considering syntax change for author-in-text citations

considering syntax change for author-in-text citations

[John MacFarlane]

I'm thinking about making a change in the citation syntax.

Currently there are three forms:

1. author-in-text:

@foo  -> Foo (2013)
@foo [p. 33] -> Foo (2013, 33)
@foo [p. 33; @bar, p. 45]  -> Foo (2013, p. 33; Bar 2015, p. 45)

2. normal:

[@foo] -> (Foo 2013)
[@foo, p. 33]  -> (Foo 2013, p. 33)
[@foo, p. 33; @bar, p. 45]  -> (Foo 2013, p. 33; Bar 2015, p. 45)

3. suppress-author:

[-@foo] ->  (2013)
[-@foo, p. 33]  -> (2013, p. 33)

The proposal would be to change the author-in-text form so that
it requires square brackets:

Old                           New
--------------------------    ---------------------------
@foo                          [/@foo]
@foo [p. 33]                  [/@foo, p. 33]
@foo [p. 33; @bar, p. 45]     [/@foo, p.33; @bar, p. 45]

Mostly this is motivated by thinking about how I'm going
to implement the citation extension for commonmark.  It's
very hard to figure out how to support the current author-in-text
syntax nicely given the way we're parsing commonmark.
A secondary motivation is that the current @foo syntax
overlaps with references to example list items.  It would
be nice to have a syntax that is unambiguously a citation.

I'm not wedded to the [/@foo]; [+@foo] would also be quite
natural given the current use of - to suppress author.  But the
`/` looks a bit better to me.  [:@foo] would be another option.

I think this change would make text look a *little* less natural:

@foo says blah  -> [/@foo] says blah.

But it would make citations stand out more clearly from other
kinds of elements; it would make it possible to handle
citations in commonmark; and it would avoid syntactic
confusion between references to example list items and
citations.

Probably what I'd do is implement the new syntax for citations
for commonmark+citations only, keeping the old one for
markdown+citations.  This could be a bit confusing, but it
would allow documents to be converted automatically to the
new form.

I'd be interested in your thoughts on this.

Re: considering syntax change for author-in-text citations

[jcr]

[-- Attachment #1.1: Type: text/plain, Size: 2960 bytes --]

I like the brackets because they mean that all citations could have 
attributes, at least with commonmark+attributes. I'm hoping that classes or 
attributes will one day give control over the formatting of the citation.

Between [/@foo], [+@foo], and [:@foo], my preference would be for [+@foo] 
because it seems more logical: I don't see how "/" or ":" suggest 
author-in-text. But I would rarely, if ever, use an author-in-text citation 
in my writing.

On Sunday, November 29, 2020 at 1:40:04 AM UTC+1 John MacFarlane wrote:

>
> I'm thinking about making a change in the citation syntax.
>
> Currently there are three forms:
>
> 1. author-in-text:
>
> @foo -> Foo (2013)
> @foo [p. 33] -> Foo (2013, 33)
> @foo [p. 33; @bar, p. 45] -> Foo (2013, p. 33; Bar 2015, p. 45)
>
> 2. normal:
>
> [@foo] -> (Foo 2013)
> [@foo, p. 33] -> (Foo 2013, p. 33)
> [@foo, p. 33; @bar, p. 45] -> (Foo 2013, p. 33; Bar 2015, p. 45)
>
> 3. suppress-author:
>
> [-@foo] -> (2013)
> [-@foo, p. 33] -> (2013, p. 33)
>
> The proposal would be to change the author-in-text form so that
> it requires square brackets:
>
> Old New
> -------------------------- ---------------------------
> @foo [/@foo]
> @foo [p. 33] [/@foo, p. 33]
> @foo [p. 33; @bar, p. 45] [/@foo, p.33; @bar, p. 45]
>
> Mostly this is motivated by thinking about how I'm going
> to implement the citation extension for commonmark. It's
> very hard to figure out how to support the current author-in-text
> syntax nicely given the way we're parsing commonmark.
> A secondary motivation is that the current @foo syntax
> overlaps with references to example list items. It would
> be nice to have a syntax that is unambiguously a citation.
>
> I'm not wedded to the [/@foo]; [+@foo] would also be quite
> natural given the current use of - to suppress author. But the
> `/` looks a bit better to me. [:@foo] would be another option.
>
> I think this change would make text look a *little* less natural:
>
> @foo says blah -> [/@foo] says blah.
>
> But it would make citations stand out more clearly from other
> kinds of elements; it would make it possible to handle
> citations in commonmark; and it would avoid syntactic
> confusion between references to example list items and
> citations.
>
> Probably what I'd do is implement the new syntax for citations
> for commonmark+citations only, keeping the old one for
> markdown+citations. This could be a bit confusing, but it
> would allow documents to be converted automatically to the
> new form.
>
> I'd be interested in your thoughts on this.
>
>

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/d00922bc-e4c7-4a0e-a7c7-a77014945e86n%40googlegroups.com.

[-- Attachment #1.2: Type: text/html, Size: 3823 bytes --]

AW: considering syntax change for author-in-text citations

[denis.maier-FfwAq0itz3ofv37vnLkPlQ]

I also think using brackets would be a good move here .

Regarding the choice if characters I'm rather agnostic. Both / and + loom a bit obscure to me. I don't think they successfully convey what they are intended to mean.

Are we restricted to those no d of prefix characters? Or could something more meaningful be used, e.g. t or other mnemonic letters?

Also, perhaps we should consider some sort of extensible syntax?



________________________________________
Von: pandoc-discuss-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <pandoc-discuss-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org> im Auftrag von jcr <ffi.appdev-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Gesendet: Sonntag, 29. November 2020 12:49:08
An: pandoc-discuss
Betreff: Re: considering syntax change for author-in-text citations

I like the brackets because they mean that all citations could have attributes, at least with commonmark+attributes. I'm hoping that classes or attributes will one day give control over the formatting of the citation.

Between [/@foo], [+@foo], and [:@foo], my preference would be for [+@foo] because it seems more logical: I don't see how "/" or ":" suggest author-in-text. But I would rarely, if ever, use an author-in-text citation in my writing.

