[sage-devel] Graph(....,format='elliptic_curve_congruence')

[Nathann Cohen]

Hello everybody,

It is possible to build a graph with the following syntax:

Graph(something, format="elliptic_curve_congruence")

I will attempt to refactor Graph.__init__ today, and I am a bit
embarassed by the corresponding block of code (~30 lines, at line
1602) in graph.py that deals with elliptic curves (no doctest in
graph.py).

Could this block of code be moved elsewhere, in a file that deals with
elliptic curves?

This Graph.__init__ function is appallingly long, and I will precisely
try to split it into subfunctions.

Thanks,

Nathann

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: The end of stopgaps

[kcrisman]

I will note that sometimes stopgaps are harder to implement than in other 
cases.  For instance, sometimes integrals returned are wrong; but then 
again, they likely always will be (and not just in Sage/Maxima) so it's 
hard to imagine having every single use of integration return a warning. 
 But I may be in the minority in this opinion.  I don't know the context 
from the ticket you refer to, of course.  That said, I think just before a 
stable release is probably the *best* time to add a stopgap, since it's 
unlikely it will be fixed in the last rc :)

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Graph(....,format='elliptic_curve_congruence')

[Nathann Cohen]

P.S.: It seems that this code is not called by *any* part of Sage:

(read|…)~/sage$ grep "elliptic_curve_congruence" . -R
./graphs/graph.py:   -  ``'elliptic_curve_congruence'`` - data must be an
./graphs/graph.py:elif format == 'elliptic_curve_congruence':

No doctest, no other function calls it.

Nathann

On 9 October 2015 at 09:23, Nathann Cohen  wrote:
> Hello everybody,
>
> It is possible to build a graph with the following syntax:
>
> Graph(something, format="elliptic_curve_congruence")
>
> I will attempt to refactor Graph.__init__ today, and I am a bit
> embarassed by the corresponding block of code (~30 lines, at line
> 1602) in graph.py that deals with elliptic curves (no doctest in
> graph.py).
>
> Could this block of code be moved elsewhere, in a file that deals with
> elliptic curves?
>
> This Graph.__init__ function is appallingly long, and I will precisely
> try to split it into subfunctions.
>
> Thanks,
>
> Nathann

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: SSLv3 and Sage 6.8

[Jori Mäntysalo]

On Fri, 14 Aug 2015, kcrisman wrote:

At http://trac.sagemath.org/ticket/17164 it seems that Sage 6.8 has not 
support for vulnerable SSLv3 anymore. But it still seems to have - at 
least on SageNB.



Dima was the one who green lighted that, I assume he can confirm it's not a
problem.


Anything new on this?

--
Jori Mäntysalo

Re: [sage-devel] Re: Graph(....,format='elliptic_curve_congruence')

[John Cremona]

Disclaimer --  I did not wrote that code and did not know it was there!

I agree that having this hidden away in graphs.py is not sensible, and
having it in src/sage/schemes/elliptic_curves/ is more sensible.

How shall we do this?  Would it work to make that a separate ticket
from anything else you are doing, Nathann?

John

On 9 October 2015 at 04:52, Nathann Cohen  wrote:
> P.S.: It seems that this code is not called by *any* part of Sage:
>
> (read|…)~/sage$ grep "elliptic_curve_congruence" . -R
> ./graphs/graph.py:   -  ``'elliptic_curve_congruence'`` - data must be an
> ./graphs/graph.py:elif format == 'elliptic_curve_congruence':
>
> No doctest, no other function calls it.
>
> Nathann
>
> On 9 October 2015 at 09:23, Nathann Cohen  wrote:
>> Hello everybody,
>>
>> It is possible to build a graph with the following syntax:
>>
>> Graph(something, format="elliptic_curve_congruence")
>>
>> I will attempt to refactor Graph.__init__ today, and I am a bit
>> embarassed by the corresponding block of code (~30 lines, at line
>> 1602) in graph.py that deals with elliptic curves (no doctest in
>> graph.py).
>>
>> Could this block of code be moved elsewhere, in a file that deals with
>> elliptic curves?
>>
>> This Graph.__init__ function is appallingly long, and I will precisely
>> try to split it into subfunctions.
>>
>> Thanks,
>>
>> Nathann
>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Graph(....,format='elliptic_curve_congruence')

[Nathann Cohen]

> Disclaimer --  I did not wrote that code and did not know it was there!

Same here :-P

> I agree that having this hidden away in graphs.py is not sensible, and
> having it in src/sage/schemes/elliptic_curves/ is more sensible.
>
> How shall we do this?  Would it work to make that a separate ticket
> from anything else you are doing, Nathann?

Yes of course. The refactoring ticket I opened will already be
sufficiently hard to review.

How do you want to see it moved there? Should we keep the syntax
Graph(,format="elliptic_curve_congruence") or would it be better
as some_object.graph()?

