question on suppress-author citation variant

29 views
Skip to first unread message

Bruce D'Arcus

unread,
Apr 23, 2021, 2:39:56 PM4/23/21
to pandoc-discuss
Org-mode is primed to add citation support, with a syntax inspired in part by pandoc's.

But I'm now wondering about a design detail it shares with pandoc, and so I thought I'd ask this community before raising it there.

So this is what pandoc's `[-@key]` would look like:

>   [cite:-@key]

If there are multiple items in the same citations, it could be this:

 [cite:-@key1;@key2]

As in, suppress author on first item, but not second, as opposed to all items within the citation.

In retrospect, does this make sense?

The org syntax will also have an optional "style" section, which applies to the whole citation; so:

 [cite/text:@key1;@key2]

... which leads me to wonder if "suppress author" is really just such a style.

Bruce

Bastien DUMONT

unread,
Apr 23, 2021, 2:51:48 PM4/23/21
to pandoc-...@googlegroups.com
> [cite:-@key1;@key2]
>
> As in, suppress author on first item, but not second, as opposed to all items
> within the citation.
>
> In retrospect, does this make sense?

One real example:

[It was already the case in the 2nd century, following the convincing analysis that Olivier Munnich proposed of the _Dialogue_ of Justin in -@MunnichJudaism;, confirmed by the analysis of Tryphon's arguments in @PouderonSource]
> --
> You received this message because you are subscribed to the Google Groups
> "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email
> to pandoc-discus...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/
> pandoc-discuss/8553e982-3a50-4c67-b94a-20796b539de1n%40googlegroups.com.

Bruce D'Arcus

unread,
Apr 23, 2021, 3:04:08 PM4/23/21
to pandoc-...@googlegroups.com
On Fri, Apr 23, 2021 at 2:51 PM Bastien DUMONT
<bastien...@posteo.net> wrote:
>
> > [cite:-@key1;@key2]
> >
> > As in, suppress author on first item, but not second, as opposed to all items
> > within the citation.
> >
> > In retrospect, does this make sense?
>
> One real example:

Thank you!

> [It was already the case in the 2nd century, following the convincing analysis that Olivier Munnich proposed of the _Dialogue_ of Justin in -@MunnichJudaism;, confirmed by the analysis of Tryphon's arguments in @PouderonSource]

I guess you use note-based styles?

Why do you have those as part of the same citation, as opposed to
putting that content in a note, with two distinct citations?

In any case, I guess more broadly:

Does it make sense to treat "suppress author" differently than other
citation variant; like, say, author-in-text (in natbib, \citet)?

Are these really the properties that should attach to the citation
items, rather than the citation overall?

For org, I'm trying to figure out if "suppress author" is really a
style, and secondarily, if it is, should it be allowed to apply to the
items within a citation as well.

Bruce

Bastien DUMONT

unread,
Apr 23, 2021, 3:28:14 PM4/23/21
to pandoc-...@googlegroups.com
> I guess you use note-based styles?

Yes!

> Why do you have those as part of the same citation, as opposed to
> putting that content in a note, with two distinct citations?

The problem is that Pandoc would assign the "author in text" output
method to both citations, which can result in unwanted behaviour in
some cases. Maybe this doesn't apply to Org-mode.

> Does it make sense to treat "suppress author" differently than other
> citation variant; like, say, author-in-text (in natbib, \citet)?

I just saw today something like "following Andrews' demonstration
(2008),...", so yes, it can be viewed as an additional property.

> Are these really the properties that should attach to the citation
> items, rather than the citation overall?

From a purely technical point of view, it depends on whether two
distinct citations can be placed inside a footnote. From a conceptual
point of view, I would say that author suppression is "micro-context-
dependant", so yes, but it is more debatable.

Bruce D'Arcus

unread,
Apr 23, 2021, 4:55:56 PM4/23/21
to pandoc-...@googlegroups.com
On Fri, Apr 23, 2021 at 3:28 PM Bastien DUMONT
<bastien...@posteo.net> wrote:
>
> > I guess you use note-based styles?
>
> Yes!
>
> > Why do you have those as part of the same citation, as opposed to
> > putting that content in a note, with two distinct citations?
>
> The problem is that Pandoc would assign the "author in text" output
> method to both citations, which can result in unwanted behaviour in
> some cases. Maybe this doesn't apply to Org-mode.

The org support is designed with configurable processor support, so in
theory, once it's merged, it should be possible to hook up John's
citeproc library with a little elisp glue.

> > Does it make sense to treat "suppress author" differently than other
> > citation variant; like, say, author-in-text (in natbib, \citet)?
>
> I just saw today something like "following Andrews' demonstration
> (2008),...", so yes, it can be viewed as an additional property.

I don't follow. Why isn't that just a simple suppress-author citation?

> > Are these really the properties that should attach to the citation
> > items, rather than the citation overall?
>
> From a purely technical point of view, it depends on whether two
> distinct citations can be placed inside a footnote. From a conceptual
> point of view, I would say that author suppression is "micro-context-
> dependant", so yes, but it is more debatable.

Bruce

Bastien DUMONT

unread,
Apr 23, 2021, 5:50:52 PM4/23/21
to pandoc-...@googlegroups.com
> >
> > I just saw today something like "following Andrews' demonstration
> > (2008),...", so yes, it can be viewed as an additional property.
>
> I don't follow. Why isn't that just a simple suppress-author citation?

