"Practice Guides" Spec

gjr musings

A "practise guide" is very similar to a taxon profile and possibly can use the same infrastructure/storage (in the Taxon Profile Toolkit). It aims to pull together a subset of terms which are suitable for use in a given context (i.e. type of work and taxonomic discpline).

"Practise guides" arose out of PA's lab chat. It became very apparent that we needed to present select sets of terms from Wallace Core for particular applications (e.g. the eFlora of Australia) and the full presentation of WC was just confusing. This useful abstraction allows WC to be record sets of terms as they are applied in different contexts.

Draft

Metadata about the use guide
*Metadata* *Description*
Taxonomic Domain e.g. plants? entomology? fauna? etc
Kind of Profile formal, a flora, or .. (see list of profile types) below
Basis of guide who; what organisation; or what collection of works if the guide is based on a survey (e.g. bird works)
Status of guide official - accepted by who? draft? from WC compilations etc

Meta Terms
metaTerms for the usage Guide Description  
Label The label that will be used in the guide.
It should be grounded in the domain and use for which the guide was made.
 
Basis "exact match" WC term  
Usage Any particular usage information that is particular to this guide.. but must not affect the WC term basis  
Comments just because it seems like there should be.  
Order Presentation Order in guide ??????  

Meeting 8/12/2010

Romping forward

  1. Information content category
  2. Metadata
Present JRC, PA, GJR

Discussed

directedgraphplugin-1.png

Discussed>labels and overall structure of content hierachy

Clarified> metadata v. content (of a profile)

Things such as Nomenclature, systematics are where the author's original text is recorded.

Subsequent linkage to nomenclature (e.g. APNI/AFD) is METADATA

Decided>

To add terms to Metadata for Nomenclatoral services. => PA

made note a few typo's and misplacements that need amendment
  • delete taxon - term => PA
  • check that all name terms are NOW in the appropiate place in the content hierachy => PA
  • check for other things which have fallen through the cracks=> PA
Changes to contents of WC.Systematics
  • create new Term Labelled tentatively "Taxon Origins" => PA
  • it should contain phylogeny (and current members of phylogeny) should be sisters=> PA
Discussed> Pratice Guides for profiles

Decided>

Confirmed this is a good idea.

We need to source and describe two more floras

  • Flora of N. America => PA to report back
  • Flora of NZ - Aaron Whitton=> PA to report back
Need to spec out "practice Guides" => GJR to report back

Discussed>GISIN

lots to ideas to use here.

bigger than anticipated

Decided>

some good grounding. and linkouts to metadata resources we can also use.=> PA & GJR to start and report back

Discussed> Profile Coverage.

PA had just completed a new profile describing Greening Australia's book of ACT birds. TaxonProfile.Definitions/BBBirds

90% of terms had counterparts in WC.

create new terms and mappings for 2 which do not .=> PA

Discussed> CritterQuery? - looks grandnice addendum to the visualisation permitted by semantic markup

https://testwiki.trin.org.au/Mangroves/SemanticLinksReport

Draws just connections to/from Aegiceras_corniculatum:

http://testwiki.trin.org.au/Mangroves/CritterQuery?Critters=Aegiceras_corniculatum

If you want all related nodes (not just connections to corniculatum) on
the "to" (left) side, at the moment you have to type all those names out
explicitly, example:

http://testwiki.trin.org.au/Mangroves/CritterQuery?Critters=Aegiceras_corniculatum&Critters=Bostrychia_calliptera&Critters=Gelidium_sp

Discussed> an extending this foo with wordle where size proportionate to number of times a relationship is used.

Could be useful wherever we code relationships

e.g. Wordle => PH to investigate and report back

PH's Report on tag-cloudification:
  • Counting stats on semantic links is "easy", but somewhat expensive (waiting on mongo smile ; general [ [links]] are not easy (until I do BacklinksPlugin).
  • Still investigating easy-to-user tag cloud visualisation (could hack existing TagMePlugin, but that thing is so '90s)

Meeting 6/12/2010

Agenda

Megalo, Megalo, Grrand Meeting.
  • new people
  • Phylowidget
  • Linked lab book
  • PDA’s
  • And more
  • TBCB
  • GBIF paper Comments?
  • server's after TRIN
  • ASAD - LLB
  • Google Info session
  • webdav madness
  • Image annotation
nna blacka montan mapparoom

Present PA, GJR. Most matters held over for another time.

Discussed

Discussed> Prototype script for showing duplicate terms as discussed pervious meeting.

GJR> looks good. If only it could be a jqgrid script.

Sorted Out>

Content relations for
  • Systematics
  • DistributionHabitat
Next meeting>

Content relations for
  • Information
  • Metadata
  • Figures

Meeting 3/12/2010

Agenda

Romping forward.
  • More definitions.
  • Tighter ever tighter.
  • And other stuff… SNOWMED

Discussed

Discussed> Snowmed

Cote RA (1986) "Architecture of SNOMED Its Contribution to Medical Language" Proc Annu Symp Comput Appl Med Care. 26: 74–80.

Tho' this is a very different application and quite old. It does entail a precident of sorts for the axes. and thus worth a quick read

Content Term Group> Tied ourselves in knotts thinking about The information coded in the nomenclature term group

Actions>

PA to
  • to investigate how we might merge terms
  • start scripting a means to identify terms and examine their characteristics simulataneously so we can find duplicates.

Meeting 30/11/2010

Agenda

nb earlier time (because of CAKE Meeting)

Romping forward.
  • More definitions.
  • Tightening up procedures.
  • GBIF Paper

Discussed

Discussed> GBIF Paper

Decided> noted that TRIN was mentioned and in a critical note about SPM. Imperative need to comment and correct

GJR to seek input from JRC?

Actions> PA, GJR to read and comments; GJR will collate and send on to GBIF

Discussed> Clarified various hierarchies and mappings in play. Came up with simpler language. NB this is not the only way that this particular cake can be cut. Chosen for clarity, simplicity of how we can move forward.

Key idea being that there are rough dependencies in that we need to have progress on 1 before we can start on 2 and so on..
  1. Content-wise = HasPart/!IsPart (current focus ≈ the parts of a taxon profile and what may be contained therein).
  2. Meaning-wise = BT/NT (will be built as we write the definitions - build as we sort out the content relationships)
  3. Concept-wise = Topic Maps etc. (will be built as we add associations - build once we sort out the definitions)
Also the content hierarchy will get quite deep and complex given the number of variants in even our small sample. This is should not be a problem as the other two mappings will allow navigation and different term taxonomies (semantic) to be preserved.

Discussed> The definition of Systematics and the terms which it "contains". nb as previously noted by JRC, there is more in this space than just well founded names. And we need to work out what we can farm to nomenclators and what can not be.

Decided> PA to sort out the many variants of names, common and notes.

Discussed> How to reduce/prune or combine terms

Decided> PA to write a script which will show terms which have a common term basis

PA to broach with PH MECHANISMS for merging terms (is it feasible? etc)

Meeting 26/11/2010

Agenda

Definitions and methods for doing them quickly and efficiently

Discussed

Begin the task re-worked definitions and content hierachy

Meeting 23/11/2010

Agenda


Rolling forward

1. Systematics?

2. Definitions
http://www.thedevilsdictionary.com/
Standard and template to be set by Abrose Bierce
  • ZOOLOGY, n. The science and history of the animal kingdom, including its king, the House Fly (Musca maledicta). The father of Zoology was Aristotle, as is universally conceded, but the name of its mother has not come down to us. Two of the science's most illustrious expounders were Buffon and Oliver Goldsmith, from both of whom we learn (L'Histoire generale des animaux and A History of Animated Nature) that the domestic cow sheds its horn every two years.
  • BOTANY, n. The science of vegetables — those that are not good to eat, as well as those that are. It deals largely with their flowers, which are commonly badly designed, inartistic in color, and ill-smelling.

Discussed

short meeting as map room computer still hasn't been fixed.

Discussed> Meaning and topic maps

Decided>

heuristic for now
  • define WC terms (that is describe and define the content holders for bits of taxon profiles)
  • with WC definitions in hand we will be able to then address semantic mappings - as we will have made the 1st pass in defining the content.
  • and then on to topic maps
Discussed> concepts and topic maps. Concept clouds associated with term, its content, the definition of the term, its basis and the work in which it appears. A complex space indeed.

Decided> consider add a new field to the Term definition to add text for "related concepts" this will seed future content and help identify/search e.g. clarifying that legislation with terms rules. => PA to investigate and implement

Meeting 18/11/2010

Agenda

You've got to latch on to the unconfirmable
eliminate the undefinable
sort out the indeterminate
don't mess with mister in between

leaping forward
  • folksonomy of taxon profile metadata terms
  • progress report on digitizing the folksonomy
  • Lets us begin to systematic definitions of terms

Discussed

Discussed> Details of typification and specimens in relation to terms created at Tuesday's workshop.

Decided> This morning decided that WC:TaxonConcept should be created above/parentOf Nomenclature as follows Folksonomised> Finished the metadata group YA. All done bar the data entry!

Discussed> The tricky delicate matter of the difference between Phylogeny and Classification which led to a lot of messy whiteboard drawings.

GJR here are some clarifying notes>

This figure illustrates the relationship of phylogeny and classification. LHS being a phylogeny and bracketed sections showing groupings of plants by Rank (i.e. the subclass (?) Malvids) viz http://en.wikipedia.org/wiki/Rosids

A good Discussion at GJR subsequent Thinking>

Given that systematics consists of four sub-disciplines
  1. Taxonomy - nomenclature
  2. Classification
  3. Phylogony
  4. Identification
does it make sense to do the following??? Discussed> Procedure and best heuristics to quickly (semi-accurately) define WC terms and co.

Proposed proceedure>

Top down Approach provides a small list of terms to refine.

Define each of these terms and then iterate onto siblings. This may help with speed as pertinent and useful words can be ported from the parten terms.

General Principles
  • Term siblings should be as distinct as possible. They may overlap but not entirely.
  • Definition of a term should be able to describe all its children.
    • if not the definition needs to be changed and the term label may need to be adjusted
    • or children need to be moved
  • The definition of a term must be independant of its parent (no references in the words so that it can stand alone)
  • The definition of a term must be consistent with its parent (an opportunity to port pertinent words from the parent)
Words in the definition of term> Definitions of WC terms describe the nature of the content that goes into a particular portion of a profile - it is important not to be confused with the definition of the words used to label the term.

e.g. WC.Rank is all about the content an author/contributor wishes to convey about the rank of a taxon. It will usually specify the rank of a taxon.. but may also include other information deemed relevant by the author. So we cannot just cut and paste the definition of rank from wikipedia. It is content about rank. This is because WC terms are on the 'has-parts' axis.

We should follow the ISO standard for definitions described below.

Meeting 16/11/2010

Agenda

Rolling forward
  • finish Folksonomy of Nomenclature Term Group

  • Discuss FLoOz - how to represent recommended Usage… How to curate recommended Usage?

Discussed

Folksonomised> Finished the nomenclature group YAY

GJR's Observations> There is a great deal of variability in the type of information and presentation of names, nomenclature and taxonomy. Some of which may not easily meld with nomenclature services such as GW's. Its a real dogs breakfast. That said, many of these terms impact on ther interpreation or coverage of the profile and need to be accounted for somehow. - perhaps Greg will have ideas? It won't be easy to do a one to one link with GNA, ALA or anything else for that matter. Definitly fuzzy at best.

Also there is information in the nomenclature term group that I do not believe is held in the name indexes.

Meeting 11/11/2010

Agenda

Marching forward
  • Discuss FLoOz - how to represent recommended Usage… How to curate recommended Usage?
  • Folksonomy of Nomenclature Term Group

Discussed

Discussed> how we were to represent WC terms based on terms from several profiles.As it provides evidence for the support of each WC term.

Decided> Enter in each term as a basis. Discuss with PaulH best means to parse these out separately

Folksonomised> 50% of the many terms in the Nomenclature Term Group (up to WC.Related species)

NOTE> more than just simple names here. We encountered terms which we are not sure have equivalents in APNI / AFD. Action consult with Greg

Meeting 9/11/2010

Agenda

Pushing forward
  • Representations (Images / figures) and messy meta data boundaries
  • Discuss FLoOz - how to represent recommended Usage… How to curate recommended Usage?
  • folksonomy on nomenclature term group

Discussed

Discused: Margaret’s GAS. Ipad, android phone etc

Discussed: Image “metadata” as outlined in previous meeting.

Question is what to do with terms in profiles which seem to be metadata.


Decided:

minimum (PH) we need is the image and any text associated with the data. Last meetings notes look like we are doing a taxon profile project just for images.

Discussed (JC) difference between Image.Caption and ImageCaption. Former treats Caption as an ‘attribute’

PlinianCore:CaptionImage1 maps to WC:ImageCaption
PlinianCore:CaptionImage2 maps to WC:ImageCaption

Decided we should avoid attributes for the moment.
Tho’ they may become important when composing new profile instances.

Decided: we need a Wallace Core term ImageCaption

Discussed: whether we should look inside the structure of a plate of figure.

Decided: When there is structure in a "plate" in a profile when there is use to encode we should encode it as part of a WC term e.g. FungiDU:Figure1a = WC:DistributionMap


Decided: (JrC) – if there is uniform structure within the figures/plates of a work with clear labels we can elect to treat the internal structure of the figure as we would in the profile text. i.e. we use the "labeled" parts of any figure/image as WC terms.

Decided: to avoid proliferation of WC terms we will use “has parts” for figure parts such as legend i.e. We have profile 21 distribution map.legend, code it as part of WC:DistributionMap; much as EMC suggested in the first place - see photocopy of whiteboard.