Or will it become an individual function that turns some input into a graph?

It's mostly up to you, given that I do not even understand what this
code is about ^^;

Nathann

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Graph(....,format='elliptic_curve_congruence')

[John Cremona]

On 9 October 2015 at 09:44, Nathann Cohen  wrote:
>> Disclaimer --  I did not wrote that code and did not know it was there!
>
> Same here :-P
>
>> I agree that having this hidden away in graphs.py is not sensible, and
>> having it in src/sage/schemes/elliptic_curves/ is more sensible.
>>
>> How shall we do this?  Would it work to make that a separate ticket
>> from anything else you are doing, Nathann?
>
> Yes of course. The refactoring ticket I opened will already be
> sufficiently hard to review.
>
> How do you want to see it moved there? Should we keep the syntax
> Graph(,format="elliptic_curve_congruence") or would it be better
> as some_object.graph()?

I can do that independently, if it will not cause conflicts with your
ongoing work.

The function's argument is a list of objects (elliptic curves over Q)
rather than one object, so it cannot be a method of some elliptic
curve class but will have to be stand-alone.  Still, it will be more
visible than at present since it will at least appear in the reference
manual in an elliptic curve page where people who might use it might
actually see it!

I will make a ticket for this now.

>
> Or will it become an individual function that turns some input into a graph?
>
> It's mostly up to you, given that I do not even understand what this
> code is about ^^;
>
> Nathann
>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Graph(....,format='elliptic_curve_congruence')

[Nathann Cohen]

> I can do that independently, if it will not cause conflicts with your
> ongoing work.

Works for me. You can base your branch on whatever you are using right
now, and I will rebase it (if needed) atop of my refactoring. There
will probably be a conflict, but very a probably one that is solved
easily.

Thanks,

Nathann

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Graph(....,format='elliptic_curve_congruence')

[John Cremona]

On 9 October 2015 at 10:00, John Cremona  wrote:
> On 9 October 2015 at 09:44, Nathann Cohen  wrote:
>>> Disclaimer --  I did not wrote that code and did not know it was there!
>>
>> Same here :-P
>>
>>> I agree that having this hidden away in graphs.py is not sensible, and
>>> having it in src/sage/schemes/elliptic_curves/ is more sensible.
>>>
>>> How shall we do this?  Would it work to make that a separate ticket
>>> from anything else you are doing, Nathann?
>>
>> Yes of course. The refactoring ticket I opened will already be
>> sufficiently hard to review.
>>
>> How do you want to see it moved there? Should we keep the syntax
>> Graph(,format="elliptic_curve_congruence") or would it be better
>> as some_object.graph()?
>
> I can do that independently, if it will not cause conflicts with your
> ongoing work.
>
> The function's argument is a list of objects (elliptic curves over Q)
> rather than one object, so it cannot be a method of some elliptic
> curve class but will have to be stand-alone.  Still, it will be more
> visible than at present since it will at least appear in the reference
> manual in an elliptic curve page where people who might use it might
> actually see it!
>
> I will make a ticket for this now.

http://trac.sagemath.org/ticket/19382


>
>>
>> Or will it become an individual function that turns some input into a graph?
>>
>> It's mostly up to you, given that I do not even understand what this
>> code is about ^^;
>>
>> Nathann
>>
>> --
>> You received this message because you are subscribed to the Google Groups 
>> "sage-devel" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to sage-devel+unsubscr...@googlegroups.com.
>> To post to this group, send email to sage-devel@googlegroups.com.
>> Visit this group at http://groups.google.com/group/sage-devel.
>> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Fwd: Re: nauty in Sage

[Dima Pasechnik]

On Thursday, 8 October 2015 11:46:34 UTC-7, Jeroen Demeyer wrote:
>
> On 2015-10-08 20:22, Dima Pasechnik wrote: 
> > if it talks about changes, this means changes are allowed, no? 
> Well, in principle I agree, but it's safer not to assume anything. 
>

Brendan replied that remaining two clauses in the license fall under Sect. 
7 of  GPL v3.

Indeed the latter does allow extra clauses to be added, no?

 

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Fwd: nauty in Sage (was: Code generating finite posets of given size (fwd))

[Dima Pasechnik]

-- Forwarded message --
From: Brendan McKay 
Date: 9 October 2015 at 02:45
Subject: Re: nauty in Sage (was: Code generating finite posets of given
size (fwd))
To: Dima Pasechnik 
Cc: Jori Mäntysalo , gunnar.brinkm...@ugent.be,
b...@cs.anu.edu.au


Dear Dima,

Please replace nauty-h.in by the attached version.  Note that it may
only work with the new version that sent you yesterday.

The two remaining caveats in the copyright notice are compatible
with GPLv3 section 7.

Cheers, Brendan.


