Re: [sage-devel] Re: Sage's references: new policy?

[Thierry]

Hi,

On Thu, Sep 22, 2016 at 05:48:15AM -0700, Ralf Stephan wrote:
> On Wednesday, September 21, 2016 at 1:03:27 AM UTC+2, John H Palmieri wrote:
> >
> > - all references should be, insofar as possible, in a standard form
> >
> 
> There is one standard form for everything, even old papers, the Google 
> Scholar cluster link.
> Guess which one will survive longer: Google or all the other schemes?

By far the other schemes, see e.g.
https://en.wikipedia.org/wiki/List_of_Google_products#Discontinued_products_and_services

We already suffered from the end of google-id support, by chance we didn't
rely on google-code, let us just hope that google-group will not end soon,
and let us not depend on yet-another closed stuff, which is not even a
bibliographical format (unless i missed something from the links).

Ciao,
Thierry

 
> Example author link: https://scholar.google.com/citations?user=M2Ky9OkJ
> Example paper cluster 
> link: 
> https://scholar.google.com/scholar?oi=bibs=en=1802735033823745065
> 
>  
> 
> -- 
> 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 https://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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[Johan S . H . Rosenkilde]

>>> As discussed in another thread [1]_ on sage-devel recently, I propose 
>>> changing our policy toward references:
>>>
>>> - all references should be put into a master bibliography file
>>>
>>
>> There is one significant drawback to this: it will mean that a lot of 
>> ticket branches will be modifying this file, so merge conflicts between 
>> tickets may become more prevalent. If we can do something to ensure that 
>> resolving these merge conflicts is likely to fall within what standard 
>> merge strategies can do automatically we should probably do that.

Will it really be that bad? The proposed master bibliography is sorted
alphabetically by first author, so conflicts should only occur when two
tickets insert/modify citations right next to each other (or with
perhaps 1 citation between them). With >1000 references right now,
that's not going to cause too many extra conflicts, I think:

If there's 1000 "equidistant" references and a single release has
tickets that create 10 new random references (this is high, I think),
then there's roughly 20% chance that 1 pair of these references will be
within 1 of each other in the existing reference list.

Dima Pasechnik writes:
> I agree - however, perhaps it's better to think of using several bibtex 
> files (which is perfectly possible in LaTeX); e.g. one for number theory, 
> one for group theory, one for the commutative algebra, etc.

Hmm, that seems complicated: Wouldn't many references naturally fall
into multiple such categories, so every time you want to add a reference
you would have to grep for it across all files.

Best,
Johan

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Sage's references: new policy?

[Dima Pasechnik]

On Thursday, September 22, 2016 at 4:32:11 AM UTC, Nils Bruin wrote:
>
> On Tuesday, September 20, 2016 at 4:03:27 PM UTC-7, John H Palmieri wrote:
>>
>> As discussed in another thread [1]_ on sage-devel recently, I propose 
>> changing our policy toward references:
>>
>> - all references should be put into a master bibliography file
>>
>
> There is one significant drawback to this: it will mean that a lot of 
> ticket branches will be modifying this file, so merge conflicts between 
> tickets may become more prevalent. If we can do something to ensure that 
> resolving these merge conflicts is likely to fall within what standard 
> merge strategies can do automatically we should probably do that.
>

I agree - however, perhaps it's better to think of using several bibtex 
files (which is perfectly possible in LaTeX); e.g. one for number theory, 
one for group theory, one for the commutative algebra, etc.

 

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Sage's references: new policy?

[Ralf Stephan]

On Wednesday, September 21, 2016 at 1:03:27 AM UTC+2, John H Palmieri wrote:
>
> - all references should be, insofar as possible, in a standard form
>

There is one standard form for everything, even old papers, the Google 
Scholar cluster link.
Guess which one will survive longer: Google or all the other schemes?

Example author link: https://scholar.google.com/citations?user=M2Ky9OkJ
Example paper cluster 
link: 
https://scholar.google.com/scholar?oi=bibs=en=1802735033823745065

 

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Sage's references: new policy?

[Eric Gourgoulhon]

Hi John,

Le mercredi 21 septembre 2016 01:03:27 UTC+2, John H Palmieri a écrit :
>
> As discussed in another thread [1]_ on sage-devel recently, I propose 
> changing our policy toward references:
>
> - all references should be put into a master bibliography file, and
>
 
What about the pdf documentation? At present, the reference manual at 
http://doc.sagemath.org/ is split in many separate pdf files and each of 
them has its bibliographic references at the end. Will the change to a 
single master file preserve this?

