# "document" item type

**URL:** <https://discourse.citationstyles.org/t/document-item-type/394>\
**Category:** CSL Development\
**Created:** [December 15, 2007, 8:43pm UTC](https://discourse.citationstyles.org/t/document-item-type/394 "2007-12-15T20:43:11Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Elena\_Razlogova](https://avatars.discourse-cdn.com/v4/letter/e/c89c15/32.png) [@Elena\_Razlogova](https://discourse.citationstyles.org/u/Elena_Razlogova)\
**Post date:** [December 15, 2007, 8:43pm UTC](https://discourse.citationstyles.org/t/document-item-type/394/1 "2007-12-15T20:43:11Z")

</div>

Bruce–

This discussion got pretty arcane on the forums so I moved it to dev  
lists:

> **[Add "Press Release" to droplist of publication/media type](https://forums.zotero.org/discussion/504/add-press-release-to-droplist-)**
>
> With MLA export to bibliography:

of-publicationmedia-type/#Item\_10

You advise:

> > > Add “Press Release” to droplist of publication/media type  
> > > Just use “document”. IMHO there’s nothing particularly special  
> > > about a press release that warrants its own type. I say this as  
> > > someone who cites them.

Users can’t currently use “document” in Zotero because it does not  
include “type” or “place” fields.

In general, users can’t currently use “document” to cite all  
documents without a dedicated Zotero item type because there is no  
way to determine whether the title is cited in italics (as  
published), or quotation marks (as unpublished).

> in the excerpt you quote above, what logical principle allows the  
> CMS to say “is treated like an unpublished source”?

Titles for published documents are usually cited in italics, titles  
for unpublished documents are usually cited in quotation marks (this  
is basically what CMS means by “is treated like an unpublished source”).

Right now, most CSL item types, as a fallback, are mapped to either  
“book” (title in italics) or “article” (title in quotation marks).  
Because you want to cite both published and unpublished documents via  
the “document” item type, you won’t be able to fallback “document” to  
either of these.

> What are the characteristics of an “unpublished source”? That we  
> call it that? Or is it something really banal like a lack of a  
> publisher?

Both CMS and MLA treat “informally published material” as  
unpublished, and I suspect most other styles do too. Moreover,  
sometimes unpublished document do include a name of an institution in  
a citation.

For example, this report citation does not include a publisher  
string, yet the title is formatted in italics:

Merrill Lynch Advisory Services Group, _Merrill Lynch Consults  
Service, Disclosure Statement_, April 2000.

> **[The Chicago Manual of Style, 17th Edition](https://www.chicagomanualofstyle.org/ch17/ch17_sec241.html)**
>
> Find it. Write it. Cite it. The Chicago Manual of Style Online is the venerable, time-tested guide to style, usage, and grammar in an accessible online format. ¶ It is the indispensable reference for writers, editors, proofreaders, indexers,...

whereas a working paper may include the name of the institution where  
it was drafted, formatted as a publisher would be, yet the title  
would be enclosed in quotation marks.

I. Clark, “Should the IMF Become More Adaptive?” Working Paper WP/  
96/11 (Washington, DC: International Monetary Fund, 1996).

> **[The Chicago Manual of Style, 17th Edition](https://www.chicagomanualofstyle.org/ch17/ch17_sec355.html)**
>
> Find it. Write it. Cite it. The Chicago Manual of Style Online is the venerable, time-tested guide to style, usage, and grammar in an accessible online format. ¶ It is the indispensable reference for writers, editors, proofreaders, indexers,...

Because of these quirks, “publisher” field by itself does not  
determine whether the document is formatted as unpublished.

> And what formatting rules trigger in that situation? Perhaps the  
> “news release” note gets added to better describe the resource?

Sure, but that doesn’t help with italics vs. quotation marks issue.  
To use “document” item type to cite “press release”, or any other  
document for which no clear item type exists, Zotero would need to  
add three fields to “document”:

1. “place” - to denote place of creation/publication

2. “type” - to enter the type of document

3. “publication status” - to denote whether the document is published  
or not.

