Open denismaier opened 4 years ago
In general, 1.1 will have support for open-ended ranges on data input (not sure we've explicitly dealt with styling yet); locators and dates.
On Wed, Sep 2, 2020, 5:29 PM Denis Maier notifications@github.com wrote:
There's been a request on pandoc-discuss https://groups.google.com/g/pandoc-discuss/c/aPl79Of7MZ4 regarding open-ended page ranges:
Is it possible to have pandoc-citeproc interpret a locator like 12ff as multiple pages? Currently, I get p. 12ff while pp. 12ff would be correct.
I don't think there's support for that at the moment, right? Also, I don't think 1.1 will add support for this yet, correct? Or, am I just missing something here?
Arguably, open-ended page-ranges are not really useful for readers, but in some traditions, it's rather common. In German language academia you'll often even find that page-ranges are always shortened to starting page + "ff.". So, "215-243" will be "215ff." (When writing in German, it's actually not a big deal since the abbreviation for "Seite" and "Seiten" is in both cases "S.").
Both citeproc-js and pandoc-citeproc seem to have some support for this already. A locator "215-" will be interpreted as an open-ended range, and the label will be adjusted accordingly, so in citeproc-js the result is "pp. 215-", with pandoc-citeproc it's "pp. 215".
I suggest we should add some support for this. Some possibilities:
- a new term psqq à la biblatex that could be used with open-ended page-ranges.
- a new term psq, and a new option to support "12-13"=>"12f." like behavior.
- a new option to support unconditional collapsing of page-ranges into first page + "ff.": "12-25" => "12ff."
Thought? @jgm https://github.com/jgm @cormacrelf https://github.com/cormacrelf @fbennett https://github.com/fbennett @andras-simonyi https://github.com/andras-simonyi @bdarcus https://github.com/bdarcus @bwiernik https://github.com/bwiernik @georgd https://github.com/georgd
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/citation-style-language/schema/issues/372, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAAI3X6FDZAM3UXS6O5M73SD22NLANCNFSM4QTZYUVQ .
Yeah, input should be safe. Question is just what should happen in styles.
Allow me first to complain that this is one of those archaic style conventions like op cit that I wish would go away. 😜
It's not really clear to me exactly how open-ended ranges addresses the data needs for 183f.
versus 183ff.
It seems almost to be unrelated.
For example, if a cited range were 183–184
, shouldn't that be collapsed to 183f.
? Similarly, if the range is 183–187
, should that be collapsed to 183ff.
? What is the number of pages where the format changes from ff.
to actually spelling out the range?
It seems more like the 183f.
format would need to be addressed using a new page range format option alongside minimal, chicago, etc.
If I'm misunderstanding, could you elaborate on when the 183ff.
format versus an enumerated range is used?
(With respect to what is the expected output for open-ended ranges in typical styles? Would it be something like pp. 183–
?)
Allow me first to complain that this is one of those archaic style conventions like op cit that I wish would go away. 😜
I agree. But in some areas this convention is as firmly established as anything. So I don’t see it going away.
It's not really clear to me exactly how open-ended ranges addresses the data needs for
183f.
versus183ff.
It seems almost to be unrelated.
Agreed, the two are unrelated in respect to the question of open-ended ranges. The former is not open-ended but simply a two-page range. They are only related with respect to the proposal by @denismaier that a new term should be added which would be singular for the two-page range and plural for any other page range.
For example, if a cited range were
183–184
, shouldn't that be collapsed to183f.
? Similarly, if the range is183–187
, should that be collapsed to183ff.
? What is the number of pages where the format changes fromff.
to actually spelling out the range?
All the styles I know of to require page-range collapsing (all of them German language, in different areas, mostly history, business, and law), do so unconditionally. In such a style, a two-page range is always 183 f., and any other page-range is always 183 ff.
Seems like adding a new page range option collapse option would be the approach then.
183f.
is indeed not open-ended. It's a two-page range, with special collapsing behavior being applied. So, yes, a new new page range format (together with an appropriate term) would be the way to go.
Regarding ff.
there are two scenarios:
183-185
will always be rendered as 183ff.
. We don't actually have an open-ended range here. So that would be just another new page range format.183-
, with the desired output 183ff.
, or maybe 183-
, which would correspond to Chicago's open ended year range format. I don't know of a style that uses pp. 183-
though. Chicago's take on this, 14.149:
Only when referring to a section for which no final number can usefully be given should ff. (“and the following pages, paragraphs, etc.”) be resorted to.
So, it seems this could boil down to this:
psq
~ following-page
with two forms:
f.
ff.
(Alternatively, the plural could be a term ~psqq
~ following-pages
on its own.)psq
~ collapse-following-page
=> collapse two-page ranges into "starting page" + "f."psqq
~ collapse-following-pages
=> same as ~psq
~ following-page
+ collapse multi-page-ranges into "starting page" + "ff."@use-psqq-with-open-ended-page-range
~ @use-following-pages-with-open-ended-page-range
(the naming might still be improved ;-)); if true, open-ended page ranges are rendered as "starting-page" + "ff."2. A user might want, for whatever reason, to occassionally use proper open-ended ranges ...
If you mean, for some locators use ff, but in others not, then not clear to me we should support this requirement.
If you mean, for some locators use ff, but in others not, then not clear to me we should support this requirement.
Yes, I mean this:
If a user enters a page-range like 183-187
, format it according to the chosen page range format.
But, if they enter 183-
, format it as 183ff.
We might not even need a new option for this. If a user really wants an open-ended range, I don't see how that should be formatted differently. (The ff.
must be localized of course.)
What's psq mean?
psq is borrowed from biblatex; it's a combination of two abbreviations: p + sq => the following page.
