Just when you thought it was safe to go in the water again …
When there are more than seven authors of a work, list the first six,
and the last, separated by an ellipsis. Looks like this wants an
option attribute on citation/bibliography. Perhaps something like
last-name-as-et-al=“true”?
Not pushing this for CSL 1.0, just noting to the list so that we have
it in the mix.
Frank
Maybe, in addition to et-al-use-first, we could add et-al-use-last (which
also accepts any positive integer value, and counts backwards). We’d still
need a way to specify the ellipsis (…), though. Maybe we can also
rename the “symbol” option of the and-attribute on cs:name to “ampersand”,
and add a new value, “ellipsis”?
RintzeOn Wed, Sep 23, 2009 at 2:01 PM, Frank Bennett <@Frank_Bennett>wrote:
Maybe, in addition to et-al-use-first, we could add et-al-use-last (which
also accepts any positive integer value, and counts backwards). We’d still
need a way to specify the ellipsis (…), though. Maybe we can also
rename the “symbol” option of the and-attribute on cs:name to “ampersand”,
and add a new value, “ellipsis”?
Do you think there will be a call ever for more than one name at the tail end?
I never thought someone would come up with this sort of scheme to
begin with. So I don’t have much confidence some other editorial group
won’t come along and say “ah, this is cool, but let’s use last two”.
Bruce
Maybe, in addition to et-al-use-first, we could add et-al-use-last (which
also accepts any positive integer value, and counts backwards). We’d still
need a way to specify the ellipsis (…), though. Maybe we can also
rename the “symbol” option of the and-attribute on cs:name to “ampersand”,
and add a new value, “ellipsis”?Do you think there will be a call ever for more than one name at the tail end?
I never thought someone would come up with this sort of scheme to
begin with. So I don’t have much confidence some other editorial group
won’t come along and say “ah, this is cool, but let’s use last two”.
I feel your pain. Rather too well, I’m afraid.
Might be worth getting someone to double-check that this is a
mandatory feature of the new APA. The explanation given in the blog
entry linked in the forum verges on psychotic raving – essentially,
“every once in awhile, a list of authors has chiastic structure, which
is why we’ve assumed that all of them do”. Even I can see the case
for boycotting this rule, if there’s a way to dodge it.
Frank
Maybe, in addition to et-al-use-first, we could add et-al-use-last
(which
also accepts any positive integer value, and counts backwards). We’d
still
need a way to specify the ellipsis (…), though. Maybe we can also
rename the “symbol” option of the and-attribute on cs:name to
“ampersand”,
and add a new value, “ellipsis”?Do you think there will be a call ever for more than one name at the
tail end?I never thought someone would come up with this sort of scheme to
begin with. So I don’t have much confidence some other editorial group
won’t come along and say “ah, this is cool, but let’s use last two”.
Might be worth getting someone to double-check that this is a
mandatory feature of the new APA.
It’s noted in other (derived) online resources:
http://libwebdev2006.fullerton.edu/Documents/APA_6th-2009-Final.pdf
(search for 6.27)
The explanation given in the blog
entry linked in the forum verges on psychotic raving – essentially,
“every once in awhile, a list of authors has chiastic structure, which
is why we’ve assumed that all of them do”.
I actually think it makes sense (even though it may not be very pretty).
But having a lot of co-authors with the principle investigator at the end is
more common in the natural sciences, I guess.
RintzeOn Wed, Sep 23, 2009 at 4:02 PM, Frank Bennett <@Frank_Bennett>wrote:
…
I actually think it makes sense
But having a lot of co-authors with the principle investigator at the end is
more common in the natural sciences, I guess.
But as I say in my comment*, there’s a big difference if the logic is
to include the PI; suggests a different solution.
Bruce
I actually think it makes sense
But having a lot of co-authors with the principle investigator at the end is
more common in the natural sciences, I guess.
I stand corrected – I had no idea this was a common practice. Learn
something everyday.
Frank
I presume their logic is that the last author is most probably the (main)
PI, and that guessing is the preferred option. I guess the alternative
you’re thinking of would be to give the PI(s) a dedicated creator type, so
that they can be formatted separately? But that would require additional
curating of the data (you would have to figure out who the PI(s) are, which
isn’t always obvious).
Rintze
…
I actually think it makes sense
pretty).
But having a lot of co-authors with the principle investigator at the
end is
more common in the natural sciences, I guess.But as I say in my comment*, there’s a big difference if the logic is
to include the PI; suggests a different solution.I presume their logic is that the last author is most probably the (main)
PI, and that guessing is the preferred option. I guess the alternative
you’re thinking of would be to give the PI(s) a dedicated creator type, so
that they can be formatted separately?
Yes; it’s one obvious alternative.
But that would require additional
curating of the data (you would have to figure out who the PI(s) are, which
isn’t always obvious).
But the “last author” approach has its own problems.
Bruce
Such as?
Rintze
If the intention is only to list the PI, and the last author is not
the PI, then it is wrong!
Bruce
As discussed in the reply to your APA blog comment, that’s not the aim.
Rintze
Coming back to this, as a ticket has been opened:
(BTW, I agree this is something for post-CSL 1.0)
So the goal is to express this:
Terracciano, A., Abdel-Khalek, A. M., Adam, N., Adamovova, L., Ahn, C.,
Ahn, H., . . . McCrae, R. R.
I think with the current schema (with expanded et-al support) it would be
sufficient to just have the earlier proposed et-al-use-last attribute. When
specified, this would include the number of authors specified with the
attribute after the et-al-term, e.g.:
The term “and others” can then be set to the ellipsis ("…").
There are of course some overlap issues here, e.g. if et-al-min=“3”,
et-al-use-first=“2” and et-al-use-last=“2”. It probably makes sense to
remove names that have appeared from the use of et-al-use-first from those
that can be outputted with et-al-use-last. There may be some delimiter
issues lingering, so perhaps we need to use affixes on the et-al term, but I
think that’s the only problem with this approach.
Rintze
As a slight modification to my earlier proposal, I was thinking we might not
need the ellipsis as a (redefined) term. Instead the CSL processor could
just implicitly add it if et-al abbreviation is active and et-al-use-last is
set.
(earlier, I was a bit worried that some other style might want an extra
comma before the last name, like: “Terracciano, A., Abdel-Khalek, A. M.,
Adam, N., Adamovova, L., Ahn, C., Ahn, H., . . ., McCrae, R. R.”. But it
might be better to worry about that later once we come across such a style)
RintzeOn Wed, Dec 16, 2009 at 8:58 AM, Rintze Zelle <@Rintze_Zelle>wrote:
So how confident are you in this proposal?
What does everyone think we ought to do about this? For context, it’s
required to fully support the new APA style.
Please register comments at ticket:
http://bitbucket.org/bdarcus/csl-schema/issue/14/list-last-authors-name-in-truncated-authors
My big question is whether we could consider this a backward
compatible change, and so fold into some minor release?
Bruce
For those only following the xbiblio mailing list, and not the Bitbucket
issue tracker: there seems to be consensus that APA’s new rule for
references with more than 7 authors can be supported with a new boolean
et-al-use-last option and a term for the ellipsis. The proposed spec entry
can be found here:
The only thing holding me back from implementing this in the schema is that
I realized we might need a names-use-last attribute on cs:key. cs:key
already has the names-min and names-use-first attributes to override the
values of et-al-min and et-al-use-first during sorting (
http://citationstyles.org/downloads/specification.html#sorting) and I think
names-use-last would be needed to give full control over name sorting. I was
curious whether everyone agreed with this analysis and proposed addition.
Rintze