Ticket #800 (closed issue: fixed)

Opened 12 years ago

Last modified 12 years ago

[M] Moles vocabulary definition redudancy

Reported by: lawrence Owned by: ko23
Priority: discussion Milestone: Reporting
Component: MOLES Version:
Keywords: Cc:

Description

Every moles parameter in dif parlance has a huge heirarchy stored in it. Is that really necessary if the final descriptive element is known to the vocabulary server, and so all the sub elements relationships are there. ie. instead of

<dgStdParameterMeasured>
                    <dgValidTerm>EARTH SCIENCE</dgValidTerm>
                    <dgValidTermID>
                        <ParentListID>http://vocab.ndg.nerc.ac.uk/term/P111</ParentListID>
                        <TermID>GCAT0001</TermID>
                    </dgValidTermID>
                    <dgValidSubterm>
                        <dgValidTerm>ATMOSPHERE</dgValidTerm>
                        <dgValidTermID>
                            <ParentListID>http://vocab.ndg.nerc.ac.uk/term/121</ParentListID>
                            <TermID>null</TermID>
                        </dgValidTermID>
                        <dgValidSubterm>
                            <dgValidTerm>AtmosphericWinds</dgValidTerm>
                            <dgValidTermID>
                                <ParentListID>http://vocab.ndg.nerc.ac.uk/term/P131</ParentListID>
                                <TermID>null</TermID>
                            </dgValidTermID>
                            <dgValidSubterm>
                                <dgValidTerm>UpperLevelWinds</dgValidTerm>
                                <dgValidTermID>
                                    <ParentListID>http://vocab.ndg.nerc.ac.uk/term/P141</ParentListID>
                                    <TermID>null</TermID>
                                </dgValidTermID>
                                <dgValidSubterm>
                                    <dgValidTerm>uzonalwindspeed</dgValidTerm>
                                    <dgValidTermID>
                                        <ParentListID>null</ParentListID>
                                        <TermID>null</TermID>
                                    </dgValidTermID>
                                    <ListLevel>4</ListLevel>
                                </dgValidSubterm>
                                <ListLevel>3</ListLevel>
                            </dgValidSubterm>
                            <ListLevel>2</ListLevel>
                        </dgValidSubterm>
                        <ListLevel>1</ListLevel>
                    </dgValidSubterm>
                    <ListLevel>0</ListLevel>
                </dgStdParameterMeasured>

Isn't all the information retrievable from the UpperLevelWinds? (plus lower level untagged name and obviously the <ParentListID> http://vocab.ndg.nerc.ac.uk/term/P141</ParentListID> element)

Can a future version of MOLES be significantly simplified?

Change History

comment:1 Changed 12 years ago by rkl

  • Status changed from new to closed
  • Resolution set to fixed

Yes. It comes down to whether we store the full parameter hierarchy or even ontology in MOLES or simply an entry point into an external resource such as the Vocabulary Server. In the example given we could either store just the base of the GCMD hierarchy (parameter - the entry from P141 in the above example)or the concatenated GCMD hierarchy from the P041 list.

The difference between these options is that in the first case multiple hierarchies are possible such as:

"EARTH SCIENCE > Biosphere > Microbiota > Chlorophyll" "EARTH SCIENCE > Biosphere > Vegetation > Chlorophyll" "EARTH SCIENCE > Hydrosphere > Water Quality/Water? Chemistry > Chlorophyll" "EARTH SCIENCE > Oceans > Ocean Chemistry > Chlorophyll"

whereas the second option nails one e.g.

"EARTH SCIENCE > Oceans > Ocean Chemistry > Chlorophyll"

How we populate this bit of MOLES is down to choice driven by a guidance document.

Note: See TracTickets for help on using tickets.