On Sunday, November 29, 2020 at 1:40:04 AM UTC+1 John MacFarlane wrote:

I'm thinking about making a change in the citation syntax.

Currently there are three forms:

1. author-in-text:

@foo -> Foo (2013)
@foo [p. 33] -> Foo (2013, 33)
@foo [p. 33; @bar, p. 45] -> Foo (2013, p. 33; Bar 2015, p. 45)

2. normal:

[@foo] -> (Foo 2013)
[@foo, p. 33] -> (Foo 2013, p. 33)
[@foo, p. 33; @bar, p. 45] -> (Foo 2013, p. 33; Bar 2015, p. 45)

3. suppress-author:

[-@foo] -> (2013)
[-@foo, p. 33] -> (2013, p. 33)

The proposal would be to change the author-in-text form so that
it requires square brackets:

Old New
-------------------------- ---------------------------
@foo [/@foo]
@foo [p. 33] [/@foo, p. 33]
@foo [p. 33; @bar, p. 45] [/@foo, p.33; @bar, p. 45]

Mostly this is motivated by thinking about how I'm going
to implement the citation extension for commonmark. It's
very hard to figure out how to support the current author-in-text
syntax nicely given the way we're parsing commonmark.
A secondary motivation is that the current @foo syntax
overlaps with references to example list items. It would
be nice to have a syntax that is unambiguously a citation.

I'm not wedded to the [/@foo]; [+@foo] would also be quite
natural given the current use of - to suppress author. But the
`/` looks a bit better to me. [:@foo] would be another option.

I think this change would make text look a *little* less natural:

@foo says blah -> [/@foo] says blah.

But it would make citations stand out more clearly from other
kinds of elements; it would make it possible to handle
citations in commonmark; and it would avoid syntactic
confusion between references to example list items and
citations.

Probably what I'd do is implement the new syntax for citations
for commonmark+citations only, keeping the old one for
markdown+citations. This could be a bit confusing, but it
would allow documents to be converted automatically to the
new form.

I'd be interested in your thoughts on this.


--
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<mailto:pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org>.
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/d00922bc-e4c7-4a0e-a7c7-a77014945e86n%40googlegroups.com<https://groups.google.com/d/msgid/pandoc-discuss/d00922bc-e4c7-4a0e-a7c7-a77014945e86n%40googlegroups.com?utm_medium=email&utm_source=footer>.

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/089fc9e359ef47c592808886466dbda0%40ub.unibe.ch.

AW: considering syntax change for author-in-text citations

[denis.maier-FfwAq0itz3ofv37vnLkPlQ]

Regarding extensibility: maybe even something like with org:

[citeX: @smith, p. 78]

Sure, that's verbose. But it's clear and extensible.

What do you think?
________________________________________
Von: pandoc-discuss-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <pandoc-discuss-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org> im Auftrag von denis.maier-FfwAq0itz3ofv37vnLkPlQ@public.gmane.org <denis.maier-FfwAq0itz3ofv37vnLkPlQ@public.gmane.org>
Gesendet: Sonntag, 29. November 2020 13:56:12
An: pandoc-discuss-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
Betreff: AW: considering syntax change for author-in-text citations

I also think using brackets would be a good move here .

Regarding the choice if characters I'm rather agnostic. Both / and + loom a bit obscure to me. I don't think they successfully convey what they are intended to mean.

Are we restricted to those no d of prefix characters? Or could something more meaningful be used, e.g. t or other mnemonic letters?

Also, perhaps we should consider some sort of extensible syntax?



________________________________________
Von: pandoc-discuss-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <pandoc-discuss-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org> im Auftrag von jcr <ffi.appdev-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Gesendet: Sonntag, 29. November 2020 12:49:08
An: pandoc-discuss
Betreff: Re: considering syntax change for author-in-text citations

I like the brackets because they mean that all citations could have attributes, at least with commonmark+attributes. I'm hoping that classes or attributes will one day give control over the formatting of the citation.

Between [/@foo], [+@foo], and [:@foo], my preference would be for [+@foo] because it seems more logical: I don't see how "/" or ":" suggest author-in-text. But I would rarely, if ever, use an author-in-text citation in my writing.

On Sunday, November 29, 2020 at 1:40:04 AM UTC+1 John MacFarlane wrote:

I'm thinking about making a change in the citation syntax.

Currently there are three forms:

1. author-in-text:

@foo -> Foo (2013)
@foo [p. 33] -> Foo (2013, 33)
@foo [p. 33; @bar, p. 45] -> Foo (2013, p. 33; Bar 2015, p. 45)

2. normal:

[@foo] -> (Foo 2013)
[@foo, p. 33] -> (Foo 2013, p. 33)
[@foo, p. 33; @bar, p. 45] -> (Foo 2013, p. 33; Bar 2015, p. 45)

3. suppress-author:

[-@foo] -> (2013)
[-@foo, p. 33] -> (2013, p. 33)

The proposal would be to change the author-in-text form so that
it requires square brackets:

Old New
-------------------------- ---------------------------
@foo [/@foo]
@foo [p. 33] [/@foo, p. 33]
@foo [p. 33; @bar, p. 45] [/@foo, p.33; @bar, p. 45]

Mostly this is motivated by thinking about how I'm going
to implement the citation extension for commonmark. It's
very hard to figure out how to support the current author-in-text
syntax nicely given the way we're parsing commonmark.
A secondary motivation is that the current @foo syntax
overlaps with references to example list items. It would
be nice to have a syntax that is unambiguously a citation.

I'm not wedded to the [/@foo]; [+@foo] would also be quite
natural given the current use of - to suppress author. But the
`/` looks a bit better to me. [:@foo] would be another option.

I think this change would make text look a *little* less natural:

@foo says blah -> [/@foo] says blah.

But it would make citations stand out more clearly from other
kinds of elements; it would make it possible to handle
citations in commonmark; and it would avoid syntactic
confusion between references to example list items and
citations.

Probably what I'd do is implement the new syntax for citations
for commonmark+citations only, keeping the old one for
markdown+citations. This could be a bit confusing, but it
would allow documents to be converted automatically to the
new form.

I'd be interested in your thoughts on this.


