noticeIdentifier structuring concern

[brianaAtWork]

I don't understand the need for creating a complex type noticeIdentifier.

It seems reasonable to get rid of the type noticeIdentifier and instead put the simple types noticeNumber (s/b id or just number), sequenceNumber, noticeYear (s/b just year), and noticeType (maybe msgType to make more generic?) inside of the notice.

Then refersToNotice (or referencesNotice) and cancelsNotice could just point to a simple type of noticeNumber (or just number or maybe just id since it is already inside the "notice" object).

[FihoFi]

The noticeIdentifier identifies a notice according to rules set up in S-4. The unique identification of a single notice is using a combination of values noticeNumber, SequenceNumber and year, according to current numbering schemes used by differnt HO:s . Also ChartProduct has a reference to noticeIdentifier.

[brianaAtWork]

Wouldn't that combination ultimately become the MRN? Perhaps we should name that complex type MRN instead of noticeIdentifier?

And also have ChartProduct reference MRN?

[FihoFi]

Maybe in a next version. We should keep in mind this datamodel is capturing an existing structure of data (Notices to Mariners), which is already well defined within S-4. Main intended usage is human reading within a (pdf) booklet or webservice.

[brianaAtWork]

The noticeIdentifier identifies a notice according to rules set up in S-4. The unique identification of a single notice is using a combination of values noticeNumber, SequenceNumber and year, according to current numbering schemes used by differnt HO:s . Also ChartProduct has a reference to noticeIdentifier.

Thanks, that helps a lot! AND now I have even MORE questions and comments! :-) (sorry...I should have just gone to the workshop!)

Just for record keeping I'm including that I've searched the S-4 (https://www.iho.int/iho_pubs/standard/S-4/S4%20V4-6-0_ENG_April16.pdf) and under section B-630.4 found the "Arrangement on NM": Arrangement of NM. The limits of oceans and seas described in IHO publication S-23 should be used as a basis for the geographical arrangement of NM editions. A geographical index and a numerical index of the charts affected should be given in each edition. The sequence in which the information is given should always be the same (although not all the following items may always be applicable), for example: a) Number of NM (see B-630.5). b) General region (normally one of the following categories): Ocean/Sea or Country name (see B-631.3). c) Sub-region (for example: coast; gulf; island; river). d) Specific location (for example: port name; terminal). e) Subject (for example: lights; depths). f) Lights List numbers (if applicable – see B-631.7). g) Authority (that is: original source upon which the NM is based – see B-631.6). This must always be stated. h) Amplifying remarks, including cancellations of (P) & (T) NMs (if required – see B-631.8). i) Chart(s) affected (see B-631.7). (Publications affected, if applicable). j) To enable the chart updater to ensure no NM has been missed, a reference to the preceding NM number may be required (see B-631.7). k) Horizontal datum (if considered necessary - see B-631.7). l) Date of establishment, alteration, etc (for example the implementation date of a routeing measure). m) Detailed description (using INT1 as a guide – see B631.4). n) Position (see B-631.5).

However, to me this doesn't look exactly like (yes similar but not exact) the data model in the zip file in this repo. And the more I dive into it the more I wonder why this data model doesn't seem to represent it the way I would have expected.

And I do see the unique number description in B-630.5:

Numbering. A standard method of numbering notices should be adopted, the arrangement being

1. a unique and sequential number of NM/
2. Year of publication, for example NM1234/09.
3. (T) or (P) must be added for temporary or preliminary NMs, for example NM1234(P)/09. Additional elements may be added to the NM number, for example:
4. the national chart number;
5. a sequential number of the update for each chart;
6. the periodical number of the NM booklet. For example, the NM number may comprise the NM periodical number and chart (or nautical publication) number which, together with the year of publication, forms a unique, sequential number, > for example 36/413/2009 [periodical/chart/year]. If this system is used, it would be normal to group all updates for one chart under the same unique NM number.

To me this shows the number used is UNIQUE AND SEQUENTIAL as one number yet in the data model it is listed as a mandatory notice number and an optional sequenceNumber.

But then 4-6 above from s-4 show optional values that are not shown in the UML datamodel.

Was it decided at the meeting to not allow the optional method?

(thanks for your patience...)

[brianaAtWork]

Maybe in a next version. (MRN)

Looks like the perfect recipe for the MRN already included in the S-4: https://www.iho.int/iho_pubs/standard/S-4/S4%20V4-6-0_ENG_April16.pdf

B-631.3 Title. The NM should be given a title which will assist the mariner in identifying the geographical location and then where on the chart the update is located. It is therefore normal to start the title with the country name (except in international waters, where the ocean or sea name should be used), followed by any sub-region, local names and a general indication of the nature of the update, for example: NEW ZEALAND – North Island – West coast – North Taranaki Bight – Marine reserve. Buoyage. Names should be in agreement with the largest scale chart. There will often be a choice of region between the relevant country and its adjacent sea or ocean; whenever possible use the country name, particularly in coastal waters.

[brianaAtWork]

The NM should be given a title

BTW.... the S-4 says NM should be given a title...but I don't see one represented in the data structure.

[brianaAtWork]

The noticeIdentifier identifies a notice according to rules set up in S-4. The unique identification of a single notice is using a combination of values noticeNumber, SequenceNumber and year, according to current numbering schemes used by differnt HO:s . Also ChartProduct has a reference to noticeIdentifier.

Still working with this one and see that for the US at the top level of the entire content of an issued/published Notice to Mariner there is a unique identifier in the form of district/week/year. The district would be equivalent to the Area and the week would be the sequence number listed in the noticeIdentifier. I'm seeing this similar to how we discussed the MRN at some point....at the top of a file would be the "base" MRN and then each item would extend it from that identifying its own unique indicator, but combining it with the base one to show it's full identifier. This would save on redundant use of many of the fields in the current noticeIdentifier.

Also, I wanted to add that the noticeIdentifier outlined in S-4 seems to be more of a publication-esque identifier...in other words it is something that could easily be programmatically created using the elements of the data file and a proper transform. (and since there are actually two ways s-4 describes having a unique identifier...it would allow the output of the transform to be the formatter of the notice identifier chosen.