On 9/10/2015 3:46 PM, Dima Pasechnik wrote:

> Dear Brendan,
>
> On Thu, Oct 08, 2015 at 05:48:08PM +1100, Brendan McKay wrote:
>
>> You can use the source code of nauty in Sage with the words "with
>> the exception of
>> sale for profit or application with nontrivial military
>> significance" removed from the
>> copyright notice.  I take it that you can keep the rest of the
>> copyright notice intact.
>>
> I didn't mean to be dense, but it turns out we'd need to modify it a bit
> more;
> namely, it's not clear on whether modifications are allowed (well, I
> presume
> that by requesting that changes are to be documented you mean that they are
> allowed), and the clause "you must document any changes that you make to
> this
> program" is not compatible with GPL.
>
> For the latter, we can assure you that the developent of Sage makes changes
> and their documentation more or less compulsory (changes are recorded
> in the publicly available git repository, and they are reviewed on a
> publicly readable server trac.sagemath.org.
> That is to say, at least the eventual changes we'd make will be traceable
> and documented.
>
> The easiest for us would be if you chose a license
> from
> http://www.gnu.org/licenses/license-list.en.html#GPLCompatibleLicenses
> e.g. http://directory.fsf.org/wiki/License:ClearBSD
> and released (a version of) nauty under it.
> Although I'd fully understand if you choose not to do so.
>
> Thanks,
> Dima
>
>
> You can take this email as official permission.
>>
>> Perhaps you can use the attached version (not much distributed yet).
>> In particular
>> utilities for files of digraphs are implemented.
>>
>> Cheers, Brendan.
>>
>>
>> On 08/10/15 07:05, Dima Pasechnik wrote:
>>
>>> Dear Brendan,
>>>
>>> Sage(math) has the nauty package up an running,  e.g
>>>
>>> sage: list(digraphs.tournaments_nauty(4))
>>> [Digraph on 4 vertices,
>>> Digraph on 4 vertices,
>>> Digraph on 4 vertices,
>>> Digraph on 4 vertices]
>>>
>>> but we want more: a better integration...
>>> (e.g. the above would only work if the user installed an optional
>>> package 'nauty')
>>>
>>> To bring this up again: the only way we could incorporate nauty
>>> into the main Sage library
>>> is by including your files with the  "exception of sale
>>> for profit or application with nontrivial military significance"
>>> part of the copyright notice removed.
>>> As Sage is released under GPL, we cannot keep it in the source.
>>> (cf. http://www.gnu.org/licenses/gpl-faq.en.html#NoMilitary)
>>>
>>> Did you mean to tell us to remove it if we must?
>>> Or rather you weren't aware that we cannot just put your files
>>> there untouched?
>>>
>>> Thanks,
>>> Dima
>>>
>>>
>>>
>>> On 7 October 2015 at 11:11, Jori Mäntysalo >> > wrote
>>>
>>> -- Forwarded message --
>>> Date: Thu, 30 Oct 2014 12:39:58 +
>>> From: Brendan McKay >> >
>>> To: Jori Mantysalo mailto:jori.mantys...@uta.fi
>>> >>,
>>> Gunnar Brinkmann >> >,
>>> "b...@cs.anu.edu.au "
>>> mailto:b...@cs.anu.edu.au>>
>>> Subject: Re: Code generating finite posets of given size
>>>
>>> Sorry, I forgot to answer.  It applies to all of the nauty files, but
>>> for incorporation into Sage you can ignore it.
>>>
>>> Brendan.
>>>
>>> 
>>> From: Jori Mantysalo >> >
>>> Sent: 30 October 2014 23:00
>>> To: Gunnar Brinkmann; b...@cs.anu.edu.au 
>>> Subject: Re: Code generating finite posets of given size
>>>
>>> On Wed, 8 Oct 2014, Gunnar Brinkmann wrote:
>>>
>>> I am now packaging it. Nauty.h contains copyright with
>>> "exception of sale
>>> for profit or application with nontrivial military
>>> significance." Does this
>>> apply to whole package? If so, I will make this an optional
>>> package, as
>>> core packages are GPL only.
>>>
>>>
>>> this is a question Brendan will answer. Good to hear that you
>>> succeeded
>>> in making it available in Sage.
>>>
>>>
>>> There has now been three weeks since I asked this. No horry, but I just
>>> ping you to be sure that this has not been forgotten.
>>>
>>> --
>>> Jori Mäntysalo
>>>
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails

[sage-devel] Re: nauty in Sage (was: Code generating finite posets of given size (fwd))

[Volker Braun]

Looks great; The two restrictions 

*   * You must not remove this section of the text, containing author   
 *
* attribution, copyright notice, and legal disclaimer. 
  *
*   * You must clearly mark modified versions of this software 
  *
* as differing from the original.   
 *

are explicitly allowed by GPLv3 section 7:

b) Requiring preservation of specified reasonable legal notices or author 
attributions in that material or in the Appropriate Legal Notices displayed 
by works containing it; or
c) Prohibiting misrepresentation of the origin of that material, or 
requiring that modified versions of such material be marked in reasonable 
ways as different from the original version; or