Best regards,

Eric.

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Sage's references: new policy?

[Nils Bruin]

On Tuesday, September 20, 2016 at 4:03:27 PM UTC-7, John H Palmieri wrote:
>
> As discussed in another thread [1]_ on sage-devel recently, I propose 
> changing our policy toward references:
>
> - all references should be put into a master bibliography file
>

There is one significant drawback to this: it will mean that a lot of 
ticket branches will be modifying this file, so merge conflicts between 
tickets may become more prevalent. If we can do something to ensure that 
resolving these merge conflicts is likely to fall within what standard 
merge strategies can do automatically we should probably do that.

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[John H Palmieri]

On Wednesday, September 21, 2016 at 9:39:28 AM UTC-7, Salvatore Stella 
wrote:
>
> * Thierry  [2016-09-21 
> 18:35:25]: 
>
> >Hi, 
> > 
> >bikeshedding for bikeshedding: 
> > 
> >- if we decide to centralize everything in a single file (but we should 
> be 
> >  aware that a backward move (e.g. for modularization) will require some 
> >  work), why not using bibtex (there must be some sphinx interface 
> >  somewhere), to that we keep all information with proper fields (might 
> >  also be good for pdf rendering) ? 
>
> Et Voilà: 
> https://sphinxcontrib-bibtex.readthedocs.io/en/latest/ 
>
>
That looks interesting. If/when we switch to a single bibliography file, 
then we could later switch to using this interface. Converting a single 
ReST bibliography file to a bibtex file would be painful but not that hard, 
and then changing all references throughout Sage from [ABC1999]_ to 
:cite:`ABC1999` could be done by a script.

-- 
John

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[Travis Scrimshaw]

On Wednesday, September 21, 2016 at 11:35:30 AM UTC-5, Thierry 
(sage-googlesucks@xxx) wrote:
>
> Hi, 
>
> bikeshedding for bikeshedding: 
>
> - if we decide to centralize everything in a single file (but we should be 
>   aware that a backward move (e.g. for modularization) will require some 
>   work), why not using bibtex (there must be some sphinx interface 
>   somewhere), to that we keep all information with proper fields (might 
>   also be good for pdf rendering) ? 
>
> - regarding the citation link, explicit is better than implicit, avoids 
>   collisions, and is not that verbose: [Milnor1958], [AuthorCoauthor2016], 
> ... 
>
>  At what point do you stop (i.e., how many authors or characters?), and 
what do you switch to? What about those really long names?

Generally speaking, [AC2016] will very likely not have any collisions and 
it saves on overall space, which is part of the reasons why we have 
citations to references (it's a fallacy, but why not just put the full 
citation in every place, be super explicit?).

Best,
Travis

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[VulK]

* Thierry  [2016-09-21 18:35:25]:


Hi,

bikeshedding for bikeshedding:

- if we decide to centralize everything in a single file (but we should be
 aware that a backward move (e.g. for modularization) will require some
 work), why not using bibtex (there must be some sphinx interface
 somewhere), to that we keep all information with proper fields (might
 also be good for pdf rendering) ?


Et Voilà:
https://sphinxcontrib-bibtex.readthedocs.io/en/latest/



- regarding the citation link, explicit is better than implicit, avoids
 collisions, and is not that verbose: [Milnor1958], [AuthorCoauthor2016], ...

My two cents,
Thierry



On Wed, Sep 21, 2016 at 08:46:29AM -0700, John H Palmieri wrote:

There may be two issues here.

- How should references be written in source code?
- How should references appear in documentation output?

The default behavior in Sphinx is to use the source code citation name also
in the output. I don't know how hard it would be to change that.

We can have discussions about the best way to format references purely in
the documentation output, and I think it is clear that we will not come to
universal agreement. More importantly, any discussion strictly about the
documentation output (for example, using [1], [2], ... -- no one is
suggesting that this is how the references should be named in the source
code, right?) is orthogonal to the issue at hand: anyone can work on
modifying Sphinx so it formats the references in another way independently
of the format in the source code. Feel free to do that and propose such a
change here. For now, the discussion should be on how to format code in the
source (= the format in the output for now, because that is Sphinx's
behavior).