--
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<mailto:pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org>.
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/d00922bc-e4c7-4a0e-a7c7-a77014945e86n%40googlegroups.com<https://groups.google.com/d/msgid/pandoc-discuss/d00922bc-e4c7-4a0e-a7c7-a77014945e86n%40googlegroups.com?utm_medium=email&utm_source=footer>.

--
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/089fc9e359ef47c592808886466dbda0%40ub.unibe.ch.

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/60acdd44a95946b491f098b48df2eda9%40ub.unibe.ch.

Re: AW: considering syntax change for author-in-text citations

[John MacFarlane]

<denis.maier-FfwAq0itz3ofv37vnLkPlQ@public.gmane.org> writes:

> Regarding extensibility: maybe even something like with org:
>
> [citeX: @smith, p. 78]

With pandoc we try to avoid using English words as syntactic
cues.  When you're writing in German, you shouldn't have
English words sprinkled in.

Hence the choice of an (admittedly more obscure) symbol like
+ or : or /.

Re: considering syntax change for author-in-text citations

[John Gabriele]

I haven't used citations, so I'm guessing that what you mean by "@foo  -> Foo (2013)" is that if I have `@foo` in my source document, it will be turned into "Foo (2013)" by Pandoc in the output.

My only comment is that the forward slash: `[/@foo]` kind of looks like a closing tag, and reminded me of [BBCode](https://en.wikipedia.org/wiki/BBCode). Maybe you had some other thinking in-mind when considering the forward slash?

-- John


On Sat, Nov 28, 2020, at 7:39 PM, John MacFarlane wrote:
> 
> I'm thinking about making a change in the citation syntax.
> 
> Currently there are three forms:
> 
> 1. author-in-text:
> 
> @foo  -> Foo (2013)
> @foo [p. 33] -> Foo (2013, 33)
> @foo [p. 33; @bar, p. 45]  -> Foo (2013, p. 33; Bar 2015, p. 45)
> 
> 2. normal:
> 
> [@foo] -> (Foo 2013)
> [@foo, p. 33]  -> (Foo 2013, p. 33)
> [@foo, p. 33; @bar, p. 45]  -> (Foo 2013, p. 33; Bar 2015, p. 45)
> 
> 3. suppress-author:
> 
> [-@foo] ->  (2013)
> [-@foo, p. 33]  -> (2013, p. 33)
> 
> The proposal would be to change the author-in-text form so that
> it requires square brackets:
> 
> Old                           New
> --------------------------    ---------------------------
> @foo                          [/@foo]
> @foo [p. 33]                  [/@foo, p. 33]
> @foo [p. 33; @bar, p. 45]     [/@foo, p.33; @bar, p. 45]
> 
> Mostly this is motivated by thinking about how I'm going
> to implement the citation extension for commonmark.  It's
> very hard to figure out how to support the current author-in-text
> syntax nicely given the way we're parsing commonmark.
> A secondary motivation is that the current @foo syntax
> overlaps with references to example list items.  It would
> be nice to have a syntax that is unambiguously a citation.
> 
> I'm not wedded to the [/@foo]; [+@foo] would also be quite
> natural given the current use of - to suppress author.  But the
> `/` looks a bit better to me.  [:@foo] would be another option.
> 
> I think this change would make text look a *little* less natural:
> 
> @foo says blah  -> [/@foo] says blah.
> 
> But it would make citations stand out more clearly from other
> kinds of elements; it would make it possible to handle
> citations in commonmark; and it would avoid syntactic
> confusion between references to example list items and
> citations.
> 
> Probably what I'd do is implement the new syntax for citations
> for commonmark+citations only, keeping the old one for
> markdown+citations.  This could be a bit confusing, but it
> would allow documents to be converted automatically to the
> new form.
> 
> I'd be interested in your thoughts on this.
> 
> -- 
> 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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/m2k0u5ov1p.fsf%40MacBook-Pro.hsd1.ca.comcast.net.
>

Re: considering syntax change for author-in-text citations

[Brandon Keith Biggs]

[-- Attachment #1: Type: text/plain, Size: 3808 bytes --]

Hello,
I'm not a big fan of the proposed syntax because it's more verbose and
programmer-oriented than the current syntax.
Is there some way where we could figure out if the [@authorname] is at a
period? I am having a hard time thinking of a time when [@authorname] is
currently not at the end of a sentence.
Perhaps the default could be [@authorname] would figure out if it's
followed by a period and if so, then it will do the (authorname, 2019).
Otherwise it will do authorname (2019).
If we wanted to force (authorname, 2019) in the middle of a sentence, then
we could do something like [+@authorname]. If we want to force the in-paper
citation, we could do something like [--@authorname]
Thanks,



Brandon Keith Biggs <http://brandonkeithbiggs.com/>


On Sat, Nov 28, 2020 at 4:40 PM John MacFarlane <jgm-TVLZxgkOlNX2fBVCVOL8/A@public.gmane.org> wrote:

>
> I'm thinking about making a change in the citation syntax.
>
> Currently there are three forms:
>
> 1. author-in-text:
>
> @foo  -> Foo (2013)
> @foo [p. 33] -> Foo (2013, 33)
> @foo [p. 33; @bar, p. 45]  -> Foo (2013, p. 33; Bar 2015, p. 45)
>
> 2. normal:
>
> [@foo] -> (Foo 2013)
> [@foo, p. 33]  -> (Foo 2013, p. 33)
> [@foo, p. 33; @bar, p. 45]  -> (Foo 2013, p. 33; Bar 2015, p. 45)
>
> 3. suppress-author:
>
> [-@foo] ->  (2013)
> [-@foo, p. 33]  -> (2013, p. 33)
>
> The proposal would be to change the author-in-text form so that
> it requires square brackets:
>
> Old                           New
> --------------------------    ---------------------------
> @foo                          [/@foo]
> @foo [p. 33]                  [/@foo, p. 33]
> @foo [p. 33; @bar, p. 45]     [/@foo, p.33; @bar, p. 45]
>
> Mostly this is motivated by thinking about how I'm going
> to implement the citation extension for commonmark.  It's
> very hard to figure out how to support the current author-in-text
> syntax nicely given the way we're parsing commonmark.
> A secondary motivation is that the current @foo syntax
> overlaps with references to example list items.  It would
> be nice to have a syntax that is unambiguously a citation.
>
> I'm not wedded to the [/@foo]; [+@foo] would also be quite
> natural given the current use of - to suppress author.  But the
> `/` looks a bit better to me.  [:@foo] would be another option.
>
> I think this change would make text look a *little* less natural:
>
> @foo says blah  -> [/@foo] says blah.
>
> But it would make citations stand out more clearly from other
> kinds of elements; it would make it possible to handle
> citations in commonmark; and it would avoid syntactic
> confusion between references to example list items and
> citations.
>
> Probably what I'd do is implement the new syntax for citations
> for commonmark+citations only, keeping the old one for
> markdown+citations.  This could be a bit confusing, but it
> would allow documents to be converted automatically to the
> new form.
>
> I'd be interested in your thoughts on this.
>
> --
> 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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/m2k0u5ov1p.fsf%40MacBook-Pro.hsd1.ca.comcast.net
> .
>