Discussed: the different kinds of images like photos, illustrations. JRC doesn’t see that there is a difference

“We got maps and we got illustrations. We assign maps to WC:DistributionMap. The others we attempt to put into other Wallace Core terms.”

Decided: dispite diffences of opinion not a central issue. let the data be our guide… viz record what we see in profiles. Have argument again if it becomes important.

Discussed: the use of triples in the mangrove profile text. Refer Previous meeting notes (28/10) for details and PaulH's plugin documentation

!JrC: what Emma and John are doing for the mangroves is a really great use case for the Wallace Core. If things get messy then this is going to bring us real life complications. We should work closely with them. They are "loose", going into the literature, using a constrained set of relationship words.

PH: has set Emma up with reciprocal terms for her relationships.
GJR: well done paul

Meeting 4/11/2010

Moving forwards

Folksonomies left

Term groups without Folksonomies
  • Notes ~<<30 (Up to Plume Moths G)
  • Illustrations 37
  • Nomenclature 191
  • Profile Metadata 33
Review
  • Definitions
  • Mappings

Discussed.

Decided> GJR & PA to review Helen's link

http://hiscom.chah.org.au/wiki/HISCOM/CHAH_Workshop:_Towards_a_common_approach_to_Australasian_Electronic_Floras#Australasian_eFloras_versus_Flora_of_Australia_schema

Folksonomies done>Notes and Illustrations

Decided> Illustrations problematic as many profiles contain what might be best treated as part of 'CONTENTS' of a term ( in our lingo the 'REPRESENTATION') including metadata to do with the contents

For Figures (which can include illustrations and images in any order), we have encounted the follwoing terms in our sample profiles....
  • labels / captions
  • scale
  • legend
  • figure
    • illustration / Diagram
    • image
  • Location
    • URN
if we can treat

* Figure.sources is separate - in that it is text about figures not contained in a profile

And we have at least several places in WC where images might be encountered being
  • GeneralFigure?
  • Maps - various
  • Morphological form
  • Colour Notes
  • Habitat
As PA, suggests - we should treat Figures as one of many potential Content Representations which can go into any WC term.

More thought is required about what we do with those components of a figure which receive individual and particular attention in a profile.

Perhaps an encoding scheme as follows. Profile21.DistributionMap.Legend = meaning Profile21 has a legend separately mentioned for a distribution map.

We could do the same for other kinds of multimedia.

Thoughts????

Meeting 3/11/2010

Moving forwards

TDWG & SPMI?

Term groups without Folksonomies
  • Illustrations 37
  • Nomenclature 191
  • Notes 51
  • Profile Metadata 33

Discussed.

Discussed>Paul A's lab chat of the morning. Which evoked a lot of discussion and was well recieved. especially with Helen T and IBIS team

Discussed> Helen T's suggestions re eflora >> GJR to contact Helen T. DONE

Discussed> Briefed EMC on current work and project status

Discussed> Collaboration with EOL re SPMI. >> GJR to follow up

Folksonomy> For term group Notes.. up to PlumeMoths?.Description

Meeting 28/10/2010

Moving forwards
  • folksonomies past
    • Review New WC terms created from Biology, Identification
  • folsonomies future
    • Biology (beyond G)
    • The remainder: Small term groups e.g. Phylogeny, Notes
Discussion of coding inter taxon relationships
  • beyond triples.
Mangroves and the mutable profiles
  • any problems? challenges?

Discussed.

Discussed> The trouble with Triples.

Paul H has come up with an elegant way to represent triples based on the semantic media wiki syntax.

A plugin for which is nearly complete.

We discussed how the citation plugin could be modified so that relationships can be linked to citations.

in the manner of cats hate dogs (Garfield 1994). In such a form, this would be very useful to the mangroves project.

Folksonomies Completed

for term groups>
  • Finished Biology (beyond G)
  • Classification
  • Phylogeny
  • Specimens
  • Status

Meeting 25/10/2010

Moving forwards
  • folksonomies past
    • Ecology term group / habitat
    • additions to knowledge and distinguishing features
  • folsonomies future
    • Biology or Identification.
Discussion of coding inter taxon relationships
  • beyond triples.

Discussed.

Discussed> Mapping inter and intra.. GJR's axes and logic (Below) at lenght

Decided> GJR needs to clarify and redraft. Axes analogy doesn't work for set minded folk ---> GJR

Folksonomies completed for Identification, Biology up to G. --> PA to create WC terms and record basis as per past practice

Meeting 19/10/2010

Agenda

  • Review new WC terms for Knowledge and Distinguishing attributes.
  • Next .. lets do one of Biology, Ecology or Identification.
Mindful that some of or all of these may be sunk into Distinguishing Attributes or Knowledge.

GJR Musings.

1. Overfitting.

To counter overfitting to our sample set, we can take additional SMALLER number of sample profiles and see how much modification is required to absorb their terms. Goals being in order of preference i) no change, ii) modification of existing terms and iii) addition of terms. We will know the Wallace Core model is suficient when > 9/10 profiles need case i) and the remainder mostly require case ii) and rarely case iii.

2. Term Basis

As a general rule all terms in WC should have a basis in a profile or external reference such as TDWG vocabs etc.

exception 1.

being those terms which are created on the basis of Wallace Core itself for instance where a term is required because of an existing term in WC to "contain it" or for symmetry ( i.e. a terms concerning Dispersal implies their is a case for corresponding terms concerning Settlement).

exception 2.

being terms created where an overly deep nesting of WC terms is rationalised or simplified e.g. a (silly but illustrative) case should there be a term such as as WC.MetatarsalLength it would be contained in WC.BodyPartLength (with no basis except what it contains). Prudence would suggest that we prune WC.MetatarsalLength as being specialised to a taxon/domain and transfer its term basis to the WC.BodyPartLength etc

Meeting 14/10/10

Agenda

Next meeting to review terms in
  • fresh folksonomies Distinguishing Attributes and Knowledge (up to H)
  • mappings between profiles
  • continue folksonomy Knowledge (beyond H)

Meeting 12/10/10

Agenda

Next meeting to review terms in
  • Knowledge Term Group
  • Distinguishing attributes Group
  • review creation WC terms based on distribution and terms encoding distributions of cyclical movements

Discussed

Discussed> WC Terms based on the Distribution Term group.

Some difficulties encountered with term basis were cleared up. ==> PA to enter missing Term basis

Discussed> difficulties with terms to do with the spatio-temporal distribution of cyclical movements.

PA & GJR came up with a folksonomy of terms for the spatio-temporal distribution of cyclical movements. ==> PA to write up and create WC terms (to be tested via mapping from sample profiles)
  • SpatioTemporalDistribution
    • SpatioTemporalMovements
      • CyclicSpatioTemporalDistribution
        • SpatioTemporalDistOfDispersal
        • SpatioTemporalDistOfSettlement
        • SpatioTemporalDistMigration
Discussed> Mappings used within and between profiles. PA has changed all internal WC mappings to be Contains-wise relationships.

nb this needs to be the case... remember it may seem that some terms are attributes of others but this is not necessarily the case for all terms that belong in a given content category (viz flower colour, and flower stamen description... both contain information about a reproductive part.. but one is an attribute the other is not).

Mappings between profiles need to be reviewed to make sure that we use only Meaning-wise profiles for relationships between profiles. ==> PA to review

Discussed> Term groups Distinguishing Attributes and Knowledge (up to H)

created a two 1st pass 'folksonomies' of these terms for Wallace core. ==> PA will document in Workshop Notes and treat as previously for entry to WC

  • Distinguishing features
    • Variation
    • Distinctive Features (Characters of Note)
    • Diagnosis
      • Taxonomically Distinct
      • Diagnostic...(Morphology, Polymorphism, Variants, Molecular data)
      • Technical Diagnosis
        • Latin Diagnosis
      • Field characters
      • ID? (Sink ID into Diagnosis?)
        • Key
          • Synoptic Key
    • Summary of Distinguishing features (Key features)
TB=TermBasis
  • Knowledge
    • Notes Remarks/Synopsis TB Weeds:Krypto, Eul:Introduction, AntsDC:OpeningComments, Mangrove:Synopsis
    • Characteristics
      • Physiology
      • Smell
      • Signs
        • Tracks
        • Scats
      • Substrate
      • Genetics
        • Chromosome No. TB FoA?:ChromosomeNo
      • Sound
      • Appearance
        • Colour
          • Colour Notes ?
        • Form/Morphology, SPMI, Eul,FungiDU
          • Size
            • Length
              • Dimorpism (?)male v female
              • Polymorphism
            • Weight
          • Whole
            • Morphs(?)/Lifesatge
              • Eggs
              • Seedlings
            • Habit
          • Parts
            • Plumage , Birds:
              • Plumage Colour
            • Reproductive Parts
              • Fruits (colour of fruit)
              • Flowers (Stamen, colour of Flowers)
            • Leaves, RainF?:
      • Behaviour
        • BehaviourAgonistic (includes Antagonistic by definition )
        • BehaviourConspecific
        • BehaviourReproduction
        • BehaviourHeterospecific
      • Genetics
    • General Description
      • Uses
      • History
        • Origin Of Name ? (Nomenclature)
GJR> While these term taxonomies seem useful.. I am concerned they veer too far towards overmodelling / overfitting the data. Of course, best thing to do is test through the application to past and future profiles.

Meeting 8/10/10

Agenda

Definition Bingo

Discussed.

Discussed> Creation of WC terms for Distribution based on lists below.

Actions> recorded basis. Marked terms in list suitable for inclusion to WC. ==> PA to create WC terms

Decided> Next meeting to review terms in

and create WC terms .

Meeting 5/10/10

Relationships

Tentative next steps. Inference and rules.

Relationship Operator Scope in which we apply it role
DCTerms:HasPart Internal Contains
DCTerms:isPartOf Internal Contains
SKOS:HasBroader Internal Meaning
SKOS:HasNarrower Internal Meaning
SKOS:BroadMatch External Meaning
SKOS:CloseMatch External Meaning
SKOS:ExactMatch External Meaning
SKOS:NarrowMatch External Meaning
SKOS:RelatedMatch External Meaning

Some statements to illustrate their validity indicated as true or false for terms which are part of two profiles A & B.
Validity Statement Why?
TRUE ProfileB.GeneralDescription HasBroader ProfileA.Description by definition
TRUE ProfileA.Description HasPart ProfileA.Size by definition and use in context of Profile A
FALSE ProfileA.Size HasBroader ProfileA.Description Cannot be true as Definition does not necessarily contain use in context
FALSE ProfileA.Size HasBroader ProfileA.GeneralDescription Cannot be true as ProfileA.Size HasBroader ProfileA.Description is FALSE
Depending on context, the meaning of elements in a particular profile can change with different profiles. Size may be either a 'Distinguishing Attribute' OR just a part of the 'General Knowledge' in two given profiles. The definition of an element should be as independant of context as possible to ensure its general utility and the broadest possible applicability. So, relationships such as 'HasPart' and 'IsPartOf' will indicate the context in which a term exists within a profile, referred (by us) to as 'CONTAINS' or 'Contents'.

Here is a cartoon to illustrate the axes we have chosen (nb not truly orthogonal but enough so for our purposes). Key idea is to separate out the application of relationships for further inductive and deductive steps and so the use of different types of relationships.

Relationship_axes.gif

With these two sets of relationship types in use, we can start to do some inference. Here are terms from three profiles mapped to Wallace Core all different but essentially to do with where a taxon may be found. Green indicates the data and labels of each profile (MEANING-wise). Blue indicates the 'CONTAINS'-wise structure of Wallace Core.

Profile 1 mapped to Wallace Core

Profile1_Geocode_Relationship.gif

Profile 2 mapped to Wallace Core

Profile2_location_Text_Relationship.gif

Profile 3 mapped to Wallace Core

Profile3_Distribution_.gif

Now some logic.*

*Givens

  • Profile3.Distribution ExactMatch WC.Distribution
  • Profile2.LocationText ExactMatch WC.LocationText
  • Profile1.Geocode ExactMatch WC.Geocode
  • WC.Geocode IsPartOf WC.SpatialDistribution
  • WC.LocationText IsPartOf WC.SpatialDistribution
  • WC.SpatialDistribution IsPartOf WC.Distribution
  • Profile1.SpatialDistribution DOES NOT EXIST
  • Profile2.SpatialDistribution DOES NOT EXIST
  • Profile1.Distribution DOES NOT EXIST
  • Profile2.Distribution DOES NOT EXIST

Definition

If a term exists in a profile then by implication the terms which it is contained within by either a 'HasPart' or 'IsPartOf' relationship triple can be said to be inferred if they are not explicitly in the said profile. These inferred terms do not exist but can be used in subsequent logic.

annotation to indicate inferred terms and relationships?
  • An inferred term inferred( term ) or i(term)
  • An inferred relationship inferred( relationship) or i(relationship)
Consequences

Now we can say
  • Profile2.LocationText ExactMatch WC.LocationText AND WC.LocationText IsPartOf WC.SpatialDistribution
Means
  • Profile2.LocationText IsPartOf inferred(Profile2.SpatialDistribution)
and subsequently
  • WC.SpatialDistribution !IsPartOf WC.Distribution
  • WC.SpatialDistribution !IsPartOf WC.DistributionAndHabitat
Means
  • Profile2.LocationText IsPartOf inferred(Profile2.Distribution)
  • Profile2.LocationText IsPartOf inferred(Profile2.DistributionAndHabitat)
and by a similar chain of inference
  • Profile1.Geocode IsPartOf inferred(Profile2.Distribution)
  • Profile1.Geocode IsPartOf inferred(Profile2.DistributionAndHabitat)
Now, we cannot say that this means Profile3.Distribution, Profile2.LocationText, Profile1.Geocode have any precise relationships in terms of meaning. But we can say something looser that they are RelatedMatches as follows. If we take an inclusive and fuzzy/broad meaning to the relationship 'RelatedMatch' we have.....
  • inferred(Profile1.Distribution) ExactMatch WC.Distribution
  • inferred(Profile2.Distribution) !ExactMatch WC.Distribution
  • inferred(Profile1.Distribution) !RelatedMatch inferred(Profile2.Distribution)
and similarly
  • inferred(Profile2.Distribution) RelatedMatch Profile3.Distribution
  • inferred(Profile1.Distribution) RelatedMatch inferred(Profile2.Distribution)
  • inferred(Profile1.Distribution) !RelatedMatch Profile3.Distribution
and we can go further and conclude through some reductive steps that
  • Profile2.LocationText !RelatedMatch Profile1.Geocode
  • Profile2.LocationText !RelatedMatch Profile3.Distribution
  • Profile1.Geocode !RelatedMatch Profile3.Distribution
etc

and we can make a nice fuzzy thesaurus which will help us relate terms in differnt profiles etc etc.

