[auscope-geosciml] Communication between GeoSciML and SafeSoftware [SEC=UNCLASSIFIED]

Boisvert, Eric Eric.Boisvert at RNCan-NRCan.gc.ca
Fri Jun 17 08:40:58 EDT 2011


> raising the vexing question that if the desired information could be transferred using a simple feature schema, what was gained by transforming it into a complex schema? I feel like a heretic!
 
simple feature is good if you have a "download all that is there" (or the BBOX case) and don't have a case when you need to filter stuff (because property values are collapsed).
We deal with a case that is does not work for us - borehole.  Several of the services we connect to have a denormalised view of the log (well parameters and lithology as a single "feature")  - it's actually something we asked for from them because complex feature WFS was not available / acceptable.  Any service that have 2 features (or more)  that are in a "composition relationship" (feature A in made of a bunch of Bs) merged into a single "feature", except the rare case where the relationship between 1..1, will show the same problem :
 
Any filter that constrains the 'many' side, will return an incomplete feature.  
 
Our typical problem.  Well -> Log.  If I ask, give me all well that have a till interval will only return the till interval (and not the whole well), so I can't reconstruct the well from the query.  
 
Ok, inner query if you have a SQL backend 
 
SELECT * WHERE A.well id IN (SELECT DISTINCT A.wellid FROM A where a.litho = 'TILL')  
(this is actually the form a SQL that represents how WFS/FES complex query would work)
 
But in WFS, you must make at least two passes.
 
a) give me all the interval that are till
b) extract all the well id
c) get all the well for those id
     - In WFS, I just need to make a massive FID= request right ? wrong ! FID are the id of the interval, not the wells. You either have to create the mother of all <ogc:Or> clause -- no 'IN' clause in Filter , which risk crashing the backend SQL size limit -- when there is a real SQL backend -- or you have to create some specific code to get the dataset in pages by slicing those <OR> statement in reasonnable chunks)
 
downloading the whole database is not an option (we are talking about 1.8 millions features here)
 
Same scenario if you have GeologicUnit - > CompositionPart.  Either you lump composition into a single field, this limitx queryability (and rendering), or you create a "composition" feature (repeating GeologicUnit properties n times) and run into another kind of queryability problem.
 
I hear in the back of the room "But WFS 1.1.0 has JOIN - couldn't you join a Well feature with a litho feature ?"  - no, same problem (and beside, this is queryign a complex model !!)
 
So, I agree, simple feature are useful if you are working with the Pareto principle (I had to write this just for François L.) , which will cover a lots of use cases - and we need to address those use cases, but it won't cover all use cases.
 
Sorry for the Friday rant..
 
Eric

 
 
________________________________

De : auscope-geosciml-bounces at lists.arcs.org.au [mailto:auscope-geosciml-bounces at lists.arcs.org.au] De la part de Stephen Richard
Envoyé : 16 juin 2011 23:47
À : auscope-geosciml at lists.arcs.org.au
Objet : Re: [auscope-geosciml] Communication between GeoSciML and SafeSoftware [SEC=UNCLASSIFIED]



Steve Grise pointed Don at me, and I sent him this:

 

Don-

It's exciting to hear you're interested complex feature XML. One of the big bottlenecks for moving forward with GeoSciML is the absence of any client support. I think it's a bootstrapping problem-no motivation to develop clients until there are services, no motivation to develop services until there are clients. ...  We (AzGS) implemented a demo WFS client for GeoSciML in ArcMap (pretty basic...) -that's the use case that I've been most interested in. The stuff (including example GeoSciML docs) is in a subversion repository: http://code.google.com/p/feature-service-toolbar/ <http://code.google.com/p/feature-service-toolbar/>   you'll want to get the source-that's the only download.

 

Our client uses v2 since that's the current tag release. We haven't pushed it forward since then, but have deployed a gsmltv service and are promoting use of that for the AASG geothermal data service. Ryan set up a little demo web page using the gsmltv services at http://services.usgin.org/ncgmp/demo. Not pretty, but... Uses WFS response to spin up legend for current view frame, click on a poly and get gsmltv content. I have to confess I was profoundly influenced by the way OneGeology Europe seems to have been engineered-most participants implemented a simple feature service that was transformed into complex features by the mediator at BRGM, raising the vexing question that if the desired information could be transferred using a simple feature schema, what was gained by transforming it into a complex schema? I feel like a heretic!