So we can discuss the best way to format references in the source code. To
some extent, of course, this is bikeshedding. Whether we use [AC2016] or
[MR234898349] or [doi:...] or something else, there will always be
arguments for doing one of the others. I personally find [Mil1958] in a
discussion of the Steenrod algebra to convey information: I know that it
refers to Milnor's 1958 paper. I would not recognize the MR number or the
doi number for this. So I personally find the format [AC2016] a good
balance between readability, brevity, and (to a large extent) unique
representation. (Also, my suggested usage would be to often include more
information than just the citation name: "In [Mil1958], Milnor showed ..."
or "Milnor showed that ... -- see [Mil1958]" or something similar. Again,
there is a balance between readability and verbosity.)

--
John

--
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 https://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 https://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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[Thierry]

Hi,

bikeshedding for bikeshedding:

- if we decide to centralize everything in a single file (but we should be
  aware that a backward move (e.g. for modularization) will require some
  work), why not using bibtex (there must be some sphinx interface
  somewhere), to that we keep all information with proper fields (might
  also be good for pdf rendering) ?

- regarding the citation link, explicit is better than implicit, avoids
  collisions, and is not that verbose: [Milnor1958], [AuthorCoauthor2016], ...

My two cents,
Thierry



On Wed, Sep 21, 2016 at 08:46:29AM -0700, John H Palmieri wrote:
> There may be two issues here.
> 
> - How should references be written in source code?
> - How should references appear in documentation output?
> 
> The default behavior in Sphinx is to use the source code citation name also 
> in the output. I don't know how hard it would be to change that.
> 
> We can have discussions about the best way to format references purely in 
> the documentation output, and I think it is clear that we will not come to 
> universal agreement. More importantly, any discussion strictly about the 
> documentation output (for example, using [1], [2], ... -- no one is 
> suggesting that this is how the references should be named in the source 
> code, right?) is orthogonal to the issue at hand: anyone can work on 
> modifying Sphinx so it formats the references in another way independently 
> of the format in the source code. Feel free to do that and propose such a 
> change here. For now, the discussion should be on how to format code in the 
> source (= the format in the output for now, because that is Sphinx's 
> behavior).
> 
> So we can discuss the best way to format references in the source code. To 
> some extent, of course, this is bikeshedding. Whether we use [AC2016] or 
> [MR234898349] or [doi:...] or something else, there will always be 
> arguments for doing one of the others. I personally find [Mil1958] in a 
> discussion of the Steenrod algebra to convey information: I know that it 
> refers to Milnor's 1958 paper. I would not recognize the MR number or the 
> doi number for this. So I personally find the format [AC2016] a good 
> balance between readability, brevity, and (to a large extent) unique 
> representation. (Also, my suggested usage would be to often include more 
> information than just the citation name: "In [Mil1958], Milnor showed ..." 
> or "Milnor showed that ... -- see [Mil1958]" or something similar. Again, 
> there is a balance between readability and verbosity.)
> 
> -- 
> John
> 
> -- 
> 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 https://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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[John H Palmieri]

There may be two issues here.

- How should references be written in source code?
- How should references appear in documentation output?

The default behavior in Sphinx is to use the source code citation name also 
in the output. I don't know how hard it would be to change that.

We can have discussions about the best way to format references purely in 
the documentation output, and I think it is clear that we will not come to 
universal agreement. More importantly, any discussion strictly about the 
documentation output (for example, using [1], [2], ... -- no one is 
suggesting that this is how the references should be named in the source 
code, right?) is orthogonal to the issue at hand: anyone can work on 
modifying Sphinx so it formats the references in another way independently 
of the format in the source code. Feel free to do that and propose such a 
change here. For now, the discussion should be on how to format code in the 
source (= the format in the output for now, because that is Sphinx's 
behavior).

So we can discuss the best way to format references in the source code. To 
some extent, of course, this is bikeshedding. Whether we use [AC2016] or 
[MR234898349] or [doi:...] or something else, there will always be 
arguments for doing one of the others. I personally find [Mil1958] in a 
discussion of the Steenrod algebra to convey information: I know that it 
refers to Milnor's 1958 paper. I would not recognize the MR number or the 
doi number for this. So I personally find the format [AC2016] a good 
balance between readability, brevity, and (to a large extent) unique 
representation. (Also, my suggested usage would be to often include more 
information than just the citation name: "In [Mil1958], Milnor showed ..." 
or "Milnor showed that ... -- see [Mil1958]" or something similar. Again, 
there is a balance between readability and verbosity.)