On Friday, October 9, 2015 at 4:40:38 PM UTC+2, Dima Pasechnik wrote:
>
>
> -- Forwarded message --
> From: Brendan McKay >
> Date: 9 October 2015 at 02:45
> Subject: Re: nauty in Sage (was: Code generating finite posets of given 
> size (fwd))
> To: Dima Pasechnik >
> Cc: Jori Mäntysalo >, gunnar.b...@ugent.be 
> , b...@cs.anu.edu.au 
>
>
> Dear Dima,
>
> Please replace nauty-h.in by the attached version.  Note that it may
> only work with the new version that sent you yesterday.
>
> The two remaining caveats in the copyright notice are compatible
> with GPLv3 section 7.
>
> Cheers, Brendan.
>
>
> On 9/10/2015 3:46 PM, Dima Pasechnik wrote:
>
>> Dear Brendan,
>>
>> On Thu, Oct 08, 2015 at 05:48:08PM +1100, Brendan McKay wrote:
>>
>>> You can use the source code of nauty in Sage with the words "with
>>> the exception of
>>> sale for profit or application with nontrivial military
>>> significance" removed from the
>>> copyright notice.  I take it that you can keep the rest of the
>>> copyright notice intact.
>>>
>> I didn't mean to be dense, but it turns out we'd need to modify it a bit 
>> more;
>> namely, it's not clear on whether modifications are allowed (well, I 
>> presume
>> that by requesting that changes are to be documented you mean that they 
>> are
>> allowed), and the clause "you must document any changes that you make to 
>> this
>> program" is not compatible with GPL.
>>
>> For the latter, we can assure you that the developent of Sage makes 
>> changes
>> and their documentation more or less compulsory (changes are recorded
>> in the publicly available git repository, and they are reviewed on a
>> publicly readable server trac.sagemath.org.
>> That is to say, at least the eventual changes we'd make will be traceable
>> and documented.
>>
>> The easiest for us would be if you chose a license
>> from 
>> http://www.gnu.org/licenses/license-list.en.html#GPLCompatibleLicenses
>> e.g. http://directory.fsf.org/wiki/License:ClearBSD
>> and released (a version of) nauty under it.
>> Although I'd fully understand if you choose not to do so.
>>
>> Thanks,
>> Dima
>>
>>
>> You can take this email as official permission.
>>>
>>> Perhaps you can use the attached version (not much distributed yet).
>>> In particular
>>> utilities for files of digraphs are implemented.
>>>
>>> Cheers, Brendan.
>>>
>>>
>>> On 08/10/15 07:05, Dima Pasechnik wrote:
>>>
 Dear Brendan,

 Sage(math) has the nauty package up an running,  e.g

 sage: list(digraphs.tournaments_nauty(4))
 [Digraph on 4 vertices,
 Digraph on 4 vertices,
 Digraph on 4 vertices,
 Digraph on 4 vertices]

 but we want more: a better integration...
 (e.g. the above would only work if the user installed an optional
 package 'nauty')

 To bring this up again: the only way we could incorporate nauty
 into the main Sage library
 is by including your files with the  "exception of sale
 for profit or application with nontrivial military significance"
 part of the copyright notice removed.
 As Sage is released under GPL, we cannot keep it in the source.
 (cf. http://www.gnu.org/licenses/gpl-faq.en.html#NoMilitary)

 Did you mean to tell us to remove it if we must?
 Or rather you weren't aware that we cannot just put your files
 there untouched?

 Thanks,
 Dima



 On 7 October 2015 at 11:11, Jori Mäntysalo >>> 
 > wrote

 -- Forwarded message --
 Date: Thu, 30 Oct 2014 12:39:58 +
 From: Brendan McKay 
 >
 To: Jori Mantysalo  >>> jori.ma...@uta.fi >>,
 Gunnar Brinkmann 
 >,
 "b...@cs.anu.edu.au  "
  >
 Subject: Re: Code generating finite posets of given size

 Sorry, I forgot to answer.  It applies to all of the nauty files, but
 for incorporation into Sage you can ignore it.

 Brendan.

 
 From: Jori Mantysalo 
 >
 Sent: 30 Octobe

Re: [sage-devel] Fwd: Re: nauty in Sage

['Julien Puydt' via sage-devel]

Hi,

Le vendredi 09 oct. 2015 ą 07:30:06 (-0700), Dima Pasechnik a écrit :
> On Thursday, 8 October 2015 11:46:34 UTC-7, Jeroen Demeyer wrote:
> >
> > On 2015-10-08 20:22, Dima Pasechnik wrote: 
> > > if it talks about changes, this means changes are allowed, no? 
> > Well, in principle I agree, but it's safer not to assume anything. 
> >
> 
> Brendan replied that remaining two clauses in the license fall under Sect. 
> 7 of  GPL v3.
> 
> Indeed the latter does allow extra clauses to be added, no?
> 

Hmmm... will he publish a full version with this copyright notice? Because
if he does, then distributions could start packaging it.