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAKAWQkUmb2SnCEwUx9s3KMS_M3gJDcWT2FOdffA_3bCZmYYMQA%40mail.gmail.com.

[-- Attachment #2: Type: text/html, Size: 5237 bytes --]

Re: considering syntax change for author-in-text citations

[Joost Kremers]

On Sun, Nov 29 2020, John MacFarlane wrote:
> The proposal would be to change the author-in-text form so that
> it requires square brackets:

I'm not a fan of this idea.

> I think this change would make text look a *little* less natural:

For exactly this reason. My cite keys usually take the form of author+year, so I
can write, e.g.,

As @Foo2013 argues, ...

which looks close enough to an actual author-year citation that you don't have
to think about it when you read the Markdown text. In fact, all three forms are
like that, which fits the original design goal of Markdown quite well.

The proposed syntax doesn't have this effect, obviously. Isn't there an
alternative that's easier to parse and doesn't use brackets?

> But it would make citations stand out more clearly from other
> kinds of elements; it would make it possible to handle
> citations in commonmark; and it would avoid syntactic
> confusion between references to example list items and
> citations.

None of these points would convince the author in me. The second point would
convince the programmer in me, but he doesn't write Markdown. ;-)

> Probably what I'd do is implement the new syntax for citations
> for commonmark+citations only, keeping the old one for
> markdown+citations.  This could be a bit confusing, but it
> would allow documents to be converted automatically to the
> new form.

Why not use +old_citations or something? That would be less confusing and
possibly a bit more future-proof (assuming that you might one day wish to use
new-style citations for markdown as well).

-- 
Joost Kremers
Life has its moments

Re: considering syntax change for author-in-text citations

[John MacFarlane]

Brandon Keith Biggs <brandonkeithbiggs-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:

> Hello,
> I'm not a big fan of the proposed syntax because it's more verbose and
> programmer-oriented than the current syntax.
> Is there some way where we could figure out if the [@authorname] is at a
> period? I am having a hard time thinking of a time when [@authorname] is
> currently not at the end of a sentence.
> Perhaps the default could be [@authorname] would figure out if it's
> followed by a period and if so, then it will do the (authorname, 2019).
> Otherwise it will do authorname (2019).

I thought about trying to do it automatically, but I don't
that is going to be reliable.  The idea would be to trigger
author-in-text form if the citation comes after certain
punctuation marks (period, comma, colon, semicolon) + space.
But this might give bad results for

- abbreviations:  blah blah etc. [@foo2002]
- people who use note styles and put the normal (note) citations
  right after the sentence:  blah blah.[@foo2002]
- and more??

Re: considering syntax change for author-in-text citations

[Julien Dutant]

[-- Attachment #1.1: Type: text/plain, Size: 2609 bytes --]

Hi,

I like the idea of making the syntax unambiguous. Another ambiguity that 
was worrying me was punctation right after the citation, e.g.:

@Doe:2012: -> Doe (2012):
@Smith.Doe:2013. -> Smith and Doe (2013). 

Pandoc's manual specifies that the keys may only contain *internal* 
punctuation, which ensures the results above. But that doesn't seem to be 
required by CSL-JSON or BibTeX. 

Would that be a more readable alternative compatible with Commonmark? 
@Doe2017 []
@Doe2017 [34-45]

I like unifying the syntax though. Of the two proposed, [/@foo] makes me 
think of an optional insert, as in "meeting (/party)". [+@foo] pairs well 
with [-@foo]. 

How would the syntax behave with (attempted) prefixes? In the 
author-suppressed format the result is normal citation with a botched 
prefix: 
[see @Doe2012, p.33] -> (see Doe 2012, p.33)
[-see @Doe2012, p.33] -> (-see Doe 2012, p.33)
[- see @Does 2012, p.33] -> (- see Doe 2012, p.33)
[\see @Doe2012, p.33] -> ? see Doe (2012, p.33)? or (\see Doe 2012, p.33)?

Best, -- J

On Monday, November 30, 2020 at 12:29:20 AM UTC John MacFarlane wrote:

> Brandon Keith Biggs <brandonk...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes: 
>
> > Hello, 
> > I'm not a big fan of the proposed syntax because it's more verbose and 
> > programmer-oriented than the current syntax. 
> > Is there some way where we could figure out if the [@authorname] is at a 
> > period? I am having a hard time thinking of a time when [@authorname] is 
> > currently not at the end of a sentence. 
> > Perhaps the default could be [@authorname] would figure out if it's 
> > followed by a period and if so, then it will do the (authorname, 2019). 
> > Otherwise it will do authorname (2019). 
>
> I thought about trying to do it automatically, but I don't 
> that is going to be reliable. The idea would be to trigger 
> author-in-text form if the citation comes after certain 
> punctuation marks (period, comma, colon, semicolon) + space. 
> But this might give bad results for 
>
> - abbreviations: blah blah etc. [@foo2002] 
> - people who use note styles and put the normal (note) citations 
> right after the sentence: blah blah.[@foo2002] 
> - and more?? 
>
>

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/50695cec-f273-4efe-9caf-3d8be60e0704n%40googlegroups.com.