-- 
John

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[Travis Scrimshaw]

>From working on stuff that involves 100+ references, even having [1] causes 
problems. Then you also have essentially random numbers that can change on 
every new version of Sage. Also, I feel doing stuff like "Foo in [1]" can 
be overly verbose to redundant at times. So I am strongly for references in 
the [AC2016] format.

Best,
Travis

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

['Martin R' via sage-devel]

Am Mittwoch, 21. September 2016 13:49:57 UTC+2 schrieb Johan S. R. Nielsen:
>
> With MR numbers, do you mean a link of the type [MR3352496]? 
>
 
Yes!  (Except, that in a compiled document such links could be transformed 
into a link such as [1].
 

> > well, for preprints clearly there is of course the arXiv number and for 
> > sciences without a good database, there is doi. 
>
> Neither arXiv nor DOI completely catalogues all publications. I don't 
> know how many such cases appear in Sage's bibliography of course. 
>

Exactly, that's why findstat uses all of them (and more), but identifies 
references with the same (up to some noise) title and authors. 
 

> > concerning readability, there is a well known justification for using 
> > sequential numbers 
>
> Can you clarify? How would sequential numbers work? The documentation of 
> Sage is never read in sequence but more like random access. 
>

What I meant is the common referencing scheme in many mathematics 
publications.  Of course, if you read something like

See [1] and [2,4] for background and recent developments.

that's terrible.  However, if the author cares at all, (s)he would have 
written:

See the textbook by Foo [1] and the recent papers by Harry and John 
[2,4] for background and recent developments.

A reference like [Tho2000] is to me much more recognisable than 
> [MR1794692]. Having two or three of the latter kind of references in a 
> text, it takes brain-effort simply to distinguish if they are different 
> or not. 
>

That is very true. 

In articles and books, [Tho2000] is a much more popular format, and I 
> guess for exactly this reason. Of course in such publication sizes the 
> scalability problems don't show well, which could be the case for 
> Sage.
>

> I just don't think so: in the current master bibliography, there's 1130 
> references. There's *2* collisions with the current naming scheme 
> (broken by appending 'a', 'b', etc.)! 
>

OK. 

> I'm not making this up, I used this to organise the references for 
> > www.findstat.org, and I'm very happy with the result. 
>
> Can you elaborate? When I look at e.g. 
>
> http://www.findstat.org/GelfandTsetlinPatterns?action=diff=66=65 
>
> then the references are [KTT04], [Lo04] and [Sta99].
>

Yes, things are more complicated than advertised.

So, part of the truth is as follows 
(see http://www.findstat.org/StatisticsDatabase/St000575 
or http://www.findstat.org/StatisticsDatabase/St000319 for examples

* Alice submits a statistic using 
http://www.findstat.org/StatisticsDatabase/NewStatistic, and in the 
reference field, she is (strongly) encouraged to type something like

[[MathSciNet:3338726]] [[arXiv:1304.4309]]

* FindStat retrieves bibtex from the mathscinet and arxiv catalogues and 
checks that they indeed refer to the same paper (by definition: authors and 
title must match).  The identifier from the most preferred catalogue is 
then used as bibtex key in our bibtex file.  If somebody references this 
paper again, the bibtex file will not be modified.

Within a statistic entry, the references are simply numbered [1], [2], [3], 
...  Currently, there is no automatic linking between numbers used in the 
description field of the statistic and the list of references.  In this 
case, I think it would be overkill.

* in case Alice did not use an identifier but rather free text, FindStat 
sends this to http://www.ams.org/mathscinet-mref and tries to obtain an MR 
number.  Failing that, the reference is left as Alice types it.

* On the webpage, we show then author, title, and identifiers.

I would have suggested to write "As shown by Chern et al. 
[[MathSciNet:3338726]] one can do this at that" because it works for me (I 
also use this in my LaTeX files, using reftex), but I also admit that I do 
not care enough to advertise it more :-)

The main advantage is that one then has automatically unique identifiers, 
which can (but need not) be transformed easily into whatever you want for 
the end user (eg., the one reading docstrings).

