[Foundation-l] A Wikisource Definition (was: RfC: A Free Content and Expression Definition)

Thu May 4 15:59:27 UTC 2006

Dear Erik Moeller (and others concerned),

The inclusion policies of the English Wikisource, at least, are rather
well-defined. We do indeed invite users to translate texts into
English, and even have a standardised title format for disambiguating
multiple translations of the same work. Annotations are welcome as
well, and editorial notes are considered standard. We are actively
expanding our category structure with discussion, ensuring simplicity
and scalability are maintained with every change. You and I seem to
generally agree over what Wikisource should host and what it should
allow.

However, disagreement arises over the question of what "free" is. You
state that any texts that are less globally free should not be
included, despite arguing in favour of fair use content on Wikipedia,
because you claim that fair use content is absolutely necessary in the
creation of a quality encyclopedia. However, fair use has a
significant impact on the freedom of the information; consider the
following quote from the template {{fair use in}}:

"However, it is believed that the use of this work in the article
'Foo': to illustrate the object in question; where no free equivalent
is available or could be created that would adequately give the same
information; *on the English-language Wikipedia*, hosted on servers
*in the United States* by the non-profit Wikimedia Foundation,
qualifies as fair use under *United States* copyright law. *Any other
uses of this image, on Wikipedia or elsewhere, may be copyright
infringement*. See Wikipedia:Fair use and Wikipedia:Copyrights."

Does that sound "free" to you?

Most English Wikisource users that have participated in discussions
have expressed favour for (or, at least, no opposition to) the hosting
of non-derivative works. These are considered free enough (whereas
many users oppose fair use, for example), because they allow every
freedom except modification. If the end-user wishes only works that
can be modified, he can simply ignore the non-derivative license
categories. To this purpose, we are actively developing a
comprehensive structure for categorising works by license. For
example, a Creative Commons license may be too restrictive for some
end-user; perhaps they want public domain works only. In that case,
they can browse or download from the public domain categories.

Can a user do that on Wikipedia? Can they easily choose to ignore any
article that isn't entirely free, or do they have to manually sift
through and examine every article? In other words, every end-user of
Wikipedia must assume that every article is as free as the most unfree
article. The end-user must, in fact, assume that every single article
provided by Wikipedia is fair use.

Non-derivative is perfectly fine for someone seeking to expand his own
archive, or print a collection of free stories; GFDL is fine for most
other uses; Public domain is best for some. Wikisource allows the user
to selectively decide on the degree and flavour of freedom they want.
We are careful to exclude the unfree licenses (noncommercial,
copyright, fair use, et cetera), but I see no reason to exclude
non-derivative works.

Yours sincerely,
Jesse Martin ([[User:Pathoschild]])

