Tuesday 1315

Report
How will your GIS data
be uploaded to the
NG9-1-1 System?
Jason Horning & Brian Rosen
NENA: The 9-1-1 Association
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
What a SIF is… Currently
A
standardized interface used between
GIS databases and NG9-1-1 databases.
 May
be a component of a larger GIS
product offering; may also stand on its
own, integrating into established GIS
systems.
 Simple,
N E N A
nothing fancy here.
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
What a SIF isn’t… Currently
A
database
A
QA/QC tool
 An
N E N A
aggregation tool
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
SIF Design
 Modeled
after the “Replication Feed”
identified within OGC 10-069r2:
GeoRSS
+ Atom + WFS/GML
 Replication
Feed is a log of changes
that have been applied to a data
provider’s data store.
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
SIF Design (continued)
 NG9-1-1
Functional Elements (ECRF,
LVF, etc.) subscribe to the SIF’s feed.
 Insert,
Update, Delete changes are
broadcast by the SIF and
incorporated into listening FEs.
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
Sample SIF Process
As an Add-on to Existing GIS System
LVF
GIS
ESInet
SIF
ECRF
Map
Animated Slide
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
Sample SIF Process
As a New GIS System
LVF
GIS
ESInet
GIS
SIF
ECRF
Map
Animated Slide
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
XML Schema and Sample Feed
 XML
Schema for SIF Provisioning is
available for review and comment.
 Sample
N E N A
Feed Example
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
The GIS to NG9-1-1 FE Process
SIF
CountyBoundary
Counties or
Equivalents
SourceofData
DateUpdated
EffectiveDate
Authoritative
UniqueID
Country
State
County
GML
SourceOfDate
DateUpdate
StateUnqID
Country
State
County
GML
Data Flow
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
Current SIF Conflicts
 I3
Architecture scope is narrower than the
scope defined within other NENA standards
 Example:
“SIF contains, maintains, and supplies
the authoritative copy of the data….” NENASTA-005 Provisioning and Maintenance of GIS
data to ECRF/LVF (DRAFT)
 i3
Architecture Scope is narrower than
Vendor’s products
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
Discussion
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
SIF Thoughts
Brian Rosen
Neustar
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
SIF from the beginning

There are no standardized interfaces from a GIS
system. Closest you get is download of a shape
file, which is proprietary and batch

We needed something standardized, and nearreal-time

In a discussion with Carl Reed at OGC, we learned
about the mechanism we currently specify

It seemed like a good match to get data out of a
GIS into an ECRF/LVF
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
The SIF is

A standardized provisioning interface for ECRF, LVF,
GCS, MCS and Map Database

Designed to maintain, in near-real-time, a set of layers
where the server is a GIS system, and a NGCS element
is the client.

Designed so an update in GIS is reflected in the ECRF
in seconds

On-line, near-real-time

It has ONE standardized interface.
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
As Originally Envisioned
 SIF
was to be built in to a GIS, integrated
into the GIS APIs to detect changes in the
layers and reflect them in the interface.
 ECRFs/LVFs,
etc. would coalesce data
when multiple sources fed one element
and detect problems between contributors
(gap/overlap)
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
And So (from here forward is my
opinion)
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
QA/QC

Someone please tell me why edits to GIS systems
should not follow INTERNAL QA/QC processes to
ensure all systems fed from it get good data?

Why are EXTERNAL QA/QC processes necessary
or desirable?

Who thinks ECRFs, or any other NGCS element
should blindly assume its provisioning is good
and gap/overlap is someone else’s problem?
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
Coalescing data is necessary
 And
 But
doing it repetitively is unnecessary
it’s not hard (and if you buy my
gap/overlap position, it’s almost free)
 But
it least makes more sense to me to at
least allow it to be done in some box
between the GIS systems and the NGCS
elements then not doing gap/overlap in
those elements
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
If we do invent new FEs


What is the interface between the GIS and the new FE?

Back to square one on standardized interfaces to GIS

Doesn’t make a lot of sense to me to make that proprietary
If gap/overlap is not retained as must in ECRF et. al., then we need a whole new
error mechanism to report a problem with the data

And everything has to be near-real-time, flow through provisioning. Think
seconds from changing a service boundary polygon to getting route changes in
the ECRF

N E N A
Not clear every layer has to be that fast, but if one does, how does that fact help anything?
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a
Which leads me to think
 Retain
SIF as the layer replication
interface from GIS to everything else
 Allow
arbitrary function SIF-in, SIF-out
boxes that do things like coalesce
 Still
think NGCF boxes should do
gap/overlap and coalescing, but it
could be fairly primitive
N E N A
D e v e l o p m e n t
C o n f e r e n c e
|
O c t o b e r
2 0 1 4
|
O r l a n d o ,
F l o r i d a

similar documents