Martin

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[Dima Pasechnik]

On Wednesday, September 21, 2016 at 11:52:57 AM UTC, Martin R wrote:
>
> Am Mittwoch, 21. September 2016 13:36:31 UTC+2 schrieb Dima Pasechnik:
>>
>>
>>
>> On Wednesday, September 21, 2016 at 9:36:06 AM UTC, Martin R wrote:
>>>
>>> well, for preprints clearly there is of course the arXiv number and for 
>>> sciences without a good database, there is doi.
>>>
>>> concerning readability, there is a well known justification for using 
>>> sequential numbers
>>>
>>
>> we talk about readability of the source code, too.
>> IMHO one should not name variables and functions just using sequential 
>> numbers :-)
>>
>> Hm, I'd say that reference identifiers in docstrings and variable names 
> are a bit different.  The argument in favour of using numeric references is 
> that it encourages writing: "In 1783, Xin and Müller [1] have shown foo" 
> instead of "In [XiMü1783] foo is shown".
>

No, we are talking about the source, not the output. You will not want to 
write \cite{ref4242} in your LaTeX, you will rather use
\cite{sqlifemean}. Similarly in Python code or in rst files.
And also writing Xin and Müller \cite{sqlifemean} 10 times is bad style, I 
think...



 

>  
>
>> Having said this, I again would argue for an option to have aliases.
>>
>> E.g. say there is a popular Arxiv preprint cited 10 times in the source, 
>> which then becomes
>> a publication. It is really unnecessary to change all these 10 citations?
>>
>
> I have implemented the following for findstat: as long as title and 
> authors coincide (using a threshold for Levenstein distance to get rid of 
> some noise such as accents and punctuation, etc.) the two entries are 
> merged, with preference (in the bibtex file) given to the MR entry.  (I use 
> MathSciNet, zbMath, arXiv and DOI for citations).
>
> Martin
>

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

['Martin R' via sage-devel]

Am Mittwoch, 21. September 2016 13:36:31 UTC+2 schrieb Dima Pasechnik:
>
>
>
> On Wednesday, September 21, 2016 at 9:36:06 AM UTC, Martin R wrote:
>>
>> well, for preprints clearly there is of course the arXiv number and for 
>> sciences without a good database, there is doi.
>>
>> concerning readability, there is a well known justification for using 
>> sequential numbers
>>
>
> we talk about readability of the source code, too.
> IMHO one should not name variables and functions just using sequential 
> numbers :-)
>
> Hm, I'd say that reference identifiers in docstrings and variable names 
are a bit different.  The argument in favour of using numeric references is 
that it encourages writing: "In 1783, Xin and Müller [1] have shown foo" 
instead of "In [XiMü1783] foo is shown".
 

> Having said this, I again would argue for an option to have aliases.
>
> E.g. say there is a popular Arxiv preprint cited 10 times in the source, 
> which then becomes
> a publication. It is really unnecessary to change all these 10 citations?
>

I have implemented the following for findstat: as long as title and authors 
coincide (using a threshold for Levenstein distance to get rid of some 
noise such as accents and punctuation, etc.) the two entries are merged, 
with preference (in the bibtex file) given to the MR entry.  (I use 
MathSciNet, zbMath, arXiv and DOI for citations).

Martin

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[Johan S . H . Rosenkilde]

> Having said this, I again would argue for an option to have aliases.
>
> E.g. say there is a popular Arxiv preprint cited 10 times in the source, 
> which then becomes
> a publication. It is really unnecessary to change all these 10 citations?

That's a good point. But does Sphinx support such aliasing out of the
box, or would we have to patch it on ourselves? If the latter is the
case, perhaps it's not *that* important?

Best,
Johan

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[Johan S . H . Rosenkilde]

With MR numbers, do you mean a link of the type [MR3352496]?

> well, for preprints clearly there is of course the arXiv number and for 
> sciences without a good database, there is doi.

Neither arXiv nor DOI completely catalogues all publications. I don't
know how many such cases appear in Sage's bibliography of course.

> concerning readability, there is a well known justification for using 
> sequential numbers

Can you clarify? How would sequential numbers work? The documentation of
Sage is never read in sequence but more like random access.

A reference like [Tho2000] is to me much more recognisable than
[MR1794692]. Having two or three of the latter kind of references in a
text, it takes brain-effort simply to distinguish if they are different
or not.

In articles and books, [Tho2000] is a much more popular format, and I
guess for exactly this reason. Of course in such publication sizes the
scalability problems don't show well, which could be the case for
Sage.

I just don't think so: in the current master bibliography, there's 1130
references. There's *2* collisions with the current naming scheme
(broken by appending 'a', 'b', etc.)!

> I'm not making this up, I used this to organise the references for 
> www.findstat.org, and I'm very happy with the result.

Can you elaborate? When I look at e.g.

http://www.findstat.org/GelfandTsetlinPatterns?action=diff=66=65

then the references are [KTT04], [Lo04] and [Sta99].

