[Requests] OGC 15-095 comments

Pat Cappelaere pat at cappelaere.com
Wed Jan 20 17:40:00 EST 2016


PART A


1. Evaluator:
        Pat Cappelaere
		Vightel Corporation
		8303 Winter Wind Ct, Ellicott City MD 21043 USA
		pat at vightel.com
		pat at cappelaere.com

2. Submission: [OpenGIS Project Document Number, Name]
	OGC 15-095 Emergency and Disaster Management Information Framework



PART B


1. Requirement: [General, #] 


2. Implementation Specification Section number: [General, #]


3. Criticality: [Major, Minor, Editorial, etc.]
	Major

4. Comments/justifications for changes: [Comments]
	
	1. Scenarios and Use Case should come first and should be expanded in a consistent manner.  Have you considered using a User Story template and User Scenario?
https://www.newfangled.com/how-to-tell-the-users-story/ <https://www.newfangled.com/how-to-tell-the-users-story/>
http://dannorth.net/whats-in-a-story/ <http://dannorth.net/whats-in-a-story/>

	2. It is fairly obvious that all of these "scenarios/use cases" call for actionable information (aka products).  Are you considering the standardization of these products by name and output so we can all agree to generate similar products on-demand when requested.  If yes, would this make sense to have that in the document?  This seems fairly critical.
	
	3. End-users need those products during an emergency.  They do not know who can provide them or what services to use to generate them.  Since we seem to have a service-oriented architecture, we have a problem.  How does the proposed framework address this issue?
	Note: Catalog services may be good for service discovery but not so much for product discovery and generation.  Catalog services are extremely hard to curate and suffer from extremely quick obsolesence.  How do you propose to address issue within a distributed framework of catalog services?
	
	4. Proposed dataset formats include NetCDF, HDF, GML... These formats are extremely heavy-duty and require major expertise/tools to use properly.  They are very appropriate for scientists and software engineers.  But do you really expect first responders to use them during an emergency?  They are not scientists.  They may not have the expertise/tools nor the bandwidth to download those large datasets.  How would this work for mobile platforms?  Woudln't we rather want to standardize on a compact open format that would be very easy to use?
	
	5. What is the recommendation for minimum metadata for disasters?  If I generate a flood map or a rainfall accumulation over the last 5 days, what would I need to specify as metadata to be useful?
	
	6. What is the security approach for this framework?  How does one sign a message?  How do you propose to authenticate users across an ad-hoc federation during disasters?
	
	7. Push architecture
		Two patterns.  OK.  Any more details?  
		Are you proposing a publish-subscribe architecture?  What are the messages/topics?  Would they be standardized?  What would the proposed underlying implementation be?
		
	8. Alert messages
		What's the proposed approach? CAP over IPAWS? That may be fine if you are FEMA or equivalent.  But we are not in that group.  What do we do?
	it seems to me that Twitter and Facebook are better used by the general population during disasters.  Any intent to leverage these existing capabilities?  How one would publish products on Facebook/Twitter so they can be discovered quickly by local community?
		
	9. INSPIRE and OGC Services
		 How does this paragraph fit with the Emgergency and Disaster Management Information Framework?
	
	10.  What are the gaps that need to be addressed?


Many thanks,

Pat Cappelaere
pat at cappelaere.com

Cell: 410 340 4868
Skype: patrice_cappelaere
gmail: cappelaere at gmail.com





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengeospatial.org/pipermail/requests/attachments/20160120/78e74fe1/attachment.html>


More information about the Requests mailing list