# xbiblio-devel Digest, Vol 34, Issue 14

**URL:** <https://discourse.citationstyles.org/t/xbiblio-devel-digest-vol-34-issue-14/641>\
**Category:** CSL Development\
**Created:** [May 24, 2009, 12:22am UTC](https://discourse.citationstyles.org/t/xbiblio-devel-digest-vol-34-issue-14/641 "2009-05-24T00:22:38Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Richard\_Karnesky](https://avatars.discourse-cdn.com/v4/letter/r/ba8739/32.png) [@Richard\_Karnesky](https://discourse.citationstyles.org/u/Richard_Karnesky)\
**Post date:** [May 24, 2009, 12:22am UTC](https://discourse.citationstyles.org/t/xbiblio-devel-digest-vol-34-issue-14/641/1 "2009-05-24T00:22:38Z")

</div>

> Requiring attribution is an unneeded PITA and  
> actually against what I think many of us believe rather strongly: that  
> the content of a citation style is not subject to copyright claims.

Attribution during redistribution seems somewhat important to me: I  
wouldn’t want an app to distribute CSL files after ripping out the  
author/contributor information.

Attribution for derivative works may get “tricky,” but note that the  
attribution clause of CC states “provide, reasonable to the medium or  
means You are utilizing.” Within a CSL file, I would think this would  
mean doing something like using the template link to point to the  
style you borrowed from and/or keeping contributors intact. Neither  
seems to be a real PITA.

I also don’t see why an attribution clause is any more counter to the  
notion that style files aren’t copyrightable than applying ANY license  
to those files in the first place.

> So for sake of argument, why not just say that all CSL styles are  
> licensed under …

I agree with Frank’s concerns and think it is important to say where a  
default license would be applied: by the schema, by a style generator,  
or by a hosted repository of styles.

I am opposed to the _schema_ assigning a “default” license. If style  
files are copyrightable in the first place, the creator of a style  
should be able to dictate whether or not he applies a license and/or  
the terms of that license.

It’d seem slightly more acceptable if style-generating _tools_  
inserted a particular license (because you could choose to use some  
other tool or modify the generated style), but even this would seem  
somewhat obnoxious to me.

If style _repositories_ wanted to mandate or encourage particular  
licenses, that would seem fine. It’d be nice if the Zotero-hosted  
repository actually did this. For the Zotero repository, I agree with  
Simon that a copyleft license makes sense. If we want CSL adoption in  
proprietary apps (and I think we do), the GPL is out. I guess LGPL  
would be in, but it seems weird to license these files like source  
code & it just seems pragmatic to use something like CC-SA.

Licensing at the hosting level is what is done for most other file  
formats: Wikipedia and sourceforge make you license your content in a  
particular way if you want it to be hosted by them. But our word  
processors and IDEs (and even MediaWiki) don’t apply a license on our  
behalf. While there are probably file formats that require a  
particular license, I know of none that are popular.

[But I also believe this isn’t “all or nothing” and that “some is  
better than none.” I’d love to see Thomson-Reuters adopt CSL in  
Endnote, even if they “locked up” the style files they distributed.]

> So maybe we just add a required:
> 
> rights-uri = element cs:right-uri { licences }

I’m a fan of having lightweight files & don’t know if we should make  
this mandatory (especially if we don’t think the content warrants  
copyright!).

Also, why not either:

 or: info-rights = element cs:rights { attribute uri { xsd:anyURI }?, attribute label { text }? }

–Rick

---

<div class="post-metadata">

**Author:** ![Bruce\_D\_Arcus1](https://avatars.discourse-cdn.com/v4/letter/b/4bbf92/32.png) [@Bruce\_D\_Arcus1](https://discourse.citationstyles.org/u/Bruce_D_Arcus1)\
**Post date:** [May 24, 2009, 5:02pm UTC](https://discourse.citationstyles.org/t/xbiblio-devel-digest-vol-34-issue-14/641/2 "2009-05-24T17:02:00Z")

</div>

> > Requiring attribution is an unneeded PITA and  
> > actually against what I think many of us believe rather strongly: that  
> > the content of a citation style is not subject to copyright claims.
> 
> Attribution during redistribution seems somewhat important to me: I  
> wouldn’t want an app to distribute CSL files after ripping out the  
> author/contributor information.
> 
> Attribution for derivative works may get “tricky,” but note that the  
> attribution clause of CC states “provide, reasonable to the medium or  
> means You are utilizing.” Within a CSL file, I would think this would  
> mean doing something like using the template link to point to the  
> style you borrowed from and/or keeping contributors intact. Neither  
> seems to be a real PITA.

Some of the issues I’m raising here relate to the MakeCSL idea, where  
the idea is you to create a style by pulling in pre-defined macros.  
One way to solve this issue is to just have a single, master, CSL  
style with these macros, and very clear licensing terms that enable  
what I’m looking for here. In the case, I’d imagine a “created with  
MakeCSL …” statement on the finished styles that includes some  
terms, but am not sure.

> I also don’t see why an attribution clause is any more counter to the  
> notion that style files aren’t copyrightable than applying ANY license  
> to those files in the first place.

INAL, so am happy if we could get some guidance from someone that is.  
It might be enough if we could just agree on what we’d like to  
achieve. For me, it is that CSL styles can be, and remain, freely  
distributable usable and remixable.

> > So for sake of argument, why not just say that all CSL styles are  
> > licensed under …
> 
> I agree with Frank’s concerns and think it is important to say where a  
> default license would be applied: by the schema, by a style generator,  
> or by a hosted repository of styles.
> 
> I am opposed to the _schema_ assigning a “default” license. If style  
> files are copyrightable in the first place, the creator of a style  
> should be able to dictate whether or not he applies a license and/or  
> the terms of that license.

But are they copyrightable? I don’t think so. I often don’t put my  
name on style mods, b/c I think it’s donkey work.

> It’d seem slightly more acceptable if style-generating _tools_  
> inserted a particular license (because you could choose to use some  
> other tool or modify the generated style), but even this would seem  
> somewhat obnoxious to me.
> 
> If style _repositories_ wanted to mandate or encourage particular  
> licenses, that would seem fine. It’d be nice if the Zotero-hosted  
> repository actually did this. For the Zotero repository, I agree with  
> Simon that a copyleft license makes sense. If we want CSL adoption in  
> proprietary apps (and I think we do), the GPL is out. I guess LGPL  
> would be in, but it seems weird to license these files like source  
> code & it just seems pragmatic to use something like CC-SA.
> 
> Licensing at the hosting level is what is done for most other file  
> formats: Wikipedia and sourceforge make you license your content in a  
> particular way if you want it to be hosted by them. But our word  
> processors and IDEs (and even MediaWiki) don’t apply a license on our  
> behalf. While there are probably file formats that require a  
> particular license, I know of none that are popular.
> 
> [But I also believe this isn’t “all or nothing” and that “some is  
> better than none.” I’d love to see Thomson-Reuters adopt CSL in  
> Endnote, even if they “locked up” the style files they distributed.]
> 
> > So maybe we just add a required:
> > 
> > rights-uri = element cs:right-uri { licences }
> 
> I’m a fan of having lightweight files & don’t know if we should make  
> this mandatory (especially if we don’t think the content warrants  
> copyright!).
> 
> Also, why not either:
> 
> or: info-rights = element cs:rights { attribute uri { xsd:anyURI }?, attribute label { text }? }

That might be work.

Bruce