Crown Copyright in ca.py

[kaplanmyrth]

If we want to add anything under Crown Copyright, then there needs to be a branch for that:

  if author = "Crown" then pdyear = pubyear + 50

(translated into actual Python code, of course :-)

[David Strauss (davidstrauss)]

I don't think a string match on the author is the right approach. Let's see what the OL people have to say about including a boolean field for Crown Copyright.

[rejon (rejon)]

Man, I wish we could compare to some other legal jurisdictions about other nuances like Crown copyright...well, I guess better to get this working first and then can open it up to more and compare.

[solrize (solrize)]

"Crown" should not be a string in the author field, but rather there should be a separate (list-valued?) field with some symbolic values indicating this type of stuff, which can be jurisdiction-specific.

[rejon (rejon)]

We need to track jurisdiction specific features for copyright like this...so, we have a generic facts at a high level, but then need a jurisdiction specific way to track features for pd/copyright like british crown copyright (yes or no) and then the same say for canadian copyright (yes or no)...

Still, not a high priority for access copyright code because none of the access copyright data has it right now....

[rejon (rejon)]

Would be good to throw an exception for this feature since its not implemented...that would be good to do...need multiple exceptions...David is writing a patch for this :)

[David Strauss (davidstrauss)]

Of course, we'll need crown copyright to appear in the data model before I can throw an exception if it's marked on an item.

[kaplanmyrth (andy-kaplan-myrth)]

Does Access Copyright have Crown Copyright in their own records, and
they just didn't give it to us thinking we didn't need it?

In Canada specifically, anything printed by the "Queen's Printer" is
under Crown Copyright. For instance, something I'd really like to read
now having found it here, published in 1972 and under Crown Copyright so
in the PD in 2022, not in 2139 as indicated:
   http://pdregistry.ca/b/Alien_animals_in_British_Columbia

But that may be too customized to one country's needs?

David Strauss wrote:
> Of course, we'll need crown copyright to appear in the data model before
> I can throw an exception if it's marked on an item.
>

--
Andy Kaplan-Myrth, M.A., LL.B.
Barrister & Solicitor
------------------------------------------------
email: <email address hidden>
web: http://kaplan-myrth.ca
------------------------------------------------
PGP Key ID 0xE9349025
------------------------------------------------

[rejon (rejon)]

Ok, so sounds like there are works under crown copyright, but not marked such in the data...thus, we need to account for crown copyright and have some way to mark this in the UI if discovered, right?

[David Strauss (davidstrauss)]

To directly respond to Andy's question, Access Copyright's data does not have a field for Crown Copyright. If we add Crown Copyright capabilities, we'll be adding data to the system.

Possibly cool idea: The Access Copyright data does contain publication dates and copyright status. It *may* be possible to make some guesses about Crown Copyright status for items that are marked as public domain in their system but could only be so under Crown Copyright.

[kaplanmyrth (andy-kaplan-myrth)]

Could we reliably figure out Crown Copyright by searching the data for
"publisher:(queen printer)"? That only turns up 29 items right now, all
of which are under Crown Copyright, but this is still the sample data,
right?

In fact, could we do that query on the data before it's used as the
database for the site, and use the output of that query to add a Crown
Copyright data field?

David Strauss wrote:
> To directly respond to Andy's question, Access Copyright's data does not
> have a field for Crown Copyright. If we add Crown Copyright
> capabilities, we'll be adding data to the system.
>
> Possibly cool idea: The Access Copyright data does contain publication
> dates and copyright status. It *may* be possible to make some guesses
> about Crown Copyright status for items that are marked as public domain
> in their system but could only be so under Crown Copyright.
>

--
Andy Kaplan-Myrth, M.A., LL.B.
Barrister & Solicitor
------------------------------------------------
email: <email address hidden>
web: http://kaplan-myrth.ca
------------------------------------------------
PGP Key ID 0xE9349025
------------------------------------------------

[rejon (rejon)]

That sounds reasonable to me...yeah, this is the full data they gave us,
but from that email from Marcus, guess they haven't given us
*everything*...

