SCR 3-1115 Vote Comments


Steve Hughes

I would vote yes. By making all but the essential keywords optional, we have defined the base_IMAGE_MAP_PROJECTION. Extensions to the base to define specific objects for specific missions is what we have always done in the past. E.g. HEADER <- FITS_HEADER. The resulting label in the archive is essentially the specific object class definition. If the nodes want strict validation then we defined the specific objects for the DD as desired. The PDS Data Model, aka the IPDA data model, explicitly defines the base classes and their subclasses this way.


Susie Slavney

Geo votes a) yes, move to implementation, with the understanding that Chris and Elizabeth will eventually make a table listing the required keywords for various projections, and that the whole map projection object will eventually be redesigned with PDS-4.


Todd King

I have to admit that I have little experience with the IMAGE_MAP_PROEJCTION object. So, I feel I should explain my vote just in case I’ve missed something in my analysis. When I look at the standards document there are examples (A.25.5, page A-103 and A.25.8, page A-107) where the keywords FIRST_STANDARD_PARALLEL and SECOND_STANDARD_PARALLEL are not included, yet example B.14.6, page B-34 does include all required keywords. In example B.14.6 the values for the keywords at issue are specified as “N/A”. When I look at the three examples, none look more cluttered or misleading than any of the others. The correct example (B-34) is simply more explicit. Since the SCR states “clutter” as the reason for the change I would vote “no”. However, if the value for these keywords are “N/A” then it appears that these keywords are not necessary to fully describe a basic IMAGE_MAP_PROJECTION. While the SCR did not describe this reason I think its fair to conclude that the keywords could be made optional without an adverse affect. So I’m voting “yes”.


Dick Simpson

The arguments in favor are not convincing. We have been drifting away from requirements that keep text files (including labels) easily readable for the average user. “Confusion” and “clutter” will not be issues for machine processing of the data.

If such keywords are required for any image map projection, the system will be simpler and easier to maintain if we go with the “N/A” option. Rather than accept a new set of required keywords as being officially optional, we should be looking more carefully (and quickly) at the ones that have incorrectly been designated optional and get them into the required category. What happens for those projections which actually DO require these keywords? PDS3 will consider them optional and not flag the error. This SCR is actually a step backward in terms of system function and integrity.

I also don’t accept the urgency argument. If HiRISE has been constructing PDS compliant labels, approval of the SCR means they can omit the keywords at their leisure sometime in the future. Denial of the SCR should change nothing. If they have not been constructing PDS compliant labels, then the products are not ready for delivery. If there’s any urgency here, it’s in getting the existing object/keyword combinations corrected.