Wikidata:Identifier migration/1

See Wikidata:Identifier migration for full history and User:ArthurPSmith/Identifiers/1 for statistics for properties on this page.

Properties in range P1 - P999: Wikidata:Identifier migration/0

Properties in range P1000-P1999: Wikidata:Identifier migration/1

Properties in range P2000-P2999: Wikidata:Identifier migration/2

Properties

Done

good to convert

Z39.5 abbreviation (P1161) - abbreviation, not identifier
Bluebook abbreviation (P1162) - abbreviation, not identifier

Both of these are external, and UIDs. Both are instances of Wikidata property for authority control for works (Q19833377). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:30, 15 August 2016 (UTC)[reply]

@Hoo man, Pigsonthewing: I agree, these two (+ Index Herbariorum code (P5858)) should be converted. --Epìdosis 12:47, 12 May 2020 (UTC)[reply]

Dewey Decimal Classification (P1036) - only 91.23% unique out of 4026 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Library of Congress Classification (P1149) - only 97.60% unique out of 541 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Regensburg Classification (P1150) - only 93.88% unique out of 98 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Universal Decimal Classification (P1190) - only 93.48% unique out of 92 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]

@Hoo man: These are library classifications, so they should be converted; see Basisklassifikation (P5748), which has been created as external-id. --Epìdosis 12:41, 12 May 2020 (UTC)[reply]

No, classification systems are NOT external identifiers, unless them promise a unique ID for every work (which I don't think any of these do). ArthurPSmith (talk) 18:04, 18 May 2020 (UTC)[reply]

@ArthurPSmith: The problem effectively is present in three of them, for which see now Wikidata:Property proposal/Library classifications' IDs for topics; in the meanwhile, @Hoo man:, Regensburg Classification (P1150) is already fine for conversion (since it isn't used anywhere for works or editions). --Epìdosis 22:25, 18 May 2020 (UTC)[reply]

Mix of good to convert and disputed

KOATUU ID (P1077) - only 98.80% unique out of 30732 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
media type (P1163) - only 91.78% unique out of 219 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
- Officially registered media types are unique. It seems to me that there may be cases where non-officially registered media types are used (for some reason). --Izno (talk) 12:20, 7 April 2016 (UTC)[reply]
OpenStreetMap tag or key (P1282) - only 93.97% unique out of 696 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Monte Carlo Particle Number (P1360) - 2 duplicates out of 61 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Linguasphere code (P1396) - 3 duplicates out of 43 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Foundational Model of Anatomy ID (P1402) - only 98.75% unique out of 3284 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Jewish Encyclopedia ID (Russian) (P1438) - only 83.05% unique out of 1363 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Patientplus ID (P1461) - only 96.77% unique out of 773 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
reporting mark (P1471) - 1 duplicate out of 43 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
UBERON ID (P1554) - only 98.70% unique out of 460 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]
Statistics Indonesia area code (P1588) - only 96.90% unique out of 46641 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]
@ArthurPSmith Should be converted because this code is unique, published by Statistics Indonesia (Q3449895), a government department responsible for statistics. Hddty (talk) 05:14, 21 June 2022 (UTC)[reply]
Pokémon index (P1685) - only 52.48% unique out of 1010 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]
- Disputed by Multichill at Wikidata:Project_chat/Archive/2016/02#upcoming_deployments.2Ffeatures
International Standard Industrial Classification code Rev.4 (P1796) - only 86.84% unique out of 38 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]
RSL scanned publication ID (P1815) - 2 duplicates out of 99 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]
Open Food Facts food additive ID (P1820) - 1 duplicate out of 16 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]

talk]]) 03:13, 15 August 2016 (UTC)

NIOSH Pocket Guide ID (P1931) - only 94.17% unique out of 720 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]
MCN code (P1987) - only 81.72% unique out of 93 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]
UCI code of cycling team (P1998) - only 47.11% unique out of 1316 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]