Snark on #sagemath

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: nauty in Sage (was: Code generating finite posets of given size (fwd))

[Dima Pasechnik]

On Friday, 9 October 2015 07:44:55 UTC-7, Volker Braun wrote:
>
> Looks great; The two restrictions 
>
> *   * You must not remove this section of the text, containing author 
>*
> * attribution, copyright notice, and legal disclaimer. 
>   *
> *   * You must clearly mark modified versions of this software 
>   *
> * as differing from the original. 
>*
>
> are explicitly allowed by GPLv3 section 7:
>
> b) Requiring preservation of specified reasonable legal notices or author 
> attributions in that material or in the Appropriate Legal Notices displayed 
> by works containing it; or
> c) Prohibiting misrepresentation of the origin of that material, or 
> requiring that modified versions of such material be marked in reasonable 
> ways as different from the original version; or
>
>
> well, we rather talk about having the following copyright:

nauty is copyright (1984-2013) Brendan McKay. Traces is copyright 
(2008-2013) Adolfo Piperno. All rights reserved. You must not remove this 
copyright notice, and you must document any changes that you make to this 
program. This software is subject to this copyright only, irrespective of 
any copyright attached to any package of which this is a part. 
 

>
> On Friday, October 9, 2015 at 4:40:38 PM UTC+2, Dima Pasechnik wrote:
>>
>>
>> -- Forwarded message --
>> From: Brendan McKay 
>> Date: 9 October 2015 at 02:45
>> Subject: Re: nauty in Sage (was: Code generating finite posets of given 
>> size (fwd))
>> To: Dima Pasechnik 
>> Cc: Jori Mäntysalo , gunnar.b...@ugent.be, 
>> b...@cs.anu.edu.au
>>
>>
>> Dear Dima,
>>
>> Please replace nauty-h.in by the attached version.  Note that it may
>> only work with the new version that sent you yesterday.
>>
>> The two remaining caveats in the copyright notice are compatible
>> with GPLv3 section 7.
>>
>> Cheers, Brendan.
>>
>>
>> On 9/10/2015 3:46 PM, Dima Pasechnik wrote:
>>
>>> Dear Brendan,
>>>
>>> On Thu, Oct 08, 2015 at 05:48:08PM +1100, Brendan McKay wrote:
>>>
 You can use the source code of nauty in Sage with the words "with
 the exception of
 sale for profit or application with nontrivial military
 significance" removed from the
 copyright notice.  I take it that you can keep the rest of the
 copyright notice intact.

>>> I didn't mean to be dense, but it turns out we'd need to modify it a bit 
>>> more;
>>> namely, it's not clear on whether modifications are allowed (well, I 
>>> presume
>>> that by requesting that changes are to be documented you mean that they 
>>> are
>>> allowed), and the clause "you must document any changes that you make to 
>>> this
>>> program" is not compatible with GPL.
>>>
>>> For the latter, we can assure you that the developent of Sage makes 
>>> changes
>>> and their documentation more or less compulsory (changes are recorded
>>> in the publicly available git repository, and they are reviewed on a
>>> publicly readable server trac.sagemath.org.
>>> That is to say, at least the eventual changes we'd make will be traceable
>>> and documented.
>>>
>>> The easiest for us would be if you chose a license
>>> from 
>>> http://www.gnu.org/licenses/license-list.en.html#GPLCompatibleLicenses
>>> e.g. http://directory.fsf.org/wiki/License:ClearBSD
>>> and released (a version of) nauty under it.
>>> Although I'd fully understand if you choose not to do so.
>>>
>>> Thanks,
>>> Dima
>>>
>>>
>>> You can take this email as official permission.

 Perhaps you can use the attached version (not much distributed yet).
 In particular
 utilities for files of digraphs are implemented.

 Cheers, Brendan.


