| View previous topic :: View next topic |
| Author |
Message |
Site Admin
Site Admin
Joined: 24 Jun 2004
Posts: 368
Location: Off at yet another ITS standards meeting
|
 Posted: Wed Oct 11, 2006 11:40 am Post subject: new XML schema repository is now active |
 |
|
The new XML schema repository is now active for people to use at:
http://www.itsware.net/ITSschemas/ReadMe.htm
I am now loading up some of the recent XML files to fill this out a bit.
So if you were looking (for example) for IM release 36, you would follow the defined naming to a place like:
http://www.itsware.net/ITSschemas/im/IM-03-00-36/
where you would find the directory.
If any deployment wants to start uploading local content, please drop me a line.
We will leave the older files in place indefinitely, but will start to link people the new pages now.
DCK for Admin |
|
| Back to top |
|
 |
|
Overwhelmed by all this detailed techie stuff? |
FHWA sponsors many
useful courses to get users up to speed,
typically taught by noted experts familiar with the many areas of
ITS
|
|
jjelliso
Joined: 10 Aug 2005
Posts: 45
Location: College Park, MD
|
| Posted: Mon Nov 13, 2006 12:47 pm Post subject: |
|
|
I am glad to see there is some movement on this badly needed resource for the ITS community. That said it needs more work or explanation. I downloaded the IEEE 1512 rev 36 schema that is linked to above. The first thing I noticed is that the schema files do not even pass basic validity testing for an xml schema. I am not sure if | Quote: | | IS NOT SUITABLE FOR USE AS SHIPPED |
from the Readme is descriptive enough to cover that. Just glancing over the schemas I see that most of the documents imported into the main IM schema, ATIS, LRMS,TMDD..., include duplicates of the global elements. Could this be a bug in the mini-edit export? Also the import statements that connect the documents are not always valid. The LRMS schema seems to be missing elements even though it includes my favorite definition of AreaLocation:
| Code: | <!-- Descriptive Name: LRMS_Area_Location -->
<xs:complexType name="AreaLocation" >
<xs:annotation>
<xs:documentation>
the ModelProfile has been removed
</xs:documentation>
</xs:annotation>
<xs:sequence>
<xs:element name="things" type="GeometryProfile" />
<xs:element name="which" type="GeographicCoordinate" />
<xs:element name="i" type="GridProfile" />
<xs:element name="do" type="LinearReferenceProfile" />
<xs:element name="not" type="ISP-Vehicle" />
<!-- was: CrossStreetsProfile -->
<xs:element name="wishTo" type="AddressProfile" />
<xs:element name="typeOut" type="PublicGridProfile" />
</xs:sequence>
</xs:complexType> |
LRMS is not unique in missing types. Pretty much every schema is missing elements in its own namespace. RSPA tries to embed another schema inside the main schema which certainly isn't allowed. It seems to me that by the time everyone fixes these errors each deployment will end up with a different schema and then what is the point of having a standard in the first place? |
|
| Back to top |
|
|
DC Kelley
Site Admin
Joined: 24 Jun 2004
Posts: 1068
Location: Lost in LA
|
| Posted: Mon Nov 13, 2006 4:12 pm Post subject: Its a start... |
|
|
All I can say in my own defense is that it is better (to me) to set this resource up where everyone can see the good and bad things, flaws and all, and then we can start correcting them. I find it annoying to fix the same flaw more than once, but with this resource I can be better able to grab the most current schema building blocks and go.
That said, I purposely started populating the sets with the best available for each standard. And that means some pretty large holes in places. But if I waited until we had perfect releases, there would be no repository. Also, frankly the FHWA and other do no place as high a value on inter-operation as I might like some days and making these things fit - while vital to any real deployment use - is still a volunteer effort of others. Some of the older ones are a snapshot in time, and method have evolved as well.
If you want a few schema sets where everything really does connect and 100% work, try the current ITIS (which all other should n turn then use in its entirety) or the AzTech ones (a subset of TMDD). I hope to be able to add ATIS to this next, it is what I am working on this week. |
|
| Back to top |
|
|
jjelliso
Joined: 10 Aug 2005
Posts: 45
Location: College Park, MD
|
| Posted: Mon Nov 13, 2006 6:17 pm Post subject: |
|
|
I am not sure what is happening, but your posting - which I just replied to (see next thread), seems to have gotten stepped on by my post.
Will see it I can restore it, but has has never happened before.
DCK |
|
| Back to top |
|
|
DC Kelley
Site Admin
Joined: 24 Jun 2004
Posts: 1068
Location: Lost in LA
|
| Posted: Mon Nov 13, 2006 9:23 pm Post subject: |
|
|
| Quote: | | I did not mean to attack your work. |
I did not take it that way, relax. I just see so much more that we could do and that frustrates me too.
Regarding what does compliance mean, there is a meeting this week in Washington DC to try and sort that one out (what the 1201 rules mean), but until they do it is a rather wide open concept. If they even decided to really do it, then I could send a ton of time building standard compliant test suites for such a things and love it. Right now, with no clear market, firms won't even invest in such a thing.
I tend to think you really can hold people to the standards if you wanted to, in that the stds are better than anythings a local deployment would come up with on its own. Another, pessimistic way to say that, is if the effort to "fix" the standard for you own needs is less then the effor to invent your own - use the stds. We passed that point a couple of years ago. Now we are at the stage where you need an expert to "profile" a unique set of the relevant standards for your local use. People like myself can do that in a week or 2 for most deployments, (but if you are smart get the Feds to pay for it). That's about where we are now. But my own hope is that soon the results and success stories from these profiled uses of standards will lead to other saying things like "we just used the schema that XYZ developed, got it off the forum" and that will further bring the cost of deployment down. I'm ranting... time to go find a nice British beer.
| Quote: | | I think I have a good copy of the LRMS standard schema that I could send you. Would that help? |
Yes it would. I also need to get the one that Richards G worked on as well and put it up too. I am not sure we in fact have an "authoritative" and adopted LRMS from SAE at this point that I could put my hands on. Please PM it to me. |
|
| Back to top |
|
|
|
FAQ
Calender
Search
Log in to check your private messages
Register
Log in
Calendar 
Queries: 19 
Generation Time: 1.10713 Seconds