[Esip-dds] Time to vote! Poll closes Thursday
PUhlir at nas.edu
Thu May 23 02:15:02 EDT 2013
Hi all: if we have enough funds or can get additional speakers we could do two panels. Nevertheless, we should keep in mind that we are not actually doing the study, but setting it up. The vision talk(s) therefore should be focused on why a decadal survey of data management research priorities is important to do and what it can accomplish. Such info would certainly be useful for a proposal and for presenting to the survey committee. The study committee, however, will do its own "vision thing".
I will try to get on the call, at least at the start. I am in Dublin and my hotel connectivity for my skype is spotty (it didn't work at all last night, for instance). I hope to talk with you later today.
From: Parsons, Mark [parsom3 at rpi.edu]
Sent: Wednesday, May 22, 2013 2:02 PM
To: Uhlir, Paul
Cc: Christopher Lenhardt; Joe Hourcle; Anne Wilson; esip-dds at lists.esipfed.org
Subject: Re: [Esip-dds] Time to vote! Poll closes Thursday
I'm late (again) to this conversation but it's quite intriguing. I think Chris is on to something. While I agree we don't just want TED talks and that we do want action, I also think it is critical that the survey explicitly address the trends that Chris outlined. I think there is an argument that the data management and info science community as a whole is not thinking creatively enough about the challenges we face. On the flip side, we are also subject to distraction by the latest shiny trend ("big data" anyone?). A *decadal* survey will need to consider that things are changing both fast (technology) and slow (academia) and will need to take the broad view Chris advocates while still making practical recommendations. I guess I'm saying, I like the two panel approach.
Regarding speakers from outside our discipline, I can see arguments both ways. I will say, however, one of the highlights of my career was giving an invited talk on polar data sharing at a mouse genomics data conference. I and I think the participants gained new insights into core problems by seeing them with new eyes. There was also a bit of the comfort of knowing that others are grappling with the same problems. The conference even resulted in a Nature article.
On May 14, 2013, at 7:51 AM, "Uhlir, Paul" <PUhlir at nas.edu> wrote:
> Chris, I like the idea of two panels, if we can swing it. If not, one visionary would be enough to set the tone, but other speakers need to be more focused on the process, since we are not solving the problems but discussing how to best approach them. The study committee itself will need to take the big picture into account.
> -----Original Message-----
> From: Christopher Lenhardt [mailto:clenhardt at renci.org]
> Sent: Tuesday, May 14, 2013 9:33 AM
> To: Uhlir, Paul; Joe Hourcle; Anne Wilson
> Cc: esip-dds at lists.esipfed.org
> Subject: Re: [Esip-dds] Time to vote! Poll closes Thursday
> Fair enough. I am always a fan of doing over talking about doing. I think what I would ultimately like to suggest is two panels. I don't know if that is do-able for the summer meeting or not.
> One panel, plenary perhaps, which gets people thinking beyond their usual frame of reference. If we don't prod people a bit, in my opinion, we'll get more of the same. That is a study that says we need more metadata standards, data interoperability, data life cycle attention, preservation etc. all in the context of their existing data streams. I would argue our profession and maybe even the funders are not truly prepared for what's coming. One might argue we're in one of Kuhn's scientific revolutions. The DOE ARM archive got a number of new 3D cloud scanning radars out the ARRA funding and they had real challenges figuring out how to deal with the something like 30K increase in data and how to integrate that with the science and that's one example. A genomic biologist might be a stretch, but the power of open source science conducted by non-scientists is not and that's what the presentation I saw was all about. Maybe panel is too strong, maybe a set of speakers on a theme.
> A second panel, also during the summer meeting, that would synthesize the two threads, I.e. high-level and the practical. That panel would have the charge to really lay out how this thing might go forward.
> Anyway, we'll have to continue this conversation over a beer some time!
> On 5/14/13 9:14 AM, "Uhlir, Paul" <PUhlir at nas.edu> wrote:
>> Thank you for this thoughtful addition, Chris. I agree that we don't
>> want to get lost in the weeds, but I also disagree with a "big picture"
>> approach that will obfuscate the practicalities as well. The purpose of
>> this session, as I understand it, is to scope out the study and discuss
>> the pros and cons of different options. I like the questions posed in
>> the abstract posted in the wiki by Anne. In my view TED talks are very
>> fun to listen to but will not move this project very far from a
>> practical standpoint. I would make a very strong pitch to have someone
>> in a senior academic role who is closely familiar with the decadal
>> survey study process. Perhaps we could have one "visionary" speaker,
>> but I would argue against having a genomic biologist on that list.