Best,
Johan



'Martin R' via sage-devel writes:

> well, for preprints clearly there is of course the arXiv number and for 
> sciences without a good database, there is doi.
>
> concerning readability, there is a well known justification for using 
> sequential numbers
>
> I'm not making this up, I used this to organise the references for 
> www.findstat.org, and I'm very happy with the result.
>
> Martin
>
> Am Mittwoch, 21. September 2016 11:10:00 UTC+2 schrieb Dima Pasechnik:
>>
>>
>>
>> On Wednesday, September 21, 2016 at 8:46:13 AM UTC, David Roe wrote:
>>>
>>> Preprints won't have MR numbers.  I also find MR numbers less readable.
>>>
>> and not all the CS-related publications make it into MR database, either.
>>  
>>
>>>
>>> We could just append letters ("a" then "b," etc) if there are collisions.
>>>
>>
>> I wonder whether it is possible to create aliases for references, i.e. 
>> make [Bla]_ and [Foo]_ both refer to [Foo].
>> This would allow less changes in the source.
>>
>>
>>  
>>
>>> David
>>>
>>> On Wed, Sep 21, 2016 at 4:38 AM, 'Martin R' via sage-devel <
>>> sage-...@googlegroups.com> wrote:
>>>
 Why not use the MR number as reference format?

 Martin


 Am Mittwoch, 21. September 2016 01:03:27 UTC+2 schrieb John H Palmieri:
>
> As discussed in another thread [1]_ on sage-devel recently, I propose 
> changing our policy toward references:
>
> - all references should be put into a master bibliography file, and
> - all references should be, insofar as possible, in a standard form: 
> for a work by a single author "Author" published in YEAR: [AutYEAR]. For 
> a 
> work published by "Author" and "Coauthor" in YEAR: [ACYEAR]. The year 
> should be four digits.
>
> The main point is the first item is to avoid conflicting 
> cross-references, and it also seems to make sense to list all references 
> in 
> one place. (The goal behind the second item is just consistency.)
>
> This is implemented at https://trac.sagemath.org/ticket/21454.
>
> Any comments?
>
> -- 
> John
>
> REFERENCES:
>
> .. [1] 
> https://groups.google.com/d/msg/sage-devel/-_kszKLhICw/SjLMs4rXCAAJ
>
> -- 
 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+...@googlegroups.com.
 To post to this group, send email to sage-...@googlegroups.com.
 Visit this group at https://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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[Dima Pasechnik]

On Wednesday, September 21, 2016 at 9:36:06 AM UTC, Martin R wrote:
>
> well, for preprints clearly there is of course the arXiv number and for 
> sciences without a good database, there is doi.
>
> concerning readability, there is a well known justification for using 
> sequential numbers
>

we talk about readability of the source code, too.
IMHO one should not name variables and functions just using sequential 
numbers :-)

Having said this, I again would argue for an option to have aliases.

E.g. say there is a popular Arxiv preprint cited 10 times in the source, 
which then becomes
a publication. It is really unnecessary to change all these 10 citations?

 

>
>
> I'm not making this up, I used this to organise the references for 
> www.findstat.org, and I'm very happy with the result.
>
> Martin
>
> Am Mittwoch, 21. September 2016 11:10:00 UTC+2 schrieb Dima Pasechnik:
>>
>>
>>
>> On Wednesday, September 21, 2016 at 8:46:13 AM UTC, David Roe wrote:
>>>
>>> Preprints won't have MR numbers.  I also find MR numbers less readable.
>>>
>> and not all the CS-related publications make it into MR database, either.
>>  
>>
>>>
>>> We could just append letters ("a" then "b," etc) if there are collisions.
>>>
>>
>> I wonder whether it is possible to create aliases for references, i.e. 
>> make [Bla]_ and [Foo]_ both refer to [Foo].
>> This would allow less changes in the source.
>>
>>
>>  
>>
>>> David
>>>
>>> On Wed, Sep 21, 2016 at 4:38 AM, 'Martin R' via sage-devel <
>>> sage-...@googlegroups.com> wrote:
>>>
 Why not use the MR number as reference format?