Description

from an email sent by GJR

"Been thinking about the rats nest that is "description".. best thing is to jettison it to be replaced with narrower concepts e.g."
  • Description
    • Distinguishing attributes
      • Circumscription
        • Identification
        • keys
        • discriptlets
      • technical description
      • Distinguishing Anatomical Attributes
      • Distinguishing Morphological Attributes
      • etc
    • Knowledge and (General) Notes
      • Physiology
      • comments
      • Morphology
"The names/ labels are not quite right but they seem a better fit with what we were grappling with last week.. Can you make an appropiate entry into the workshop notes AND see if these conceptual models fit better with the sample profiles etc"

Meeting 23/9/10

Agenda

  1. terms from cyndy parr
  2. term taxonomy for Distribution
  3. (part) term taxonomy for Description

Discussed

Discussed> Cyndy has sent on a list of terms which she's accrued in the course of wrangling content for the EOL

Decided> Paul will create a simple (restricted access) wiki page with a table of triples containing potential matches to WC terms as he can identify them. For discussion by group => PA

Discussed> The terms assigned to the Distribution Term Group

'WC:Distribution and Habitat' is a broad term which contains subcategories.

Some of the proposed WC terms names need work. Indents implies is contained in...

Draft Term WC Terms (term names may need finessing).
  • WC:Distribution_ And_ Habitat
    • WC:Habitat!
    • WC:Distribution
      • WC:Range + Dispersal
      • WC:NaturalDistribution
      • WC:LocationsInCaptivity
      • WC:DistributionOfIdentifiedPopulations
      • WC:Distribution_within_focus_region_of_interest_to_Profile
      • WC:extralimital_distribution
      • WC:SpatioTemporalDistribution
        • WC:TemporalDistribution
          • WC:PastDistribution
          • WC:CurrentDistribution
      • WC:Range
        • WC:GeographicRange
        • WC:Range_within__focus_region_of_interest_to_Profile
        • WC:extralimital_range
      • WC:DistributionOfMovements
        • WC:PeriodicMovements
        • WC:DistrubutionofMigratoryMovements - other broader details of Migration belong in Behavior
          • WC:DistrubutionofMigratoryMovementsTiming
          • WC:DistrubutionofMigratoryMovementsSpatial
        • WC:DistrubutionofDispersal - other broader details of Disperal belong in Behavior
          • WC:DistrubutionofDispersal-Timing
          • WC:DistrubutionofDispersa-Spatial
        • WC:DistrubutionofSettlement - other broader details of settle belong in Behavior
      • WC:Original
        • WC:OriginDistribution
      • WC:ObservedLocations - disambiguate from e.g. broader details to do with observations of behaviour
      • WC:SpecimenLocalities
      • WC:DistributionByRegion text containing lists where taxon reported e.g. state, country or a contolled vacabulary such as IBRA regions
      • WC:DistrubutionMap
      • WC:DistrubutionSurvey
Also here are some terms which are part of ecology

=> PA to create above terms and map from term group

  • WC:Ecology should include
    • WC:PopulationDetails
    • WC:HomeRange
    • WC:SurveyEcology
=> PA to create above terms and map from term group

Discussed> quick categorisation of terms in the Description term group

  • WC:Description
    • WC:DescriptionExtendedPhenotypes -distinguishing elements in form and structure of the extended phenotyoe http://en.wikipedia.org/wiki/The_Extended_Phenotype
      • WC:DescriptionNests
        • WC:WC:DescriptionNestsTime
        • WC:WC:DescriptionNestsForm
    • WC:Comments
    • WC:DescriptionBehaviouralCharacters - solely to do with distinguishing elements of behaviour (as opposed to details of behavoiur/ethology)
      • WC:Description_of_Vocalizations
      • WC:RenderingOfVocalizations
      • WC:RecordingOfVocalizations
    • WC:Phsyiology - solely to to with distinguishing elements of Physiology (as opposed to details of physiology)
    • WC:DescriptionMorphology
      • WC:DesriptionOfMorphologicalElement - of note e.g. {fungidu.head, fungidu.gills etc }
    • WC:DescriptionForm
      • WC:DescriptionFormThroughLifecycle - details of how the form varies through the lifecycyle of the taxon
      • WC:DescriptionOfHabit
      • WC:ColourNotes
        • WC:Flower
      • WC:DesciptionOfPlumage
      • WC:ExperientalDescription
        • WC:DescriptionFeel
        • WC:DescriptionSmell
      • WC:ChromosomeCount
=> PA to create above terms and map from term group

I think we need a term WC:uses if we don't have one.

  • WC:Uses
    • WC:Food
      • WC:PortionUsedForFood
    • WC:Artefacts
OK and for terms to do with how we identify taxa

  • WC:Identification
    • WC:Diagnosis
    • WC:Key
      • WC:dichotomous_key
      • WC:matrix
      • WC:discriplet - viz descriptlets as developed by IdentifyLife
=> PA to create above terms and map from term group

Meeting 16/9/10

Agenda.

  1. PA will present full mappings of common names and scientific names to WC
  2. We will wrangle, finanlise and consider workflow.
  3. Any business to do with EOL and Delivery Plan Finalisation

Discussed

Discussed> Namespace Categories of terms as shown below.

Decided> to rename namespace Category "Specification" to "Taxon profile Model" => PA

Discussed> The recording of example profiles or profiles yet to be described (see "Profile Example Database" below)

Decided> to record such example and casually encountered desired profiles in the WC Taxon Profile Namespace but for clarity we will add an additional ttribute to namespaces "Status" (maybe a better name)

This "Status" attribute will be a check list of states including
  • not described (in WC)
  • don't describe
  • to describe
  • described
  • under review
  • Retired
We can then used as a workflow, collect profiles into one system and then record easily what we choose to do.

Scripts will need to be adjusted => PA and PH

Discussed> Common Names

Paul had mapped all Common Names encountered in Profiles to WC.

We tested the mappings of these names against the categories for common names the 7/9/10

Mapping good. Paul has subsequently created WC terms for each of these types of common names (bar two). => PA

Discussed> Scientific Names.

Paul had mapped all Scientific Names encountered in Profiles to WC.

It turned out to be as complex as the arrangement encountered in common names .. and not as clean a match as we would hope for that used by nomenclatures such as APNI

Decided>

"Encoding what we see is a separate matter from later reccommendations about best practice in taxon profiles."

in this case best practice would be to use linked data or a GUID.. to a nomenclator such as APNI.

WC.Names - should be a broad category for those profiles which mix scientific and common names all into one container

which contains (hasparts)
  • --- WC.Scientific Name
  • ---- WC.Common Name
WC.Nomenclature = a container for text to do with scientific names

which contains (hasparts)
  • Taxon Name with authority taken from a named source - may be named for the 'parent' work - Taxon Names include binomial names but are any nomenclaturally correct latinate form for a taxon. Typically Author of the taxon is not specified just the taxon name.
  • Accepted Taxon Name - Taxon Names include binomial names but are any nomenclaturally correct latinate form for a taxon - accepted by an authority which is usually named in the work or profile or may be inferred by the context in which the profile was compiled.
  • Voucher Name - .... e.g. Acacia Sp1243 - a name for a potential taxon not yet formally described
  • Taxon Name + Author - Taxon Names include binomial names but are any nomenclaturally correct latinate form for a taxon. Typically year of the of publication is not specified.
  • Taxon Name + Author + year - Taxon Names include binomial names but are any nomenclaturally correct latinate form for a taxon. with sufficient details to identify the citation. In best practice it conforms with nomenclatural rules.
  • Taxon Name + Citation Used for Taxon names where the publication citation of the description is included.
  • Genus - Name of the genus of taxon in profile
  • epithet - epithet ( ) of the genus of taxon in profile
  • taxon author – the scientist who first published the name (and concept)
  • WC.Rank - describes the rank of the taxon .. often just a single word .. very rarely may also describe the concepts and reasoning behind the assignation of rank
  • WC.NomenclaturalNameStatus - a text of narrative which describes the status of the name, any related debates or concepts
  • WC.GUID - link to nomenclator.. something about what a GUID is...
  • WC.Synonyms - formerly or otherwise known as....
WC.Phylogeny - should contain
  • WC.sub-taxa - text which describes sub taxa which belong to the taxon under consideration typically sub-species.. may just be a list of names and as such can be confused with nomenclature... and in a profound way is nomenclatural in that there are nomenclatural tools to indicate subspecies etc.
Actions => PA to create WC terms and map examples as term basis. Best practice will be to use name services such as APNI and AFD so best not to over egg these descriptions. Term definitions can probably be sourced from either TDWG or GBIF. 'nuff said.

Taxon Profile Models

I've been reading the SPMI wiki. Here are some links to check for loose ends from a quick read of the SPMI entries on the TDWG wiki. Have we dealt with the following from http://wiki.tdwg.org/twiki/bin/view/SPM/BackgroundReading and http://wiki.tdwg.org/twiki/bin/view/SPM/SpeciesProfileModel ? Also note Nicole Coombs (who I think you met last week) contributed a comparison of commentary on SPRAT and SPMI.

Lets discuss. Maybe add what's missing to the wiki as commentary if it doesn't merit a formal treatment. Question: what merits a formal treatment? opinions?

G

This suggests a new category of Taxon Profile Name Space = Taxon Profile Model

Which should contain material drawn ( as deemed relevant ) from We have a lot of this in train already. So not so scary.

Profile Example Database

We need a means to capture examples of profiles of note. (not described or mapped to WC but for interest for variety or completeness etc). And that we add to this page as much as possible. As a matter of practice, we should archive the examples. here is a start

Meeting 7/9/10

Agenda

  1. Awesome sprat
  2. review of field day
  3. review of demo material - what more do we need for showing off
  4. bingo
  5. EOL - collaboration.
Note PA on leave Wed 8th - Returning to work 14th

Discussed

Discussed> Marvelled at SPRAT - as a model for some well defined elements to do with locality and distribution.

Decided> Name of Workspace should change EPBC -> SPRAT. => PA

Discussed> potential collaboration with EOL

Garry owes Cyndy Parr a creme caramel. Jim will take responsibility.

Discussed> Terms yet to be finished in Fungu DU and in Wallace Core

Decided> Common Names need Clarification. PA to create Common Names in WC along the following manner => PA

Common Names - is a broad content category which containing a number of better defined sub-terms.
  • Recognized Common Name - by some authority (or by the author in a particular work)
  • Assigned Common Names - a name assigned in a work with the intent that it be used as a common name
  • Other common Names - may need a better label - other names in common usage which are assosciated in a strong enough manner that they are used for this taxon.
  • Cultivar Common Names - a special nomenclatural name used for cultivars
  • Trade Names - Names by which this taxon is known commercially
  • Legal Names - a proscribed common name for a taxon in a given jurisdiction for regulation and control (may be same as Trade Names)
  • Popular Names - a name in wide usage associated with this taxon which may also apply to many taxa
  • Indigenous Names - a special class of common names - may include reference to the context and methods used to verify use, validity, language group, locality details and taxon coverage. In principle, these methods could be applied to all common names (who said, how do we know it, what means did we take to verify its application and our understanding of this particular vernacular). In practice, this validation is only used for indigenous language
Discussed> definition of FungiDU.Family. GJR had a brain melt.

Decided> FungiDU?.Family is a case of a term who's definition can be reasonably sourced from This is the case for many terms in WC . And where this is so we should source words from TDWG, the codes, or GBIF => PA

Discussed> Workflow. Split brain and processess are likely to cause group thrombosis.