“Publication status” string could then determine how the title is  
formatted.

This may be a possible solution for CSL (although I don’t think so)  
but it seems to me way too general for Zotero item types.

Currently in Zotero, odd published documents can be formatted as  
“report” item type (fallback to CSL “book”) and odd unpublished  
documents can be formatted as “manuscript” item type (fallback to CSL  
“article”). It works great in CSL and needs only minor tweaking with  
Zotero fields and CSL schema.

Unpublished documents are a pretty big category and need their own  
item type for convenient entry, just as “journal article” and  
“magazine article” need their own types right now instead of a  
generic “article” and a special field to denote “type of periodical”.  
Even in hierarchical item types I would suggest separating journal  
and magazine articles for convenience. For the same reason, I would  
suggest keeping “manuscript” for unpublished documents and “report”  
for odd published sources.

> I “cite press releases and other odd unpublished documents today”  
> as well, so to the extend we have different opinions on this, it  
> can’t be reduced to that.

You must not cite these types through Zotero/CSL then because you  
keep suggesting on forums that people use “document” to cite these  
despite the fact that currently (as you note yourself below) there is  
no way to cite items like press releases correctly:

> But the Document UI could use a “type” field so that users can  
> enter “press release.”

That’s just bad advice.

Best,  
Elena

---

<div class="post-metadata">

**Author:** ![Bruce\_D\_Arcus1](https://avatars.discourse-cdn.com/v4/letter/b/4bbf92/32.png) [@Bruce\_D\_Arcus1](https://discourse.citationstyles.org/u/Bruce_D_Arcus1)\
**Post date:** [December 15, 2007, 9:51pm UTC](https://discourse.citationstyles.org/t/document-item-type/394/2 "2007-12-15T21:51:53Z")

</div>

Elena Razlogova wrote:

> You advise:
> 
> > > > Add “Press Release” to droplist of publication/media type  
> > > > Just use “document”. IMHO there’s nothing particularly special  
> > > > about a press release that warrants its own type. I say this as  
> > > > someone who cites them.
> 
> Users can’t currently use “document” in Zotero because it does not  
> include “type” or “place” fields.
> 
> In general, users can’t currently use “document” to cite all  
> documents without a dedicated Zotero item type because there is no  
> way to determine whether the title is cited in italics (as  
> published), or quotation marks (as unpublished).

I’ll come back to this below …

> > in the excerpt you quote above, what logical principle allows the  
> > CMS to say “is treated like an unpublished source”?
> 
> Titles for published documents are usually cited in italics, titles  
> for unpublished documents are usually cited in quotation marks (this  
> is basically what CMS means by “is treated like an unpublished source”).

Italics vs. quoted is not a logical principle though. It’s just a  
convention some style uses to distinguish among different classes of  
resources. I’m asking for why they say this; what is it they’re  
distinguishing, and how?

> Right now, most CSL item types, as a fallback, are mapped to either  
> “book” (title in italics) or “article” (title in quotation marks).

Yes, and the larger structural/logical principle here is there are  
resources that are published as part of other resources, and there are  
those that stand alone. This explains the fallback behavior, and ends up  
working for most resources.

> Because you want to cite both published and unpublished documents via  
> the “document” item type, you won’t be able to fallback “document” to  
> either of these.

It’s true CSL makes no assumptions about published vs. unpublished. That  
distinction is an increasingly fuzzy one in the 21s century in any case.

> > What are the characteristics of an “unpublished source”? That we  
> > call it that? Or is it something really banal like a lack of a  
> > publisher?
> 
> Both CMS and MLA treat “informally published material” as  
> unpublished, and I suspect most other styles do too. Moreover,  
> sometimes unpublished document do include a name of an institution in  
> a citation.

Right, but the institution is not then a “publisher.”

