Date: Fri, 13 May 2011 13:49:07 -0400
From: Rintze Zelle <@Rintze_Zelle>
Subject: Re: [xbiblio-devel] use of CSL styles in Peaya with changed
attribution
In related news, I recently updated the CSL schema to disallow the cs:rights
element in cs:info for dependent styles, as I think we shouldn’t bother with
licensing those (none of the existing dependent styles include a cs:rights
element).
Rintze contacted me off-list because (contrary to the above) dependent
styles I wrote did use this and other recently removed elements. I’ll
just throw in my two cents: I think that it was useful for dependent
styles to have authors/contributors that have apparently been removed
from the schema recently. Rintze also noticed that two of the
AIP-dependent styles I wrote used the same title & was able to contact
me for clarification. A point of contact is useful for many things.
If I see a dependent style doesn’t seem quite right for a paper I’m
writing, do I fork the style into a dependent one? Do I modify the
style it is dependent on, changing the output for all other dependent
styles? Or perhaps I should just add an ISSN or other identifying
information to let me know that the dependent style applies to some
other journal whose name happens to be similar to the name of some
different journal I’m actually writing to. Being able to ask for
clarification seems like a good thing. I’d like to see the author
elements restored.
Regarding rights: I know that Bruce has taken the position that ALL
CSL styles are not copyrightable. Neither of us are lawyers, but I’ve
talked to lawyers about this issue. Lawyers never give you a straight
answer, but the informal consensus I got was that the US copyright
laws were so all-encompassing that they probably are subject to US
copyright & that we should act as if they are, as we don’t really lose
anything if we do.
The dependent styles are much less expressive than the independent
styles & much less is lost if they are somehow used contrary to
CC-BY-SA. But I don’t know whether this, alone, is reason to remove
the rights tag. I suppose I just don’t see what real benefit doing
that will bring.
I’ll restore cs:author, cs:contributor and cs:rights to dependent styles
(unless somebody disagrees with Rick). Removing them was a maybe not so
thought-out decision when I split the schema for dependent and independent
styles.
RintzeOn Tue, May 17, 2011 at 10:18 AM, Richard Karnesky <@Richard_Karnesky>wrote:
Date: Fri, 13 May 2011 13:49:07 -0400
From: Rintze Zelle <@Rintze_Zelle>
Subject: Re: [xbiblio-devel] use of CSL styles in Peaya with changed
attribution
In related news, I recently updated the CSL schema to disallow the cs:rights
element in cs:info for dependent styles, as I think we shouldn’t bother with
licensing those (none of the existing dependent styles include a cs:rights
element).
Rintze contacted me off-list because (contrary to the above) dependent
styles I wrote did use this and other recently removed elements. I’ll
just throw in my two cents: I think that it was useful for dependent
styles to have authors/contributors that have apparently been removed
from the schema recently. Rintze also noticed that two of the
AIP-dependent styles I wrote used the same title & was able to contact
me for clarification. A point of contact is useful for many things.
If I see a dependent style doesn’t seem quite right for a paper I’m
writing, do I fork the style into a dependent one? Do I modify the
style it is dependent on, changing the output for all other dependent
styles? Or perhaps I should just add an ISSN or other identifying
information to let me know that the dependent style applies to some
other journal whose name happens to be similar to the name of some
different journal I’m actually writing to. Being able to ask for
clarification seems like a good thing. I’d like to see the author
elements restored.
That seems reasonable to me, but I have no strong opinion either way.
Regarding rights: I know that Bruce has taken the position that ALL
CSL styles are not copyrightable. Neither of us are lawyers, but I’ve
talked to lawyers about this issue. Lawyers never give you a straight
answer, but the informal consensus I got was that the US copyright
laws were so all-encompassing that they probably are subject to US
copyright & that we should act as if they are, as we don’t really lose
anything if we do.
Just to be clear, I have not taken any position on this, since IANAL.
I have only raised questions, most specifically about what a “BY”
clause on an individual style would mean practically (the larger
question is whether this is really a creative work, and thus whether
CC licenses are most appropriate at all).
The dependent styles are much less expressive than the independent
styles & much less is lost if they are somehow used contrary to
CC-BY-SA. But I don’t know whether this, alone, is reason to remove
the rights tag. I suppose I just don’t see what real benefit doing
that will bring.
I have a feeling Rintze is more worrying about future hypothetical
issues. Are we OK with someone submitting a dependent style to the
repo that has rights like “all rights reserved; contact author for
permission to use”?
It would be nice if we could clear up the legal issues how, and then
adopt policies accordingly.
The dependent styles are much less expressive than the independent
styles & much less is lost if they are somehow used contrary to
CC-BY-SA. But I don’t know whether this, alone, is reason to remove
the rights tag. I suppose I just don’t see what real benefit doing
that will bring.
I have a feeling Rintze is more worrying about future hypothetical
issues.
Well, dependent styles are basically just bookmarks with a little bit of
metadata, so I thought it would make little sense to apply a license.
Are we OK with someone submitting a dependent style to the
repo that has rights like “all rights reserved; contact author for
permission to use”?
