Prepare registry and repository solutions

[DavidFatDavidF]

Is your feature request related to a problem? Please describe. Based on the task force and plenary discussions so far, it seems clear that the standard will be in need of at least one registry and of at least one CLDR-like repository.

Describe the solution you'd like The group should start looking for stable, public but controllable places that could host the required registry or registries and the repository. Ideally both the registry and repository would be hosted by Unicode. The standard should reuse CLDR data where available, consider adding new data categories to CLDR where viable, and put data categories not accommodated by CLDR into an MFWG maintained repository..

Registry or registries should hold both Unicode MFWG guaranteed registry items and items guaranteed by other authorities who wish to register their registry items with Unicode MFWG.

Describe why your solution should shape the standard The standard should make a clear distinction between and among the following:

1. core data model (provisions in the standard)
2. registry driven options and values
3. language specific lists of admissible registry values and other language specific data collections required for message parsing and formatting
4. extension points for user defined customization

The standard should also define a registration process for users to make their customizations publicly available through the registry. The standard should make clear that user customizations must not be used for features available through the registries or the standard's core. Even in cases the core or registry guaranteed functionality is smaller than the intended user feature, users must reuse core and the MFWG guaranteed registry in the largest possible extent and only provide the net difference of the intended functionality as their customization to maximize core and registry guaranteed interoperability.

Additional context or examples

Example 1

grammatical case registry item

"grammatical_case" :
[
"nominative",
"genitive",
...
"vocative",
"ablative",
...
]

Example

repository data for case registry values in Czech

"cs-CZ" :

[
"nominative",
"genitive",
"dative",
"accusative",
"vocative",
"locative",
"instrumental"
]