Date: Fri Sep 12, 2003  2:30:42 PM US/Pacific

To: (John Shalf)

Subject: Re: DiVA Survey (Please return by Sept 10!)


John Shalf writes:


On Friday, September 12, 2003, at 11:57 AM, wrote:

Obviously, this could pose a signficant amount of overhead on the


system, and the effort may not be justified if the DOE doesn't have

great need for dealing with instrument acquired data. I only added the

point as a discussion topic as it is fairly important to us. At the

very least, I would hope to have the flexibility to hack support

for missing data if it was not integral to the core framework.


It should be included.  I regard the framework as an opportunity to

encode best-practices for things we *do* know and that are not well

supported by existing systems.  While I can't see making it a

requirement for *all* components, it could certainly be defined in such

a way that such information could be forwarded through a vis/analysis

pipeline opaquely as "optional information" for algorithms that can

actually understand it.


So from a requirements standpoint, it should be a requirement that any

design does not preclude incorporation and transparent transport of

this kind of information.  From an implementation standpoint, we'll

have to figure out how such information can be propagated by components

that have no interest or understanding of it (otherwise, each

requirement/desire must be understood by *all* components... not a

practical situation).


Agreed on both counts!