 Martin


 Am Mittwoch, 21. September 2016 01:03:27 UTC+2 schrieb John H Palmieri:
>
> As discussed in another thread [1]_ on sage-devel recently, I propose 
> changing our policy toward references:
>
> - all references should be put into a master bibliography file, and
> - all references should be, insofar as possible, in a standard form: 
> for a work by a single author "Author" published in YEAR: [AutYEAR]. For 
> a 
> work published by "Author" and "Coauthor" in YEAR: [ACYEAR]. The year 
> should be four digits.
>
> The main point is the first item is to avoid conflicting 
> cross-references, and it also seems to make sense to list all references 
> in 
> one place. (The goal behind the second item is just consistency.)
>
> This is implemented at https://trac.sagemath.org/ticket/21454.
>
> Any comments?
>
> -- 
> John
>
> REFERENCES:
>
> .. [1] 
> https://groups.google.com/d/msg/sage-devel/-_kszKLhICw/SjLMs4rXCAAJ
>
> -- 
 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+...@googlegroups.com.
 To post to this group, send email to sage-...@googlegroups.com.
 Visit this group at https://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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

['Martin R' via sage-devel]

well, for preprints clearly there is of course the arXiv number and for 
sciences without a good database, there is doi.

concerning readability, there is a well known justification for using 
sequential numbers

I'm not making this up, I used this to organise the references for 
www.findstat.org, and I'm very happy with the result.

Martin

Am Mittwoch, 21. September 2016 11:10:00 UTC+2 schrieb Dima Pasechnik:
>
>
>
> On Wednesday, September 21, 2016 at 8:46:13 AM UTC, David Roe wrote:
>>
>> Preprints won't have MR numbers.  I also find MR numbers less readable.
>>
> and not all the CS-related publications make it into MR database, either.
>  
>
>>
>> We could just append letters ("a" then "b," etc) if there are collisions.
>>
>
> I wonder whether it is possible to create aliases for references, i.e. 
> make [Bla]_ and [Foo]_ both refer to [Foo].
> This would allow less changes in the source.
>
>
>  
>
>> David
>>
>> On Wed, Sep 21, 2016 at 4:38 AM, 'Martin R' via sage-devel <
>> sage-...@googlegroups.com> wrote:
>>
>>> Why not use the MR number as reference format?
>>>
>>> Martin
>>>
>>>
>>> Am Mittwoch, 21. September 2016 01:03:27 UTC+2 schrieb John H Palmieri:

 As discussed in another thread [1]_ on sage-devel recently, I propose 
 changing our policy toward references:

 - all references should be put into a master bibliography file, and
 - all references should be, insofar as possible, in a standard form: 
 for a work by a single author "Author" published in YEAR: [AutYEAR]. For a 
 work published by "Author" and "Coauthor" in YEAR: [ACYEAR]. The year 
 should be four digits.

 The main point is the first item is to avoid conflicting 
 cross-references, and it also seems to make sense to list all references 
 in 
 one place. (The goal behind the second item is just consistency.)

 This is implemented at https://trac.sagemath.org/ticket/21454.

 Any comments?

 -- 
 John

 REFERENCES:

 .. [1] 
 https://groups.google.com/d/msg/sage-devel/-_kszKLhICw/SjLMs4rXCAAJ

 -- 
>>> 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+...@googlegroups.com.
>>> To post to this group, send email to sage-...@googlegroups.com.
>>> Visit this group at https://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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[Dima Pasechnik]

On Wednesday, September 21, 2016 at 8:46:13 AM UTC, David Roe wrote:
>
> Preprints won't have MR numbers.  I also find MR numbers less readable.
>
and not all the CS-related publications make it into MR database, either.
 

>
> We could just append letters ("a" then "b," etc) if there are collisions.
>

I wonder whether it is possible to create aliases for references, i.e. make 
[Bla]_ and [Foo]_ both refer to [Foo].
This would allow less changes in the source.


 