> For example, this report citation does not include a publisher  
> string, yet the title is formatted in italics:
> 
> Merrill Lynch Advisory Services Group, _Merrill Lynch Consults  
> Service, Disclosure Statement_, April 2000.  
> [Page Not Found](http://www.chicagomanualofstyle.org/ch17/ch17_sec241.html)

Right; it’s treated “like a book” (to quote the CMS) because of its  
structural characteristics: that it’s a standalone document.

> whereas a working paper may include the name of the institution where  
> it was drafted, formatted as a publisher would be, yet the title  
> would be enclosed in quotation marks.
> 
> I. Clark, “Should the IMF Become More Adaptive?” Working Paper WP/  
> 96/11 (Washington, DC: International Monetary Fund, 1996).  
> [Page Not Found](http://www.chicagomanualofstyle.org/ch17/ch17_sec355.html)

I’m looking at this section of CMS and it’s not explaining this. So I  
can’t really comment on it one way or the other.

> Because of these quirks, “publisher” field by itself does not  
> determine whether the document is formatted as unpublished.

Only because you’re wanting to say that the “authorizing body” = a  
“publisher”. But it’s not, since it’s not published. Perhaps that indeed  
explains the different formatting in this case.

> > And what formatting rules trigger in that situation? Perhaps the  
> > “news release” note gets added to better describe the resource?
> 
> Sure, but that doesn’t help with italics vs. quotation marks issue.  
> To use “document” item type to cite “press release”, or any other  
> document for which no clear item type exists, Zotero would need to  
> add three fields to “document”:
> 
> 1. “place” - to denote place of creation/publication
> 
> 2. “type” - to enter the type of document
> 
> 3. “publication status” - to denote whether the document is published  
> or not.
> 
> “Publication status” string could then determine how the title is  
> formatted.
> 
> This may be a possible solution for CSL (although I don’t think so)  
> but it seems to me way too general for Zotero item types.

If a) it works, and b) it accurately describes the resource, what’s the  
problem?

> Currently in Zotero, odd published documents can be formatted as  
> “report” item type (fallback to CSL “book”) and odd unpublished  
> documents can be formatted as “manuscript” item type (fallback to CSL  
> “article”). It works great in CSL and needs only minor tweaking with  
> Zotero fields and CSL schema.

But IT"S WRONG! A working paper isn’t a report. A press release issued  
by the White House or the ACLU and available on the internet, is NOT a  
manuscript.

> Unpublished documents are a pretty big category and need their own  
> item type for convenient entry, just as “journal article” and  
> “magazine article” need their own types right now instead of a  
> generic “article” and a special field to denote “type of periodical”.  
> Even in hierarchical item types I would suggest separating journal  
> and magazine articles for convenience. For the same reason, I would  
> suggest keeping “manuscript” for unpublished documents and “report”  
> for odd published sources.

As above: I don’t think that’s a solution.

OTOH, perhaps breaking apart Document into Published Document and  
Unpublished Document might be a possibility, and then, for example,  
treating a Manuscript as a subclass of the latter. But that might  
introduce unforeseen problems (like for resources that can be either  
published or unpublished).

> > I “cite press releases and other odd unpublished documents today”  
> > as well, so to the extend we have different opinions on this, it  
> > can’t be reduced to that.
> 
> You must not cite these types through Zotero/CSL then because you  
> keep suggesting on forums that people use “document” to cite these  
> despite the fact that currently (as you note yourself below) there is  
> no way to cite items like press releases correctly:

I’m suggesting that every time someone has some problem that the correct  
solution is not to add a new type. If something like Document doesn’t  
fully work, then that’s a bug that needs fixing. Document in Zotero  
should allow me to fully store the funky stuff I run across; I shouldn’t  
need to ask the Zotero devs to add a new type.

> > But the Document UI could use a “type” field so that users can  
> > enter “press release.”
> 
> That’s just bad advice.

