Open ppKrauss opened 8 years ago
Some infos are available in this topic : http://www.naturalearthdata.com/forums/topic/thematic-codes/
A complete list would be very usefull :)
Thanks @brunob for the feedback (!).
Hum... The problem is not only the information existence (in some hidden point of the universe), but the cost to find information... Today the best practice is to publish data and metadata together, in a standard way. There are two good open standards:
PS: for a quick fix... Your link is not showing the "DESCRIPTION??" column, can you fill here the description?
An example of abbrev versus postal is:
name: California abbrev: Calif. postal: CA iso_a2: US (which is the country code, not region code)
On Wed, Aug 10, 2016 at 8:00 AM, b_b notifications@github.com wrote:
Some infos are available in this topic : http://www.naturalearthdata. com/forums/topic/thematic-codes/
A complete list would be very usefull :)
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/nvkelso/natural-earth-vector/issues/153#issuecomment-238894724, or mute the thread https://github.com/notifications/unsubscribe-auth/AA0EO0e47KUQIGmHw5-nzQrn093gtDwxks5qeeeegaJpZM4GVQm8 .
One of the stated goals of Natural Earth is not to get bogged down in mind numbing XML metadata, but I hear that these field names aren't making sense to you. Keep the questions coming and I'll answer them, though.
On Wed, Aug 10, 2016 at 9:46 AM, Peter notifications@github.com wrote:
Thanks @brunob https://github.com/brunob for the feedback (!).
Hum... The problem is not only the information existence (in some hidden point of the universe), but the cost to find information... Today the best practice is to publish data and metadata together, in a standard way. There are two good open standards:
PS: for a quick fix... Your link is not showing the "DESCRIPTION??" column, can you fill here the description?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/nvkelso/natural-earth-vector/issues/153#issuecomment-238928257, or mute the thread https://github.com/notifications/unsubscribe-auth/AA0EOzijGHlRTz3w99sWCIoORoRCOnPlks5qegB2gaJpZM4GVQm8 .
Maybe a simple csv file describing the different fields in each zip could do the job, or a web page (on the repo wiki or on the official website) linked in every readme files ?
Anyway, @nvkelso no hurry, and thanks a lot for these super useful datasets :)
@brunob yes, a JSON or a CSV file with descriptions will be perfect!
@nvkelso Some obvious names as "abbrev" and "name" make sense, but many others as "geou_dif" and "su_a3" no sense... we need some "translation" ;-)
About goals: hum... this will be other discussion, and perhaps more ideology and personal point of view ... I will try to explain.
Today _Open Data and "not to get bogged down in mind numbing XML metadata"_ are not compatible. Low corruption and transparency needs clarity and good semantic (metadata), the basic standars are the minimal for clarity.
Natural Earth are good and build with good people (!), but the corrupts are hiding in the lack of explanation... Today, in some "more serious use", we are blocked from using Natural Earth because there are not good explanations.
Primarily the confusing field names are optional / rarely used. It's a once or twice a year question over millions of downloads, shrug. Send a PR if you feel super strongly about it.
geou_diff = Geo unit is different than its parent administrative unit, true false. This is helpful for labeling maps, and to not repeat labels.
su_a3 = administrative sub-unit Natural Earth alpha 3 character code, like ISO A3 codes, but different.
On Aug 10, 2016, at 11:44, Peter notifications@github.com wrote:
@brunob yes, a JSON or a CSV file with descriptions will be perfect!
@nvkelso Some obvious names as "abbrev" and "name" make sense, but many others as "geou_dif" and "su_a3" no sense... we need some "translation" ;-)
About goals: hum... this will be other discussion, and perhaps more ideology and personal point of view ... I will try to explain.
Today Open Data and "not to get bogged down in mind numbing XML metadata" are not compatible. Low corruption and transparency needs clarity and good semantic (metadata), the basic standars are the minimal for clarity.
Natural Earth are good and build with good people (!), but the corrupts are hiding in the lack of explanation... Today, in some "more serious use", we are blocked from using Natural Earth because there are not good explanations.
― You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.
From @jalbertbowden on Apr 10, 2015:
See:
i've pieced together a list of links with some info, and i think labeled
a few correctly for 110m cultural boundaries, but a) there's alot of gaps
and b) i'm not 100% on the ones i have.
if you can fill in the gaps, or add to the list, that would be sweet.
trying to format it the data, but its hard without the definitions
https://gist.github.com/jalbertbowden/1a94aa339682eabfdc6a
Was there any update on this? It's mind boggling that the schema is not annotated and I don't understand what name_zh
field is, is it traditional chinese or simplified?