Yes, we need a way to track crown copyright in the system, and then we
can parse the data....so, how can we best add this to the OL system as
described in this bug?

David, need your assistance in this...

Jon

On Mon, 2008-02-11 at 16:32 +0000, kaplanmyrth wrote:
> Could we reliably figure out Crown Copyright by searching the data for
> "publisher:(queen printer)"? That only turns up 29 items right now, all
> of which are under Crown Copyright, but this is still the sample data,
> right?
>
> In fact, could we do that query on the data before it's used as the
> database for the site, and use the output of that query to add a Crown
> Copyright data field?
>
>
> David Strauss wrote:
> > To directly respond to Andy's question, Access Copyright's data does not
> > have a field for Crown Copyright. If we add Crown Copyright
> > capabilities, we'll be adding data to the system.
> >
> > Possibly cool idea: The Access Copyright data does contain publication
> > dates and copyright status. It *may* be possible to make some guesses
> > about Crown Copyright status for items that are marked as public domain
> > in their system but could only be so under Crown Copyright.
> >
>
> --
> Andy Kaplan-Myrth, M.A., LL.B.
> Barrister & Solicitor
> ------------------------------------------------
> email: <email address hidden>
> web: http://kaplan-myrth.ca
> ------------------------------------------------
> PGP Key ID 0xE9349025
> ------------------------------------------------
>
--
Jon Phillips

San Francisco, CA
CHINA PH [phone]
<email address hidden>
http://www.rejon.org

MSN, AIM, Yahoo Chat, Skype: kidproto
Jabber Chat: <email address hidden>
IRC: <email address hidden>

Inkscape (http://inkscape.org)
Open Clip Art Library (www.openclipart.org)
Creative Commons (www.creativecommons.org)
San Francisco Art Institute (www.sfai.edu)

[David Strauss (davidstrauss)]

Does OL have separate publisher records? The best way to add crown copyright to the system is to add a "crown" field to publisher records. We can load the related publisher records when analyzing the copyright status of a work. It would be badly denormalized to put the publisher name and the crown publisher status on the actual work.

[kaplanmyrth (andy-kaplan-myrth)]

Whatever works -- Just don't replace existing publisher info with new
"crown" as publisher. In Canada at least, anything published by the
Federal Queen's Printer or any province's Queen's Printer is under Crown
Copyright, but you don't want to lose the record of which jurisdiction
published it.

David Strauss wrote:
> Does OL have separate publisher records? The best way to add crown
> copyright to the system is to add a "crown" field to publisher records.
> We can load the related publisher records when analyzing the copyright
> status of a work. It would be badly denormalized to put the publisher
> name and the crown publisher status on the actual work.
>

--
Andy Kaplan-Myrth, M.A., LL.B.
Barrister & Solicitor
------------------------------------------------
email: <email address hidden>
web: http://kaplan-myrth.ca
------------------------------------------------
PGP Key ID 0xE9349025
------------------------------------------------

[David Strauss (davidstrauss)]

"In Canada at least, anything published by the Federal Queen's Printer or any province's Queen's Printer is under Crown Copyright, but you don't want to lose the record of which jurisdiction published it."

That's exactly why publishers should merely be marked as Crown publishers. They can remain independent in the system, yet everything considered published by them is automatically subject to Crown Copyright rules.

[rejon (rejon)]

Yes, lets do this, but I also want to make sure that we are tracking the
jurisdiction that this is relevant in, as well as the jurisdiction
published in, for the most relevant copyright assessment...

Paul, what is the best way to do this...I feel like we need to make sure
that the more complex structure for tracking copyright is in place that
we discussed before...by jurisdiction....

Jon

On Thu, 2008-02-14 at 19:49 +0000, David Strauss wrote:
> "In Canada at least, anything published by the Federal Queen's Printer
> or any province's Queen's Printer is under Crown Copyright, but you
> don't want to lose the record of which jurisdiction published it."
>
> That's exactly why publishers should merely be marked as Crown
> publishers. They can remain independent in the system, yet everything
> considered published by them is automatically subject to Crown Copyright
> rules.
>
--
Jon Phillips

San Francisco, CA
CHINA PH [phone]
<email address hidden>
http://www.rejon.org

MSN, AIM, Yahoo Chat, Skype: kidproto
Jabber Chat: <email address hidden>
IRC: <email address hidden>

Inkscape (http://inkscape.org)
Open Clip Art Library (www.openclipart.org)
Creative Commons (www.creativecommons.org)
San Francisco Art Institute (www.sfai.edu)

[Greg Grossmeier (greg.grossmeier)]

David seems to have a good idea on how to implement it.