I think it equally “bad advice” to be saying I ought to call something a  
“manuscript” when it’s not.

You read one style guide Elena; we have to support thousands of  
potential styles, with often quite different lists of types\*, and even  
ways to distinguish them (media vs. intellectual content vs. ???). We  
can’t do that without linking specific style guidelines to general  
principles. And those principles, I believe, are not really about types.

Bruce

- Here’s a challenge for someone with too much time on their hands: take  
the top 10 style guides from across the sciences, humanities, social  
sciences, and law. Itemize every type mentioned in those guides. You’ll  
probably end up with hundreds. Now, try to condense them into a  
consolidated list that removes duplicates.

---

<div class="post-metadata">

**Author:** ![Elena\_Razlogova](https://avatars.discourse-cdn.com/v4/letter/e/c89c15/32.png) [@Elena\_Razlogova](https://discourse.citationstyles.org/u/Elena_Razlogova)\
**Post date:** [December 15, 2007, 11:54pm UTC](https://discourse.citationstyles.org/t/document-item-type/394/3 "2007-12-15T23:54:56Z")

</div>

> > To use “document” item type to cite “press release”, or any other  
> > document for which no clear item type exists, Zotero would need to  
> > add three fields to “document”:
> > 
> > 1. “place” - to denote place of creation/publication
> > 
> > 2. “type” - to enter the type of document
> > 
> > 3. “publication status” - to denote whether the document is published  
> > or not.
> > 
> > “Publication status” string could then determine how the title is  
> > formatted.
> > 
> > This may be a possible solution for CSL (although I don’t think so)  
> > but it seems to me way too general for Zotero item types.
> 
> If a) it works, and b) it accurately describes the resource, what’s  
> the  
> problem?

It may work fine for CSL, but Zotero needs a separate item type for  
unpublished documents for reasons I’ve already outlined.

Try this then in CSL if you think it might work. Don’t advise on  
forums that it works already. It’s misleading.

> > Currently in Zotero, odd published documents can be formatted as  
> > “report” item type (fallback to CSL “book”) and odd unpublished  
> > documents can be formatted as “manuscript” item type (fallback to CSL  
> > “article”). It works great in CSL and needs only minor tweaking with  
> > Zotero fields and CSL schema.
> 
> But IT"S WRONG! A working paper isn’t a report. A press release issued  
> by the White House or the ACLU and available on the internet, is NOT a  
> manuscript.

Whatever, rename them as “published” and “unpublished” source. The  
formatting will still work.

> > OTOH, perhaps breaking apart Document into Published Document and  
> > Unpublished Document might be a possibility, and then, for example,  
> > treating a Manuscript as a subclass of the latter. But that might  
> > introduce unforeseen problems (like for resources that can be either  
> > published or unpublished).

Yes, that would make sense (if you had an “unpublished document” type  
you wouldn’t need “manuscript” at all). I don’t see any problems with  
this solution but would be happy to look at an example of a potential  
problem.

> > > I’m suggesting that every time someone has some problem that the  
> > > correct  
> > > solution is not to add a new type. If something like Document doesn’t  
> > > fully work, then that’s a bug that needs fixing. Document in Zotero  
> > > should allow me to fully store the funky stuff I run across; I  
> > > shouldn’t  
> > > need to ask the Zotero devs to add a new type.

I’m not proposing to add any new types at all. Both Zotero and CSL  
have all the types I need–the name of the item type doesn’t matter  
to me, only the function.

Zotero developers did add a new item type–“document”–at your  
request. “Document” wasn’t there originally; users can’t use it to  
cite anything; and you’re presenting it on forums as a working type  
when it is not. Why is it in Zotero interface? I’m still mystified  
about this, and somewhat concerned that this type will travel to the  
new Zotero item type structure without editing.

A transparent taxonomy and a set of general principals for CSL sound  
great–I look forward to that. In the meantime, though, I’m going to  
stick with imperfect solutions that work.

Best,  
Elena