[-- Attachment #1.2: Type: text/html, Size: 3544 bytes --]

Re: considering syntax change for author-in-text citations

[Brandon Keith Biggs]

[-- Attachment #1: Type: text/plain, Size: 1738 bytes --]

Couldn't you put [@authorname.]
Rather than [@authorname].
Thanks,

Brandon Keith Biggs <http://brandonkeithbiggs.com/>


On Sun, Nov 29, 2020 at 4:29 PM John MacFarlane <jgm-TVLZxgkOlNX2fBVCVOL8/A@public.gmane.org> wrote:

> Brandon Keith Biggs <brandonkeithbiggs-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:
>
> > Hello,
> > I'm not a big fan of the proposed syntax because it's more verbose and
> > programmer-oriented than the current syntax.
> > Is there some way where we could figure out if the [@authorname] is at a
> > period? I am having a hard time thinking of a time when [@authorname] is
> > currently not at the end of a sentence.
> > Perhaps the default could be [@authorname] would figure out if it's
> > followed by a period and if so, then it will do the (authorname, 2019).
> > Otherwise it will do authorname (2019).
>
> I thought about trying to do it automatically, but I don't
> that is going to be reliable.  The idea would be to trigger
> author-in-text form if the citation comes after certain
> punctuation marks (period, comma, colon, semicolon) + space.
> But this might give bad results for
>
> - abbreviations:  blah blah etc. [@foo2002]
> - people who use note styles and put the normal (note) citations
>   right after the sentence:  blah blah.[@foo2002]
> - and more??
>
>

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAKAWQkVZO0Pb%3DzNRtTA72esKH2z-0V6%2BNvEMcyokSutzJNaUmw%40mail.gmail.com.

[-- Attachment #2: Type: text/html, Size: 2734 bytes --]

Re: considering syntax change for author-in-text citations

[John MacFarlane]

Julien Dutant <julien.dutant-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:

> Hi,
>
> I like the idea of making the syntax unambiguous. Another ambiguity that 
> was worrying me was punctation right after the citation, e.g.:
>
> @Doe:2012: -> Doe (2012):
> @Smith.Doe:2013. -> Smith and Doe (2013). 
>
> Pandoc's manual specifies that the keys may only contain *internal* 
> punctuation, which ensures the results above. But that doesn't seem to be 
> required by CSL-JSON or BibTeX. 

This proposal isn't going to fix that issue.  After all, in
normal bracketed citations you still have to figure out whether
the final colon here is part of the cite key or the prefix:

[@Doe:2013: 3]

> How would the syntax behave with (attempted) prefixes? In the 
> author-suppressed format the result is normal citation with a botched 
> prefix: 
> [see @Doe2012, p.33] -> (see Doe 2012, p.33)
> [-see @Doe2012, p.33] -> (-see Doe 2012, p.33)

That's not how you do it:  rather,

[see -@Doe2012, p. 33]

You can't add prefixes to the author-in-text citations
with the current syntax.  With the bracketed + syntax, it would
become syntactically possible to add a prefix, which is okay I
guess.

Re: considering syntax change for author-in-text citations

[BPJ]

[-- Attachment #1: Type: text/plain, Size: 4452 bytes --]

I agree that [/@foo] looks like some kind of closing tag. I would prefer
[+@foo], to the extent that it can be construed as the opposite of [-@foo],
but can it really?

-- 
Better --help|less than helpless

Den sön 29 nov. 2020 20:46John Gabriele <jgabriele-97jfqw80gc6171pxa8y+qA@public.gmane.org> skrev:

> I haven't used citations, so I'm guessing that what you mean by "@foo  ->
> Foo (2013)" is that if I have `@foo` in my source document, it will be
> turned into "Foo (2013)" by Pandoc in the output.
>
> My only comment is that the forward slash: `[/@foo]` kind of looks like a
> closing tag, and reminded me of [BBCode](
> https://en.wikipedia.org/wiki/BBCode). Maybe you had some other thinking
> in-mind when considering the forward slash?
>
> -- John
>
>
> On Sat, Nov 28, 2020, at 7:39 PM, John MacFarlane wrote:
> >
> > I'm thinking about making a change in the citation syntax.
> >
> > Currently there are three forms:
> >
> > 1. author-in-text:
> >
> > @foo  -> Foo (2013)
> > @foo [p. 33] -> Foo (2013, 33)
> > @foo [p. 33; @bar, p. 45]  -> Foo (2013, p. 33; Bar 2015, p. 45)
> >
> > 2. normal:
> >
> > [@foo] -> (Foo 2013)
> > [@foo, p. 33]  -> (Foo 2013, p. 33)
> > [@foo, p. 33; @bar, p. 45]  -> (Foo 2013, p. 33; Bar 2015, p. 45)
> >
> > 3. suppress-author:
> >
> > [-@foo] ->  (2013)
> > [-@foo, p. 33]  -> (2013, p. 33)
> >
> > The proposal would be to change the author-in-text form so that
> > it requires square brackets:
> >
> > Old                           New
> > --------------------------    ---------------------------
> > @foo                          [/@foo]
> > @foo [p. 33]                  [/@foo, p. 33]
> > @foo [p. 33; @bar, p. 45]     [/@foo, p.33; @bar, p. 45]
> >
> > Mostly this is motivated by thinking about how I'm going
> > to implement the citation extension for commonmark.  It's
> > very hard to figure out how to support the current author-in-text
> > syntax nicely given the way we're parsing commonmark.
> > A secondary motivation is that the current @foo syntax
> > overlaps with references to example list items.  It would
> > be nice to have a syntax that is unambiguously a citation.
> >
> > I'm not wedded to the [/@foo]; [+@foo] would also be quite
> > natural given the current use of - to suppress author.  But the
> > `/` looks a bit better to me.  [:@foo] would be another option.
> >
> > I think this change would make text look a *little* less natural:
> >
> > @foo says blah  -> [/@foo] says blah.
> >
> > But it would make citations stand out more clearly from other
> > kinds of elements; it would make it possible to handle
> > citations in commonmark; and it would avoid syntactic
> > confusion between references to example list items and
> > citations.
> >
> > Probably what I'd do is implement the new syntax for citations
> > for commonmark+citations only, keeping the old one for
> > markdown+citations.  This could be a bit confusing, but it
> > would allow documents to be converted automatically to the
> > new form.
> >
> > I'd be interested in your thoughts on this.
> >
> > --
> > 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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> > To view this discussion on the web visit
> >
> https://groups.google.com/d/msgid/pandoc-discuss/m2k0u5ov1p.fsf%40MacBook-Pro.hsd1.ca.comcast.net
> .
> >
>
> --
> 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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/d32d39dd-8b51-4240-9fab-6fcbad083de8%40www.fastmail.com
> .
>

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CADAJKhCbFs1_s-opDZjD4V_Gk-qV5n9o7G_GSnrNddfD7mKvjg%40mail.gmail.com.