 On 08/10/15 07:05, Dima Pasechnik wrote:

> Dear Brendan,
>
> Sage(math) has the nauty package up an running,  e.g
>
> sage: list(digraphs.tournaments_nauty(4))
> [Digraph on 4 vertices,
> Digraph on 4 vertices,
> Digraph on 4 vertices,
> Digraph on 4 vertices]
>
> but we want more: a better integration...
> (e.g. the above would only work if the user installed an optional
> package 'nauty')
>
> To bring this up again: the only way we could incorporate nauty
> into the main Sage library
> is by including your files with the  "exception of sale
> for profit or application with nontrivial military significance"
> part of the copyright notice removed.
> As Sage is released under GPL, we cannot keep it in the source.
> (cf. http://www.gnu.org/licenses/gpl-faq.en.html#NoMilitary)
>
> Did you mean to tell us to remove it if we must?
> Or rather you weren't aware that we cannot just put your files
> there untouched?
>
> Thanks,
> Dima
>
>
>
> On 7 October 2015 at 11:11, Jori Mäntysalo  > wrote
>
> -- Forwarded message --
> Date: Thu, 30 Oct 2014 12:39:58 +
>

Re: [sage-devel] Fwd: Re: nauty in Sage

[Dima Pasechnik]

On Friday, 9 October 2015 09:12:51 UTC-7, Snark wrote:
>
> Hi, 
>
> Le vendredi 09 oct. 2015 ą 07:30:06 (-0700), Dima Pasechnik a écrit : 
> > On Thursday, 8 October 2015 11:46:34 UTC-7, Jeroen Demeyer wrote: 
> > > 
> > > On 2015-10-08 20:22, Dima Pasechnik wrote: 
> > > > if it talks about changes, this means changes are allowed, no? 
> > > Well, in principle I agree, but it's safer not to assume anything. 
> > > 
> > 
> > Brendan replied that remaining two clauses in the license fall under 
> Sect. 
> > 7 of  GPL v3. 
> > 
> > Indeed the latter does allow extra clauses to be added, no? 
> > 
>
> Hmmm... will he publish a full version with this copyright notice? Because 
> if he does, then distributions could start packaging it. 
>

we talk about nauty being included into Sage - the permission is being 
given for this, and only this, AFAIU...
 

>
> Snark on #sagemath 
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Fwd: Re: nauty in Sage

[Nathann Cohen]

> we talk about nauty being included into Sage - the permission is being given
> for this, and only this, AFAIU...

I don't think that " + must only be used in Sage" is GPL-compatible.

Nathann

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Fwd: Re: nauty in Sage

[Dima Pasechnik]

On Friday, 9 October 2015 09:26:48 UTC-7, Nathann Cohen wrote:
>
> > we talk about nauty being included into Sage - the permission is being 
> given 
> > for this, and only this, AFAIU... 
>
> I don't think that " + must only be used in Sage" is 
> GPL-compatible. 
>

As long we we do not write this in... 
I suppose I should make it clear for Brendan that someone will be able to 
create a Sage fork consisting of nauty 
under essentially GPL.



> Nathann 
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: nauty in Sage (was: Code generating finite posets of given size (fwd))

[William Stein]

On Fri, Oct 9, 2015 at 9:19 AM, Dima Pasechnik  wrote:
>
>
> On Friday, 9 October 2015 07:44:55 UTC-7, Volker Braun wrote:
>>
>> Looks great; The two restrictions
>>
>> *   * You must not remove this section of the text, containing author
>> *
>> * attribution, copyright notice, and legal disclaimer.
>> *
>> *   * You must clearly mark modified versions of this software
>> *
>> * as differing from the original.
>> *
>>
>> are explicitly allowed by GPLv3 section 7:
>>
>> b) Requiring preservation of specified reasonable legal notices or author
>> attributions in that material or in the Appropriate Legal Notices displayed
>> by works containing it; or
>> c) Prohibiting misrepresentation of the origin of that material, or
>> requiring that modified versions of such material be marked in reasonable
>> ways as different from the original version; or
>>
>>
> well, we rather talk about having the following copyright:
>
> nauty is copyright (1984-2013) Brendan McKay. Traces is copyright
> (2008-2013) Adolfo Piperno. All rights reserved. You must not remove this
> copyright notice, and you must document any changes that you make to this
> program. This software is subject to this copyright only, irrespective of
> any copyright attached to any package of which this is a part.

Unfortunately this license doesn't even grant permission to use the
software.   For example, the BSD license starts: "Redistribution and
use in source and binary forms are permitted ...".   By default,
without an explicit *grant* of rights, all rights are reserved.   In
fact, the copyright notice above very explicitly states that "all
rights reserved", which means: "It indicates that the copyright holder
reserves, or holds for their own use, all the rights provided by
copyright law."

I hope you can convince Brendan to pick a standard license, or just
guide him into choosing BSD 2-clause:

   https://opensource.org/licenses/BSD-2-Clause

It's basically what he wants, except for "you must document any
changes that you make".   I still have absolutely no idea what that
statement is supposed to mean.   However, it's either trivial (due to
revision control), so should not be included, or it actually means
something, so it is GPL incompatible.

During the first few years of Sage I got around 10 projects to change
their licenses.  I never ever considered having any of them make up a
new license.  That's a difficult path, since it is copyright law and
we are not lawyers.