Decided> to seek simplicity. Next meeting will focus on two WC terms which have complete mappings from the sample profiles to do with Scientific Names and Common Names. => PA

Decided> graphs to be used a workflow tool, ditto termgroups (as previously) => PA to get graphs working

Decided> All scripts as far as practical to indicate if they have the status of being used as the BASIS of a WC term => PA

Notes 2/9/10

Name space Metadata

  • broad taxa (e.g. birds, plants etc ) [already in place on NS metadata form]
    • contains a simple plain language description of the taxa covered by the profile (Birds, Plants, Eremophila etc)
  • types of works (i.e. types of profiles) [already in place on NS metadata form] is labelled Category
    • already refined to a check list & contolled vocab
  • coverage [to be added to NS metadata form]
    • a simple and brief plain language describes the non taxonomic attributes group of organisms chosen for the work
      e.g. Marine, Invasive, Rare, Pests, subtropical, Queensland
  • Relevance [to be added to NS metadata form]
    • why was this work chosen to be described? Why is it so?
      e.g. chosen to extend coverag of WC sample to insects or
      "This is a well formed profile with some interesting new terms" or
      "Created as part of the Plume Moth Wiki"
  • usage [to be added to NS metadata form]
    • field guide, Encyclopaedia
      also a summary of notes and remarks giving guidance as to how the author thinks the work may be used
  • citation [to be added to NS metadata form]
    • COMPULSORY.
    • an unambigious reference that identifies the work
    • [with nested forms this maybe a GUID for the work/resource] =>PA / PH (ISBN or DOI)
  • intent [to be added to NS metadata form]
    • distinct from usage in that it describes briefly REASONS why the work exists
    • e.g. "to assist land care ", "Border control", "convey new information"
    • this field has the weakest case for inclusion *=>PJA I believe that including this attribute will provide a way to eliminate the need document atomized data (often put in Descriptions), and raise the importance of works like: Identification field guides and texts on bushtucker that place less importance on the Taxon Names. An example is; there are many taxon profiles that have the intent of cataloguing/accessioning a collection of specimens/treatments where the name is important, and others are compilations a bush plants and their medicinal, industrial or uses as food. *
  • Audience [to be added to NS metadata form]
    • COMPULSORY
    • brief comments as to whom the work appears to be targeted
    • e.g. "General", "Technical", "Biological", "Medical", etc
PA to implement ASAP.

Meeting 26/8/10

Agenda

  1. PA is to choose and map at least 3 moderately sized profiles to map to WC prior to meeting
    ..suggestions
    - Eremophila
    - Fungidu
    - crusty
    - Weeds
  2. Field day prep - BRIEFLY Review prep.
    - remote microscope
    - wiki
    - taxon profiles
    - keyserver
    - ??

Discussed

Discussed> The TRIN Field Day

Discussed> Wiki and the Field day

There are new and coming capabilities in the wiki - neat skunkworks stuff.

PH Demonstated some of the new wiki capabilities that will come when jqgrid is fully deployed.
  • including a copy of Lauries Latin Dictionary, and neat things with Taxon Profiles, Wallace Core and Profile Terms & co.
  • A neato way to export from the wiki - select fields from a taxon profile.
  • Some of these will be available to demo in a preliminary form for the field day
PH will bring his workstation. for bestest effect at the field day.

Discussed> Taxon Profile, Wallace Core and the Field Day

Decided>

We need some canned material. If needs to compliment the poster and if possible be a step by step sequence.

This should be compiled in a page in the Taxon Profile Wiki as a child of the WebHome under with a link appearing under the heading Qickstart in the webhome text.=> PA
  • Terms and Profiles. Some Profiles and terms which "look good". meaning - need to convey impression of care and rigour they someways towards being described, they may be incomplete, but must be at least with original definitions and mappings. to check over and choose
  • Relationships. Similarly need some canned examples.. again do not need to be complete but certainly need to convey a real impression of care and rigour
  • A profile, preferably some profiles, that we have described.
  • Link to a term which demonstrates hwo we are going about definitions including the basis of the term and collaborative QC provided by the happiness index
  • A link to sample mappings, relationships and the happiness index
  • A examplar WC term including the lovely graphviz terms
  • a link to the list of terms in WC
  • ditto to the relationships.
  • hand over to PH for demo of content hosting for taxon profiles.
Discussed> Select Taxon Profile terms and mappings. Including Distribution. Made changes. Argued about Distribution and maps.

Clarification 25/8/10

PA and GJR Discussed Nested Profiles (See problem raised previous meeting)

Decided> Clarity is possible.

GJR's opinion is

We have two cases.
  1. A work (web page, book etc) contains more than one profile - often applied at different levels of classification
  2. A work has only one profile that is applied with minimal variation (e.g. in part at different times) at different levels of classification
  • An example of 1 is Macdonald, J. D. (1973) "Birds of Australia"
  • An example of 2 is .... ... possibly "Mammals of Australia" - but it can be argued that some of the nomenclatural terms with the same label are really different
Key thing is that for #2 to be apropiate then ALL TERMS must mean the same thing precisely at all levels in the nested profile.

If a term has the same name but a different meaning (at different parts of the work) then it is more than one profile.

** NOW. The important thing to decide is whether there is any UTILITY for our project (meaning in the breadth of our sample or in the quality of the sample) in describing minor profiles.

I contend we don't need to describe all the profiles in a work. Just those which broaden or add value in some other way to the basis for WC.

Opinions?

Meeting 24/8/10

Agenda.

  1. Prep of Talking points and examples for Field day
  2. Sparkly new scripts PA has written for wiki
  3. Definition Bingo
  4. new WC terms - from poster See below

Discussed

Discussed> Prep for Field Day

In this wiki prepare a page in the Taxon profile web with text and links to examples of the various elements of taxon profiles including Taxon Profile. This is also a good documentation for later on. => PA

Examples to include links to
  • Exemplar Profiles (see other action items below)
  • Wallace Core
  • Mappings to WC
Wik. Similar idea. Prepare a pageon wiki with links to various examples and uses we want to show off
  • wiki hosting
  • Profiles
  • taxon and society business
We also workshopped talking points and thoughts about the goals of our projec. Following quotes adapted from from EMC's notes

Overall goals of WC and TPT
  • to make descriptive information
  • accessible
  • reusable
* No standardisation is possible

WC
  • a translation devices between multiple profiles allowing similar (like) data to be recognised and then analysed, interrogated
  • Not arbitrarily making claims but basing them on past and current usage with various taxon profile structures sampled (namespaces in our lingo)
different degrees of similiarity (relationships)

relationships - to express exactly concepts like 'exact match', 'near match', 'is part of' etc

Fluidity, rigour, and quality control documented by the use of 'happiness indices' at all levels

capture information on the wiki - move it aroundto other sites, databases etc

Wiki Message
  • Collaboration
  • Independance
  • wiki is the 'database' without a database. Structured through a document-object data model allows users to create arbitrary data structures early and evolve them.
  • PH will supply words from abstract prepared earlier
  • Platform for operating on Taxon Profiles.

Decided> may approach SD for assistance. Keep on with current plans.

Discussed> Ajax-ified versions of fix term 'scripts'.

looks great. Needs to be re-factored to take account of the new workflow and capabilities that ajax enables.

For the field day... not a priority. They just need to work so that they can be driven as required. => PA & PH

Discussed> the encoding of work with nested profiles. To recap we had discussed this in part in an earlier meeting.

quote from Meeting 2 / 7/ 10

"Clarification > Where profiles are nested, and the fields vary with nesting, these are to be described as separate profiles distinguished by the profile name e.g. BirdsOfNewZealand?.Family vs BirdsOfNewZealand?.Species and so forth. => PA"

Contrariwise in this meeting we said> "where a profile contains different terms or a hierachy of terms it's ok" as long as "it needs to be really clear that in the original work" i.e. "in the definition or comments of the term it is indicated that in the original work that the taxonomic how it was applied"

in this case we used the Mammals of Australia as a reference example of nested profiles.

Decision> ???

GJR> Clearly we are confused. And more data is required. Suggest we wait and see whether this is a problem of any significance. Also, we must always be aware that we are coding works as they are... and WC can be our own child to 'suite' our designs. Faithful replication is the thing for our source profiles. Meaning must be clear. With that spray in mind, I suggest the following

Decision>
  1. It must always be REALLY clear that the context in which the term was used in the original work
  2. This should be indicated in the definition of the term. With an example as ever of its usage
  3. When we have sufficient DATA (i.e. examples) to make a better determination as to the best way to go we may refactor previously encoded terms to match ... so they must be easily found. Suggest entering a standard markup so that they can searched for. Perhaps "multi-profile"
=> PA

Discussed> White space on forms and scripts

  • PH has plans in hand with coming neat tricks based on AJAX, nested forms, reworking the wysiwyg editor and JQuery spreadsheet.
  • In the meantime, perhaps some CSS reworking.
  • we will need another developer or two if we want to stretch to iphones, ipads and such like
Next Meeting>

PA is to choose and map at least 3 moderately sized profiles to map to WC prior to meeting => PA

..suggestions
  • Eremophila
  • Fungidu
  • crusty
  • Weeds

Discussed> Proposed New WC terms

Some new terms to add to WC => PA

WC:Status
WC:Impacts
WC:Dispersal / and its counterpart WC:Recruitment
WC:Origin
WC:Overview aka 'Key Points'

Decided> Status to be created with Definition based on words used in the equivalent term group "status" - spatiotemporal basis

basis of terms two weed profiles referenced in which have already been described.

Next meeting PA to bring list of terms with original definitions (such as they are) in place for review.

Meeting 19/8/10

Agenda

  1. An idea 'bout how to encode images, plates, and illustrations in profiles; relationship ops
  2. Definition bingo…
  3. may be taken up with prep for field day

Discussed

Discussed> The poster for the TRIN field day

Decided> on content. Simplicity. few words. and other changes. to be submitted to SD Friday => PA

Discussed> Images, illustrations, plates and other multimedia type objects that may be in a profile.

There was a vigorous discussion about how multi-purpose content could be encoded and existing schemes that could be adopted to that end.

Also that we risked going too far into the detail of profiles.

Decided> to keep it simple. That said a little detail is needed to be clear.
  1. Where content in illustration/image/drawing/etc can be unambigiously attributed to one Profile Term. Then it should be described as such. An example would be a distribution map - as discussed previously.
  2. Where an image contains hybrid content and is a portmanteau of many content categories e.g. it is a composite containing a map, a line drawing and a picture then a separate term will be created for the image. If a profile in a work has more than one image then it will either be a repeated element or separate elements - depending on whether the images differ in content, constitution, or meaning -as the case may be.
  3. Only when it is tractable and worthwhile, we will record that the content has parts as topics - an example follows.

NS1:Illustration = a topic describing the image (not the content or format etc just as per other topics)

NS1:!DistributionMap ispartof NS1:Illustration

NS1:!MorphologicalIllustration ispartof NS1:illustration

and so on. but only insofar as it acts to describe useful information about the categories and taxonomy of information in the profile being described. => PA / PH

Discussed> WC:GeographicalDistribution.

Decided> Refined its definition. Terms which are mapped to WC:GeographicalDistribution need to be reviewed.=> PA

Decided> PC:Territory belongs to behaviour => PA

Discussed> Nomenclatural elements in Marine benthic flora and elsewhere.

Even though nomenclatural elements are handled by service such as APNI etc we need to be able to represent profiles as they are as well as we wish them to be.

Decided> That terms in WC need to be created for all typical nomenclatural elements .=> PA

Decided> add Typification to WC. Basis would be SPMI or maybe somewhere in the TDWG voc=> PA

Meeting 17/8/10

Agenda


Paul will have hand crafted a whole heap of terms (with source definitions & co. ) for us to argue about

Discussed

Discussed> additional capabilities added to the stat scripts

Now enumerates the number each kind of relationship. good stuff.

Decided> it would be nice to be able to get a clickable list of terms with each relationship=> PA

Discussed /Argued > about a portmanteau term in Eremophila 'Distribution/Ecology'

2 issues to be address. i) works which use terms in a non standard way need disambiguation

Decided> To add a 'disambiguation label' to term description to complement our derived definition. This term is to be sparingly used. Scripts will be adapted (as appropiate to use the disambiguation term if it exists => PA & PH

ii) 2nd issue arises. Mapping the portmanteau term in Eremophila 'Distribution/Ecology' into WC

Decided> to create a corresponding portmanteau term in WC for 'Distribution/Ecology' and map as follows

WC:DistributionHabitat haspart WC:Distribution

WC:DistributionHabitat haspart WC:Habitat

WC:DistributionHabitat ExactMatch? Ere:DistributionHabitat_

ERE:DistributionHabitatEcology haspart ERE:Distribution

ERE:DistributionHabitatEcology haspart ERE:Habitat_

17082010001.jpg

General Principle is this is to be done sparingly and will be on a case by case basis.

If a profile has such a portmanteau term then we may consider atomising it in our description. Again a case by case. If so, the distinction must be maintained throughout the work (if only typographically).

Discussed> Maps and other illustrations and plates.

Decided > In eremophila , the map is sufficiently clear that it should be coded as a separate item in the Distribution.. Ere:Map ispart of WC:Map ispart of WC:Distribution (or similar)

Decided> This may not apply to all pictures and images in profiles. Certainly not reliably in pictures or plates.

It may be cleared for line illustrations as in the Eremophila but even there authors often compose complicated multipurpose illustrations which will be difficult to code.

Decided> Where it is possible to code images and illustrations so that it can be analysed. .. but not too much.

Meeting 12/8/10

Agenda

  1. following on from Tuesday
  2. seeking more definitions

Discussed

Discussed> addition of a new type of Profile type "Species Complex"

Decided> not yet. In principle, any new class/category should be on the base of discussion of data(examples) before creation

Discussed> use of happiness index.

Discussed> Ideas for notification of changes to terms, and notification of implied change through linked terms i.e. change to a linked term means that a review may be required depending on the nature of the link.