Has there been any update from Paul on the best way to handle this?

[rejon (rejon)]

David, can you implement this? This is very high priority for Access
Copyright. ETA on this and best direct line of sight to get this done.

Jon

On Tue, 2008-07-15 at 18:57 +0000, Greg Grossmeier wrote:
> David seems to have a good idea on how to implement it.
>
> Has there been any update from Paul on the best way to handle this?
>
--
Jon Phillips
San Francisco, CA + Guangzhou + Beijing
GLOBAL +1.415.830.3884
CHINA +86.1.360.282.8624
<email address hidden>
http://www.rejon.org
IM/skype: kidproto
Jabber: <email address hidden>
IRC: <email address hidden>

[David Strauss (davidstrauss)]

It's mostly an issue of storing the data in an OL book/author record. I think you can just edit the template to do that, but I'll look into it.

----- "rejon" <email address hidden> wrote:

> David, can you implement this? This is very high priority for Access
> Copyright. ETA on this and best direct line of sight to get this
> done.
>
> Jon
>
> On Tue, 2008-07-15 at 18:57 +0000, Greg Grossmeier wrote:
> > David seems to have a good idea on how to implement it.
> >
> > Has there been any update from Paul on the best way to handle this?
> >
> --
> Jon Phillips
> San Francisco, CA + Guangzhou + Beijing
> GLOBAL +1.415.830.3884
> CHINA +86.1.360.282.8624
> <email address hidden>
> http://www.rejon.org
> IM/skype: kidproto
> Jabber: <email address hidden>
> IRC: <email address hidden>
>
> --
> Crown Copyright in ca.py
> https://bugs.launchpad.net/bugs/183052
> You received this bug notification because you are a member of
> Pdregistry.ca-bugteam, which is the registrant for PDregistry.ca.
>
> Status in A public domain registry for Canada.: Confirmed
>
> Bug description:
> If we want to add anything under Crown Copyright, then there needs to
> be a branch for that:
>
> if author = "Crown" then pdyear = pubyear + 50
>
> (translated into actual Python code, of course :-)

[solrize (solrize)]

We can add a boolean field, the main issue is accurately identifying which books should have it, but that would be up to you folks to figure out.

[David Strauss (davidstrauss)]

Or should the boolean field be on the author? It also couldn't be just "crown copyright" because it's not clear how Canadian law would handle crown copyright held by another country's government.

----- "solrize" <email address hidden> wrote:

> We can add a boolean field, the main issue is accurately identifying
> which books should have it, but that would be up to you folks to
> figure
> out.
>
> --
> Crown Copyright in ca.py
> https://bugs.launchpad.net/bugs/183052
> You received this bug notification because you are a member of
> Pdregistry.ca-bugteam, which is the registrant for PDregistry.ca.
>
> Status in A public domain registry for Canada.: Confirmed
>
> Bug description:
> If we want to add anything under Crown Copyright, then there needs to
> be a branch for that:
>
> if author = "Crown" then pdyear = pubyear + 50
>
> (translated into actual Python code, of course :-)

[rejon (rejon)]

Ok, I think we need to have whatever field is set to crown copyright to link
to that this is for canadian determination. David and Solrize, best practice
on that?

Jon

----- Original Message -----
From: "solrize" <email address hidden>
To: <email address hidden>
Sent: Tuesday, July 15, 2008 7:21 PM
Subject: [Bug 183052] Re: Crown Copyright in ca.py

We can add a boolean field, the main issue is accurately identifying
which books should have it, but that would be up to you folks to figure
out.

--
Crown Copyright in ca.py
https://bugs.launchpad.net/bugs/183052
You received this bug notification because you are subscribed to
PDregistry.ca.