 -- William



-- 
William (http://wstein.org)

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: nauty in Sage (was: Code generating finite posets of given size (fwd))

[Volker Braun]

On Friday, October 9, 2015 at 7:08:19 PM UTC+2, William wrote:
>
> the copyright notice above very explicitly states that "all 
> rights reserved"


As you said, thats a no-op: 
http://www.iusmentis.com/copyright/allrightsreserved/

The modified nauty-h.in does give permission to everything except the two 
clauses. The two clauses are fine with the GPLv3 so its obviously 
compatible.

Since clearly nobody read it ;-) here is the license part of the new 
nauty-h.in:

/*
*   
 *
*AUTHOR: Brendan D. McKay   
 *
*Research School of Computer Science   
  *
*Australian National University 
 *
*Canberra, ACT 2601, Australia 
  *
*phone:  +61 2 6125 3845   
  *
*email:  Brendan.McKay@xxx*
*   
 *
*   Nauty is copyright (1984-2015) Brendan McKay.  All rights reserved. 
 *
*   Permission is hereby given for the use, distribution and modification   
 *
*   of this software subject to the following. 
  *
*   * You must not remove this section of the text, containing author   
 *
* attribution, copyright notice, and legal disclaimer. 
  *
*   * You must clearly mark modified versions of this software 
  *
* as differing from the original.   
 *
*   
 *
*   This program is only provided "as is".  No responsibility will be taken 
 *
*   by the author, his employer or his pet rabbit* for any misfortune which 
 *
*   befalls you because of its use.  I don't think it will delete all your 
  *
*   files, burn down your computer room or turn your children against you, 
  *
*   but if it does: stiff cheddar.  On the other hand, I very much welcome 
  *
*   bug reports, or at least I would if there were any bugs.   
  *
*   * RIP, 1989 
 *
*   Traces is copyright Adolfo Piperno (2011-). 
 *
*   
 *
**

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: nauty in Sage (was: Code generating finite posets of given size (fwd))

[William Stein]

On Fri, Oct 9, 2015 at 10:36 AM, Volker Braun  wrote:
> On Friday, October 9, 2015 at 7:08:19 PM UTC+2, William wrote:
>>
>> the copyright notice above very explicitly states that "all
>> rights reserved"
>
>
> As you said, thats a no-op:
> http://www.iusmentis.com/copyright/allrightsreserved/
>
> The modified nauty-h.in does give permission to everything except the two
> clauses. The two clauses are fine with the GPLv3 so its obviously
> compatible.
>

+1  Nice -- and thanks for posting the actual license!

William

> Since clearly nobody read it ;-) here is the license part of the new
> nauty-h.in:
>
> /*
> *
> *
> *AUTHOR: Brendan D. McKay
> *
> *Research School of Computer Science
> *
> *Australian National University
> *
> *Canberra, ACT 2601, Australia
> *
> *phone:  +61 2 6125 3845
> *
> *email:  Brendan.McKay@xxx*
> *
> *
> *   Nauty is copyright (1984-2015) Brendan McKay.  All rights reserved.
> *
> *   Permission is hereby given for the use, distribution and modification
> *
> *   of this software subject to the following.
> *
> *   * You must not remove this section of the text, containing author
> *
> * attribution, copyright notice, and legal disclaimer.
> *
> *   * You must clearly mark modified versions of this software
> *
> * as differing from the original.
> *
> *
> *
> *   This program is only provided "as is".  No responsibility will be taken
> *
> *   by the author, his employer or his pet rabbit* for any misfortune which
> *
> *   befalls you because of its use.  I don't think it will delete all your
> *
> *   files, burn down your computer room or turn your children against you,
> *
> *   but if it does: stiff cheddar.  On the other hand, I very much welcome
> *
> *   bug reports, or at least I would if there were any bugs.
> *
> *   * RIP, 1989
> *
> *   Traces is copyright Adolfo Piperno (2011-).
> *
> *
> *
> **
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/d/optout.



-- 
William (http://wstein.org)

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Fwd: Re: nauty in Sage

['Julien Puydt' via sage-devel]

Le vendredi 09 oct. 2015 à 09:43:31 (-0700), Dima Pasechnik a écrit :
>
> On Friday, 9 October 2015 09:26:48 UTC-7, Nathann Cohen wrote:
> >
> > > we talk about nauty being included into Sage - the permission is being 
> > given 
> > > for this, and only this, AFAIU... 
> >
> > I don't think that " + must only be used in Sage" is 
> > GPL-compatible. 
> >
> 
> As long we we do not write this in... 
> I suppose I should make it clear for Brendan that someone will be able to 
> create a Sage fork consisting of nauty 
> under essentially GPL.

William had a point when he said : stick to a known license, don't try
to invent your own.

The author might actually be interested by BSD-3-clause, in fact.