[-- Attachment #2: Type: text/html, Size: 6171 bytes --]

Re: considering syntax change for author-in-text citations

[Joost Kremers]

On Mon, Nov 30 2020, BPJ wrote:
> I agree that [/@foo] looks like some kind of closing tag. I would prefer
> [+@foo], to the extent that it can be construed as the opposite of [-@foo],
> but can it really?

IMHO no, it can't. [@foo] is "(Foo 2013)", so [-@foo] being "(2013)" makes kinda
sense, at least intuitively, because something is taken away, or "subtracted".
But "Foo (2013)" doesn't add anything compared to "(Foo 2013)".

That doesn't mean that [+@foo] would be a terrible choice, though. Given the
constraint that brackets are required, [+@foo] is probably as good an option as
any.


-- 
Joost Kremers
Life has its moments

Extended example list syntax? (was Re: considering syntax change for author-in-text citations)

[BPJ]

[-- Attachment #1: Type: text/plain, Size: 6545 bytes --]

Oh I forgot to mention that I think it would be very good if example list
references and citation references be clearly disambiguated. Especially if
that would mean that we could have namespaces for example lists like
`@foo:bar` vs. `@baz:bar`, with `foo` and `baz` being two different
counters. A while back I needed to reference items in a list from rows in a
table (i.e. the list was notes on the table, only they weren't footnotes)
like this:

``````markdown
| This | That | see
|------|------|------
| Blah | Blah | @blah

@blah) Note on Blah.
``````

except that since numbered examples were also needed I had to rely on some
rather fragile system with markers like `` `blah`{.tr} `` which I copied in
order using a Vim function, so that the notes ended up in the right order
based on first mention in the table, and then insert a div as an anchor

``````markdown
#)  ::: {#tr-blah}
``````

and finally a filter doing one pass "collecting" the list numbers and
another pass inserting links with the right number into the table. It ought
not to be quite that hard! Come to think of it it would also be good if one
could indicate the kind of list style one wants for a named/example list
item/reference, something like

-   `@foo:bar.1` decimal
-   `@foo:bar.i` lower roman
-   `@foo:bar.A` upper alpha

NB the character after the period should always be one of `1 i I a A` (or
perhaps `1 x X a A` to better distinguish between decimal and roman) as a
shorthand for the desired style, and *not* used to indicate the actual
start number used in the list. Come to think of it maybe the
counter/numbering style could be declared beforehand (at the top of the
document) through a syntax like `@@foo.x`, or even in a defaults file or
metadata, so that it could easily be changed globally for a given counter.

-- 
Better --help|less than helpless

Den mån 30 nov. 2020 09:03BPJ <bpj-J3H7GcXPSITLoDKTGw+V6w@public.gmane.org> skrev:

> I agree that [/@foo] looks like some kind of closing tag. I would prefer
> [+@foo], to the extent that it can be construed as the opposite of [-@foo],
> but can it really?
>
> --
> Better --help|less than helpless
>
> Den sön 29 nov. 2020 20:46John Gabriele <jgabriele-97jfqw80gc6171pxa8y+qA@public.gmane.org> skrev:
>
>> I haven't used citations, so I'm guessing that what you mean by "@foo  ->
>> Foo (2013)" is that if I have `@foo` in my source document, it will be
>> turned into "Foo (2013)" by Pandoc in the output.
>>
>> My only comment is that the forward slash: `[/@foo]` kind of looks like a
>> closing tag, and reminded me of [BBCode](
>> https://en.wikipedia.org/wiki/BBCode). Maybe you had some other thinking
>> in-mind when considering the forward slash?
>>
>> -- John
>>
>>
>> On Sat, Nov 28, 2020, at 7:39 PM, John MacFarlane wrote:
>> >
>> > I'm thinking about making a change in the citation syntax.
>> >
>> > Currently there are three forms:
>> >
>> > 1. author-in-text:
>> >
>> > @foo  -> Foo (2013)
>> > @foo [p. 33] -> Foo (2013, 33)
>> > @foo [p. 33; @bar, p. 45]  -> Foo (2013, p. 33; Bar 2015, p. 45)
>> >
>> > 2. normal:
>> >
>> > [@foo] -> (Foo 2013)
>> > [@foo, p. 33]  -> (Foo 2013, p. 33)
>> > [@foo, p. 33; @bar, p. 45]  -> (Foo 2013, p. 33; Bar 2015, p. 45)
>> >
>> > 3. suppress-author:
>> >
>> > [-@foo] ->  (2013)
>> > [-@foo, p. 33]  -> (2013, p. 33)
>> >
>> > The proposal would be to change the author-in-text form so that
>> > it requires square brackets:
>> >
>> > Old                           New
>> > --------------------------    ---------------------------
>> > @foo                          [/@foo]
>> > @foo [p. 33]                  [/@foo, p. 33]
>> > @foo [p. 33; @bar, p. 45]     [/@foo, p.33; @bar, p. 45]
>> >
>> > Mostly this is motivated by thinking about how I'm going
>> > to implement the citation extension for commonmark.  It's
>> > very hard to figure out how to support the current author-in-text
>> > syntax nicely given the way we're parsing commonmark.
>> > A secondary motivation is that the current @foo syntax
>> > overlaps with references to example list items.  It would
>> > be nice to have a syntax that is unambiguously a citation.
>> >
>> > I'm not wedded to the [/@foo]; [+@foo] would also be quite
>> > natural given the current use of - to suppress author.  But the
>> > `/` looks a bit better to me.  [:@foo] would be another option.
>> >
>> > I think this change would make text look a *little* less natural:
>> >
>> > @foo says blah  -> [/@foo] says blah.
>> >
>> > But it would make citations stand out more clearly from other
>> > kinds of elements; it would make it possible to handle
>> > citations in commonmark; and it would avoid syntactic
>> > confusion between references to example list items and
>> > citations.
>> >
>> > Probably what I'd do is implement the new syntax for citations
>> > for commonmark+citations only, keeping the old one for
>> > markdown+citations.  This could be a bit confusing, but it
>> > would allow documents to be converted automatically to the
>> > new form.
>> >
>> > I'd be interested in your thoughts on this.
>> >
>> > --
>> > 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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>> > To view this discussion on the web visit
>> >
>> https://groups.google.com/d/msgid/pandoc-discuss/m2k0u5ov1p.fsf%40MacBook-Pro.hsd1.ca.comcast.net
>> .
>> >
>>
>> --
>> 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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/pandoc-discuss/d32d39dd-8b51-4240-9fab-6fcbad083de8%40www.fastmail.com
>> .
>>
>

