I am against trying to use values that mix concepts - in this case conflating the definition of "sex" with definitions of multiple objects bearing different sexes.  Such an approach is only going to make it more difficult for end users to understand what they are getting.

Best, Rob

On Thu, Feb 5, 2015 at 12:18 PM, Chuck Miller <Chuck.Miller@mobot.org> wrote:

Another term instead of “mixed”, could be “multi”.

 

Chuck

 

From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Bryan
Sent: Thursday, February 05, 2015 1:13 PM
To: John Wieczorek
Cc: TDWG Content Mailing List
Subject: Re: [tdwg-content] Darwin Core terms sex and lifeStage with mixed content

 

I have one suggestion for conversation because the idea is incomplete. "Sex: null" might not reflect the current knowledge about multiple sex identified specimen individuals. Null seems to connote that the sex is unknown or just not filled in. An option might be to add an item to the controlled vocabulary such as "mixed" as in a mixture of sexes. Such a designation could be a signal to look in dynamicProperties for additional information on counts. I am not happy with the word "mixed" since it might be read in rare cases to be a gender chimera but not hermaphrodite. 

 

--bryan 

 

On Thu, Feb 5, 2015 at 8:09 AM, John Wieczorek <tuco@berkeley.edu> wrote:

Dear all,

 

There is a body of topics surrounding Darwin Core terms for which it is tempting to overload the content in an effort to provide richer data than the standard would seem to support. We have a pair of older open issues on this general topic in the Darwin Core issue tracker.

 

 

You can see in the discussions on those issues that there is some consensus that the content of Darwin Core terms should remain consistent with the definitions and not mix concepts.

 

I've had recent questions in particular about the fields 'sex' and 'individualCount'. Here is an example. How should one populate Darwin Core terms for a record of 2 males and 5 females in a single lot?

 

I propose that part of this is easy - the individualCount should be 7.

 

The definition of the sex term (without looking for clues in the Comments) might be seen as a little ambiguous, "The sex of the biological individual(s) represented in the Occurrence. Recommended best practice is to use a controlled vocabulary."

Strictly speaking, there is no single sex that matches the biological individuals in this example lot Occurrence record, and so the sex fields should be left blank.

 

There are a number of terms in Darwin Core whose definitions suggest that a list (formatted with ' | ' as a separator) be used to match multiple values. Though it might seem that sex could be one of these terms, it isn't currently. Again this argues for a single value for sex from a controlled vocabulary.

 

The biggest problem is with the Comment on the sex term, which gives an example with different semantics from what the definition says - namely "8 males, 4 females". I pose that this is an error and must be corrected.

 

Yet, all hope of retaining information in the Occurrence record is not lost. To capture the richness of the information in the multi-sex lot example, I would recommend the use of dynamicProperties. 

 

How should one populate Darwin Core terms for a record of 2 males and 5 females in a single lot? I would do this:

 

sex = null

individualCount = 7

dynamicProperties= { "count of males":2, "count of females":5 }

 

Hope that makes sense.

 

Since this topic has arisen from multiple independent sources, I would like to formally propose changes to the sex and lifeStage terms to remove the spurious examples. Today begins a minimum 30-day public commentary period to close on 5 March 2015 if no dissenting opinions are made in this public forum. 

 

Here is how the new terms would appear if the changes are accepted:

 

Label: Sex

Definition: The sex of the biological individual(s) represented in the Occurrence. Recommended best practice is to use a controlled vocabulary.

Comment: Examples: "female", "hermaphrodite", "male". For discussion see http://terms.tdwg.org/wiki/dwc:sex

Refines:

Status: recommended

Date Issued: 2008-11-19

Date Modified: 2015-02-05

Has Domain:

Has Range:

Version: sex-2015-02-05

Replaces: Sex-2009-04-24

Is Replaced By:

ABCD 2.06: DataSets/DataSet/Units/Unit/Sex

 

Label: Life Stage

Definition: The age class or life stage of the biological individual(s) at the time the Occurrence was recorded. Recommended best practice is to use a controlled vocabulary.

Comment: Examples: "egg", "eft", "juvenile", "adult". For discussion see http://terms.tdwg.org/wiki/dwc:lifeStage

Refines:

Status: recommended

Date Issued: 2008-11-19

Date Modified: 2015-02-05

Has Domain:

Has Range:

Version: lifeStage-2015-02-05

Replaces: LifeStage-2009-04-24

Is Replaced By:

ABCD 2.06: DataSets/DataSet/Units/Unit/MycologicalUnit/MycologicalSexualStage or DataSets/DataSet/Units/Unit/MycologicalUnit/MycologicalLiveStages/MycologicalLiveStage (Note DwC spec uses ”MycologicalLifeStage” or DataSets/DataSet/Units/Unit/ZoologicalUnit/PhasesOrStages/PhaseOrStage

 

 

Cheers,

 

John


_______________________________________________
tdwg-content mailing list
tdwg-content@lists.tdwg.org
http://lists.tdwg.org/mailman/listinfo/tdwg-content



 

--

Bryan Heidorn
University of Arizona
http://www.sirls.arizona.edu/heidorn


_______________________________________________
tdwg-content mailing list
tdwg-content@lists.tdwg.org
http://lists.tdwg.org/mailman/listinfo/tdwg-content