We wouldn’t be okay with that. I think an important question is whether, if
we decide that CC-BY-SA is a good license for dependent and independent
styles, it would make sense to make it the mandatory license for styles
hosted in the github repo. As I mentioned here ( http://xbiblio-devel.2463403.n2.nabble.com/use-of-CSL-styles-in-Peaya-with-changed-attribution-td5267001.html#a6360339),
that would make it much easier for CSL adopters to comply with the licensing
terms (currently, they’d theoretically have to check each individual style
to see which license it uses).
We wouldn’t be okay with that. I think an important question is whether, if
we decide that CC-BY-SA is a good license for dependent and independent
styles, it would make sense to make it the mandatory license for styles
hosted in the github repo. As I mentioned here ( http://xbiblio-devel.2463403.n2.nabble.com/use-of-CSL-styles-in-Peaya-with-changed-attribution-td5267001.html#a6360339),
that would make it much easier for CSL adopters to comply with the licensing
terms (currently, they’d theoretically have to check each individual style
to see which license it uses).
I don’t have much of an opinion on most aspects of this - but this part I
agree with 100% - I think we should push for the repo to be on a single,
open license.–
Sebastian Karcher
Ph.D. Candidate
Department of Political Science
Northwestern University
Do any of the other parties aside from Zotero that use CSL styles (i.e.
Mendeley, Mekentosj/Papers) have any legal advice on whether CC-BY-SA would
be a suitable license?
In that case, I’m pretty sure that should be fine. We display the
original author (as specified in the CSL file) in the style browser
interface in the Mendeley client, we don’t currently display any
licensing information there but I’m happy to add that.
Or the unported license (http://creativecommons.org/licenses/by-sa/3.0/).
The main question is whether CC-licenses are applicable at all to something
like CSL styles, or whether we should pick a software license. If everybody
is comfortable with CC-BY-SA, it’d like to make it the requirement for
styles hosted in the github repo.
We touched on this before, but we never came to a clear conclusion. We could
really use some advice from somebody legally schooled.
RintzeOn Mon, May 30, 2011 at 6:37 PM, Robert Knight <@Robert_Knight>wrote:
In that case, I’m pretty sure that should be fine. We display the
original author (as specified in the CSL file) in the style browser
interface in the Mendeley client, we don’t currently display any
licensing information there but I’m happy to add that.
See, this is the awkward thing about all this. I don’t necessarily
want my name to be displayed in clients. When I do put my name as a
contributor on a style, it’s only for convenience (so someone knows
who to contact); not to assert any claim to the style.
Rintze followed up with:
The main question is whether CC-licenses are applicable at all to something like
CSL styles, or whether we should pick a software license. If everybody is
comfortable with CC-BY-SA, it’d like to make it the requirement for styles hosted
in the github repo.
We touched on this before, but we never came to a clear conclusion. We could really
use some advice from somebody legally schooled.
Yes; and preferably someone with experience with these specific
licenses, and understanding of copyright law.
Authorship assignments are rather messy in the domain of CSL styles, as
styles almost always based on other styles, and sometimes receive
significant updates after their first release, so the contribution by the
primary author may be limited. I agree with Bruce that it’s probably better
to not include author/contributor names in clients.
I added the license to styles that did not carry one, updated older
versions of the license, and replaced ported variants by the unported
version. I got permission of the style authors who had released their
styles under by-nc-sa licenses to switch to by-sa.
I am all for it. It seems very reasonable, and while it may keep a few styles out (?), it makes it much easier to use those styles without extra friction.
I also updated the license conditions for the style repository, which
seemed necessary, since the CC BY-SA license specifies that “[y]ou must
attribute the work in the manner specified by the author or licensor”.
In short, I think we should demand that any project using CSL styles from
the repo should publicly acknowledge this, and that
authorship/contributorship information shouldn’t be stripped from
distributed styles. Let me know if you think the text needs changes/polish.
In short, I think we should demand that any project using CSL styles from the repo should publicly acknowledge this,
and that authorship/contributorship information shouldn’t be stripped from distributed styles.
Let me know if you think the text needs changes/polish.
I would suggest putting the licensing details under a separate heading
in the readme from the general information.
The attribution requirements sound fine to me. We list the author
next to each style and credit the CSL language / citeproc-js in the
About dialog but not citationstyles.org itself. I’ve filed a ticket
to resolve that.
Creative Commons just released their 4.0 licenses (see e.g. https://twitter.com/dancohen/status/405331406622306304 ). IANAL, but
it is my understanding that we’re free to upgrade all existing styles
and locales from CC BY-SA 3.0 to this license without having to
consult the original contributors. See http://wiki.creativecommons.org/License_Versions#Compatibility_mechanism_in_BY-SA_licenses
, which says “Starting with the release of the 2.x license suites, CC
expanded compatibility by allowing contributions to adapted material
to be created under the same or later version of the original license,
including other ported versions of the same or later version of the
license”.
I’m not really sure there are big advantages for us to upgrading to
4.0, but I think it would be a simple change to make if we wanted to.