 

We've been pretty focused on simple feature services for the geothermal data system, currently deployed using ArcGIS server, so there's WMS, WFS, KML and ESRI rest services. ArcGIS can connect and use the simple feature services out of the box (if the interoperability extension is installed-requires custom install). I haven't been able to keep up with the conversations on v3; at this point I'm of the opinion that we've got to get a tag posted, so we need to lock with what we've got and focus on demonstrating value in utilization of the complex schema. 

 

steve

 

Stephen M. Richard

Arizona Geological Survey

416 W. Congress St., #100

Tucson, Arizona, 85701   USA

phone: 520 209-4127

AZGS Main: (520) 770-3500.  FAX: (520) 770-3505

email: steve.richard at azgs.az.gov

 

From: auscope-geosciml-bounces at lists.arcs.org.au [mailto:auscope-geosciml-bounces at lists.arcs.org.au] On Behalf Of Oliver.Raymond at ga.gov.au
Sent: Thursday, June 16, 2011 5:52 PM
To: auscope-geosciml at lists.arcs.org.au
Subject: [auscope-geosciml] Communication between GeoSciML and Safe Software [SEC=UNCLASSIFIED]

 

Hi all,

 

I need to bring everyone in on a conversation initiated yesterday by Safe Software (the makers of FME).  They are trying to upgrade their software's capabilities with XML and are looking for examples for importing/exporting GeoSciML.  Rob Woodcock and I will be having an introductory teleconference with the Canadian CEO, Don Murray, and one of their Australia retailers next Wednesday, and then Don wants to talk face-to-face at Edinburgh.

 

We probably need to get a couple of the relevant IWG people together with Don at Edinburgh to coordinate collaboration  - John, Bruce, etc?  Let me know who the appropriate people are and I'll do some scheduling for a meeting in Edinburgh.

 

Cheers,

Ollie

 

_______________________________________________________________________

 

Ollie Raymond

 

Project Leader

National Geological Maps and Data Standards Project <http://www.ga.gov.au/minerals/projects/current-projects/geological-maps-standards.html> 

Geoscience Australia

 

Interoperability Working Group <https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/InteroperabilityWG> 

IUGS Commission for the Management and Application of Geoscience Information

 

Address: GPO Box 378, Canberra, ACT, 2601, Australia | ABN: 80 091 799 039

Ph: +61 2 62499575  |  Fax: +61 2 62479992  |  Email: oliver.raymond at ga.gov.au <mailto:oliver.raymond at ga.gov.au>   |  Google Map <http://maps.google.com/maps?f=q&hl=en&geocode=&q=canberra+australia&ie=UTF8&ll=-35.344028,149.158362&spn=0.007684,0.016404&t=h&z=17&iwloc=addr&om=1>  

_______________________________________________________________________

 

--- This message was created with 100% recycled electrons ---

 

________________________________

From: Raymond Oliver 
Sent: Friday, 17 June 2011 10:42 AM
To: 'Don Murray'
Cc: Brett Madsen @ Lagen Spatial; Kensington) Woodcock Robert (E&M; Woodcock Robert; Preeti Sharma
Subject: RE: GeoSciML [SEC=UNCLASSIFIED]

 

 

Hi Don,

 

I have arranged for the teleconference facilities for us for next Tuesday 10pm Vancouver time.  The number for you to dial in is +61 2 6249 9905

 

Regarding Edinburgh, I'll get a few of my colleagues together and let you know as soon as I can.  What day do you leave Edinburgh?

 

Cheers,

Ollie

 

________________________________

From: Don Murray [mailto:dcm at safe.com] 
Sent: Friday, 17 June 2011 10:10 AM
To: Raymond Oliver
Cc: Brett Madsen @ Lagen Spatial; Kensington) Woodcock Robert (E&M; Woodcock Robert; Preeti Sharma
Subject: Re: GeoSciML [SEC=UNCLASSIFIED]

 

Oliver, Brett,

 

I can make the call next week  (10 PM Tuesday).  Can you provide me the call details?

