Taxonomic hierarchy in SDD

Kevin Thiele kevin.thiele at BIGPOND.COM
Wed Nov 28 12:38:51 CET 2001

Can I propose that we note but set aside the Taxonomic Hierarchy thread for
the moment. In the challenge cases, Challenge 7 will begin to address this
issue (in this challenge we will attempt to represent descriptions from a
nested set of descriptions e.g. family (genus (species (specimens)))). When
this comes up, we will need to address both hierarchy as a way of handling
inheritance, and also alternate hierarchies (differing taxonomic opinions).
But surely we need to work out how to represent a single description before
we tackle a nested set.

Concerning Steve's exemplar drawn from a DELTA butterfly treatment: I've
added a few challenges to the challenge cases (number 9 in the attached)
specifically designed to address representation of DELTA datasets in the new
standard. Personally, I don't think we should start with a DELTA data set as
our first challenge, for several reasons:

1. We have agreed several times that it's important not to be canalized by
DELTA or any other existing representation, but to keep the existing
representations in mind while working. Starting with XDELTA seems to me to
increase the risk of canalization
2. Of all the descriptions in the world, 99.9999999% of them are not in
DELTA. Probably 99.999% of them are textual (natural language) descriptions.
Surely this should form the basis of our first challenge, methinks.
3. Related to 2 above, many (though by no means all) DELTA datasets are
already abstractions from the source (a set of natural language
descriptions). We should start with the source.

This time I've attached the attachment. Sometime soon we'll put this up on
the web instead of flying it around as an attachment.

Cheers - k

Content-Type: text/html;
        name="SDD TDWG 2001.htm"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
        filename="SDD TDWG 2001.htm"


<meta http-equiv=3D"Content-Type" content=3D"text/html; =
<meta name=3D"GENERATOR" content=3D"Microsoft FrontPage 4.0">
<meta name=3D"ProgId" content=3D"FrontPage.Editor.Document">
<title>Basic description of two taxa</title>


<p><b><font size=3D"3">SDD Challenge Cases</font></b></p>
<p><font size=3D"2">The challenge cases below represent specific classes =
problems that the SDD Standard needs to address. The exemplar for each =
provides a chunk or fragment of descriptive data that minimally =
exemplifies the
problem; the challenge is to validly represent the data under the =
standard. The challenges are listed in the approximate order in which =
they will
be worked through.</font></p>
<font SIZE=3D"2">
<p><b><i>Challenge 1: Represent basic descriptions of two taxa from one =
language) source</i></b></p>
<p><font color=3D"#0000FF"><i>Discaria pubescens</i><br>
Rigid, spreading shrub to c. 1m high and wide; stems glabrous. Leaves =
deciduous (particularly on older plants), +/- oblong, (4-)6-10(-15) mm =
long, 2-3 mm wide, apex obtuse or minutely mucronate within an apical
notch, glabrous or a few hairs present near tip; stipules dark =
reddish-brown, c. 1 mm
long, often shallowly joined around the node; spines stout, 1.5-4 cm =
<i>Discaria nitida</i><br>
Slender shrub, to 5 m high; stems glabrous. Leaves persistent (rarely =
deciduous), elliptic to obovate,
(8-)10-20(-30) mm long, 3-7 mm wide, glabrous, shining; spines not =
developed at each node, to c. 1 cm long.</font></p>
  <li>missing data (e.g. stipule characters for <i>D. nitida</i>)</li>
  <li>modifiers (e.g. leaves of <i>D. nitida</i> rarely deciduous)</li>
  <li>freeform comments (e.g. <i>D. pubescens</i> &quot;particularly on =
  <li>structured data vs marked-up text (e.g. let's say that stipule =
    will not be a character - how to retain the data &quot;stipules =
    shallowly joined around the node&quot;)</li>
<p><b><i>2. Basic description of one taxon, two sources, basic =
<p><b><i>3. One treatment, several contributing authors</i></b></p>
  <li>flagging of data</li>
<p><b><i>4. Define basic character list</i></b></p>
  <li>structure of character lists</li>
  <li>character types</li>
<p><b><i>5. Remote vs local character list</i></b></p>
<p><b><i>6. Remote vs local taxon lists</i></b></p>
  <li>master lists</li>
<p><b><i>7. Several descriptions at different levels (e.g. family (genus =
  <li>nesting of taxon(/character?) lists</li>
  <li>data collation</li>
  <li>data inheritance</li>
<p><b><font size=3D"2"><i>8. Universal treatment, several outputs =
language, NEXUS, interactive key)</i></font></b></p>
  <li><font size=3D"2">flagging/layering of data elements</font></li>

<p><b><font size=3D"2"><i>9. Representation of legacy structured =
descriptive data</i></font></b></p>
  <li><font size=3D"2">DELTA data sets</font></li>
  <li><font size=3D"2">Lucid data sets</font></li>
  <li><font size=3D"2">DELTA Access data sets</font></li>
  <li><font size=3D"2">other formats</font></li>



More information about the tdwg-content mailing list