Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Questions while reviewing candidate standard #159

Open
PeterParslow opened this issue Jan 6, 2025 · 5 comments
Open

Questions while reviewing candidate standard #159

PeterParslow opened this issue Jan 6, 2025 · 5 comments

Comments

@PeterParslow
Copy link

Apologies if this is the wrong place, but it looked a good place to ask a few questions about the POI candidate standard.

  1. data type for contactInfo:
    why does this use CI_Responsibility rather than CI_Contact? I can see confusion arising because "contactInfo" is a property within CI_Responsibility, and there the type is CI_Contact. CI_Responsibility (in ISO 19115-1:2014) is a collection of parties performing a particular role (potentially within a specific extent). If that complexity is needed, then I suggest renaming the POI attribute to something like "responsibility".

I'm aware of some call for a CI_Responsibility to have more than one role.

If only "simple" contactInfo is required, then switch to CI_Contact.

  1. rights / constraints

One of the inputs to the revision of ISO 19115-1:2014 that might start soon is that the current MD_Constraints class is a poor implementation of 'rights'. Creative Commons and DCAT both do a bit better: rights are about permissions & obligations as well as constraints. See discussion at ISO-TC211/StandardsTracker#516, but meanwhile, POI could consider whether they actually want to model "rights" (to use the data?) or constraints. Requirement 18 gives me no clue what you expect this attribute to be used for (or intriguingly why you allow 0, 1, or 2 values)

  1. ISO 19109

Be aware that a new edition of ISO 19109 is likely during 2025 or 2026. ISO/TC 211 are resolving the comments which were made on its public review (June/July 2024). The General Feature Model isn't likely to change, but the expectation that a temporal attribute will use ISO 19108's TM_Object will go. This may not actually have any impact on the POI standard.

Overall, there are a few 'really useful classes' in here that I think could be of much wider use than POI. For example, the subset of ISO 19107 geometry you define captures what I've always wanted from (very) "simple features" geometry!

But also overall, there is little in the document that gives me clues as to how each of the attributes is expected to be used. Will it follow CityGML and have a separate User Guide to explain how to use the model/standard?

@geofizzydrink
Copy link
Contributor

geofizzydrink commented Jan 16, 2025

Response to question 1: data type for contactInfo:

CI_Responsibility was selected because it has a more complete data model behind it, allowing the publisher to expose multiple roles or types of contact.

@geofizzydrink
Copy link
Contributor

geofizzydrink commented Jan 16, 2025

Response to question 2: rights / constraints

1 - A POI does not model constraints. It imports the model from ISO 19115-1.
2 - If and when ISO 191115-1 is update it will be a trivial process to update the POI Standard by referencing the new version when it is published and implemented.
3 - We will confirm whether the 0, 1 or 2 cardinality is as defined in ISO 19115-1

@geofizzydrink
Copy link
Contributor

geofizzydrink commented Jan 16, 2025

Response to question 3: ISO 19109

We must use the standards that are published as of today.

If and when ISO 19109 is updated, it will be a trivial process to update the POI data model to reference the new version when it is published and implemented.

@geofizzydrink
Copy link
Contributor

geofizzydrink commented Jan 16, 2025

Response to question 4: Users Guide

Yes. There is a User Guide under development that will be submitted for publication with the POI Standard.

It has become a common practice for OGC Standards to separate Normative content in the Standard from informative content in an associated User Guide.

@PeterParslow
Copy link
Author

Concerning my question 2, it wasn't so much that there will one day be a new edition of ISO 19115-1.
My main point was that there are better ways to model "rights" than are available in ISO 19115-1:2014

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants