[Sigsti-l] Ideas for SIG-STI panels for 2013

[Joe Hourcle]

See the earlier message that I sent for what the actual call said.

We had a few ideas (some fully fleshed out, some just nuggets) for
possible panels, and so I'm attempting to figure out if people
think they'd be worthwhile (and if so, possible contribute or
otherwise help out with putting together the proposal).

So, please read the other message first -- as it might give you
an idea for a panel ... and then read this list of some of the
ideas that people have mentioned so far.

If you're interested in helping out with one of these ideas, let
me know, and we can put together groups to work on them.  And
you don't have to be an expert, you just have to have an interest
(and then as a group figure out who the experts are and try to
convince them to come and talk about it).

... and if you have ideas for something that might not fit as a
panel, but could be a workshop / local event / webinar / something
else, send me those, too.  (or CC the list to discuss further)


-Joe


...


1. Data Provenance -- Elliott Hauser has already written up a draft
   on this one

2. Get a bunch of scientists in a similar field, and have them
   compare / contrast their different data / information management
   techniques.  (this was also proposed by Elliott, and might be
   one component of the provenance panel, but I thought I'd mention
   it separately as I think they're both strong enough ideas to
   stand on their own)

3. "Everyone's data is special" -- actually started out as a
   comment I made about scientists building yet another system
   from scratch without caring about the implications on archiving
   or reusing existing systems or components, and someone at the
   planning meeting suggested it'd make a good title.
   (I think the full comment was something like 'they say it's
   because their data is special, but everyone's data is special')

4. problems with 'the cloud' -- because of the 'Beyond the Cloud'
   theme (see earlier e-mail), I mentioned that I'm better at
   complaining about things ... and I hate the pointy-haired-boss
   buzzword that 'the cloud' has become, so I was thinking of
   trying to see if we can come up with a panel on when/where
   'the cloud' is a problem.  eg, if you have info that needs
   to be access controlled (HIPPA, FERPA, ITAR, IRB, etc) you
   have to worry about who has access, what country it's stored
   in, if it's stored for more than 180 days, if you're notified
   of potential security lapses, etc.

.... and whatever anyone else can think of.


For #4, see
	http://www.cisco.com/web/about/doing_business/legal/privacy_compliance/docs/CloudPrimer.pdf
	http://gigaom.com/cloud/fighting-fud-cloud-players-try-to-make-sense-of-european-data-protection-laws/
	
	also, some of the cost models are really horrible for the 'long tail' of information.  (what my boss has called 'write once, read never')