[Greg Grossmeier (greg.grossmeier)]

Any update on this? (Crown Copyright special case in pdregistry)

[rejon (rejon)]

need to talk to solrize about today

[rejon (rejon)]

David, please setup a test environment and provide a patch for this to the main codebase. Need asap.

[rejon (rejon)]

URL for test setup/environment enough to fix this up:

http://openlibrary.org/dev/docs/setup

[kaplanmyrth (andy-kaplan-myrth)]

I'm sending an email around about this, but thought I should ping the thread too -- what is the update on this? Has a boolean field been added for government publications?

[Marcus Bornfreund (marcus-lawshare)]

Proposed solution for Crown Copyright in ca.py

Step 1. Tag works as <Crown> Copyrighted by searching for particular terms in dataset <Publisher> field, eg. "Queen's Printer".

Step 2. For copyright determination, where <Author> = <Crown>, follow steps in flowchart. Where <Author> = <Crown> + <Co-Author>, copyright status is determined on basis of <Co-Author> copyright held.

Does this make sense? If so, please implement. Andy, can you please input the Crown + Co-Author step in the PD Flowchart and repost?

Crown Copyright > Published? > Yes > Co-Author? > Yes = go to Joint Authorship, No = PD at publication + 50.

[rejon (rejon)]

Cool, David, can you update the patch to do this please.

Jon

On Wed, 2008-12-10 at 03:00 +0000, Marcus Bornfreund wrote:
> Proposed solution for Crown Copyright in ca.py
>
> Step 1. Tag works as <Crown> Copyrighted by searching for particular
> terms in dataset <Publisher> field, eg. "Queen's Printer".
>
> Step 2. For copyright determination, where <Author> = <Crown>, follow
> steps in flowchart. Where <Author> = <Crown> + <Co-Author>, copyright
> status is determined on basis of <Co-Author> copyright held.
>
> Does this make sense? If so, please implement. Andy, can you please input the Crown + Co-Author step in the PD Flowchart and repost?
>
> Crown Copyright > Published? > Yes > Co-Author? > Yes = go to Joint Authorship, No = PD at publication + 50.
>
--
Jon Phillips
San Francisco + Beijing
GLOBAL +1.415.830.3884
CHINA +86.1.360.282.8624
http://rejon.org
IM/skype: kidproto
Jabber: <email address hidden>

[Marcus Bornfreund (marcus-lawshare)]

From: Savitha Thampi <email address hidden>
Date: December 12, 2008 17:24:52 EST (CA)
To: 'Marcus Bornfreund' <email address hidden>
Subject: Re: Crown Copyright

Hi Marcus,

Thanks for putting up a posting on the Crown Copyright issue to the bug website. I actually want to propose another way to do it:
How about:

“Crown Copyright?”
 If “Yes” then [i.e. at least 1 author = “Queen’s Printer”, etc.]
   “Any Co-Authors?”
    If “Yes” then go to “Pseudonomysly or Anonymously Published?” (and continue as if not a Crown Copyright publication)

 If “No” then
  “Published?” (and it will continue as Crown Copyright is currently set up in the flow chart)
     If “No” then go to “Pseudonymously or Anonymously Published”

Basically, I don't think we should create a new category - if there is a co-author that isn't "Queen's Printer", then it should go back to the default flow chart (as if its not a Crown Publication at all), because the other assessments still need to be worked through.

Do you have any objections to the above flow-through?

Thanks,
Savitha

Savitha Thampi
Associate Legal Counsel
Access Copyright
The Canadian Copyright Licensing Agency
One Yonge Street, Suite 800
Toronto, ON M5E 1E5

Phone: [phone] X 254
Toll Free: 1-800-893-5777 X 254
Fax: [phone]
<email address hidden>

This message, including any attachments, is confidential and may be privileged. If you are not the intended recipient of this message, or have otherwise received this message in error, please notify us immediately by return e-mail and be advised that the use, disclosure or copying of any portion of this message is unauthorized and may be unlawful. Please permanently delete the original message, including any attachments, without making a copy.