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CADAJKhBSMs-cPmakJn5V8C1xOeHVf8fg-F3GqZoVqLVODGF3%2BA%40mail.gmail.com.

[-- Attachment #2: Type: text/html, Size: 9069 bytes --]

Re: considering syntax change for author-in-text citations

[BPJ]

[-- Attachment #1: Type: text/plain, Size: 1786 bytes --]

Agree on both counts: they are not opposites, but [+@foo] is probably still
not a bad choice.

-- 
Better --help|less than helpless

Den mån 30 nov. 2020 09:33Joost Kremers <joostkremers-97jfqw80gc6171pxa8y+qA@public.gmane.org> skrev:

>
> On Mon, Nov 30 2020, BPJ wrote:
> > I agree that [/@foo] looks like some kind of closing tag. I would prefer
> > [+@foo], to the extent that it can be construed as the opposite of
> [-@foo],
> > but can it really?
>
> IMHO no, it can't. [@foo] is "(Foo 2013)", so [-@foo] being "(2013)" makes
> kinda
> sense, at least intuitively, because something is taken away, or
> "subtracted".
> But "Foo (2013)" doesn't add anything compared to "(Foo 2013)".
>
> That doesn't mean that [+@foo] would be a terrible choice, though. Given
> the
> constraint that brackets are required, [+@foo] is probably as good an
> option as
> any.
>
>
> --
> Joost Kremers
> Life has its moments
>
> --
> 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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/87h7p7nt2a.fsf%40fastmail.fm
> .
>

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CADAJKhDXKUVRVOo2vYNjrx_gkWHcZCX5GQYTx1sQyzvo6%2BHRMQ%40mail.gmail.com.

[-- Attachment #2: Type: text/html, Size: 2749 bytes --]

Re: considering syntax change for author-in-text citations

[Joseph Reagle]

On 11/28/20 7:39 PM, John MacFarlane wrote:
> I'm not wedded to the [/@foo]; [+@foo] would also be quite natural given the current use of - to suppress author. 

I don't like the slash, at a glance I naively read it as some sort of escaping -- though I know it's a forward slash. If `-` means suppress the author, `+` can be explained as include the author in prose.

I'm also wondering about my many files that use author-in-text. Is the expectation that pandoc will *always* use markdown (rather than commonmark) as the default input format even when commonmark becomes dominant?

Re: considering syntax change for author-in-text citations

[Albert Krewinkel]

I'm in favor of the change, but I'd certainly miss the old syntax.
Authors using non-English keyboard layouts may find this a little
annoying, as square brackets are often less convenient to reach (at
least on a German QUERTZ keyboard).

Would a switch which enables the current syntax while disabling
example lists make parsing any easier? It would be nice if there was
an extension to restore the present citation syntax, especially
given how pandoc-crossref etc. plug into it.

John MacFarlane writes:

> I'm thinking about making a change in the citation syntax.
>
> Currently there are three forms:
>
> 1. author-in-text:
>
> @foo  -> Foo (2013)
> @foo [p. 33] -> Foo (2013, 33)
> @foo [p. 33; @bar, p. 45]  -> Foo (2013, p. 33; Bar 2015, p. 45)
>
> 2. normal:
>
> [@foo] -> (Foo 2013)
> [@foo, p. 33]  -> (Foo 2013, p. 33)
> [@foo, p. 33; @bar, p. 45]  -> (Foo 2013, p. 33; Bar 2015, p. 45)
>
> 3. suppress-author:
>
> [-@foo] ->  (2013)
> [-@foo, p. 33]  -> (2013, p. 33)
>
> The proposal would be to change the author-in-text form so that
> it requires square brackets:
>
> Old                           New
> --------------------------    ---------------------------
> @foo                          [/@foo]
> @foo [p. 33]                  [/@foo, p. 33]
> @foo [p. 33; @bar, p. 45]     [/@foo, p.33; @bar, p. 45]
>
> Mostly this is motivated by thinking about how I'm going
> to implement the citation extension for commonmark.  It's
> very hard to figure out how to support the current author-in-text
> syntax nicely given the way we're parsing commonmark.
> A secondary motivation is that the current @foo syntax
> overlaps with references to example list items.  It would
> be nice to have a syntax that is unambiguously a citation.
>
> I'm not wedded to the [/@foo]; [+@foo] would also be quite
> natural given the current use of - to suppress author.  But the
> `/` looks a bit better to me.  [:@foo] would be another option.
>
> I think this change would make text look a *little* less natural:
>
> @foo says blah  -> [/@foo] says blah.
>
> But it would make citations stand out more clearly from other
> kinds of elements; it would make it possible to handle
> citations in commonmark; and it would avoid syntactic
> confusion between references to example list items and
> citations.
>
> Probably what I'd do is implement the new syntax for citations
> for commonmark+citations only, keeping the old one for
> markdown+citations.  This could be a bit confusing, but it
> would allow documents to be converted automatically to the
> new form.
>
> I'd be interested in your thoughts on this.


