# Empty cites

**URL:** <https://discourse.citationstyles.org/t/empty-cites/808>\
**Category:** CSL Development\
**Created:** [January 5, 2010, 12:02am UTC](https://discourse.citationstyles.org/t/empty-cites/808 "2010-01-05T00:02:33Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Frank\_Bennett](https://yyz1.discourse-cdn.com/flex029/user_avatar/discourse.citationstyles.org/frank_bennett/32/198_2.png) [@Frank\_Bennett](https://discourse.citationstyles.org/u/Frank_Bennett)\
**Post date:** [January 5, 2010, 12:02am UTC](https://discourse.citationstyles.org/t/empty-cites/808/1 "2010-01-05T00:02:33Z")

</div>

An integrator has come across webpage references in his test data that  
produce no output in the Chicago author-date style, because they are  
unauthored, the “anonymous” string is rendered as a locale term  
grouped with the date, and the date is absent in the sample data.

Zotero currently issues a warning popup for empty citations, but if  
the citation has multiple references, and some of them will render,  
the citation goes in with no warning. No output is produced for the  
problem cites, and there is no visual hint that an intended reference  
has gone missing. If the citation contains other references that  
produce output, no warning popup is presented: the problem cites are  
just dropped from the output.

There is a small Zotero issue here – the popup should be triggered by  
an empty cite, rather than an empty citation – but there is a  
CSL/processor implementation issue as well. For server-side and  
batch-processing systems currently in development, there is not an  
opportunity for user interaction, and so no hint of trouble in the  
finished document.

Would it be acceptable to have the processor replace empty cites with  
a warning string? It might be unsightly, but it might be better than  
seamlessly dropping content from the finished output. Any thoughts on  
what ought to be done for this? If it is not felt to be a CSL issue,  
I can liaise with the citeproc-js group and implement whatever  
solution people there are comfortable with.

Frank

---

<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:** [January 5, 2010, 12:19am UTC](https://discourse.citationstyles.org/t/empty-cites/808/2 "2010-01-05T00:19:04Z")

</div>

Isn’t this a bug with the style that can and should be corrected?

Bruce

---

<div class="post-metadata">

**Author:** ![Frank\_Bennett](https://yyz1.discourse-cdn.com/flex029/user_avatar/discourse.citationstyles.org/frank_bennett/32/198_2.png) [@Frank\_Bennett](https://discourse.citationstyles.org/u/Frank_Bennett)\
**Post date:** [January 5, 2010, 1:01am UTC](https://discourse.citationstyles.org/t/empty-cites/808/3 "2010-01-05T01:01:13Z")

</div>

> > An integrator has come across webpage references in his test data that  
> > produce no output in the Chicago author-date style, because they are  
> > unauthored, the “anonymous” string is rendered as a locale term  
> > grouped with the date, and the date is absent in the sample data.
> 
> Isn’t this a bug with the style that can and should be corrected?

I’m not sure CMS permits a fallback to something more sensible than a  
bare (Anon.) key. Maybe the title … this isn’t really my  
department. Is Elena around?

Frank