[Marcus Bornfreund (marcus-lawshare)]

Good suggestion, Savitha. That makes perfect sense to me. If no one objects, we should consider this bug closed - once the algorithm and flowchart have been updated. David.. Andy.. ?

[rejon (rejon)]

Any luck checking this out on your test setup david? Simple fix.

[kaplanmyrth (andy-kaplan-myrth)]

I've updated the flowchart and attached here the image and the Dia file.

I believe the situation at law is that where a work is coauthored by a
government department and a non-government author, "copyright in the
work shall, subject to any agreement with the author, belong to Her
Majesty" (s. 12 of the Act). The flowchart does not reflect this because
we won't know about those agreements. Instead, it assumes an agreement
is in place so that copyright term will depend on the author, not the
Crown. That's the more conservative way to do this.

Also, Savitha said in the analysis provided said:
> “Any Co-Authors?”
> If “Yes” then go to “Pseudonomysly or Anonymously Published?” (and continue as if not a Crown Copyright publication)
>
> If “No” then
> “Published?” (and it will continue as Crown Copyright is currently set up in the flow chart)
> If “No” then go to “Pseudonymously or Anonymously Published”

I don't think this is quite right. As far as I know, if there are only
Crown authors and the work is unpublished, then it never enters the
public domain in Canada. The attached flowchart reflects this with the
end point "Not in PD".

Any comments on this?

Cheers,
Andy

--
Andy Kaplan-Myrth, M.A., LL.B.
Barrister & Solicitor
------------------------------------------------
email: <email address hidden>
web: http://kaplan-myrth.ca
blog: http://blog.kaplan-myrth.ca
------------------------------------------------

[rejon (rejon)]

Yes, lets talk about on the call.

Jon

On Tue, 2008-12-16 at 18:19 +0000, kaplanmyrth wrote:
> I've updated the flowchart and attached here the image and the Dia file.
>
> I believe the situation at law is that where a work is coauthored by a
> government department and a non-government author, "copyright in the
> work shall, subject to any agreement with the author, belong to Her
> Majesty" (s. 12 of the Act). The flowchart does not reflect this because
> we won't know about those agreements. Instead, it assumes an agreement
> is in place so that copyright term will depend on the author, not the
> Crown. That's the more conservative way to do this.
>
> Also, Savitha said in the analysis provided said:
> > “Any Co-Authors?”
> > If “Yes” then go to “Pseudonomysly or Anonymously Published?” (and continue as if not a Crown Copyright publication)
> >
> > If “No” then
> > “Published?” (and it will continue as Crown Copyright is currently set up in the flow chart)
> > If “No” then go to “Pseudonymously or Anonymously Published”
>
> I don't think this is quite right. As far as I know, if there are only
> Crown authors and the work is unpublished, then it never enters the
> public domain in Canada. The attached flowchart reflects this with the
> end point "Not in PD".
>
> Any comments on this?
>
> Cheers,
> Andy
>
>
> --
> Andy Kaplan-Myrth, M.A., LL.B.
> Barrister & Solicitor
> ------------------------------------------------
> email: <email address hidden>
> web: http://kaplan-myrth.ca
> blog: http://blog.kaplan-myrth.ca
> ------------------------------------------------
>
>
> ** Attachment added: "Public Domain.5.png"
> http://launchpadlibrarian.net/20472598/Public%20Domain.5.png
>
> ** Attachment added: "Public Domain.5.dia"
> http://launchpadlibrarian.net/20472599/Public%20Domain.5.dia
>
--
Jon Phillips
San Francisco + Beijing
GLOBAL +1.415.830.3884
CHINA +86.1.360.282.8624
http://rejon.org
IM/skype: kidproto
Jabber: <email address hidden>

[rejon (rejon)]

Andy, there hasn't been any comments. David, please implement Andy's suggestion. No comments equals we move ahead.

[Marcus (marcus-creativecommons)]

There has been extensive discussion: https://bugs.launchpad.net/
pdregistry.ca/+bug/183052 but consensus was reached and implemented
by Andy into the flowchart - now just needs to be coded.

On 22-Jan-09, at 00:34, rejon wrote:

Andy, there hasn't been any comments. David, please implement Andy's
suggestion. No comments equals we move ahead.

