Proposal: remove some entries

[mledoze]

After further thinking about #57 and #77, I would like to propose the removal of the following entries, so that each entry in the dataset represent only one entity:

• Saint Helena, Ascension and Tristan da Cunha (group of islands)
• Bonaire

Bonaire for example, share the same ISO codes as Sint Eustatius and Saba (which are not listed). The same goes for the three islands of Saint Helena, Ascension and Tristan da Cunha.

Let me know what you think.
Thank you.

Edit: removed South Georgia and the South Sandwich Islands, United States Minor Outlying Islands and Svalbard and Jan Mayen.

[ruimarinho]

CLDR data separates subdivisions by using exceptionally reserved ISO 3166 codes when there are significant differences between them (e.g. country calling code - +290 for Saint Helena, +247 for Ascension Island). Without this distinction, we would have to append all the known country calling codes to the callingCode array, losing much of its value.

While the ISO 3166 standard defines a single code for the groups listed (e.g. https://www.iso.org/obp/ui/#iso:code:3166:SH) it does have them listed on the standard as geographical entities. What about distributing a separate list with all geographical entities flattened?

[mledoze]

@ruimarinho

What about distributing a separate list with all geographical entities flattened?

I think that it is the solution here. I would rather not have ISO 3166-2 codes in the main file because it would only concern a few entries.

We could have something like data/shn.divisions.json with Saint Helena, Ascension and Tristan da Cunha listed separately there. It would make sense because it already exists a GeoJSON file for it: data/shn.geo.json. The same is true for Bonaire.

[BaseMax]

why removing saint helena?

On 27/12/2014, Mohammed Le Doze [email protected] wrote:

@ruimarinho

What about distributing a separate list with all geographical entities
flattened?

I think that it is the solution here. I would rather not have ISO 3166-2
codes in the main file because it would only concern a few entries.

We could have something like data/shn.divisions.json with Saint Helena,
Ascension and Tristan da Cunha listed separately there. It would make
sense because it already exists a GeoJSON file for it: data/shn.geo.json.
The same is true for Bonaire.

---

Reply to this email directly or view it on GitHub:
#93 (comment)

[mledoze]

@hebrew878

why removing saint helena?

Because Saint Helena, Ascension and Tristan da Cunha share the same ISO 3166-1 codes and this is not coherent with the rest of the entries. It will not be entirely deleted, but rather moved in the file data/shn.divisions.json. That is what I propose in my last comment.

[ruimarinho]

Sounds good to me. +1 for consistency. Maybe this can prepare future work for #40?

[yackermann]

@mledoze Whats wrong with United States Minor Outlying Islands?

[mledoze]

@herrniemand nothing. I edited my first #93 (comment).

[yackermann]

@mledoze Ok. I see *)

[yackermann]

@mledoze How do you reference specific comment?

[mledoze]

You just have to copy its SHA-1 hash, like so 8ec9d7a

[yackermann]

@mledoze I mean actual message/comment in the discussion, like you've done previously.

[yackermann]

@mledoze

@herrniemand nothing. I edited my first #93 (comment).

[mledoze]

Select text, then press r. More info https://github.com/blog/1399-quick-quotes.

[yackermann]

Select text, then press r. More info https://github.com/blog/1399-quick-quotes.

Testing

[yackermann]

@mledoze Found it LOL