ICD-9-CM (P1692) - many single value violations
NCI Thesaurus ID (P1748) - many single value violations

phone number (P1329) - only 98.03% unique out of 254 uses (also does not qualify on other grounds - same number can be reassigned to another entity, etc.) ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Disputed by Multichill at Wikidata:Project_chat/Archive/2016/02#upcoming_deployments.2Ffeatures
Phone number needs to be combined both with area code and nation code to be even remotely unique, and that's besides the fact that these can and are reassigned. Moved to not doing. --Izno (talk) 12:20, 7 April 2016 (UTC)[reply]
Format actually requires country code. I think that would easily work as an identifier.
--- Jura 16:47, 19 May 2016 (UTC)[reply]

depicts Iconclass notation (P1257) - only 50.23% unique out of 2373 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]

That's not meant to be an identifier-type property (Iconclass notation (P1256) is). --Zolo (talk) 11:13, 23 March 2016 (UTC)[reply]
Shouldn't we convert this if P1256 is?
--- Jura 10:02, 14 August 2016 (UTC)[reply]
- Nope, Iconclass notation (P1256) is used to make the one on one mapping (distinct and unique) between iconclass and Wikidata. depicts Iconclass notation (P1257) is a weird property that we should probably get rid of at one point. Take for example The Adoration of the Magi (Q17322022), it has depicts (P180) -> adoration of the Magi (Q488841) and Q488841 has Iconclass notation (P1256) -> "73B57", but The Adoration of the Magi (Q17322022) also has depicts Iconclass notation (P1257) -> "73B57". We have more where that came from. Multichill (talk) 19:32, 15 August 2016 (UTC)[reply]
  - I thought the idea of external-id was to provide links between Wikidata and other resources of the web. P1257 just works in the opposite way of P1256. So it's clear that if one has unique values, the other one doesn't.
    --- Jura 18:34, 16 August 2016 (UTC)[reply]
    - @Jura1: my understanding of the external identifier is that it is a foreign key that connects a concept here with a concept in some external source. That's a one-on-on relationship so unique and distinct. That's not the case here. Multichill (talk) 19:55, 16 August 2016 (UTC)[reply]

Not doing

Hornbostel-Sachs classification (P1762) - This is a classification, not an identifier. It's not assigned to an instrument, it's used to describe the instrument, and any number of instruments can have the same classification. - Nikki (talk) 17:03, 12 February 2016 (UTC)[reply]

Note only 50.00% unique out of 318 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]

Agree with non-conversion. Merely a codified way of describing characteristics. --- Jura 19:45, 16 February 2016 (UTC)[reply]

P1905 (P1905) (datatype: monolingual string) - not a property with string datatype. --- Jura 11:41, 3 February 2016 (UTC)[reply]

Comment Now here's one that looks a lot like an identifier to me - maybe the "string datatype only" requirement should be relaxed to allow this? Or is there some other reason it shouldn't be on the list? ArthurPSmith (talk) 18:47, 4 February 2016 (UTC)[reply]

I think it doesn't meet the most recent criterion you proposed [1]. So finally, you think it's an non-identifier identifier?
--- Jura 13:35, 7 February 2016 (UTC)[reply]

Yes, you're right on this one - I was thinking this was the identifier, but it's just the name - looks like they use DOI for the identifier according to this list. So this definitely wouldn't be eligible as an identifier. ArthurPSmith (talk) 16:09, 8 February 2016 (UTC)[reply]

Comment this is now deleted. --Liuxinyu970226 (talk) 12:16, 5 February 2018 (UTC)[reply]