--
Crown Copyright in ca.py
https://bugs.launchpad.net/bugs/183052
You received this bug notification because you are a member of
Pdregistry.ca-bugteam, which is the registrant for PDregistry.ca.

Status in A public domain registry for Canada.: Confirmed

Bug description:
If we want to add anything under Crown Copyright, then there needs to
be a branch for that:

   if author = "Crown" then pdyear = pubyear + 50

(translated into actual Python code, of course :-)

[David Strauss (davidstrauss)]

Here's the crown copyright patch.

One warning I thought about when I reviewed this work today: this patch is consistent with the flowchart in that it completely ignores the crown author if other authors are present. The alternative is to still treat pubyear + 50 as a minimum for the pdyear, even if there are non-Crown authors.

This special case only comes up when maxauthordeath (for non-Crown authors) is defined as less than pubyear, and the pubyear is at least 1999.

For example, the work could have authors Crown and David. If David's death date is 1980 and the work was published in 2000, pdyear would be 1980 + 50 in the current patch. The "alternative" method I posted would result in pdyear 2000 + 50.

[David Strauss (davidstrauss)]

To use the "alternative" method *also* apply this patch (after the one in the previous comment).

[David Strauss (davidstrauss)]

Ignore that "alternative" patch. I somehow included the patch in the patch.

Here's the chunk that needs to change versus 0001-Fix-183052.patch:
    death_year = None
    if aname == 'Crown':
      """We set death_year to pubyear for Crown authorship because items with sole Crown authorship be pdyear >= death_year + 50."""
      death_year = pubyear
      assume("We're assuming this item is under Crown copyright.")
    else:

[kaplanmyrth (andy-kaplan-myrth)]

David Strauss wrote:
> Ignore that "alternative" patch. I somehow included the patch in the
> patch.

So the PDR works for Crown Copyright now?

[David Strauss (davidstrauss)]

I can't apply the patches. I can only post them.

[rejon (rejon)]

We need IA friends to apply...I will talk with Solrize if he is awake.

Jon
--
Jon Phillips
http://rejon.org/

[solrize (solrize)]

Hi, what's up? I tend not to watch these bugs very closely these days. If I'm not responsive on irc then it's best to email me.

[rejon (rejon)]

Anand, can you apply this patch for us please? Thanks...we want to get
into the main tree.

On Mon, Mar 9, 2009 at 4:38 PM, solrize <email address hidden> wrote:
> Hi, what's up?  I tend not to watch these bugs very closely these days.
> If I'm not responsive on irc then it's best to email me.
>
> --
> Crown Copyright in ca.py
> https://bugs.launchpad.net/bugs/183052
> You received this bug notification because you are a direct subscriber
> of the bug.
>

--
Jon Phillips
http://rejon.org/

[Anand Chitipothu (anandology)]

On Tue, Mar 10, 2009 at 2:22 PM, Jon Phillips <email address hidden> wrote:
> Anand, can you apply this patch for us please? Thanks...we want to get
> into the main tree.

main tree has lot of other changes and pdregistry db may not be able
to work with those changes.
Shall I create a branch for pdregistry and add there?

[rejon (rejon)]

Mainly want to get applied to the main tree, and then could apply to
current PDR if not too much trouble...that would be great...are you in
irc? me i am :)

Jon

On Tue, Mar 10, 2009 at 1:58 AM, Anand Chitipothu <email address hidden> wrote:
> On Tue, Mar 10, 2009 at 2:22 PM, Jon Phillips <email address hidden> wrote:
>> Anand, can you apply this patch for us please? Thanks...we want to get
>> into the main tree.
>
> main tree has lot of other changes and pdregistry db may not be able
> to work with those changes.
> Shall I create a branch for pdregistry and add there?
>

--
Jon Phillips
http://rejon.org/

[rejon (rejon)]

Thanks to Anand, this patch is applied and appears to be working. Andy and David you might test this on PDR.

This is applied to our tagged tree: http://github.com/openlibrary/openlibrary/tree/pdregistry.ca and to OL master. Great!