>> -----Original Message-----
>> From: esip-dds-bounces at lists.esipfed.org
>> [mailto:esip-dds-bounces at lists.esipfed.org] On Behalf Of Christopher
>> Sent: Tuesday, May 14, 2013 9:00 AM
>> To: Joe Hourcle; Anne Wilson
>> Cc: esip-dds at lists.esipfed.org
>> Subject: Re: [Esip-dds] Time to vote! Poll closes Thursday
>> In the spirit of influencing votes, I'd like to reiterate that at this
>> stage I would argue we need to make sure we are thinking at the right
>> I would suggest that for the summer meeting panel we should be trying
>> to focus on creating an outcome that will reveal the scope of the
>> problem, the types of challenges ahead. I think we are still tending to
>> get down into the weeds like talking data preservation, life cycle and
>> things like that. Sure those are important.
>> Instead, what I think we need to get a better picture on in order to
>> frame a data study is to identify what are the big trends likely to
>> affect things like data preservation. In this instance I would point
>> to things like unstructured versus structured data, the increasing
>> ubiquitousness of sensors, citizen science, the use of UAVs and drones
>> to collect data, integration of data across scales from nano to
>> continental and beyond, the failure of information science and
>> informatics writ large to support decision-making, the impact of open
>> science on data, open publishing etc.
>> What are the implications of connected world where our phones,
>> computers, cars, etc. are all communicating and sending data. Maybe a
>> stretch in the earth science context, but not so much if you start
>> thinking about environmental health issues. These are the major trends
>> I see in play right now that will have huge impacts on how science is
>> conducted and by extension on the data challenges in the very near
>> future. For example, in a world of unstructured data, it's not
>> inconceivable that we might not be arguing over metadata standards like
>> ISO 19115, instead we'll be arguing over how to encode metadata in
>> unstructured data; a related but different question.
>> ESIP can help to shine the light on these trends and begin to discuss
>> the implications of these trends for what ESIP members and beyond
>> should be preparing to do in the not too distant future. So I would
>> suggest that the panel we put together at this stage be designed to
>> highlight these kinds of issues.
>> I would suggest speakers, if we can get them, like Michael Tieman,
>> Stephen Friend, John Wilbanks (again), Stan Ahalt. We need a set of
>> data TED talks for the panel, I think.
>> Thus I would see the following scenario, I think I pitched this before.
>> 1. Summer meeting panel to set the high level context for a data study.
>> 2. One suggested outcome from the summer meeting would be to create a
>> framework to have a short workshop in the future, late summer, early
>> fall to actually go into much more depth on what the study might look
>> like and who should be involved; a short white paper perhaps. The
>> workshop could be hosted by the Academy, cosponsored by ESIP?
>> 3. Present workshop results to wider community, e.g. program manager
>> types and broad agency at winter meeting.
>> OK, my .02.
>> W. Christopher Lenhardt
>> Domain Scientist, Environmental Data Sciences and Systems RENCI
>> University of North Carolina at Chapel Hill
>> 100 Europa Drive, Suite 540
>> Chapel Hill, NC 27517-7583
>> Office: 919-445-0480
>> On 5/14/13 7:14 AM, "Joe Hourcle" <oneiros at grace.nascom.nasa.gov> wrote:
>>> On May 13, 2013, at 8:03 PM, Anne Wilson wrote:
>>>> Hi DDSers,
>>>> It's time for us to make our list of who to actually invite to serve
>>>> on the panel. I made a quick search for a tool that allowed voting
>>>> for multiple candidates but couldn't find one. Nor could I see a way
>>>> to make Doodle work for that purpose.
>>>> Besides, 'It doesn't matter who votes, it matters who counts the
>>>> votes.' - Joseph Stalin
>>>> So, please just send me your list of your top 4 candidates to serve
>>>> on the panel, like Paul did below. I will use some secret algorithm,
>>>> er, sensible method to organize the votes into a final list.
>>>> There's still time to lobby for someone in particular - if you feel
>>>> someone is particularly good, let us know. (IMO, Dan Baker would be
>>>> very good!)
>>>> Please send your votes to me by COB Thursday, 5/16.
>>> I'm pretty sure that Google Docs has a way to make a form feed into a
>>> spreadsheet so that you can use it as a survey ... I've just never
>>> done it, and it's probably too late if we only have 3 days of voting.
>>> (and I'm abstaining, as I'm trying to get permission to go to the SPD
>>> meeting, which conflicts w/ the ESIP meeting ... and is more
>>> expensive, too ... flights to Montana might be more expensive than
>>> last year's Alaska trip)
>>> ... but I still reserve the right to try to influence other people's
>>> Esip-dds mailing list
>>> Esip-dds at lists.esipfed.org
>> Esip-dds mailing list
>> Esip-dds at lists.esipfed.org
> Esip-dds mailing list
> Esip-dds at lists.esipfed.org
More information about the Esip-dds