Snark on #sagemath

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: nauty in Sage (was: Code generating finite posets of given size (fwd))

[Dima Pasechnik]

On Friday, 9 October 2015 10:36:35 UTC-7, Volker Braun wrote:
>
> On Friday, October 9, 2015 at 7:08:19 PM UTC+2, William wrote:
>>
>> the copyright notice above very explicitly states that "all 
>> rights reserved"
>
>
> As you said, thats a no-op: 
> http://www.iusmentis.com/copyright/allrightsreserved/
>
> The modified nauty-h.in does give permission to everything except the two 
> clauses. The two clauses are fine with the GPLv3 so its obviously 
> compatible.
>
> Since clearly nobody read it ;-) here is the license part of the new 
> nauty-h.in:
>

thanks, Volker, for scrolling down the file (I only looked at the beginning 
- stiff cheddar :-)...)
It didn't occur to me it contains an amended copyright notice.

Good to go? I suppose we need a formal vote to make the package standard, 
right?



>
> /*
> * 
>*
> *AUTHOR: Brendan D. McKay 
>*
> *Research School of Computer Science   
>   *
> *Australian National University   
>*
> *Canberra, ACT 2601, Australia 
>   *
> *phone:  +61 2 6125 3845   
>   *
> *email:  Brendan.McKay@xxx*
> * 
>*
> *   Nauty is copyright (1984-2015) Brendan McKay.  All rights reserved.   
>*
> *   Permission is hereby given for the use, distribution and modification 
>*
> *   of this software subject to the following. 
>   *
> *   * You must not remove this section of the text, containing author 
>*
> * attribution, copyright notice, and legal disclaimer. 
>   *
> *   * You must clearly mark modified versions of this software 
>   *
> * as differing from the original. 
>*
> * 
>*
> *   This program is only provided "as is".  No responsibility will be 
> taken  *
> *   by the author, his employer or his pet rabbit* for any misfortune 
> which  *
> *   befalls you because of its use.  I don't think it will delete all your 
>   *
> *   files, burn down your computer room or turn your children against you, 
>   *
> *   but if it does: stiff cheddar.  On the other hand, I very much welcome 
>   *
> *   bug reports, or at least I would if there were any bugs.   
>   *
> *   * RIP, 1989   
>*
> *   Traces is copyright Adolfo Piperno (2011-).   
>*
> * 
>*
>
> **
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: The end of stopgaps

[Jakob Kroeker]

At least a user could look at the list of bugs silently producing wrong 
answers:

http://trac.sagemath.org/report/79

I updated the list by manually looking at open tickets from #1 to #13345 
and adding a stopgap marker 'todo' if required.

Remaining task (I will do that): 
check open tickets #13346 to #19383 for issues silently producing wrong 
answers
and add them to the list http://trac.sagemath.org/report/79 


Jakob


Am Donnerstag, 8. Oktober 2015 19:05:23 UTC+2 schrieb Nathann Cohen:
>
> Just to keep everybody aware of what is going on here: 
>
> - I tried to add a stopgap on a code that returns wrong results 
>
> - The stopgap was refused because it was "just before a stable 
> release" and people did not want to alarm the users by telling them 
> that Sage returns wrong results (even though it does) 
>
> - Jeroen noticed that if we add the stopgap now, the same thing may 
> happen for the next release cycle (as we are not sure that it will be 
> fixed in the meantime) 
>
> - Thus, they now decide that we will not add the stopgap at all. Ever. 
>
> So, no stopgap. Let us not tell the users that we return wrong 
> results. Let them deal with the consequences. 
>
> This enlightening conversation showing how trying to preserve our 
> public image can ruin a good software is available there: 
>
> http://trac.sagemath.org/ticket/19302#comment:27 
>
> Nathann 
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: The end of stopgaps

[Nils Bruin]

On Friday, October 9, 2015 at 1:29:14 AM UTC-7, kcrisman wrote:
>
> I think just before a stable release is probably the *best* time to add a 
> stopgap, since it's unlikely it will be fixed in the last rc :)
>

Indeed, making an invasive change in an rc which probably causes widespread 
and hard to predict changes in behaviour (there is a lot of behaviour in 
sage that depends on enumeration order of dictionaries and hence on exact 
hash values) is probably not a smart move.

A stopgap message should have some diagnostic value, however. In this case, 
the message would probably be triggered quite early on, in an unavoidable 
case, where the current hash is fine to use. A subsequent use of the hash 
in a case where its use should really be disallowed would remain unmarked. 
We might as well add to the banner "disclaimer: sage still has bugs".

That said, Vincent and I are basically done with removing the default hash 
(we identified the problem with the last doctest), so I expect 6.10 will 
not have this problem anymore--unless unexpected problems surface. It's 
nice that people care about proper procedure to let sage "do the right 
thing", but it's worth keeping in mind that effort spent on procedure is 
effort not spent on actual progress.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.