Decided> to see if we can automate notification. One idea is that when a user changes a term/ topic/ metadata/ relationship they are automatically subscribed to receive notifcations of subsequent changes to that term/relationship/topic => PA / PH to investigate

Decided> to add a 'review tag' to mark terms which need review after any change (EMC> "on dirty")... idea being that when a term is altered all terms which are either "Same As" or "Has Part" etc (a limited list of equivalance relationships) would be marked as requiring review => PA / PH to investigate

These changes will require scripts which list terms requiring review.=> PA / PH to investigate

Discussed> The descriptors: Example and 'Intent and Usage' which are attached to the termAttributeForm. https://wiki.trin.org.au/TaxonProfile/Builder/TermAttributeForm. TermAttribute:Definition and TermAttribute:IntentAndUsage serve a similar function to our use of Term:Definition with Term:OriginalDefinition

Decided> to add an "examples" field for the description of terms. This field is to contain a verbatim example or examples of how the term was used in the work being described.

Discussed> metadata for Namespaces.

Decided> add an audience field. For now which will contain a textual description of the intended audience and use of a given profile. We find structure in this field over time and post-hoc add more metadata fields to describe profiles=> PA

Decided> The inclusion of definitions for fields in forms was grand. Would be nice to make the font smaller (via css?)=> PA & PH

Discussed> next meeting

Decided> as a priority, PA is to have lots of terms ready for definition and arguement. => PA

Meeting 10/8/10

Agenda



1. Lets take some time to review where we are at
- numbers described, & co.
2. Definition Bingo - with the two profiles PA has prepared earlier

Discussed

Meeting Crashed by Siobhan and Kathryn

Discussed> Field Day

There is to be a Taxon Profile Stand & a Taxon Process Stand/poster

Siobhann requires content for Posters for both - Pictures and Text. by week ending 20th August. GJR wil lcompile with PA => GJR/EMC/PA

Discussed> a script developed by PA which gives counts of terms Term Counter based on their labels

This method shows great promise... not only as an indicator of what we have but as to how far we need to go. Makes a great worklist

Decided> Could be usefully applied to terms based on mappings. and Namespaces => PA / PH

Added> a new category to Profile Type for Specification = "Taxon Profile Specification, e.g. SPM"

Decided> to see if same script can be applied to namespaces to give stats as per Term Counter => PA

Discussed> Terms in the "Unassigned Term Group"

Added> a new term group for "Status" terms. PA may suggest an alternate name for us to discuss.

Added> Assigned new terms to Term group Status

Discussed> xml export structure for WC and Mappings

Decided> to seek advice from ALA and IBIS to ensure our WC data and mappings can be encoded without loss/formallly etc in XML or RDF

Meeting 5/8/2010

Agenda



1. definition Bingo…
2. Review of scripts
3. middle layer documentation & de-skunkification
4. Use case documentation

Discussed

Discussed>Terms mapped to WC.Description from other profiles. Did a lot of wrangling and occassionally came to a consenus

Changes noted in Comments fields and Happiness ratings adjusted accordingly. Well done PA

Discussed> New scripts suggested by all the fun above.

scripts for
  • listing terms mapped to a given WC term from which we can adjust Definitions or change mappings. would be similar to Create Relationship =>PA
  • Listing the terms in a group and WC mappings (GJR's notes unclear ?) =>PA
Review> Definition of WC.Reproduction needs work. indeed term probably no longer required. Action =>PA

Discussed> the utility of graphs shown in WC terms

Decided> to extend these graphs - to be scripts so that for a given WC term (and maybe others?) we can group mapped terms by NameSpace (i.e. the work from which they are drawn) are grouped (boxed like with like) OR select only term mapped where from particular kinds of works (i.e. NameSpace Attributes)

e.g. show mappings of terms from particular =>PA with PH
  • broad taxa (e.g. birds, plants etc )
  • types of works (i.e. types of profiles)
  • coverage
Using the NameSpace metadata described in the previous meeting. =>PA with PH

Meeting 3/8/2010

Agenda

  1. a plan
  2. Definition Bingo (on 2 profiles prepared earlier)

Discussed

Its APA season Again!

Discussed> Tasks & output of WC from list GJR is developing as part of Delivery Plan and for APA's

(some details here arise from further discussion with PH & PH folowing day 4/8)
Tasks Group Output
Storage/Search for TP Terms, mappings, content WC WC Reference Implementation
Documentation: Guide, explanations & Overview to TPT & WC WC

Disambiguation Service
- webservice returning (inc. near) matches for a given term

including near terms
and terms organised by practice, application and purpose

Scripts WC Crosswalking tools for Profiles
Mappings (esp. to WC terms)   Analysis tools for profile structure (particularly mappings)
WC term definition (dictionary)    

Profiles to fully describe + map

  • Reference Profiles
    (SPMI, EOL, PC, Edit etc)
  • Sampled Profiles ~ 25

Toolkit

Tools for making bespoke profile and/or reccomendations (including templates)

  • for new profiles
  • for practice
  • standards for taxonomic profiles

Recommended practice guides

Publish WC and Review of field WC

Tools for

  • Managing mappings
  • term definitions
  • analysis of profiles vs WC

Analysis

  • Conguency measures
  • Coverage measures
WC WC managed by Societies and Organizations representing those who make conetent
TPT packaged Toolkit Hosting of content
  Toolkit Post hoc - retrofitting of WC to other profile
  Toolkit

TPT in a distributable form

TPT in the cloud

a means to push TPT content to a central service

03082010.jpg

Decided> more about authors & metadata ..

Require additional "Meta data" fields for describing Works (i.e. books or webpages) containing profiles including
  • Relevance
  • usage
  • citation [with nested forms this maybe a GUID for the work/resource] =>PA / PH
  • intent
Action =>PA

Discussed>The create Relationship script Create Relationship

Decided> Move the Definitions of WC terms (in group under consideration) to top of the page

add anchor

Discussed> WC.Description

Decided> needs to be broader more inclusive for non Morphological attributes and traits.=>PA

Table of contents

Meeting 29/07/2010

Agenda

  1. Matters arising Tuesday
  2. Paul will have two more profiles ready for the team.
BINGO

Discussed

Discussed> the big picture at some length.

Decided> need to seek guidance from above. GJR to send next version TPT delivery plan to EMC

Discussed> Authors. Need further disambiguation

viz structure of Flora of Australia

Flora of Australia editor

(Helen Thompson)

Is the creator of the flora of australia profile structure Meta data about NameSpace?

Acacia Section Editor/Author

(Bruce Maslin)

is an editor of a section probably not recorded at present

Profile Contributor

Joe Miller

contributes information/content metadata about profile = ProfileAuthor?
** though it is frequently the case that a single author/s will be in all roles in this table.

Decided> need a metadata field for namespaces capable of recording the Citation of the work from which the namespace is derived. Authorship & co. not necessary to record. as will be part of the citation or inferrable. NamespaceCitation =>PA

Discussed> Fields in the benthic flora of SA. PA is doing a second pass at this name space

Decided> Thallus, Reproduction, Sphophyte etc are the encoded separately. To preserve the decodability of the text. They are to be defined as descriptive elements as in the mushrooms of oz. And mapped as part of to WC.Description =>PA

Meeting 27/07/2010

Agenda

  1. Review of scripts & tasks
  2. Finish off the Fungi of Australia
  3. Paul is preparing 2 profiles for the team to work through i.e. original content (definitions and co). Team to supply words & co.

Discussed

reviewed> The relationship Script i.e. Create Relationship

needs to be reworked when the storage of relationships are in their final form (refactored as a topic per relationship)

Decided> Until refactoring done, a quick jump to topic which holds the relationship under question to be implemented as a link or button =>PA

Discussed> A new script for making/refining relationships.

Script would show relationship under question as a triple + associated happiness rating

Underneath it would show definitions of terms on both sides of the triple and defitintion of the relationship.

Action =>PA to prototype/ implement (depending on the status of the refactoring of the representation of relationships)

rigourously Discussed> Authors in their various guises.

Decided> DISAMBIGUATION required of WC to distinguish nomenclatural authors from Profile or other content authors.

JRC > WC.ProfileAuthor to be same as (based on/exact match ) dc.creator = author of the work =>PA

Decided> need WC.Authority (in profile 'metadata' group ? ) an example of which would be CHAH.. refers/indicates a work has a status of an authority

Authors

of a name nomenclatural
of a paper/citation a treatment, article or citation
of this profile / 'metadata' group
Decided> that WC.Author needs to be disambiguated.

WC.Author replaced by

WC.NameNameAuthor

WC.ProfileAuthor

and maybe by WC.CitationAuthor (for citations within a profile)=>PA

DEcided> Benthic Flora. Author to be examined to see if it should be renamed to NameAuthor =>PA

Decided> Need a new field in all profile terms for usage to be recorded =>PA

Decided> that it is possible tho not desirable in general for the headings (labels) used in a profile to be different from the term name we use.

Decided> PA to examine Authors in citation and meta groups to see how they match disambiguation decided above =>PA

Meeting 22/07/2010

Agenda

  1. A spiffing round of definition poker.
  2. Relationship bingo
  3. identify which relationships which are suitable for indicating hierarchy within WC (internal only relationships)... And tools

Discussed

Discussed> definitions of skunkworks per http://rossmyers.com/?p=73 .... "As a generic term, it dates from the 1960s. One definition is very much that of the original and the one you describe: a small group of experts who drop out of the mainstream of a company’s operations in order to develop some experimental technology or new application in secrecy or at speed, unhampered by bureaucracy or the strict application of regulations "

which led to kick-a-poo joy juice.... "The original was Skonk Works (skonk is a dialect variant of skunk), the place of unexplained function near which Lonesome Polecat and Hairless Joe brewed their highly illicit bootleg Kickapoo Joy Juice from ingredients such as old shoes and dead skunks"

Discussed >Script built as per last meetings brainstorming

i.e. Create Relationship

some small amount of work to do before it is fully commissioned. Looks good.

Discussed> whether the list of possible relationships should be circumsribed to ONLY the (currently assiged) term group .

Decided >

relationships could be made to WC terms outside the (currently assiged) term group. =>PA to consult and implement.

that this would on principle be not the preferred end point BUT is necessary to record where further work is required.

Discussed> the WC definitions at bottom of page Create Relationship should be populated from
  • only WC terms within the group to which mappings have been made OR
  • a list of any WC terms to which mappings have been made irrespective of which group they belong
GJR cannot recall what meeting decided but prefers later option. =>PA to consult and implement.

Discussed> Null fields for definitions and relationships.

Decided> There should be no BLANK text fields instead

Definitions should contain standard entries indicating state such as either
  • 'Not mapped'
  • 'Not defined'
  • 'None Supplied' =>PA to implement AND RETROFIT existing blank fields.
Discussed> adding a button to Create Relationship to create a new terms as required if there is no corresponding WC term at the time of mapping. Initially, content to be copied on creation. And mapping may be an exact match. But both content and mapping will NEED TO BE REVIEWED. So the new WC term will have a happiness rating of 1 or 0. Thus allowing the newly created term to be put into a list for review. =>PA & PH

Definition Bingo> focussed on a volume of the fungi of Australia. good progress to be continued.

Note> The definition tool might be enhanced by a create term button =>PA to investigate

Clarified > WC will be hierachical. a suitable relator should emerge with time. GJR would like this to be so that it is simple to create triple searches (without additional qualifiers). PH thinks it may be one of the SKOS relationships. =>PA to watch and advise

Meeting 20/07/2010

Agenda

  1. A spiffing round of definition poker.
  2. Discussion of what should be hidden away..
  3. New profiles - a place to park them? Or a status (not yet)?

New Suggestions.

1. per Donald

NZ Arthropod Collection

The new internet factsheets, Interesting Insects and other Invertebrates, are now available on line at http://nzacfactsheets.landcareresearch.co.nz/Index.html The production of the first 18 factsheets was funded by TFBIS, a DoC?? operated Fund. It was a joint project with Landcare Research who did the IT work. The material in the factsheets can be used by people for their own projects. We welcome feedback about what you like about the factsheets and where improvements can be made. In a few months time we hope to invite members to write their own factsheets and so contribute to the series. Nicholas Martin Entomologist, Honorary Fellow, Plant & Food Research,

2. per David Yeates

http://diptera.myspecies.info/content/flytree

This is an Edit scratchpad page. And certainly something we should look to describing. Also, flies.. So politic to do so. Can you create somewhere on the wiki to keep such suggestions (perhaps just as an entry to the Workshop notes)?

http://diptera.myspecies.info/category/diptera-classification/bibionidae

http://diptera.myspecies.info/category/diptera-classification/keroplatidae

3. also per David Yeates

http://www.lifedesks.org/newfeatures/

Here is some more information on another set of species pages developed by my collaborator Torsten Dikow.

Best

Concerning the Mydidae species pages, I have two ways of how to provide these in the projects of the grant. One is to publish manuscripts in ZooKeys?? so that the journal automatically provides the species descriptions to EOL (as well as the specimen data to GBIF, nomenclatural changes to ZooBank?? etc.) and therefore species pages are produced. I have submitted a manuscript to ZooKeys?? in May with some new Mydidae from eastern Africa and the Oriental Region and this is also the way that my summer intern Katie will provide species pages for Ectyphus and Parectyphus from Africa.

Another approach is to make a LifeDesk?? for Mydidae and upload information about the species. This is what I have done with some Leptogastrinae species on which I have published previously (see a species page in the LifeDesk?? http://leptogastrinae.lifedesks.org/pages/1797

and on the EOL http://www.eol.org/pages/77090

http://www.eol.org/pages/770905.

GJR> of note is sparce content on lifedesk.. consequence of de novo model?

GJR> note opportunity for WC?


Zookeys is the best of the etaxonomy journals that Zoologists use. Automatic Zoobank rego (not sure about LSID’s), species page generation, links to EOL – read about it in their special issue.


http://pensoftonline.net/zookeys/index.php/journal/index

4. per John Hook / Donald

Sponge DB.