@pronebird
I don't understand what name_zh field is, is it traditional chinese or simplified?
name_zh
is a name from wikidata label - and a postfix language code ( _zh
) == wikidata language code : https://www.wikidata.org/wiki/Help:Wikimedia_language_codes/lists/all
If you need Simplified Chinese
you have to import/merge the wikidata zh-hans
labels manually ( or "Traditional Chinese" = wikidata zh-hant
label )
If you interested in the technical details:
@ImreSamu thanks for the pointer. Is there any reason why traditional chinese is not included in the data dump by default?
Is there any reason why traditional chinese is not included in the data dump by default?
As I know:
wikidataid
wikidataid
from the release notes: "Want to see more name translations in the next Natural Earth release? Go edit Wikidata!"NOW: If you need any other language translation ( like "Traditional Chinese" )
you can do now - with a minimal scripting knowledge:
wikidataid
column.Just have a quick question about gid
. Can I assume that these values stay the same in between updates and could therefore be used as foreign keys? Cheers!
No, use ne_id
No, use ne_id
Thanks!
Was about to create a new issue with the content below until I found this one.
One of the stated goals of Natural Earth is not to get bogged down in mind numbing XML metadata
Isn't that... kind of all of it though?The geometry data is pretty mind numbing and tedious too isn't it? Why draw the line here, when so much data is there and work already done.
Don't get me wrong, this whole project is a monumental task that I would never want to do myself, but it seems silly to not even include a best effort attempt at annotating the properties
. Or even just putting a link this discussion prominently in the repo or on the website. It took me forever to find anything about this. And as another commenter said, the lack of clarity makes this info (and indeed maybe the whole dataset, to some) not prudent to use.
The issue I was going to post, for extra keywords for other peoples' searches:
Is there documentation or explanation of all the different fields that are in a Feature's properties
? There's a lot of different but similar looking values, and their keys are usually short and ambiguous. As a geography/GIS novice, I have no idea which ones I should use, what the variability is, what the source actually is, or even what the full name of the property is (e.g. I assume BRK_NAME
is an abbreviation but what's the full name).
For example, what's the difference between ISO_A3
and ISO_A3_EH
? I can't find any info on google for "iso a3 eh". What is GEOUNIT
? BRK_NAME
? FCLASS
? WOE
?
Would be great to have at least just a single sentence for each field here (or one for each group of related fields e.g. ADM0_A3_EN
, ADM0_A3_IT
, ADM0_A3_JP
).
I volunteer to add a metadata table (3 Columns: abbreviation, type and description) to the corresponding HTML Pages (like e.g. https://www.naturalearthdata.com/downloads/10m-cultural-vectors/10m-populated-places/) if @nvkelso accepts such Pull Requests.
Yes, PRs accepted.
For example, what's the difference between ISO_A3 and ISO_A3_EH?
I too would love to know what the "_EH" suffix is used for, and thus the difference between iso_a2
and iso_a2_eh
fields. (Why is France's iso_a2
=-99
while its iso_a2_eh
is FR
?)
Per Urban Dictionary:
Eh
– Meaning confused - or used at the end of a sentence to show it is a question (Canadian English) or "approximately right" (California English)
Natural Earth and ISO have slightly different understandings of the world... and some NE admin 0 "levels" (eg country versus map unit) match more exactly with ISOs versus approximately.
I volunteer to add a metadata table (3 Columns: abbreviation, type and description) to the corresponding HTML Pages (like e.g. https://www.naturalearthdata.com/downloads/10m-cultural-vectors/10m-populated-places/) if @nvkelso accepts such Pull Requests.
Any update on this? Is metadata available, yet?
@philipshirk sorry, no time and too little knowledge yet. I am only an amateur.
I have been looking for explanations of the attribute table column headers for a while now and haven't been able to find anything. It's very frustrating to not have documentation for naming conventions to make the best choice on analyzing information and displaying results. Is there a reason the READMEs take us back to the download page? If there is a working document for Natural Earth Metadata, where is it?
There has been a start in https://github.com/nvkelso/natural-earth-vector/pull/861.
I have a few questions:
This is a bit of a struggle 😅 Many thanks.
Edit: I think I got part of my answer for the third question by looking at the map units data tables. Sovereign country has a tentative control over other independent countries, like China -> Tawian. I'm still not sure what Sovereignty means on its own for Cuba & Kazakhstan, though.
The README of 10m-admin-0-countries is this HTML... Where the complete data specifications?
There are no table/field description? Example: what the difference between
postal
andiso_a2
? when must be equal and when not? Another table description, "the NULL is -99", where it is specified?PS: other "not explained" cases that I wold like understand by peakbagger as suggested.