 

Oliver, I will also be in Edinburgh.  I will be there for INSPIRE.  As luck has it I will be staying a few days longer to vacation with my family.  It would be great to meet you while I am there and discuss GeoSciML as part of our XML technology push.

 

We should set up a time to connect.   I have added Preeti to this who manages my schedule. 

 

Warm regards, 

 

Don

________________________________

 

From: Brett Madsen [mailto:brett at lagenspatial.com.au] 
Sent: Friday, 17 June 2011 9:58 AM
To: Raymond Oliver
Cc: dcm at safe.com; Robert.Woodcock at csiro.au; Woodcock Robert
Subject: Re: GeoSciML [SEC=UNCLASSIFIED]

 

Ollie

Confirming next Wednesday.

We'll wait for a response from Don on his availability (I had sent him an email 30seconds before receiving yours)

 

REgards

BRett

 

________________________________

From: Raymond Oliver 
Sent: Friday, 17 June 2011 9:55 AM
To: 'Don Murray'
Cc: brett at lagenspatial.com.au; 'Robert.Woodcock at csiro.au'; Woodcock Robert
Subject: RE: GeoSciML [SEC=UNCLASSIFIED]

 

Hi Don,

 

I have proposed a meeting with Brett and Rob Woodcock at Geoscience Australia next Wednesday at 3pm.  That's 10pm Tuesday Vancouver time which may be difficult for you to phone conference?  Otherwise I'm pretty free over the next week for a phone call.

 

I'll be in Edinburgh, along with the entire GeoSciML development team between July 4 and July 8 for our annual meeting.  That GeoSciML meeting is immediately after the INSPIRE conference and the OneGeology management meeting also being held in Edinburgh (27 June - 1 July).  While I won't be at those prior meetings, there will be many of the GeoSciML team there, so there will be plenty of GeoSciML people that you could talk to if you were there.

 

Cheers,

Ollie

 

_______________________________________________________________________

 

Ollie Raymond

 

Project Leader

National Geological Maps and Data Standards Project <http://www.ga.gov.au/minerals/projects/current-projects/geological-maps-standards.html> 

Geoscience Australia

 

Interoperability Working Group <https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/InteroperabilityWG> 

IUGS Commission for the Management and Application of Geoscience Information

 

Address: GPO Box 378, Canberra, ACT, 2601, Australia | ABN: 80 091 799 039

Ph: +61 2 62499575  |  Fax: +61 2 62479992  |  Email: oliver.raymond at ga.gov.au <mailto:oliver.raymond at ga.gov.au>   |  Google Map <http://maps.google.com/maps?f=q&hl=en&geocode=&q=canberra+australia&ie=UTF8&ll=-35.344028,149.158362&spn=0.007684,0.016404&t=h&z=17&iwloc=addr&om=1>  

_______________________________________________________________________

 

--- This message was created with 100% recycled electrons ---

 

________________________________

From: Don Murray [mailto:dcm at safe.com] 
Sent: Friday, 17 June 2011 9:34 AM
To: Raymond Oliver
Subject: GeoSciML

 

Oliver,

 

I was passed your contact information by Brett Madsen.   We are working on improving our XML support and have made great strides with it for our upcoming FME 2012 product.

 

I have found some GeoSciML examples on the web and have had success working with them.

 

It would be great to connect with you and discuss various GeoSciML scenarios etc. for both reading and writing of GeoSciML.

 

Is there a time next week that works for you?    Are you attending any conferences in the near future in Europe or North America?  If so perhaps we could meet up in person.

