Language proposal policy/2019-04 proposed revision

Introduction

LangCom is preparing to revise the Language proposal policy. These are the rules that determine if a new proposed project is first (a) eligible for eventual approval, and later (b) actually ready for approval.

In particular, the proposal is to revise the requisites for eligibility, not the requirements for final approval. (Please note, though, that this drives a couple of changes in the "Specific issues" section, too.)

Brief summary of changes

1. We are not going to move away from the reliance on ISO 639, and particularly ISO 639-3, as the standard for defining what we do and don't consider a language. But when the current LangCom voting policy was established in 2017, it created a procedure for an exception to this rule. The current changes just clarify that procedure further:
   • No exceptions will be considered unless some effort was made to get an ISO 639 code first. We encourage working within the standard, not getting around it.
   • If the previous point is overcome, the language request must be able to be coded with a language code-plus-BCP 47 tag to be considered. (Nearly any legitimate language variant can be, so we don't anticipate this would be a major obstacle.)

   It should be noted that a request under this exception still has to overcome rule #3 (that the language be sufficiently different to merit its own project) before it is ruled eligible.

2. Having projects in both macrolanguages and their constituent languages would frequently run afoul of rule #3. The change here clarifies when the constituent languages are more appropriate for new projects and when the macrolanguage is more appropriate.
3. The change in language from "Native speakers" to "Fluent contributors" is partly semantic, in that there is substantial overlap between these two groups. But ultimately, it matters less here whether a person is a native speaker of a language than if the person can contribute fluently and constructively to a project. That is reflected here.
4. Details have been added around some criteria LangCom uses to determine when an artificial/constructed language project will be allowed. Ultimately, LangCom still determines these on a case-by-case basis.
5. A note has been added concerning Unicode. The existence of a Unicode encoding for a new project's language/script pair is not a policy requirement. However, it may be impossible to ensure that Wikimedia's servers would serve the new project correctly if such an encoding does not exist.

Discussion and !votes

Discussion and !votes will be held on the talk page. (The proposal page itself will remain protected.)

Text of the proposed revision to LPP

Markups: additions ... moves ... deletions

Requisites for eligibility

Current	Proposed
(this text to be added to the main heading of "Requisites", after what is already there)	Projects that existed before the initial version of this policy was adopted (November 2006) do not have to meet these requirements. Please do not use them as an argument that your project should be ruled "eligible". The fact that such projects exist, even when they do not meet the eligibility requirements below, is completely irrelevant to current eligibility discussions.
1. The proposal is to open a new language edition of an existing Wikimedia project that does not already exist (see the complete list of Wikimedia projects or the SiteMatrix).	1. The proposal is to open a new language edition of an existing Wikimedia project that does not already exist (see the complete list of Wikimedia projects or the SiteMatrix). (The appropriate place to go if you are proposing a completely new project is Proposals for new projects.)
2. The language must have a valid ISO 639 1–3 code (search). If there is no valid ISO 639 code, you must obtain one. The Wikimedia Foundation does not seek to develop new linguistic entities; there must be an extensive body of works in that language. The information that distinguishes this language from another must be sufficient to convince standards organizations to create an ISO 639 code.	2. The language must ordinarily have a valid ISO 639 1–3 code (search). (search link updated) If there is no valid ISO 639 code, you must try to obtain one. The Wikimedia Foundation does not seek to develop new linguistic entities; there must be an extensive body of works in that language. The information that distinguishes this language from another must ordinarily be sufficient to convince standards organizations to create an ISO 639 code. Under very unusual circumstances, and provided there has first been an effort to obtain an ISO 639–3 code, the Language Committee may be willing to consider projects in languages having a valid BCP 47 language tag. Languages having neither an ISO 639 code nor a BCP 47 tag will not be considered under any circumstances.
3. The language must be sufficiently unique that it could not coexist on a more general wiki. In most cases, this excludes regional dialects and different written forms of the same language. The degree of difference required is considered on a case-by-case basis. The committee does not consider political differences, since the Wikimedia Foundation's goal is to give every single person free, unbiased access to the sum of all human knowledge, rather than information from the viewpoint of individual political communities.	3. The language must be sufficiently unique that it could not coexist on a more general wiki. In most cases, this excludes regional dialects and different written forms of the same language. The degree of difference required is considered on a case-by-case basis. The committee does not consider political differences, since the Wikimedia Foundation's goal is to give every single person free, unbiased access to the sum of all human knowledge, rather than information from the viewpoint of individual political communities. Projects coded as macrolanguages in ISO 639–3 are possible in some cases. New projects with a macrolanguage code are generally allowed if other projects with the identical code already exist in Wikimedia. In all other cases, these are decided on a case-by-case basis. There must be a reason why the project should be in the macrolanguage and not in one of the languages within the macrolanguage. Codes for collections of languages are part of ISO 639–2, not ISO 639–3. These are no longer valid codes for new Wikimedia projects. A few existing projects continue to use such codes, though the plan is to move them all.
4. The proposal has a sufficient number of living native speakers to form a viable community and audience (Wikisource wikis are allowed in languages with no native speakers, although these should be on a wiki for the modern form of the language if possible). If the proposal is for an artificial language such as Esperanto, it must have a reasonable degree of recognition as determined by discussion (this requirement is being discussed by the Language committee ).	4. The proposal has a sufficient number of fluent contributors to form a viable community and audience. The language of the proposal has a sufficient number of fluent users to form a viable contributor community and an audience for the content. Wikisource wikis are allowed in languages with no native speakers, although these should be on a wiki for the modern form of the language, when possible. If the proposal is for an artificial language such as Esperanto, it must have a reasonable degree of recognition as determined by discussion. (Some recognition criteria include, but are not limited to: independently proved number of speakers, use as an auxiliary language outside of online communities created solely for the purpose, usage outside of Wikimedia, publication of works in the language for general sale.) Notwithstanding the existence of an ISO 639–3 code, fictional languages (link) are not eligible for projects. Proto-language reconstructions are not eligible for ISO 639-3 codes, so are not eligible for projects.

From: Specific issues

Ancient or historical languages Only Wikisource wikis in ancient or historical languages are accepted, because there are no living native communities to use other resources. Where possible, such languages should be bundled with the modern equivalent (such as Old English with English), though that is not required.	Ancient or historical languages Only Wikisource wikis in ancient or historical languages are accepted, because resources in such languages continue to be important to the world, even in the absence of native, living speakers of those languages. Where possible, such languages should be bundled with the modern equivalent Wikisource project (such as Old English with English), though that is not required.
Artificial languages Yes, there can be wikis in artificial languages. There are already wikis available in Esperanto, Ido, Interlingua, Interlingue, Lojban, Volapük and Novial. See the relevant note under the prerequisites concerning fictional languages.	Artificial languages Yes, there can be wikis in artificial languages. There are already wikis available in Esperanto, Ido, Interlingua, Interlingue, Lojban, Volapük, Novial and Lingua Franca Nova. See the relevant note (link added) under the prerequisites concerning fictional languages and reconstructed proto-languages.
(new)	Unicode encoding Strictly speaking, Unicode encoding is not required by policy in order for a project to be approved. In practice, however, it may be impossible to ensure that the Wikimedia servers serve the project correctly if the language and script do not have a standard Unicode encoding.