On 5/4/06, Erik Moeller <eloquence at gmail.com> wrote:
> On 5/3/06, Andre Engels <andreengels at gmail.com> wrote:
> > But might it not be a good idea to have the text of a license like the
> > GPL even if we don't have any material under it? Would not that be
> > well under the purpose of Wikisource?
>
> This is an excellent question.  It leads to another one: Why does
> Wikisource, as a project, exist? People will have many different
> answers to that. Here's mine.
>
> 1) All material is ready to use, because we follow a strict standard
> of freedom. Derivative works, commercial use, and so forth, are all
> permitted. This parallels the equally strict standards of Commons (I
> still think a case could be made for them to be merged, but that
> decision has long been made), and distinguishes Wikisource from other
> archives.
>
> 2) Metadata. Wikis are getting better at storing a wide range of
> relations and associations with the pages they contain. For now we
> have categories, templates, interwiki links, and regular links
> (enriched with the "What links here" feature). Projects like Semantic
> MediaWiki or Wikidata will eventually add even more functionality, and
> the collaborative editing approach makes it possible to develop
> reasonable "folksonomies" (yech). Wikisource will probably have the
> best metadata of all the source text libraries.
>
> I count annotations as metadata. There is a wide range of NPOV
> annotations that are possible, especially for classical texts.
>
> 3) Translate and collaborate. I cannot emphasize enough how strongly I
> feel translations should become part of Wikisource's core mission.
> This is where wikis, with some additional functionality (easier
> processes for managing documents and assignments), could really shine.
> There is tremendous value in free translations. Many, many books which
> are in the free Wikisource archive are not available as free
> translations even in languages like German, Italian, French, let alone
> Russian, Farsi, or Japanese.
>
> Wikis are well-suited for this kind of work, because you can both
> split the work into packages, and collaborate on refining the
> consistency of the end result. The same is true for proofreading
> scanned documents, but here, the "Distributed Proofreaders" project is
> already doing an admirable job. We'd have to do a lot of work on
> further software extensions to compete with them.
>
> 4) Limited scope archive. We cannot possibly archive every single
> document that might be of interest to someone in the future. Similarly
> to Wikimedia Commons, we need to develop criteria of usefulness. One
> such criterion is freedom of the content. This already drastically
> reduces the scope to a much more manageable amount. The material
> should also have been published at some point and meet  general
> criteria of notability.
>
> 5) Incentivize freedom. Through 2) and 3), I hope that we can create a
> real incentive for authors to release published works freely,
> especially after they have gone out of print. I have decided to put
> the first edition of my own book under a GFDL/CC-BY-SA dual license. I
> did so with the hope that it might be archived and translated on
> Wikisource. However, de.wikisource.org has neither decided whether it
> wants to do translations, nor whether it wants modern texts.
>
> I see no principal reason why Wikisource should not archive many
> different ''kinds'' of material as long as they meet criteria as
> defined in 4). For instance, I think it would be great if Wikisource
> became an archive for "open access" scientific content (and even data)
> that meets the free content definition.
>
> But with the exception of 2), all of the points above suggest
> implementing a strict standard of freedom on Wikisource. Then, in
> answer to your above question, it follows logically that license texts
> that are not used as resources are, unless they are free content,
> inappropriate on Wikisource. What do we gain by archiving them? Due to
> their very nature, only armageddon could wipe out the record of the
> most popular licenses. If we cannot translate them, if others cannot
> derive new licenses from them, if we do not use them -- then we should
> not host them.
>
> But, you might answer, aren't these documents in themselves
> philosophically compatible with our core ideas? You might make an
> equally strong case for mirroring all of Richard Stallman's
> philosophical essays. However, unless they are published, and unless
> they are free content, we should not do so.
>
> Now, a library of free licenses that others can use as modular
> building blocks to create their own, that would be a very interesting
> project indeed.
>
> All of our projects will eventually need clear definitions. There is
> some need for Board oversight here, or there will be what we call
> "semantic drift" in the WiktionaryZ project: people developing their
> own meanings, and implementing them as they see fit. Some will take
> the project away from its free content nature. Others will be too
> strict in limiting the scope of documents. Some will argue that a
> collaborative translation is a form of "original research" and should
> not be allowed. Again others might see annotations as unacceptable
> alterations of the source material.
>
> We have seen this with Wikibooks. Intended as a place to
> collaboratively write textbooks, this definition clashes with a much
> more inclusive practice that has long tolerated materials such as game
> guides, jokes, or dating tips. How much do we know about the way the
> meaning of Wikibooks or Wikisource is interpreted in other languages
> than English, when we don't have a shared definition of its mission
> which itself is literally translated into these languages?
>
> I'm glad that we did write and translate a mission statement for
> Wikinews. There was never any confusion in a local Wikinews edition
> about whether or not original reporting is allowable, for instance. So
> volunteers could immediately start working on policies for it. These
> _policies_ differ from language to language, but the core goals do
> not.
>
> Volunteers like Birgitte can be forgiven for being frustrated when
> their own ideas clash with those which are seemingly well-understood
> by a small group of people who have little to do with the project
> itself, ideas which are not well-communicated to its editors. It needs
> to be clear why Wikisource exists, and what core policies it should
> follow. Certainly such a definition can be developed through a process
> of community consultation (as we're doing with the FCD), but it still
> has to be done.
>
> Erik
> _______________________________________________
> foundation-l mailing list
> foundation-l at wikimedia.org
> http://mail.wikipedia.org/mailman/listinfo/foundation-l
>