002/1a5776ba/attachment.html>