What's the sq stand for?
sq = lat. sequens; the following one see https://www.collinsdictionary.com/dictionary/english/sq sqq is the plural
We might as well decide to not use latin abbreviations if there are suggestions for a different name.
following-page
?
following-page
?
No please. That would be way too obvious.
I've updated my suggestion above.
- Add a new attribute ~
@use-psqq-with-open-ended-page-range
~@use-following-pages-with-open-ended-page-range
(the naming might still be improved ;-)); if true, open-ended page ranges are rendered as "starting-page" + "ff."
Rather than a boolean with a specific name, how about a more general name with a specific value; say @page-open-range-format="following"
, with idea open ranges in general might just default to 1900-
.
new page range options:
- ~
psq
~collapse-following-page
=> collapse two-page ranges into "starting page" + "f."- ~
psqq
~collapse-following-pages
=> same as ~psq
~following-page
+ collapse multi-page-ranges into "starting page" + "ff."- Add a new attribute ~
@use-psqq-with-open-ended-page-range
~@use-following-pages-with-open-ended-page-range
(the naming might still be improved ;-)); if true, open-ended page ranges are rendered as "starting-page" + "ff."
@denismaier I understand that with this setup you suggest that only in certain cases, an open-ended page-range will be rendered with "starting-page" + "ff.". IMO, it should be the other way around. When collapse-following-page
is set, open-ended page-ranges always should be rendered with "ff." (to me, "S. 123 ff." looks less confusing than "S. 123-"). At least, I’ve never seen such open-ended range with anything but "ff." [edit] in styles that use this kind of collapsing.
Maybe we can simplify the terms even more:
we could just call it following
. Then, this term could also be used for open-ended year ranges. We can then have two attributes: @page-open-range-format="following"
.
@year-open-range-format="following"
.
@denismaier I understand that with this setup you suggest that only in certain cases, an open-ended page-range will be rendered with "starting-page" + "ff.". IMO, it should be the other way around. When
collapse-following-page
is set, open-ended page-ranges always should be rendered with "ff." (to me, "S. 123 ff." looks less confusing than "S. 123-"). At least, I’ve never seen such open-ended range with anything but "ff." [edit] in styles that use this kind of collapsing.
Good point. OTOH, you might want to use this kind of formatting only if the user supplies an open-ended range, but you might not want to collapse "183-187".
Perhaps collapse-following-pages
could automatically set @page-open-range-format="following"
. That would be what you describe. But you could still use @page-open-range-format="following"
with other page formats as well.
@denismaier I understand that with this setup you suggest that only in certain cases, an open-ended page-range will be rendered with "starting-page" + "ff.". IMO, it should be the other way around. When
collapse-following-page
is set, open-ended page-ranges always should be rendered with "ff." (to me, "S. 123 ff." looks less confusing than "S. 123-"). At least, I’ve never seen such open-ended range with anything but "ff." [edit] in styles that use this kind of collapsing.Good point. OTOH, you might want to use this kind of formatting only if the user supplies an open-ended range, but you might not want to collapse "183-187".
That portion of my answer got lost. I wanted to state that the closed page-range with more than two pages might be seen as the "marked" case which would not be collapsed in certain circumstances.
Perhaps
collapse-following-pages
could automatically set@page-open-range-format="following"
. That would be what you describe. But you could still use@page-open-range-format="following"
with other page formats as well.
I think, collapse-following-page
and collapse-following-pages
both should implicitly set @page-open-range-format="following"
(and perhaps collapse-following-pages
should set collapse-following-page
as well?)
(and perhaps
collapse-following-pages
should setcollapse-following-page
as well?)
Sure. These are not booleans. They are selected as page-range-format="collapse-following-page"
or page-range-format="collapse-following-pages"
.
page-range-format="collapse-following-pages"
would collapse two-page ranges to "starting page" + "f."; longer finite page ranges will trigger the "ff." collapsing.
(and perhaps
collapse-following-pages
should setcollapse-following-page
as well?)Sure. These are not booleans. They are selected as
page-range-format="collapse-following-page"
orpage-range-format="collapse-following-pages"
.
page-range-format="collapse-following-pages"
would collapse two-page ranges to "starting page" + "f."; longer finite page ranges will trigger the "ff." collapsing.
I see. Well, I think that’s a good setting.
I'm wondering if we really need two mechanisms here (page-range-format
and page-open-range-format
). If we added page-range-format="collapse-following-pages"
(which would use f.
or ff.
as appropriate) and added the open-range formatting specification to page-range-format="chicago"
, would that suffice?
It seems like the 18ff.
style is very uncommon, so I'm not sure we need to provide separate control of the open page range formatting.
There's been a request on pandoc-discuss regarding open-ended page ranges:
I don't think there's support for that at the moment, right? Also, I don't think 1.1 will add support for this yet, correct? Or, am I just missing something here?
Arguably, open-ended page-ranges are not really useful for readers, but in some traditions, it's rather common. In German language academia you'll often even find that page-ranges are always shortened to starting page + "ff.". So, "215-243" will be "215ff." (When writing in German, it's actually not a big deal since the abbreviation for "Seite" and "Seiten" is in both cases "S.").
Both citeproc-js and pandoc-citeproc seem to have some support for this already. A locator "215-" will be interpreted as an open-ended range, and the label will be adjusted accordingly, so in citeproc-js the result is "pp. 215-", with pandoc-citeproc it's "pp. 215".
I suggest we should add some support for this. Some possibilities:
psqq
à la biblatex that could be used with open-ended page-ranges.psq
, and a new option to support "12-13"=>"12f." like behavior.Thought? @jgm @cormacrelf @fbennett @andras-simonyi @bdarcus @bwiernik @georgd