5. Zookeys per DY

>this details the enhancements to etaxonomy that Zookeys has initiated.

Thanks David. pdf seems to have been mangled enroute. Was it from Lyubo's special issue of Zookeys http://pensoftonline.net/zookeys/index.php/journal/issue/view/52 ?

which has been the cause of much talk here.

G

GJR> note publishing workflow
GJR> also note collaboration & their model of a TP / treatment.

GJR> also suggestive of a workflow for science and publication of treatments.

play together or compete?

Discussed

Discussed> New profiles

Decided> to build a suggestions box. @ highest level of Taxon Profile wiki. Details to be kept in child pages of the suggestions box. => PA (to create to suit workflow)

Discussed> workflow for making relationships. Noted the need to open many tabs to make relationships.

Discussed> Simplifying (de skunkifying TPT wiki)

Decided> good idea

GJR wants "data wikis to be visible only to a small group".... and the scripts we are building to become the principle means by which informationa is viewed and changed. => GJR, PA, PH to discuss / plan / trial

Decided> To build a script with functions something like the illustration below. Idea is to provide an interactive view on relationships with a view of currently defined relationships for other terms in the same group. => PA with PH

Nb this gives Term Groups a very specific role. They are a filter to see like terms from other Name Spaces and relationships from WC

nb forgot to include happiness index in this diagram... should be there for relationships and WC definitions.

Relationship_definition_mapping_tool.gif

Discussed> whether terms could be mapped to more than one WC term.

Decided> Sloppy mapping is useful. Terms may be mapped to more than one WC term... as an interim measure to represent terms as they are received.... and then corrected by the creation of a WC term sufficiently broad to be a one to one mapping

Decided> we will need a script which lists terms which map to more than one WC term so that the situation can be remedied. => PA with PH

Discussed> the implied necessity (from above) of a hierachial WC

Decided> this is ok. As we have a separate class of relationships which are suitable for this (internal only relationships)... to be briefly discuss next meeting. => PA wihth PH

Meeting 17/7/2010

Agenda

  1. Review of Progress towards definitions.
    - scripts
    - difficulties encountered
  2. Definitions

Discussed

NOTE> EMC overseas from 19th september

Discussed> Scripts listed at TaxonProfile.Builder/FixTerms and the happiness index.

Decided> Looks great. Well done.

Decided> better use of field placement to reduce white space & thus fit more on screen etc. => PA with PH

Admired> Nifty graphviz based visualisation tool implemented for WC definitions.

Decided> move graph above table in WC terms => PA

Decided> add a Guide/Purpose of Scripts to described in TaxonProfile.Builder/FixTerms => PA

Discussed> Definitions.

Decided> PA to change happiness entry for term that is currently 9 to 5 => PA

Reiterated> Consensus (> 2 people) required for happiness to go above 5. This to be noted in style guide (see below). => PA

Decided> Original definitions to be colour coded. So as to be quickly discerned => PA

Decided> Style guide should be developed on basis of ISO definitions and co. It should be on a page in the root Taxon Profile directory.=> PA

Decided> Style guide should also include notes that definitions need to be written to reflect the context in which they are used and seen with respect to the taxon being described => PA to add note on style guide

Decided> copy original definition if it is a good match and then word smith it to style sheet. (note in style guide?) => PA to add note on style guide

Decided> on entering definitions sourced elsewhere. That we need to explicitly indicate where no definition was given for a term. => PA to add note on style guide
  • blank => not entered
  • "None" => the source work did not define the term
Worked> through definitions for fungi down under

via script https://wiki.trin.org.au/TaxonProfile/Builder/CompareDefinitions?TermGroup=Description&Namespace=FungiDU

Excellent progress. Managed to complete all 10 in an hour. Anticipate it will be even quicker next time.

Decided> that these definitions were now at a stage where they could be shown to a fungi-ologist wink e.g. Heino Lepp ANBG.

Meeting 6/7/10

Agenda

  1. Show and tell
  2. Scripts
  3. definitions.
  4. more…..

Discussed

Discussed> Meeting with Tony Rees and Daniel to talk about the use of TPT for fish information to be contributed to the ALA

spoke about getting information in and out.

Indexed by CAAB number; Tony & Daniel to discuss. Interested. Will require CAAB numbers to be exposed in some sort of webservice.

GJR is very happy with the speed and efficacy with which 2 profiles were created. Only need to make it usable for non-initiates.

Discussed> Taxon Profile Categories

looks great.

Decided> that a line is needed to clarify and define. Possibly include example of what sort of profiles goes into each category. => PA

Discussed> Definitions

Decided> to create a style guide page that can be referenced via a link where-ever definitions and links are made. (Ideally should open in another page). This page should include guidlines as written in the ISO guidelines and others we come up with. And eventually, to be used by other people who are adding content. => PA

Discussed> The definition 'script' page

Decided> That 'Original Definition' to be coded as follows => PA
  • blank = not yet entered.
  • 'None given' = no definition
  • the defintion from the orignal work
Decided> 'Happiness' index to be applied to Relationships. => PA

Discussed> Coding The SPRAT profile

Jeff Trantor = may provide a SPRAT profile schema => PA

Decided>There may be a need for a WC term 'status' to code legal status of weeds, endangered status etc . => PA to consider

Decided>Paul to aim to enter original defintions and preliminary definitions for terms in profiles on file. => PA

Discussed> Insects of Australia

Decided>That it is complex. We would leave describing it for later. As a test case

Discussed> happiness index (as a aworking strategy) and determined a rationing rules for happiness. . => PA

0 = not looked at

1 = looked at but unhappy

5 = Paul is Happy, but needs consensus

Decided > We need a script to sort terms and relationships by happiness, where happiness can be adjusted. => PA

Next meeting Thursday week to look at definitions.

Meeting 2 / 7/ 10

Agenda

  1. Review of past action items
  2. Definition widget. demo
  3. Mapping widget… design
  4. Definitions, Dust, blood and words

Discussed

Discussed> Questions arising from last meeting

Clarification> Metadata to each profile so that each can be roughly categorised as (provisionally) a formal treatment, a technical compilation, a compendium or a general profile. (as detailed in previous meeting) => PA

Clarification> It is not clear if the organisation (e.g. alphabetical, by taxa ) of profiles within a work is useful. Resolved: along with addition of a field for textually describing the coverage of a profile provision to be made for textual description of organisation and other matters in profile metadata (RTF field). => PA

Clarification > Where profiles are nested, and the fields vary with nesting, these are to be described as separate profiles distinguished by the profile name e.g. BirdsOfNewZealand.Family vs BirdsOfNewZealand.Species and so forth. => PA

Deciced> Where profiles are nested, the link to the profile content to be described by a new field (metadata) in the profile which contains a description of the link and a link to the profile i.e. the term ChildProfile in BirdsOfNewZealand.Family contains a link to the BirdsOfNewZealand.Species profile => PA

Similar process should there be a ParentProfile relationship to describe. Just as with other Profile terms these are only used when required as they are part of a profile explicitly or by inference in the layout.. => PA

Discussed> Rain Forest key Profile

Turns out to be a very simple profile indeed. Most of the fields are in access for another purpose

Decided > PA to describe the Rain Forest key profile. => PA

Discussed> Profile Coverage- have we got a sufficient sample for now of profiles?

Decided > Coverage is sufficient for now.

Discussed> Profile Coverage- two additional profiles suggested by Murray Fagg. DEWHA Weeds and SPRAT

Decided > useful and politically smart to describe both. Should also be quick. Details as follows.... => PA

Weeds - DEWHA profile

and SPRAT http://www.environment.gov.au/cgi-tmp/publiclistchanges.b0834c317f3f3b5d3f3f.html

http://www.environment.gov.au/cgi-bin/sprat/public/publicspecies.pl?taxon_id=299

http://www.environment.gov.au/cgi-bin/sprat/public/publicspecies.pl?taxon_id=64877

Discussed > Evidence & meta data for online profiles

Decided > to record screen shot of profile examples (in the manner in which we are doing so for printed profiles) => PA

Decided > PA to write an overview of the ISO standards for Definitions as a Style guide for definitions

Reviewed> Definitions Script tool

Clarification >Tool to contain a the R/O definition of the WC term to which a relationship has been asserted. With a link to the WC term should its definition need to be reworked. => PA

Discussed > Relationships

it may not be always possible to make an exactmatch or a iscontainedin match.

Decided > need to always document a reason why there is not an exact match in comments

Discussed > Scripts and other workflow enhancements.

Decided > Following requried=> PA

(Wiki topics containing) Data Review Scripts

for a profile
  • show a list of terms for a specified relationship
  • list terms which are not an exact match
For a given WC term
  • list terms in other profiles which are an exact match
  • list terms which are not an exact match but is part of
  • list terms which map to more than one wc term?
For a given term (particularly WC) list relationships with other terms
  • exact match
  • near matches

Discussed> Rules for making relationships

Decided> that tho' it would be useful cannot do it at moment and that they will evolve. Best to keep it loose& document later.

Discussed> workflow for iterating through terms

Decided> need an additional meta data element for terms and relationships. Happiness. Ranging from 0-9 i.e. unknown to bad to perfect=> PA

also a script/page which sorts terms and relationships by Happiness. => PA

Discussed> adding provision to describe labels used with terms as well as names (reflecting the common practice where the term is known as something else)

Decided> to wait and see wrt to labels.

Meeting 26/6/10

Agenda

  1. Progress.
  2. Mapping of plant profiles (thanks Murray)
  3. Coverage? What's left on the list? Do we have enough?
  4. Problems that Paul A. has encountered.
  5. Groups
  6. Workflow for tightening up definitions.
  7. Mapping

Discussed

Decided > Rainforest key descriptions to be added. PA to approach frank while he is here (leaving next week). => PA

decided>to add a scanned page of each profile (where possible) and preface (where possible) to each profile described. . => PA

Discussed> issues brought up in the works PA has mapped this week

Decided> to code relationships with nested profile info as meta data.(an example of which would be Dyrandra species profiles which contain details about cultivation of more than one sub-species). Text field = meta data about where a profile sits and what other profiles it may contain. => PA

Discussed> extending metadata attached to Profiles...

Its clear (at least for our own sakes) we should clearly distinguish different classes of taxon profiles. Perhaps this can be metadata. Suggestions that spring to mind are

  • formal taxon treatments - often but maybe not always new taxa
  • Technical compilations (targeted at nomenclaturalists and others in the field) - such as the Flora of Australia/ Marine Benthic Flora etc - principle purpose being to summarise taxonomic information about a group
  • Taxon Compendiums - summaries of taxonomically organised information - compiled for a particular technical ends... Aboriginal food plants, Poisonous Snakes of Australia etc
  • General Species/Taxon Pages - Compendiums targeted at a general audience- recognition, identification, pertinent general facts - often seen in popular works in print and online
I'm attempting to add this type of metadata for each namespace as it surfaces. We can add use/purpose and broadness of taxonomic coverage as characteristics of the profile arrangement .

"taxonomically organised information", others alphabetically, and "Rather than family or genus...arranged by a simple visual key system.".

Decided> to add metadata for both. => PA (To clarify. Metadata will be added to the profile Form? or A new form element called Metadata will be added to the Namespace form PJA to document interprofile relationships. {This is what I was trying to say in the quote above})

Discussed> workflow for defining terms and tightening up mappings.

Decided>

to build a script to implement a form something as follows. It will allow the definition to be progressively refined. Once a mapping is made to a WC term the defintion of the term will be visible on this page (R/O) and help to refine the mapping, the terms definition or indeed (for those who can) invite refinement of the WC term. => PA to liase with PH (The diagram below implies a new Term:Attribute will be created to accommodate the original text from preface, might this be dcterms:abstract )

wc_def_and_mappingn_workflow.gifMeta Data

22nd June 2010

EMC took notes which PA is acting on

17th June 2010

Agenda

Any and all of the following as takes the fancy of the Brains trust

1. More brains trust sorting into groups.
- Paul to mop blood
- whereforto place endemic, habitat, abundance, distribution… seem to belong to a supercategory
- perhaps EMC will know

2. Review of Past actions
- new metadata
- new profiles for the mill?

3. Lets look at definitions….
- For stable term groups, can we refine the definition of relevant WC terms?

Discussed

Discussed> Groups. and terms seemingly that fit in more than one term group

Decided> initially these may go into a special multi-term group that Paul has created. But, ultimately they are to go into only one group. And groups are to be redefined if they are not sufficiently broad. => PA

Decided> to rename 'Distribution' with a suitable definition and name so that it can encompass endemic, habitat, abundance, distribution etc => PA

Discussed> Definitions of terms.

Decided> to use workflow to help. Definitions to be crafted for WC terms by looking at terms in a group and seeking words to encompass them in WC terms => PA to with PH

Discussed> scripting that PA has done with assistance from PH. Looks good

Decided> to add a flag to indicate terms which are new.=> PA with PH

Decided> that pictures, illustrations, and gfx are difficult. Park them in their own category (with notes) for later consideration=> PA

Decided> on lots of term to be moved groups or revised. PA has a detailed list.=> PA
  • Rename TermGroup? WC_Distribution to WC_Extent to be more inclusive
  • Create new TermGroup? WC_Illustrations to include images and image captions
  • Terms are regrouped
  • Notes added to terms that have been discussed. PC_MolecularData to stay in WC_Description TermGroup?

15th June 2010

Agenda


Not necessarily in order.

  1. the 1st order sort continues
    - Paul to adjucate and mop the floor of blood as required
  2. additional proposed metadata 'coverage' or similar to record how a given profile was/may be applied (to what taxa; to what purpose)
  3. discuss next steps in workflow
    - after grouping …. What next?
    WC term relationship mapping + WC term refinement
  4. applicable scripts for #3?