 

Warm Regards,

 

Don

---------------------------------------------------

Don Murray | President

Safe Software Inc.
Suite 2017, 7445 - 132nd Street, Surrey, BC Canada V3W 1J8
T 604.501.9985  | F 604.501.9965

Twitter: donatsafe


don.murray at safe.com | www.safe.com <http://www.safe.com/> 

---------------------------------------------------

 

________________________________

From: Brett Madsen [mailto:brett at lagenspatial.com.au] 
Sent: Thursday, 16 June 2011 3:42 PM
To: Raymond Oliver
Cc: Robert.Woodcock at csiro.au
Subject: Re: RE: GeoSciML SEC=UNCLASSIFIED [SEC=UNCLASSIFIED]

 

Hi
They are mainly looking at format conversion so probably so sample datsets that may have been created, outside of those that are already published.
But allowing also people to write to geoscjml from other formats or the reverse.
I'll be in canberra next tuesday / wednesday by which time I can probably get some more info for you.
Let me know if you may have some time free.

Cheers
Brett
Sent from my Samsung Galaxy S

________________________________


On 16/06/2011 2:58 PM, <Oliver.Raymond at ga.gov.au> wrote:

Hi Brett,

Could you be more specific?  What exactly do you want in a "case scenario"?  What exactly do you want to test?

Cheers,
Ollie

________________________________

From: Brett Madsen [mailto:brett at lagenspatial.com.au] 
Sent: Thursday, 16 June 2011 2:03 PM
To: Wyborn Lesley
Cc: Hudson David; Robert.Woodcock at csiro.au; Raymond Oliver
Subject: Re: GeoSciML SEC=UNCLASSIFIED

Thanks.
Just an fyi.
Safe Software who build FME  are after some use case scenarios for product testing.
If you could help it would be appreciated.

Cheers
Brett
Sent from my Samsung Galaxy S

________________________________

From: Raymond Oliver 
Sent: Thursday, 16 June 2011 9:29 AM
To: 'brett at lagenspatial.com.au'
Cc: Wyborn Lesley; Hudson David
Subject: RE: GeoSciML SEC=UNCLASSIFIED [SEC=UNCLASSIFIED]

 

Hi Brett,

 

I am the lead editor of the GeoSciML Data Model (part of the IUGS-CGI Interoperability Working Group <https://www.seegrid.csiro.au/wiki/CGIModel/InteroperabilityWG> ).  Feel free to ask me any questions.

 

The current GeoSciML xml schemas (current version is v2.1) are at www.geosciml.org <http://www.geosciml.org/> .  We are currently working on version 3.0 which will be a considerably expanded, and in places reworked, data model.  We aim to have version 3.0 released within the next 6 months.

 

Cheers,

Ollie

 

_______________________________________________________________________

 

Ollie Raymond

 

Project Leader

National Geological Maps and Data Standards Project <http://www.ga.gov.au/minerals/projects/current-projects/geological-maps-standards.html> 

Geoscience Australia

 

Interoperability Working Group <https://www.seegrid.csiro.au/wiki/bin/view/CGIModel/InteroperabilityWG> 

IUGS Commission for the Management and Application of Geoscience Information

 

Address: GPO Box 378, Canberra, ACT, 2601, Australia | ABN: 80 091 799 039

Ph: +61 2 62499575  |  Fax: +61 2 62479992  |  Email: oliver.raymond at ga.gov.au <mailto:oliver.raymond at ga.gov.au>   |  Google Map <http://maps.google.com/maps?f=q&hl=en&geocode=&q=canberra+australia&ie=UTF8&ll=-35.344028,149.158362&spn=0.007684,0.016404&t=h&z=17&iwloc=addr&om=1>  

_______________________________________________________________________

 

--- This message was created with 100% recycled electrons ---

________________________________

From: Brett Madsen <brett at lagenspatial.com.au>
To: Hudson David
Sent: Thu Jun 16 09:06:12 2011
Subject: GeoSciML

Dave

Do you know of anyone at GA who has been doing work with GeoSciML?

 

Safe (FME) are looking for some scenario testing 

 

cheers

-- 

Brett Madsen

Chief Evangelist Officer

 

L A G E N S P A T I A L

Create Discover Connect Deliver

Switch: +61 2 9527 9592
Fax: +61 2 9527 9590
Mobile: +61 (0)417 064 677

PO BOX 1474 Cronulla NSW 2230

www.lagenspatial.com.au <http://www.lagenspatial.com.au/> 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengeospatial.org/pipermail/geosciml/attachments/20110617/f84c114f/attachment.htm>


More information about the GeoSciML mailing list