CDB X Proposed Path Forward

[MichalaH]

All, Thank you for your patience as the facilitator team and SOCOM took time to meet and better document our objectives. Attached here is a slide deck that outlines our proposed path for this tech sprint. It is a "proposed" path because while this is the best path forward in our perspective, we are open to suggestions and modifications.

The list of "desired capabilities" is long, and overall this task is large. We don't expect that all of these capabilities will be met, or that all the profiles will be built. We would like to work on the pieces we can, as much as we can. What work we are conducting now matches to which pieces of the proposed way forward? What can we do in the short amount of time we have? In the Engineering Report, we expect to be explicit about what was done, and what wasn't.

Please also consider how the different pieces of our proposed set of profiles work together or don't, what pieces overlap or don't, parts of the current CDB that remain unchanged, or that need changed, work we are doing now and where it fits, work that needs to be done and the order and dependencies, and terminology. Let us know if this proposal adversely affects your ability to experiment. We know some work will not be affected but some might.

As with all our GitHub discussions, points made here could end up in the ER as lessons learned, future topics for experimentation, design details, etc.

Note that all terminology used here is in flux and may be amended later. We have to put terms on things in order to talk about them but we are open to suggestions.

Thank you all for your help!

CDB-X_v4.pptx

NEW SCHEDULE: • Experimentation and Engineering Report Phases will now run concurrently until September 30 • October will be focused on finishing the Engineering Report • The CDB SWG has offered to take on the task of writing the specification for the new version of CDB, which is yet to be named • Please contact Crystal at crystal.l.nedik.ctr@socom.mil with any questions about funding, business reports, or anything related to SOFWERX

[DavidFLOR]

Slide 8, the last bullet is an unfinished bullet fragment: 'Under Logical Rules, the arrow indicates what the new rule.......'

[MichalaH]

Slide 8, the last bullet is an unfinished bullet fragment: 'Under Logical Rules, the arrow indicates what the new rule.......'

Mad Libs time!!

Either remove the word "what" or add "is". Should read "Under Logical Rules, the arrow indicates the new rule"
Those aren't strictly rules perhaps but was unsure what else to call them. Conditions? Things? Items? Was mostly trying to point out the change. :) Thank you David!

[DavidFLOR]

OK, thanks!

[cnreediii]

I have not yet had time to really digest all of the content in the slides. However, my immediate reaction is that there are way too many profiles!!! Very difficult to develop and maintain that many profiles. Perhaps two things: 1.) we will be able to collapse these use cases/profiles into a smaller set and 2.) perhaps we need to remember that a profile is a restricted subset of a standard - not an extension. I will cogitate on the slides this weekend and provide more input early next week.

Have a good weekend everyone!

[MichalaH]

One caveat to this proposal that I may not have been as clear on as I needed to be..... SOCOM does not expect that all the capabilities on the list will be met, or that all the profiles will be created with this effort. We need your feedback on what capabilities are being met with current work, what might be easier or harder to meet, what items may have only part of the work completed by the end of this, what parts of the profiles overlap, dependencies, etc. This is an ambitious list. Any progress we make in this effort is positive forward movement towards a modernized CDB. Many of you are already doing excellent work and experimentation, and we need to show where and how that fits into the overall picture we present here (it could go in several places, potentially). In the ER, we would like to paint a picture about what we were able to do and what yet needs done, why we chose certain prototypes, formats, methods over others so that anyone moving forward know where we've been.

Thank you all for your efforts!

[cnreediii]

Thanks, Michala!

[cnreediii]

All -

Another point. Let's not use the term "projection". The OGC, ISO, IETF, OASIS and other have adopted the common usage of the terms "coordinate system" and "coordinate reference system". A projected systems assumes projection of coordinates to a 2D plane. Such projections always cause distortion/errors. So, there is no such thing as an "undistorted projection". If we wish to work in a system that minimizes any distortion, then we need to work in an earth based spherical coordinate system.

Geographic coordinate systems are based on a spheroid and utilize angular units (degrees). Projected coordinate systems are based on a plane (the spheroid projected onto a 2D surface) and utilize linear units (feet, meters, etc.).

More detail can be found in the ISO/OGC Abstract Spec Referencing By coordinates: http://docs.opengeospatial.org/as/18-005r4/18-005r4.html#33

[cnreediii]

I forgot to add - Please let's not even think of Web Mercator as a candidate CRS :-)

[cnreediii]

I have been editing the Engineering Report (ER) based on some of the content provided in Michala's slides. The more recent changes are the Terms & Definitions section and the Models section (section 6). In section 6, I added a section on what the heck is foundation data and a proposal for what could be foundation data in CDB-X based on CDB 1.2. I also provided a draft definition for foundation data.

I have also added a use case sub-section based on the content of slide 6. In order to complete the ER, we actually need a few sentences describing each use case. I also reorganized the use cases (profiles?) a bit as some are actually stated as requirements, one is a general note, and one is probably long term future work. Obviously this is all drafty.

Finally, I took a look at the draft "conceptual model" this is actually a composite of conceptual, logical, and physical models. We will need to separate the different models and flush them out. We need to be careful to not specify implementation or platform specific content in the conceptual/logical model. This is necessary to meet the requirement/use case for an abstract configuration model.

[cnreediii]

Finally made it to the last slide. Lots of good content to chew on!

On the last slide, there is a statement about agreeing on one metadata standard and one attributes standard. From a conceptual and logical model perspective, this would be a mistake. Even in a core CDB-X standard, there is no reason to mandate a specific metadata and/or attribute standard. In the current CDB standard, anyone of 7 (if I remember correctly) metadata standards may be specified. CDB 1.1 and after recommends a set of metadata elements that are supported in all of the 7 allowed metadata standards. In CDB X, we should mandate a minimal set of metadata elements and then also specify a set of recommended elements. This is what a number of existing OGC standards do. Also, a number of OGC interoperability initiatives have taken the same approach, although the list of elements is much longer/deeper than in the current CDB standard. If you want to make clear requirements in the core CDB-X standard, then mandate how time is expressed and how character sets are expressed.

The same arguments can be made in relation to mandating only one attribute standard. "One ring does not rule them all" :-)

I also need to make a note regarding how there appears to be confusion/conflation of ideas related to structure and data. More on that later.

[MichalaH]

Here is an updated slide set that corrects what were slides 5 and 6, with listed desired capabilities and proposed profiles. A few people contributed to this slide set so things got mixed up there and I apologize. CDB-X_v5.pptx

[MichalaH]

"......my immediate reaction is that there are way too many profiles!!! "

There were a lot of things listed under the profiles slide that should not have been there. I fixed and attached updated slide set as CDB-X_v5 above.

[MichalaH]

Project complete, closing admin items.