Discussed

Discussed> Rank in Profiles.

Decided> where not explicitly listed in a profile Rank is metadata. And (especially in written works) it may be important to record the implied rank of a profile and the rank of the next treatment 'up'/ containing profile. As it is the context in which the text was written. Rank and phylogeny from nameservices may change as thoughts are updated & not be the original context in which it was compiled.

Action> add metadata to profiles for Rank as discussed above=> PA

Discussed> Hierachy in Wallace Core.

Decided> Hierachy required .. need to be kept as flat/shallow as possible.

Discussed> Adding 'coverage' to profiles as described above

Decided> to add the provision for multiple sentences for 'coverage' as metadata for a profile. to be tidied and systematised as required. => PA

Discussed> Taxon profile group Distribution at great length

Decided> Ants. currently 2 works combined namespace needs to be split . => PA

Decided> Need to consider a Habitat group => PA?

Decided> also may need to add hierachy.. endemic, habitat, abundance, distribution... seem to belong to a supercategory.. => PA

Decided> remove / change / revise fleuristics from Cultivar profile. => PA

10th June 2010

Agenda

1. examing terms which have been placed into groups.

(aka beginning the taxonimic sorting of profile terms sampled to date)

2. Other pressing matters concerning taxon profiles

Discussed

Most of the work in this meeting was done directly into the wiki. A great success. Terms were effortlessly grouped using a script that PaulH? had knocked together.

Discussed> Cultivar Profile.

Decided> Fleuristics is unfamiliar to Jim. Presumably something to do with flowers. possibly a typo => PA to investigate and correct.

Decided> textual definition of phylogeny gp & WC term to use the definition from SPMI for evolution => PA to amend appropiately

Decided> to delete the CERF profile => PA

Erratum> Fungi-image moved into Ecology Group. => PA to tack down & amend.

8th June 2010

Agenda

  • Open discussion
  • driven by examination of current state of term lists

Discussed

Discussed>Plinian core.

Noted the use of the term 'Target Audience' in Plinian Core

Decided> to add target audience as a META DATA field to all profiles & potentially everything else (which ever is most efficient) should be multivalue=>PA to liase with PH

Discussed> Illustrations in profiles. and whether illustrations were content in themselves and allocated a specific term or part of the content that could be in any term. Probably case by case.

Decided> that it would be worth liasing with Cyndy Parr to see how she concieves of the term target audience. => GJR???

proved to be a useful source of terms and concepts not yet mapped. => PA to continue map (2nd priority)

Discussed> Ants. Componutus Discription found in the library. JRC suggested that Camponotus would be worthy of a formal description as part of a TRIN project. => PA

Discussed> Workflow and our loose term Groupings.

Decided> Term groupings were a key ingredient in the workflow.

Decided> terms not prefixed in WC or formal descriptions will be taken to mean their common english forms.

ACTION> as a priority to review existing terms which are ungrouped for discussion NEXT MEETING => PA

ACTION> Need a means to indicate visually whether a wiki page is auto generated or hand editable =>GJR to liase with PH

(could be a visual cue; removal of the edit icon; moving the edit icon ?)

ACTION> alter template to add defintion of groups. ditto in other cases =>GJR to liase with PH

ACTION> clean groups pages to match the editing in the Biology grp page. =>GJR to liase with PH

gjr> more cleaning to come as we de-skunkify.- GJR> removing person last edited from tables and forms - often redundant & misleading.

Discussed> reinstating PH's nifty graphicical graphviz based relationship browser for i) profile relationship definition and ii) wizzard.

2nd June 2010

Agenda

  1. Reviewing what came of Friday (Acacia / FOA)
  2. status of Actions from past workshops
  3. Multiple attribute meltdown.
  4. Discussion of Environmental Attributes & co.
  5. useful Reference and web service.http://bionomenclature-glossary.gbif.org/
  6. ANDS - 'CMD' = Classify my Data Service
  7. Publishing WC

Discussed

Discussed> Multiple attributes (of same type ) for elements

Decided> Not a problem as previously thought. PH will do refactoring to make it possible. it will be useful. More uniform code and easier export.

PH> plan is that attributes will have a prefix which indicate the meta "layer" e.g. term_label, profile_label & such

Action> GJR to prepare draft MongoDB business case by HubRIS meeting 15th June (Monday is a public holiday), when the team will review it and get it ready to be submitted to the TRIN Management Team. Short inc. executive summary & What it will enable. => GJR

Discussed> Coding of relationships in Wiki WC. Currently a table of triples. OK for now. No forseeable problems.

Discussed> Registration of the Wallace Core. GJR suggested potential to register it with the ANDS Classify my Data Service - they are using SKOS.

Decided> Keep a watching brief on their technology and see how they're doing it. The preferred home for the Wallace Core would be somewhere under the auspices of TDWG.

Discussed> Outcomes of yesterdays APA with PA

Discussed> Profiles to map to the Wallace Core

Actions> PA to formally describe new profiles to increase coverage of profiles formally described and mapped to WC => PA
  • Profiles to use:
    • FoA
    • SPMI
    • Plinian Core (see action below)
    • EoL
    • Fungi of Australia
    • Handbook of Australia and NZ whatevers
    • Benthic Flora
    • Snakes of Australia
    • Sharks of Australia
    • Flies
    • Thrips
  • seek also to formally describe a profile for each of the TRIN Taxonomic Research projects
    • Mayflies
    • Mangrove project - which includes profiles for several taxa
    • Ants of Australia
    • Mammals of Australia
Action> PA to subsequently choose 5-10 additional profiles from web and library and prepare assay of coverage. A simple table. * "We are seeking an asymptote." "We'll know coverage is good when very few terms require the addition of a new supercategory to WC" => PA

Action> EMC to encourage PA and remind him regularly that there is life after term mapping => EMC

Discussed> encoding Ecological Relationships.

For example, the Term might be WC:associatedSpecies, with association relationships needing to be coded in a hierarchy e.g.
  • Eats
  • EatenBy?
  • ParasiteOf?
  • ParasitisedBy?
  • GrowsWith?
  • PollinatedBy?
  • etc.
Decided> this was a secondary problem and should be the matter of further research. The approach should be to collect data from a variety of profiles in sentences and derive the relationships from the data rather than imagine possible relationships beforehand and try to coerce the data into the relationships - "let the data drive the specifications". We should also have a discussion of elements in which we can indicate our preferences for relationships e.g. "if unencumbered by data we would do it like this".

Action> engage with Emma (Mangroves) as a model of the problem. Seek solutions driven by her needs and encourage her to use the "best" relationships whatever we all decide these might be. Keep a tally (on wiki) of similar matters for subsequent. => PA

Discussed> ANDS... CMD service. Possibly useful. chances of adoption? Note use of SKOS. Watch developments.

Discussed> Plinian Core


http://www.pliniancore.org/

GBIF has this compilation of related resources:
http://wiki.gbif.org/dadiwiki/wikka.php?wakka=SpeciesDataModel

A search on Plinian Core reveals other resources of various ages,
saying much the same.

It does not look too complicated - the categories are in many cases
broader than ours, and in some cases there are terms that seem to make
sense we may not have encountered yet.

worth a look... no - we really need to account for all of PC if we
are to avoid flak down the line

Action> Add Plinian core to the list of profiles that are to be formally described. => PA
Discussed> Frequency of meetings.

JRC would like to keep things moving. Twice weekly meeting attended by at least GJR and PA => GJR to schedule.

Discussed> direction of the project. JRC wants it to come to a point where content of meeting are only about contentious or interesting mappings into WC and there is no need for notes as app dev and co. are stable.

Discussed> http://bionomenclature-glossary.gbif.org/ its application to WC as a URI 'ground truth'

On Thu, Jun 3, 2010 at 9:08 AM, <Garry.Jolley-Rogers@csiro.au> wrote:
> so we should use it in the TP? Supplemental to Dwc

Short answer, yes... it is one of those TDWGish standards that we will lift from for a wc:term if we can not find a defined term elsewhere...

jim

Discussed> Using TP toolkit as a means to publish primary taxonomy. Steve Donnellan and others may be interested esp in light of changes to the Zoological code (fingers held in scout's salute). => GJR to investigate.

Friday 21st May 2010

Agenda.

  1. prep for EMC
  2. Multiple attribute meltdown.
    1. Multiple Defintions
      • WT?
      • Usage
    2. Options for Implementation of Multiple attributes
  3. Discussion of Environmental Attributes & co.
    1. Project scope.
    2. How much can we and should we record about species associations.
    3. Relevance of EML?
  4. Review of status of Actions from past workshops
      1. Review https://wiki.trin.org.au/TaxonProfile/WorkshopRecordsDec10#Attributes_AN3 and attribute forms
BEER (well only wishful thinking)

Discussed.

EMC's presentation for the CERF conference.

Agenda 18 th May 2010




*Landing pages and Nav in main Taxon Profile web => PA; GJR

* Workflow for adding, describing and reviewing WC


GJR's suggestions...

Understand something like as follows (subject to change etc)

Build a Profile
  1. define terms (===> coming.. facility to pick and choose from Existing ones)
  2. build forms
Forking to two separate (relatively independant tasks)

fork 1. Apply profile (if it is to be used for the compilation of information about taxa)
  • add data;
  • use in a wiki form (a la Plume Moths)
fork2.

Further Mapping of profile to WC
  • less necessary - if previous profiles used a template (need reduced once the facility to pick and choose from WC and friends)
  • essential - if being used to extend WC as a MetaProfile?


* multiple attributes are required for WC terms. => PA; JRC

* see also field to relationships form.

* EMC's prep

* Short Names?

* Claiming WC in semantic world => PA to report on what can be done

Record Tuesday 18/5/2010

Wikipedia Taxon Profiles

Discussed: Wikipedia Community profile projects http://en.wikipedia.org/wiki/Wikipedia:WikiProject_Animals

Decided:
  • Analyse. check that profiles under development will map to WC terms. => PA
  • Monitor projects which are part of the Tree of Life Community Profile Projects => PA; GJR; JRC

Action: TRIN Mangroves. Require mappings of various profiles to be completed and mapped to WC => PA

Consultations

Action: Build a list of domain experts to consult with as WC *& associated applications) matures => PA

  • Helen T.
  • Dave M. / Donald
  • Sensible & tame Zoologists
  • Sensible & tame entomologists- should they exist smile

Attributes

Discussed: Attributes. Arrived at a general position that we should have the capacity for multiple values for attributes on pretty much any WC term. With some reservations about definition. To some extent this need is driven by our data may be coded in RDF and XML

Decided: to dicsuss with PH re options and cost of retrofitting to existing data. => GJR; (DONE. to discuss Friday)

WC in the semantic World

Discussed: 'Claiming WC' in the semantic world => PA; GJR
  • to be modelled on the way that Darwin Core with EXPLANATORY plain language english
Action: create a URL for WC ... on taxonomy.org.au/WC or taxonomy.org.au/WallaceCore. for now an automatic referral to the TRIN wiki WC homepage (which is?)

Name Spaces

Discussed: NameSpaces?

Decided: Add category to distinguish Namespaces made for different purposes => PH; PA

Categories
  • Reference = are not profiles but included because they are refered to
    • e.g. SKOS, DCMI. SKOS, FOAF, DCTerms, WC etc
  • Instance = taxon profiles which have been described especially in the process of building WC OR as the basis of Taxon Profile content in the wiki
    • SPMI is special in that it plays both roles as an instance and reference but is best placed as an Instance
    • FOA; BenthicFlora?; PlumeMoths? etc

Actions: REview namespaces. Clear away junk & remnants (Biology, etc)=> PA

Decided: Terms in a Reference Profile should as a rule have a means to link to the external source. (& if possible directly to the term under consideration)

Decided: Basis if term reflects the thought processes which led to the WC term being created. There can be multiple entries. Entries will be generally on the basis that they are a CLOSEMATCH for the WC term.

Other

Action: Create a search so that instances of TaxaCreator? in the wiki are listed on the taxon profile wiki. => PA

Links

Just a quick documentation of useful links as current - to be reviewed as project matures (of course)

A profile

HubRIS/TaxaCreator

xml of Homo Nitpickerus Minor http://wiki.trin.org.au/HubRIS/HomoNitpickerusMinor?template=TaxonProfile/Builder.XmlView;contenttype=application/xml

Definition of CERF profile


TaxonProfile.Definitions/CERF?

How to specify the contents of a profile (a simple table for the form fields)

TaxonProfile.Definitions/CERF_StandardProfile_Form?

and a template for rendering the form

TaxonProfile.Definitions/CERF_StandardProfile_ViewTemplate?

definition of the term


TaxonProfile/Definitions/CERF_AssociatedWith?

A view of the Attributes of this term

http://wiki.trin.org.au/TaxonProfile/Definitions/CERF_AssociatedWith?raw=on

Wallace Core


TaxonProfile.Definitions/WC

A related term in WC

[[http://wiki.trin.org.au/TaxonProfile/Definitions/WC_AssociatedSpecies?raw=on

Wallace Core Attributes.

TaxonProfile/Definitions.WC_Attributes?===>;;; BAD

TaxonProfile/Builder/TermAttributesOverview

WC: attribute definition


TaxonProfile.Builder/Creator

WC: Relationships with Profiles


TaxonProfile.Relationships/BenthicFloraWallaceCore

TaxonProfile.Relationships/CERFWallaceCore?

WC Short Names.

PH

'The "short name" of each namespace is already encoded as the name of the
topic holding the metdata about a namespace (including the verbose labels).