Sorry, I got lost between citation classes and modes. Yes, you are right.

denis...@ub.unibe.ch

unread,
Apr 23, 2021, 6:03:29 PM4/23/21
to pandoc-...@googlegroups.com

> [cite:-@key1;@key2]

> As in, suppress author on first item, but not second, as opposed to all items within the citation.

> In retrospect, does this make sense?

 

I’d say yes, there may be things such as

«In his mail to the mailing list, D’Arcus (2021; against Doe 2020) has had a couple of questions about citation syntax.»

 

Not quite pretty, but reasonable, isn’t it ?

 

Regarding the style vs minus question : in pandoc you can’t have a thing such as this syntax :  [cite/text:@key1;@key2]. So using symbols is the only possibility. Org is different in this regard. Possibly you could use a special style for that. But the symbol based syntax can give you a nice shorthand.

 

Denis

 

 

 

 

--

You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to


To view this discussion on the web visit

Bruce D'Arcus

unread,
Apr 23, 2021, 6:15:30 PM4/23/21
to pandoc-...@googlegroups.com
On Fri, Apr 23, 2021 at 6:03 PM <denis...@ub.unibe.ch> wrote:
>
> > [cite:-@key1;@key2]
>
> > As in, suppress author on first item, but not second, as opposed to all items within the citation.
>
> > In retrospect, does this make sense?
>
> I’d say yes, there may be things such as
>
> «In his mail to the mailing list, D’Arcus (2021; against Doe 2020) has had a couple of questions about citation syntax.»
>
> Not quite pretty, but reasonable, isn’t it ?

That's debatable :-)

> Regarding the style vs minus question : in pandoc you can’t have a thing such as this syntax : [cite/text:@key1;@key2]. So using symbols is the only possibility. Org is different in this regard. Possibly you could use a special style for that. But the symbol based syntax can give you a nice shorthand.

The minus symbol somewhat confuses the issue in org.

What prompted this thread was me wondering whether suppress-author
should just be a style on the citation, so that you have:

[cite/suppress-author:@key1;@key2].

... or even:

[cite/-:@key1;@key2].

This is how it works in the bibtex/biblatex world; right?

Bruce

John MacFarlane

unread,
Apr 24, 2021, 12:22:31 AM4/24/21
to Bruce D'Arcus, pandoc-...@googlegroups.com

I do think that conceptually, suppress-author makes more sense
on the whole citation rather than on individual citation-items.

The way pandoc does things is derived from ancient lore from
pandoc-citeproc, and I think some of the thinking behind it
may be hard to recover now.

We reduce an "in-text" citation like "Jones (1989, p. 33)" to a
combination

author-only + suppress-author

But really that whole combination should be the thing you
specify on the citation.

Bruce D'Arcus

unread,
Apr 24, 2021, 6:29:38 AM4/24/21
to John MacFarlane, pandoc-...@googlegroups.com
On Sat, Apr 24, 2021 at 12:22 AM John MacFarlane <j...@berkeley.edu> wrote:

> I do think that conceptually, suppress-author makes more sense
> on the whole citation rather than on individual citation-items.
>
> The way pandoc does things is derived from ancient lore from
> pandoc-citeproc, and I think some of the thinking behind it
> may be hard to recover now.

TBH, I think this was informed by a decision originally made in Zotero.

> We reduce an "in-text" citation like "Jones (1989, p. 33)" to a
> combination
>
> author-only + suppress-author
>
> But really that whole combination should be the thing you
> specify on the citation.

As in, that example is best specified as a modifier on the citation as
a whole; so in the org-cite syntax:

[cite/text:@jones89 p33]

... but the pandoc syntax doesn't currently support such styles?

Bruce

denis...@ub.unibe.ch

unread,
Apr 24, 2021, 7:11:21 AM4/24/21
to j...@berkeley.edu, pandoc-...@googlegroups.com
The syntax for an in-text citation is: @doe2021 [p. 34].

Denis

________________________________________
Von: pandoc-...@googlegroups.com <pandoc-...@googlegroups.com> im Auftrag von Bruce D'Arcus <bda...@gmail.com>
Gesendet: Samstag, 24. April 2021 12:29:26
An: John MacFarlane
Cc: pandoc-...@googlegroups.com
Betreff: Re: question on suppress-author citation variant
--
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discus...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAF-FPGPoHunF1zB_QrLXr5M9GwPZ8-Quk9PYFT6SCYshaFZ2rg%40mail.gmail.com.

Bruce D'Arcus

unread,
Apr 24, 2021, 7:50:43 AM4/24/21
to pandoc-...@googlegroups.com
On Sat, Apr 24, 2021 at 7:11 AM <denis...@ub.unibe.ch> wrote:
>
> The syntax for an in-text citation is: @doe2021 [p. 34].

I thought @doe2021 would just give you a bare citation: "some sentence
with Doe 2021"; not "some sentence with Doe (2021)"?

Bruce

denis...@ub.unibe.ch

unread,
Apr 24, 2021, 8:04:39 AM4/24/21
to pandoc-...@googlegroups.com
There's no such thing as a bare citation, so no.
"@doe argues ... " => "Doe (2021) argues ..."

Denis
________________________________________
Von: pandoc-...@googlegroups.com <pandoc-...@googlegroups.com> im Auftrag von Bruce D'Arcus <bda...@gmail.com>
Gesendet: Samstag, 24. April 2021 13:50:32
An: pandoc-...@googlegroups.com
Betreff: Re: question on suppress-author citation variant

--
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discus...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAF-FPGOQ0o6N9Q4UjXcrdrFydoz_emBMmbMQKZm936y5YoKcAA%40mail.gmail.com.
Reply all
Reply to author
Forward
0 new messages