4. You currently check on medium if genre isn't there, right? That seems like the right idea to me

The place where this matters if for things like this:

Album recorded by P. McCartney
Song recorded by P. McCartney
Letter to P. Richardson
Review of the book "Goodnight moon"

So, genre and/or medium would contain "Album", "Song", "Letter", or "Review of the book".

Currently, I am assuming that such information is stored in genre and that medium is only used for describing the medium (e.g., "Photograph", "Tape recording", etc.). I am not currently substituting medium if genre is absent.

I had this thought for three reasons:

1. It reduces the number and complexity of <choose> elements. For example,

                  <choose>
                    <if variable="number" match="none">
                      <choose>
                        <if variable="genre">
                          <text variable="genre" text-case="capitalize-first"/>
                        </if>
                        <else>
                          <text term="letter" form="short" text-case="capitalize-first"/>
                        </else>
                      </choose>
                    </if>
                  </choose>
                  <text variable="medium" text-case="capitalize-first"/>

instead of

                  <choose>
                    <if variable="number" match="none">
                      <choose>
                        <if variable="genre">
                          <text variable="genre" text-case="capitalize-first"/>
                        </if>
                        <else-if variable="medium">
                          <text variable="medium" text-case="capitalize-first"/>
                        </else-if>
                        <else>
                          <text term="letter" form="short" text-case="capitalize-first"/>
                        </else>
                      </choose>
                    </if>
                    <else>
                      <choose>
                        <if variable="medium">
                          <text variable="medium" text-case="capitalize-first"/>
                        </if>
                        <else>
                          <text term="letter" form="short" text-case="capitalize-first"/>
                        </else>
                      </choose>
                    </else>
                  </choose>
                  <choose>
                    <if variable="genre" match="none">
                      <text variable="medium" text-case="capitalize-first"/>
                    </if>
                  </choose>

2. For some cases, it permits better style logic. For example, if an interview item doesn't have a specified title or genre, but does specify a medium of "Tape recording" (see example 7 here), the style can produce:

[Interview by C. A. Kiesler; Tape recording]

rather than:

[Tape recording by C. A. Kiesler]

Obviously, the user could enter "Interview" for genre, but permitting the style to generate this if needed seems nice.

3. Generally, I would like to see more consistent use of genre and medium. Assuming that medium really refers to recording or artistic medium rather than item subtypes seems to encourage good practice?

OK on medium vs. genre -- we can just test this out for a little and if we get a lot of protest, rethink.

See one remaining comment where I think we talked past each other because of where it sits -- otherwise this is as good as I think this is reviewable for me and we should go ahead.

I would like to start a conversation on agreeing to a test-suite for highly complex styles, otherwise these become increasingly impossible to edit.

😃 Your submission passed all our automated tests.

😃 Your submission passed all our automated tests.

I'm reconsidering medium vs genre. It seems likely that users would store "Album", "EP", etc. in Zotero's Format field for Audio Recording.

Regarding test suites, yes, it's been my intention to build one for this style. I was planning on using the 230 test items from the APA manual as a start. Did you have other thoughts on that?
(Sorry for not responding to this one yesterday; ran out of time.)

Regarding test suites, yes, it's been my intention to build one for this style. I was planning on using the 230 test items from the APA manual as a start. Did you have other thoughts on that?
(Sorry for not responding to this one yesterday; ran out of time.)

Yes, that's exactly what I had in mind. Do you have those in the test library already? Otherwise, I wonder if we could somehow pay someone to do this out of CSL funds. Same with Chicago Manual.

I have the manual items in the test library.

This is gtg from my end -- I'll merge once you tell me the other styles are ready

Want to check APA's new official annotated bibliography guidance.

😟 There are some issues with your submission.

Please check the test report for details.

😟 There are some issues with your submission.

expected: []
     got: ["citation-locator"]

(compared using ==)

Please check the test report for details.

😃 Your submission passed all our automated tests.

Okay, these should be good to go. The only thing I can't accommodate per the manual's annotated bibliography style is that, if the annotations span multiple paragraphs, paragraphs 2+ should be indented. I don't think that's possible in the current CSL spec and is too rare to worry about trying to automate.

(@adam3smith, I'll merge this with some small metadata changes today; I want to e.g. use non-abbreviated "edition" in the style titles and likely restore some original contributors to the style)

@bwiernik, I understand you did a comprehensive rewrite, but unless you started from scratch I prefer to leave the original contributors.

(never mind, this all looks good)

Thanks! 👏 👏 👏 🎆

@rmzelle You were the one who deleted the contributors. I am fine with whatever arrangement—the contributors all obviously contributed to the current product.

You were the one who deleted the contributors. I am fine with whatever arrangement—the contributors all obviously contributed to the current product.

I forgot we listed the 6th edition as template. We typically delete all authors/contributors that came from the template, so it's good as is.

Congratulations!! Happy to see this merged :)