> David
>
> On Wed, Sep 21, 2016 at 4:38 AM, 'Martin R' via sage-devel <
> sage-...@googlegroups.com > wrote:
>
>> Why not use the MR number as reference format?
>>
>> Martin
>>
>>
>> Am Mittwoch, 21. September 2016 01:03:27 UTC+2 schrieb John H Palmieri:
>>>
>>> As discussed in another thread [1]_ on sage-devel recently, I propose 
>>> changing our policy toward references:
>>>
>>> - all references should be put into a master bibliography file, and
>>> - all references should be, insofar as possible, in a standard form: for 
>>> a work by a single author "Author" published in YEAR: [AutYEAR]. For a work 
>>> published by "Author" and "Coauthor" in YEAR: [ACYEAR]. The year should be 
>>> four digits.
>>>
>>> The main point is the first item is to avoid conflicting 
>>> cross-references, and it also seems to make sense to list all references in 
>>> one place. (The goal behind the second item is just consistency.)
>>>
>>> This is implemented at https://trac.sagemath.org/ticket/21454.
>>>
>>> Any comments?
>>>
>>> -- 
>>> John
>>>
>>> REFERENCES:
>>>
>>> .. [1] 
>>> https://groups.google.com/d/msg/sage-devel/-_kszKLhICw/SjLMs4rXCAAJ
>>>
>>> -- 
>> 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+...@googlegroups.com .
>> To post to this group, send email to sage-...@googlegroups.com 
>> .
>> Visit this group at https://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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Re: Sage's references: new policy?

[David Roe]

Preprints won't have MR numbers.  I also find MR numbers less readable.

We could just append letters ("a" then "b," etc) if there are collisions.
David

On Wed, Sep 21, 2016 at 4:38 AM, 'Martin R' via sage-devel <
sage-devel@googlegroups.com> wrote:

> Why not use the MR number as reference format?
>
> Martin
>
>
> Am Mittwoch, 21. September 2016 01:03:27 UTC+2 schrieb John H Palmieri:
>>
>> As discussed in another thread [1]_ on sage-devel recently, I propose
>> changing our policy toward references:
>>
>> - all references should be put into a master bibliography file, and
>> - all references should be, insofar as possible, in a standard form: for
>> a work by a single author "Author" published in YEAR: [AutYEAR]. For a work
>> published by "Author" and "Coauthor" in YEAR: [ACYEAR]. The year should be
>> four digits.
>>
>> The main point is the first item is to avoid conflicting
>> cross-references, and it also seems to make sense to list all references in
>> one place. (The goal behind the second item is just consistency.)
>>
>> This is implemented at https://trac.sagemath.org/ticket/21454.
>>
>> Any comments?
>>
>> --
>> John
>>
>> REFERENCES:
>>
>> .. [1] https://groups.google.com/d/msg/sage-devel/-_kszKLhICw/SjLMs
>> 4rXCAAJ
>>
>> --
> 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 https://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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Sage's references: new policy?

['Martin R' via sage-devel]

Why not use the MR number as reference format?

Martin

Am Mittwoch, 21. September 2016 01:03:27 UTC+2 schrieb John H Palmieri:
>
> As discussed in another thread [1]_ on sage-devel recently, I propose 
> changing our policy toward references:
>
> - all references should be put into a master bibliography file, and
> - all references should be, insofar as possible, in a standard form: for a 
> work by a single author "Author" published in YEAR: [AutYEAR]. For a work 
> published by "Author" and "Coauthor" in YEAR: [ACYEAR]. The year should be 
> four digits.
>
> The main point is the first item is to avoid conflicting cross-references, 
> and it also seems to make sense to list all references in one place. (The 
> goal behind the second item is just consistency.)
>
> This is implemented at https://trac.sagemath.org/ticket/21454.
>
> Any comments?
>
> -- 
> John
>
> REFERENCES:
>
> .. [1] https://groups.google.com/d/msg/sage-devel/-_kszKLhICw/SjLMs4rXCAAJ
>
>

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Sage's references: new policy?

[David Coudert]

What if we have two papers by "Author" and "Coauthor" in 2016?
How to distinguish between a paper by say "R. Thomas" in 2000 and another 
by "C. Thomassen" in 2000 ?


Le mercredi 21 septembre 2016 01:03:27 UTC+2, John H Palmieri a écrit :
>
> As discussed in another thread [1]_ on sage-devel recently, I propose 
> changing our policy toward references:
>
> - all references should be put into a master bibliography file, and
> - all references should be, insofar as possible, in a standard form: for a 
> work by a single author "Author" published in YEAR: [AutYEAR]. For a work 
> published by "Author" and "Coauthor" in YEAR: [ACYEAR]. The year should be 
> four digits.
>
> The main point is the first item is to avoid conflicting cross-references, 
> and it also seems to make sense to list all references in one place. (The 
> goal behind the second item is just consistency.)
>
> This is implemented at https://trac.sagemath.org/ticket/21454.
>
> Any comments?
>
> -- 
> John
>
> REFERENCES:
>
> .. [1] https://groups.google.com/d/msg/sage-devel/-_kszKLhICw/SjLMs4rXCAAJ
>
>

-- 
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 https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.