Comments received from systematic review of ISO 19101-2:2018

[JanHjelmager]

Below are the comments received during the systematik review of ISO 19101-2:2018 (see also document N5941)

Some adjustments should be made given the evolution of the normative references on which it relies or may now rely:

• New release of ISO 19123 (ISO 19123-1:2023) and new ISO 19123-3:2023- New ISO 19124 series about post-launch calibration and validation of data
• Evolution of imagery encoding standards (GMLJP2 v2.1, GeoTIFF 1.1, High Throughput JPEG and BPJ2K 2.1, …)
• Upcoming Analysis Ready Data standard (ISO 19176 series)
• ISO 19101-2 should be revised togheter with ISO 19101-1. We see no reason to continue with two reference models, one generic and one for imagery and gridded data. They should be merged into one document. This new document should reflect the objectives in the business plan and the integration of ISO/TC 211 standards into mainstream IT standards (digital transformation, e-government) and the technological trends (AI, ++). The trick here is that ISO 19101-2 should be revised, but not as a revised ISO 19101-2 standard.
• This International Standard should be revised in order to check if correction are needed due to changes (if any) in related standards and due to publication of the new standards and other technical documents.

[wilkesg]

The integration of ISO 19101-1 and ISO 19101-2 presents a complex issue, particularly considering the specific needs and recent updates relevant to users of ISO 19101-2. Here is a summary of some considerations of the merger:

Relevance of Recent Revisions: ISO 19101-2 was recently revised in 2018, reflecting current needs and technological advancements in imagery and remote sensing. And within this short period of time new requirements and topics have emerged including CAL/VAL, and ARD. This means it is already tailored to its specific field, but the field is rapidly evolving, potentially making a merger with ISO 19101-1, which is more generic, less beneficial for users focused on imagery and gridded data.

Reference for Active Standards: ISO 19101-2 serves as a reference for other active ISO/TC 211 standards, particularly ISO/TS 19163-1:2016. A merger could complicate these reference relationships.

Size: 19101-2 is big, upwards of 70 pages. Merging the two would likely make one much larger, harder to manage standard document. Part 1 is 48 pages.

Evolving Standards and Technological Trends: The field of imagery and remote sensing is rapidly evolving, as indicated by the new releases like ISO 19123-1:2023, ISO 19123-3:2023, and the ISO 19124 series. These developments might not align neatly with the more general focus of ISO 19101-1.

Complexity in Addressing Specific Needs: The specific requirements of imagery and remote sensing, including aspects like post-launch calibration, validation of data, and evolving encoding standards (e.g., GMLJP2 v2.1, GeoTIFF 1.1), may be diluted or inadequately addressed in a merged standard.

Benefits of Keeping ISO 19101-2 Distinct: Clarity and Focus for Working Group 6 (Imagery): Keeping ISO 19101-2 separate ensures that the unique needs of imagery and remote sensing are directly addressed, benefiting those working specifically in this field.

Expertise and Resource Allocation: A distinct standard makes it easier to identify and manage expertise specifically in the field of imagery and remote sensing. This can aid in more efficient allocation of resources and experts' time. They may not as easily allocate their time to a general standard.

Alignment with Technological Advancements: The specific focus of ISO 19101-2 allows it to more rapidly adapt to technological changes and advancements in the field, such as the integration of AI and new data standards like the upcoming ISO 19176 series.

Other Considerations:

Potential for a Revised ISO 19101-2: There is recognition that ISO 19101-2 should be revised from working group 6, but not necessarily merged with ISO 19101-1. This suggests a path forward where ISO 19101-2 is updated independently to reflect new standards and technological trends.

Practicality and Efficiency: The decision to merge should also consider the practicality and efficiency of implementation. It could make a much bigger document, counter to the trend of parts in TC 211. A single, comprehensive standard might be more streamlined but could also become overly complex and less user-friendly for specialists in certain areas.

[PeterParslow]

Perhaps it would be helpful to have a common introduction acknowledging that there are two "top level models" of how to look at (classify) the world: "vector" & "raster" (for want of better labels) - although perhaps there are more? "Split" by whether the phenomenon does actually divide the world into discrete objects or is necessarily modelled as a coverage. I assume that is independent of whether we discuss things in terms of RM-ODP or some other top level (IT) architecture.

After that common introduction, perhaps two parts remains appropriate so that they can be maintained by experts in those two "ways of looking at the world".

My intention when I advocate "bringing the two together" is that people approaching geo from outside are made aware that there are these two distinct approaches to modelling the world in data.

I understand that some aspects, like ISO 19156 OMS, cross over the two. And that some see point clouds as distinct from other coverages. And so on. I'm not tied to "just two" but believe we need to make sure "mainstream IT" people understand that not everything fits the same model.