[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

BOUNCE address@hidden: Non-member submission from [Linda Miller <address@hidden>]



From address@hidden Mon Jun  7 07:43:16 2004
Received: from happy.unidata.ucar.edu (happy.unidata.ucar.edu [128.117.140.5])
        by unidata.ucar.edu (UCAR/Unidata) with ESMTP id i57DhGtK029536;
        Mon, 7 Jun 2004 07:43:16 -0600 (MDT)
Organization: UCAR/Unidata
Keywords: 200406071343.i57DhGtK029536
Date: Mon, 07 Jun 2004 07:43:15 -0600
From: Linda Miller <address@hidden>
To: Ben Domenico <address@hidden>,
        THREDDS Development List <address@hidden>,
        address@hidden
Subject: Re: An alternate approach to constructing LEAD
Message-ID: <address@hidden>
In-Reply-To: <83814599.1086519237@[10.0.0.3]>
References:  <83814599.1086519237@[10.0.0.3]>
X-Mailer: Mulberry/2.2.1 (Win32)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

Yeah, Ben!

Though I'm not directly involved in LEAD, I have been trying to follow the 
activities, and it sure seems like there's a lot of planning going on without 
much 
testing, evaluation, etc.  I thought it was just me, because I had missed 
something, 
but according to your note it sounds like it's time to move ahead!

Hopefully the LEAD meeting in Alabama will be a great place to launch these 
ideas! 
I'm sure Beth will be ready to launch her ideas too!!! :-)

Linda

--On Sunday, June 06, 2004 10:53 AM -0600 Ben Domenico <address@hidden> wrote:

> Hi,
>
> In our Unidata LEAD meetings and some email interactions with Sara Graves, I'm
> getting uneasy about all the time being spent by our staff in areas which are 
> not
> the ones in which Unidata can contribute most effectively.
>
> Comments by Tom and Doug bring to mind an analogy of a bunch of folks trying 
> to
> build a house.  In that context, the LEAD discussions started out with very
> abstract, high-level architectural design for a house that would serve any 
> use the
> homeowner might dream up.  And now we're engaged in a lengthy analysis of 
> bricks,
> nails, two-by-fours, copper wire, shingles, and pipe fittings (aka "atomic
> tasks").  These discussions may be extremely important and valuable in the 
> long
> run.  But another view of the project is that we have in the LEAD community an
> existing set of pre-fab rooms that Unidata would like to connect in some 
> fashion
> to see if we can erect a (perhaps temporary) dwelling that's functional and
> livable while the larger LEAD team assembles the ultimate house from first
> principles using atoms as building blocks.
>
> But for starters, it would be good to see if we can do things like connect the
> kitchen to the dining room so we can get food on the table.  Then at least we
> won't starve while others are working on the house that senses when we're 
> hungry,
> mixes perfect martinis, serves a multi-course meal that's both delicious and
> nutritious, then cleans up after us.
>
> So in my view, the key LEAD pre-fab components are:
>
> -- Real time data systems (starting with LDM/IDD/Decoders)
>
> -- Data assimilation (starting with ADAS)
>
> -- Forecast models (WRF)
>
> -- Data mining (starting with ADAM)
>
> -- Data access and visualization tools (IDV)
>
> -- Data and catalog services
>    (e.g., OPeNDAP/ADDE/THREDDS)
>
> -- Data forms and representations
>    (e.g., netCDF, AREA files, ESML, NcML, GML)
>
> While this list is certainly challenging, I think it's feasible to learn 
> about the
> input, output, and configuration requirements of these systems and figure out 
> how
> we can connect them to one another to form an end to end system.
>
> I'd like to work these ideas into something we can present to the larger LEAD 
> team
> as an area where we can begin to contribute if we can identify technical 
> people at
> the other existing tools sites to work with and spend a little less time
> discussing topics that don't get us closer to the objective of getting these 
> tools
> working with one another.
>
> -- Ben
>
>



Linda Miller - address@hidden
External Liaison, Unidata
University Corporation for Atmospheric Research
P.O. Box 3000
Boulder, CO 80307-3000
303-497-8646 fax: 303-497-8690