Difference between revisions of "Documentation Guidance Methods"

From Earth Science Information Partners (ESIP)
 
(22 intermediate revisions by 2 users not shown)
Line 1: Line 1:
The goal of this work is to provide useful guidance for metadata for resource discoverability, accessibility, usability, and understanding to communities that have a wide variety of metadata requirements, a variety of existing metadata dialects and a variety of management systems. In order to achieve this, the guidance must be created at a conceptual level and connected to a variety of dialects. In addition, users must be able to consider a variety of existing recommendations and how those recommendations can be implemented in their dialects of choice.
+
Deciding which metadata dialect is most appropriate for the metadata in question can be approached in two ways:
  
In order to achieve this, we have implemented an ensemble approach to generating metadata guidance that is illustrated schematically for three recommendations (ACDD, CSW, and DIF) in Figure 1. In the first selection scenario, a user needs guidance that combines the Attribute Convention for Data Discovery (ACDD) and Open Geospatial Consortium Catalog Services for the Web (CSW). There are five concepts that are included in those two recommendations (A-E in this schematic). Of those, four are in one of the recommendations (A and B in ACDD, D and E in CSW) and one (C) is in two recommendations. Thus, for selection scenario #1, A, B, D, and E have scores of 1 and C has a score of 2.
 
  
[[File:EnsembleMethod.png]]
+
== Option A - No Agreed Upon Best Practice==
 +
If there is no agreed upon “best practice” metadata dialect within the community that will be using the data, best approach is to evaluate which dialect contains the maximum required, recommended, and optional concepts as your metadata. This method minimizes the amount of your metadata that has to be changed for compatibility and therefore conserves time and effort.  (See the Figure below for a visual example)
  
In the second selection scenario, a user needs guidance that combines the NASA Directory Interchange Format (DIF) with the previous two recommendations. This brings in one additional concept (F), and changes the scores for each concept as shown in this Table:
+
[[File:No best practice.png]]
    <table border="1" cellpadding="2" width="100%">
+
 
      <tr>
+
''Note:  CSW has 5 out of 6 concepts matching your metadata, whereas ACCD and DIF have only 2 matching concepts. Thus, the preferred choice in this case would be CSW, as only 1 concept requires modification in order to achieve compatibility.''
        <td></td>
+
 
        <td colspan="2" align="center">Scores</td>
+
== Option B - Agreed Upon Best Pratice==
        <td colspan="2" align="center">ACDD</td>
+
If there is an agreed upon “best practice” metadata dialect within the community that will be using the data, that dialect should be chosen in order to remain consistent with accepted standards.  In this case, the best approach is to evaluate which concepts your metadata already contains that are required, recommended, or optional for the mandatory metadata dialect and modify all others to achieve compatibility (See the Figure below for a visual example)
        <td colspan="2" align="center">CSW</td>
+
 
        <td colspan="2" align="center">DIF</td>
+
[[File:best practice.png]]
      </tr>
+
 
      <tr>
+
''Note:  DIF is the standard “best practice” metadata dialect within the community that will be using the data.  2 out of the 6 DIF concepts match your metadata, but the 1st, 3rd, 4th, and 5th concepts will have to be modified to achieve compatibility.''
        <td>Scenario</td>
+
 
        <td align="center">1</td>
+
 
        <td align="center">2</td>
+
[[Let's Start at the Beginning]]
        <td align="center">1</td>
 
        <td align="center">2</td>
 
        <td align="center">1</td>
 
        <td align="center">2</td>
 
        <td align="center">1</td>
 
        <td align="center">2</td>
 
      </tr>
 
      <tr>
 
        <td align="center">A</td>
 
        <td align="center">1</td>
 
        <td align="center">1</td>
 
        <td align="center">x</td>
 
        <td align="center">x</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
      </tr>
 
      <tr>
 
        <td align="center">B</td>
 
        <td align="center">1</td>
 
        <td align="center">1</td>
 
        <td align="center">x</td>
 
        <td align="center">x</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
      </tr>
 
      <tr>
 
        <td align="center">C</td>
 
        <td align="center">2</td>
 
        <td align="center">3</td>
 
        <td align="center">x</td>
 
        <td align="center">x</td>
 
        <td align="center">x</td>
 
        <td align="center">x</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">x</td>
 
      </tr>
 
      <tr>
 
        <td align="center">D</td>
 
        <td align="center">1</td>
 
        <td align="center">2</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">x</td>
 
        <td align="center">x</td>
 
        <td align="center"></td>
 
        <td align="center">x</td>
 
      </tr>
 
      <tr>
 
        <td align="center">E</td>
 
        <td align="center">1</td>
 
        <td align="center">2</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">x</td>
 
        <td align="center">x</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">x</td>
 
      </tr>
 
      <tr>
 
        <td align="center">F</td>
 
        <td align="center">0</td>
 
        <td align="center">1</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">&nbsp;</td>
 
        <td align="center">x</td>
 
      </tr>
 
    </table>
 

Latest revision as of 11:40, July 29, 2015

Deciding which metadata dialect is most appropriate for the metadata in question can be approached in two ways:


Option A - No Agreed Upon Best Practice

If there is no agreed upon “best practice” metadata dialect within the community that will be using the data, best approach is to evaluate which dialect contains the maximum required, recommended, and optional concepts as your metadata. This method minimizes the amount of your metadata that has to be changed for compatibility and therefore conserves time and effort. (See the Figure below for a visual example)

No best practice.png

Note: CSW has 5 out of 6 concepts matching your metadata, whereas ACCD and DIF have only 2 matching concepts. Thus, the preferred choice in this case would be CSW, as only 1 concept requires modification in order to achieve compatibility.

Option B - Agreed Upon Best Pratice

If there is an agreed upon “best practice” metadata dialect within the community that will be using the data, that dialect should be chosen in order to remain consistent with accepted standards. In this case, the best approach is to evaluate which concepts your metadata already contains that are required, recommended, or optional for the mandatory metadata dialect and modify all others to achieve compatibility (See the Figure below for a visual example)

Best practice.png

Note: DIF is the standard “best practice” metadata dialect within the community that will be using the data. 2 out of the 6 DIF concepts match your metadata, but the 1st, 3rd, 4th, and 5th concepts will have to be modified to achieve compatibility.


Let's Start at the Beginning