--
Albert Krewinkel
GPG: 8eed e3e2 e8c5 6f18 81fe  e836 388d c0b2 1f63 1124

Re: considering syntax change for author-in-text citations

[FI Apps]

[-- Attachment #1: Type: text/plain, Size: 1574 bytes --]

On Mon, Nov 30, 2020 at 9:32 AM Joost Kremers <joostkremers-97jfqw80gc6171pxa8y+qA@public.gmane.org>
wrote:

>
> On Mon, Nov 30 2020, BPJ wrote:
> > I agree that [/@foo] looks like some kind of closing tag. I would prefer
> > [+@foo], to the extent that it can be construed as the opposite of
> [-@foo],
> > but can it really?
>
> IMHO no, it can't. [@foo] is "(Foo 2013)", so [-@foo] being "(2013)" makes
> kinda
> sense, at least intuitively, because something is taken away, or
> "subtracted".
> But "Foo (2013)" doesn't add anything compared to "(Foo 2013)".
>
> That doesn't mean that [+@foo] would be a terrible choice, though. Given
> the
> constraint that brackets are required, [+@foo] is probably as good an
> option as
> any.
>

No-author and author-in-text aren't exact opposites, but there is a partial
opposition in that the former suppresses the author's name, while the
latter ensures the author's name will appear. In a citation style that
doesn't use ibid., this is less evident, but in those that do, it adds the
author's name, producing "Foo, ibid., 125" or "Foo (ibid., 125)" instead of
"ibid., 125". This is why I suggested "+".

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAGOSsdkP7iN3Y5RdQkDCdXZPKGP-fHa_HGHWTRmxiwxG4z9HBQ%40mail.gmail.com.

[-- Attachment #2: Type: text/html, Size: 2401 bytes --]

Re: considering syntax change for author-in-text citations

[John MacFarlane]

Joost Kremers <joostkremers-97jfqw80gc6171pxa8y+qA@public.gmane.org> writes:

> On Mon, Nov 30 2020, BPJ wrote:
>> I agree that [/@foo] looks like some kind of closing tag. I would prefer
>> [+@foo], to the extent that it can be construed as the opposite of [-@foo],
>> but can it really?
>
> IMHO no, it can't. [@foo] is "(Foo 2013)", so [-@foo] being
> "(2013)" makes kinda sense, at least intuitively, because
> something is taken away, or "subtracted".  But "Foo (2013)"
> doesn't add anything compared to "(Foo 2013)".

True.  I suppose one could use something like
[<-@foo2013] which would indicate that a part of the citation
is moving outside the brackets.

But that starts to look even uglier.

I agree with earlier comments that the bracketed syntax doesn't
look as natural, so I might try harder to find a way around it.

Re: considering syntax change for author-in-text citations

[John MacFarlane]

Joseph Reagle <joseph.2011-T1oY19WcHSwdnm+yROfE0A@public.gmane.org> writes:

> I'm also wondering about my many files that use author-in-text. Is the expectation that pandoc will *always* use markdown (rather than commonmark) as the default input format even when commonmark becomes dominant?

Well, my plan is eventually to make extended commonmark the default
"pandoc markdown."  I'd keep the legacy parser around for a while
after that, to allow people to use it to transition. (Automatic
conversion of yoru documents should be possible.)

Re: considering syntax change for author-in-text citations

[John MacFarlane]

Albert Krewinkel <albert+pandoc-9EawChwDxG8hFhg+JK9F0w@public.gmane.org> writes:

> Would a switch which enables the current syntax while disabling
> example lists make parsing any easier? It would be nice if there was
> an extension to restore the present citation syntax, especially
> given how pandoc-crossref etc. plug into it.

I hadn't thought about that aspect, but that's a serious strike
against the proposal, at least until we can reproduce
pandoc-crossref's functionality in pandoc itself.

Re: considering syntax change for author-in-text citations

[Joseph Reagle]

On 11/30/20 12:15 PM, John MacFarlane wrote:
> Well, my plan is eventually to make extended commonmark the default
> "pandoc markdown."  I'd keep the legacy parser around for a while
> after that, to allow people to use it to transition. (Automatic
> conversion of yoru documents should be possible.)

If so, that gets back to my earlier question about a capability to preserve as much of the existing options/extensions as possible when doing such a translation. It probably won't be a biggie, but when pandoc moves to commonmark I'll have to give some thought to migrating content.

-- 
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-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a812d519-3927-f041-2b22-3929489b5ac2%40reagle.org.

AW: considering syntax change for author-in-text citations

[denis.maier-FfwAq0itz3ofv37vnLkPlQ]

Anything new here?

> Joost Kremers <joostkremers-97jfqw80gc6171pxa8y+qA@public.gmane.org> writes:
> 
> > On Mon, Nov 30 2020, BPJ wrote:
> >> I agree that [/@foo] looks like some kind of closing tag. I would
> >> prefer [+@foo], to the extent that it can be construed as the
> >> opposite of [-@foo], but can it really?
> >
> > IMHO no, it can't. [@foo] is "(Foo 2013)", so [-@foo] being "(2013)"
> > makes kinda sense, at least intuitively, because something is taken
> > away, or "subtracted".  But "Foo (2013)"
> > doesn't add anything compared to "(Foo 2013)".
> 
> True.  I suppose one could use something like [<-@foo2013] which would
> indicate that a part of the citation is moving outside the brackets.

[+(-@foo2013], i.e. no author inside the parens,

Or we could even make it explicit:

@foo2013 [-@foe2013]
That would be a citation composed of two individual citations. 
@foo2013 -> renders only the author name in text. (like \citeauthor)
[-@foo2013] -> renders the citations in parentheses with author name suppressed.
The obvious advantage of this is that you can have constructs like this one:

Foo argues that "asdf asdf asdf" (2013, p. 89).
@foo2013 argues that "asdf asdf asdf" [-@foo2013, 89].

On the other hand, the simpler construct "Foe (2013, 89)" is getting a lot more verbose, so a simple shorthand might be nice.

> 
> But that starts to look even uglier.

Yep.

And I don't know if any of that is useful regarding commonmark parsing.

> 
> I agree with earlier comments that the bracketed syntax doesn't look as
> natural, so I might try harder to find a way around it.

Do you have any new suggestions or ideas on this?