Commons Creator page (P1472) - These are wiki page names, see P839 on page 0. - Nikki (talk) 20:29, 17 February 2016 (UTC)[reply]
Commons Institution page (P1612) - These are wiki page names, see P839 on page 0. - Nikki (talk) 20:29, 17 February 2016 (UTC)[reply]
legal citation of this text (P1031)
- Disputed by Multichill at Wikidata:Project_chat/Archive/2016/02#upcoming_deployments.2Ffeatures
anti-virus alias (P1845)
- Disputed by Multichill at Wikidata:Project_chat/Archive/2016/02#upcoming_deployments.2Ffeatures
street key (P1945) - never used ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]
- Disputed by Multichill at Wikidata:Project_chat/Archive/2016/02#upcoming_deployments.2Ffeatures
road number (P1824) - only 73.06% unique out of 2383 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]
- Disputed by Multichill at Wikidata:Project_chat/Archive/2016/02#upcoming_deployments.2Ffeatures
route number (P1671) - only 84.44% unique out of 90 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]
- Disputed by Multichill at Wikidata:Project_chat/Archive/2016/02#upcoming_deployments.2Ffeatures
Danish Bibliometric Research Indicator level (P1240) - only 2 values among 13746 uses (0.01% unique) ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
file extension (P1195) - Identifier? Multichill (talk) 21:44, 1 February 2016 (UTC)[reply]
note only 90.23% unique out of 532 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Likely to be duplicates since there is no central management. Suggest we don't convert this. --Izno (talk) 16:57, 20 March 2016 (UTC)[reply]

Disputed by Multichill at Wikidata:Project_chat/Archive/2016/02#upcoming_deployments.2Ffeatures

series ordinal (P1545) - we seem to be all agreed at least that this is not an external identifier: it is primarily used as a qualifier, it is locally determined in each case it's used so there's no overall assigning authority, it's definitely not unique (pretty much every series has at least a '1') etc. etc. ArthurPSmith (talk) 20:53, 8 February 2016 (UTC)[reply]
note only 19.03% unique out of 1708 uses as a non-qualifier. ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]

Disputed by Multichill at Wikidata:Project_chat/Archive/2016/02#upcoming_deployments.2Ffeatures

Agree that this shouldn't be converted. This property has too many different uses. --- Jura 06:38, 17 February 2016 (UTC)[reply]
formatter URL (P1630) - not sure how this would count as an identifier in any way, it's a string saying how to generate a URL... - Nikki (talk) 17:48, 3 February 2016 (UTC)[reply]
Note only 96.56% unique out of 813 uses ArthurPSmith (talk) 15:57, 15 February 2016 (UTC)[reply]

I think this should actually have URL datatype, not sure why it's a string in the first place.
--- Jura 07:03, 17 February 2016 (UTC)[reply]
I think string is right. The actual values for this property are not expected to be valid URLs. - Nikki (talk) 19:31, 17 February 2016 (UTC)[reply]
Isn't it an URI? and for URI we are to use URL-datatype?
--- Jura 15:32, 5 March 2016 (UTC)[reply]
It isn't really a URI - in a meaning that it does not specify location of any object. It is a template from which URI can be built. ---Laboramus (talk) 00:42, 13 March 2016 (UTC)[reply]
light characteristic (P1030) - only 92.00% unique out of 50 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
Clearly not an ID. --Izno (talk) 12:24, 7 April 2016 (UTC)[reply]
Gewässerkennzahl (P1183) - only 97.23% unique out of 4291 uses ArthurPSmith (talk) 15:28, 15 February 2016 (UTC)[reply]
- Disputed by Multichill at Wikidata:Project_chat/Archive/2016/02#upcoming_deployments.2Ffeatures
- I'm in the process of putting together a set of proposals for splitting this because it appears to be multiple independent identifiers (see property talk page for more details), so it's probably not worth converting. - Nikki (talk) 20:09, 13 May 2016 (UTC)[reply]

Requests

AlgaeBase URL (P1348) It's an identifier like Curlie ID (P998). —Justin (koavf)❤T☮C☺M☯ 19:04, 7 September 2016 (UTC)[reply]

@Koavf: I don't believe we have a method for converting URL types to external id's. If you want to make that change in this case you would need to propose a new property with the appropriate formatter URL. ArthurPSmith (talk) 23:06, 7 January 2017 (UTC)[reply]

IUCN taxon ID (P627) Also an identifier, not a statement. —Justin (koavf)❤T☮C☺M☯ 06:59, 7 January 2017 (UTC)[reply]

@Koavf: No, it was decided not to convert this one for several reasons - see the discussion at the bottom of the page here. ArthurPSmith (talk) 23:04, 7 January 2017 (UTC)[reply]