For example, The http://trin.org.au/TaxonProfile/Definitions/WC topic'

Another Challenge and Parrallel Project

Mapping the varying wikipedia species profiles.... such as A different approach to taxon profiles. Community developed on Wikipedia

Agenda Friday 14/5

Agenda

  1. Review gains and implementation of the past few days
  2. application development-stability
  3. & MAC-applification
    Explaining it to others
    - what do we need to do?
    - any changes nav & UI?
  4. Discuss: Documenting guidelines for application of mappings / relationships.
  5. Discuss: levels of meta ness relationships ..
    Any case for more than one set (GJR thinks NO).
    linked data not - multiplicity and resolvability.. Care?
  6. The Challenge…. Vacellia nilotica.. Worth doing?

Clarifications

Discussed: clarifing the application of names, labels and terms to Attribute and Relationship Data Dictionaries.

Changed heading of columns in Attribute Table

Decided: to add a short explanation to Attributes and relationship Data Dictionary to clarify terminology=> PA

App-ification

Discussed App-ification:
  • Paul H is working on Dashboards for each 'app' being => PH review this list for accuracy
  • TermBuilder?
  • Relationship
  • Terms
  • Attributes
  • Mappings
  • etc
Decided: Review landing pages and Nav in main Taxon Profile web => PA; GJR Monday => Next Meeting.
  • Deeper Hierachy; fewer items on menu Make sure nav Menu is short; add tool tips.
  • links to data dictionary; attributes; terms and builders etc
Check with PH re Tooltips & menu

Workflow

Decided: Workflow for adding, describing and reviewing WC needs to be discussed => Next Meeting.

Re-iterated: links to definitions need to be on Form fields => PH is working on this already in the form template

Discussed: comments and proceedures.

Decided: Need a way to flag decisions, attributes, and mappings that need clarification. Possibly tags; or just a clever search on the wiki => PA to discuss with PH

WC Attributes

Discussed: whether multiple attributes are required for WC terms. Possible through nested forms but needs to resolved before much more work is done => JRC & PA to review ASAP; => Next Meeting.

Relationships

Decided: to add a simple see also field to relationships form. This will include text that will list related; similar or different relationships. It is to be kept simple. and should not require any thing else to implement => PA Monday => Next Meeting.

ALA

Decided: after nav and landing pages cleared up => PA will email donald re the straw-man; Monday

A challenge..Vacellia nilotica

Came across an interesting case study whilst perusing the World Wide Wattle

http://www.worldwidewattle.com/speciesgallery/species-intro.php?id=20756

One taxon..

Acacia nilotica / Vachellia nilotica

many profiles... all different... and complementary

http://www.fao.org/docrep/v5360e/v5360e0b.htm#4.1%20acacia%20nilotica

http://www.fao.org/ag/agp/agpc/doc/gbase/data/pf000124.htm

http://florabase.calm.wa.gov.au/browse/profile/20756

http://ccg.biology.uiowa.edu/jmiller/factsheets/Vachellia/nilotica.htm

http://www.kew.org/plants-fungi/Acacia-nilotica.htm

Can we map it? (into WC)

Record Tuesday 11/5/2010

Present Paul Alexander, Paul Harvey; Jim Croft; Margaret Cawsey & Garry Jolley-Rogers

Reviewed: 'state of the profile' for Margaret.

Discussed: Margaret's upcoming presentation for the CERF Forum. EMC will be coming to seek content, clarification and pretty pictures.

Data Dictionary

Discussed:'Data Dictionary' - some confusion on the application of this terminology.

Action: GJR will address with Paul Alexander in the introductory documentation => PA; GJR

Decided: Need to clarify Column headings in table view of the data dictionary. Group, Term, Label + new column Definition. => PA

Documentation of Proceedures

Decided: PA is to write a guide (on the wiki ) detailing how to make changes to Walllace Core. At both content and meta level i.e. i) how to add content to definitions of WC terms and ii) how to add additional attributes and relationships to WC term framework. Goal being so that Jim can get started adding details. Broader goal to provide a giude for other collaborators. => PA

Decided: Tables of WC attributes and Relationships to be applied to WC terms MUST have the same headings & co. => PA

Forms

Decided: PH to adjust edit template ( when time allows) to add => PH

Discussed: PH's workload. Comment <=== this is dealt with ELSEWHERE see 1, 2 and 3. Action: GJR to write up notes of last Wednesday's meeting with PH and place on 2 => GJR

Resolved: to start using Google Wave 1 again for defects & desiderata

NOTE: prepending a namespace + a colon to a term e.g. FOA:Description will be automatically expanded ( using an interwiki link) to link to the outside namespace (in principle/ eventually etc). Rex Croft's implementation of the Benthic Flora might be useful as a test case if Rex makes the labels available (currently, believed to be implemented as a part of Perl scripts etc)

Actions: PA to report on alignment of Taxon Profile attributes decided upon previously with those seen on the forms. Is there full correspondence now. => PA

Next Meeting to be scheduled. Perhaps coming Thursday.

Record Friday 7/5/2010

Attributes

Decisions: forms to use label for WC attributes
  • labels to be used on form
  • labels to be as Wiki Words. Presented with spaces using a Macro
  • Forms to contain links to full discussion of Attributes.
  • Namespace of attribute to be displayed on FORM
Decisons: Auto generated table of WC Attributes to be sortable by column=> PA

Added Notes column to attributes table => PA

Decision: Add notes column to the table of Relationships applied in WC => PA

Decided to Review: Pete de Vries tweets and URI-burned data on profiles. => PA

Meta & wiki TP Implementation

discussed: 4 levels of Meta.to review next meeting with PH => ***** next meeting?

Require: explanatory notes on how to edit profiles

Decided: Groups to be implemented Simply. NOT as hierarchy, nor as terms. Using a simplified structure. An arbitrary assosciation for presentation. Ordering 'like with like' => PA

Groups

Reworked Canonical form on front page as a basis for groups ... something like Canonical Form of a profile

Relationships

Decided to re-work Relationship table: One row per relationship. Column to indicate whether it is to be applied within profiles, across profiles or both. => PA

  • Relationships BT / NT - JRC 'guide mapping of terms when there is a decision THAT MATTERS'
  • Thesaurus like relationships to be used for disambiguation
  • Too many options for relationships. Need Guidelines => ***** next meeting?

Data Dictionary

Decided Data Dictionary => PA

Decisions
  • should be a table ordered by group
  • columns Group, term, label
Decided: PA to work up term of choice for discussion on Tuesday => PA

Linkages to Other Profiles

Decided: to describe SPM - with a view to encorporating all SPM terms and also to building a WC -> SPM exporter => PA

Record Thursday 6/5/2010

Documentation

GJR working from top down on document; PA working from bottom up.

Require page defining anatomy of a WC term.

on https://wiki.trin.org.au/TaxonProfile/Definitions.WC_Attributes
  • added columns about WC attributes & Relations so allow Definitions AND 'intent and usage'
  • Clarify: Need to clarify WC term attribute rdfs:comment. definition seems appropiate but is confusing nonetheless (for its current usage herein). Is there a better attribute?
  • Add a column to WC attributes table to record references alternatives for WC label attributes
  • need a comments attribute (distinct from rdfs:comment which is for a different purpose = name/label of term)
  • A new attribute WCTermGroup?. contains another WC term. => PA
Decided: Separate provision required to record details of releases for each WC term (as opposed to edit-versions). Technical nature Release strucure to be clarified.

Decided: check terms suggested by DH incorporated in to profile => PA

DH's list
  • Classification (includes Names, Phylogeny)
  • Description (includes Identification)
  • Biology (includes Ecology)
  • Distribution (includes Specimens/Observations, Population)
  • Molecular
  • ....
GJR to respond to Donald

Reviewed Forms

for entering WC TERMS. Name Spaces must always be displayed in labels and forms. => PA

WC terms to be grouped. No more than one level deep. => PA

A new attribute WCTermGroup?. contains another WC term. => PA

Data Dictionary Page

2 forms
  • comprehensive with page per term showing all attributes => PA
  • report form. listing terms & some relevant attributes in tabular form => PA
  • Terms to be grouped by via WCTermGroup? in at least one view of the Data Dictionary. => PA

Record Tuesday 4/5/2010

Discussed documentation of taxon profiles and WC on the wiki

Decisions

work to be divided as follows
  • GJR will use presentations and co. to start high level restructuring of Taxon profile wiki. (Please put your 0.02c in as required)
  • PA will pull together summary documents which describe
    • the Framework used to specify Wallace core terms (aka the logical framework of WC )
      • an overview of attributes & relationships and modus operandi
    • A wallace core data dictionary
      • including framing and orientating text as per DC.
    • To do this Paul will review the workshop notes and places in the wiki where WC is discussed and compile a review of the current form of WC.
Progress to be reported on Thursday.

Record 30/4/2010

Present: Paul Alexander, Garry Jolley-Rogers, Jim Croft

Scribe: GJR

Issues and decision arising from discussion of Profiles in SA benthic Flora

A reading of the prologue in Volume one gives a very simple structure.

SA benthic floras structure includes
  • typification separately encoded
  • Description (in the technical sense a GBIF:circumscription) contains one to three paragraphs which are typographically distinguished by three keywords. These keywords indicate structure that we might later on ( usefully) describe as sub elements of a description
    • Thallus
    • Reproduction
    • Sporophyte (sp?)

General principles

  • Say what the Author said in the prologue clarify only it is unclear or absent.
  • Rex Croft's OCR of the Benthic flora is a NEW WORK distinct in principle from the original published work.
  • Provision should be made in WC for such Distribution maps included by Rex Croft. These need to indicate that these are POST HOC and not part of the original work.
  • Provision needs to be made to place GBIF:typification and species examined in profiles AS WRITTEN in the original work. Otherwise we are re-writting the work. Of course, profiles once digital will be linked in with services such as APNI and AFD which also provide typification details. It is important to preserve what the author wrote.

Canononical Form of a profile

(as listed quickly by JRC - some parts ommitted as GJR's writing illigible)
  • Name
  • typification
  • Synonyms
  • description (in a circumscriptive sense) - we need an alternative word for this as it is overused with different meanings.
  • Specimens Examined
  • Distribution
  • Aftermarket maps & other value adds

Other matters

  • PaulA?> commented that his workspace in the namespaces.
  • decided relationships of a profile to WC encoded in the profile not WC. (such as 'SABF.Thallus' ispartof 'WC.Description' )
  • decided to move notes on structure of TP wiki documentation to TP wiki (DONE)
  • Some comments from JRC on wiki documentaion of TP added to these notes. TaxonProfileWikiDoc
    • Jim would prefer to come up with a logical division of topics - a typical website architecture.
    • Suggested a workshop is held to come up with this architecture.
  • Need to dig out and compare SPM with WC.

Record 29/4/2010

Present: Paul Alexander, Garry Jolley-Rogers

Discussed
  • actions to be taken to clear up documentation and wiki web for Taxon profiles. Notes are in TaxonProfileWikiDoc
  • PA to do-convolve SA Vascular plants and Algea - fruits and infleurescnce etc
  • Examined form of SA benthic flora. Decided that it was really very simple e.g. Thallus, reproduction etc were really just subsets of technical description/circumsription. They are CONTAINED within a description
  • PA to finish description nof benthic flora on wiki
  • talked about types of specimens listings in a taxon profile being type, specimens examined, and co. Decided these are essentially different in function/purpose if not content. and should be treated separatly
  • Decided: If a given profile has a term which we think is contained within an element of WC (but not equivalent) e.g. Thallus. We will use the relation 'Contains'
  • Decided: WC.term 'Contains' means that the term is wholly within the term. i.e. the WC term is a superset.
  • Decided: use of element.name; element.source; element.value to be deprecated.
  • Undecided: whether 'ispartof' is requried. or whether it is exactly complementary to 'Contains'. If so, we will discard it. However, it may be useful in another context.
  • Next: assess what work is required to get a polished process for defining WC terms? what work is required to document this process

Record 3/03/2010

Present: Jim Croft, Paul Alexander, Garry Jolley-Rogers

Summary

Actions
  • Annotate Attibute list below with explicatory comments as to how each attribute is to be used. ( Paul Alexander)
  • Add Attributes to current wiki forms ( Paul Alexander)
  • Document relationships between WWW & FOAF in wiki ( Paul Alexander)
  • Implementation of relation operators ( Paul Alexander, GarryJR , Paul Harvey to discuss best means)
  • Next waypoint (just after Forum)?

FOAF and World Wide Wattle

Compared contents and structure.

near 100% correspondence. Excepting that labels are different and Presentation. Need to map any significant differences in labels. Not significant here. Nonetheless, worth mapping for acacia project and as a 1st exercise

General Principle : for the moment NOT to go for a complete survey of extant profiles. Rather look for adequacy of coverage of WC terms for profiles of interest. An adequate survey may be something we aim to achieve later on.

Notes

  • Our expectation is that profiles for Higher level taxon should match Species Profiles.
  • Nomenclatural entries in extant profiles to be documented BUT not to be implemented in Wallace Core.
  • Similarly for dichotomous keys, Name References, and protologues.
  • Species examined is to be treated as a block of text because it should reflect what the researcher wrote.
  • Essential principle is that WC should not go into any depth specifying things which are handled elsewhere especially services exist (APNI / APC/ APII/ Identify Life etc) e.g. If the nomenclature or the synonymy differs the authors should correct it in APNI
  • Explicitly make DC.title and rdfs.label synonymous so as to make it easier to treat WC terms as resources
  • List below is close to the order of presentation
  • All are attributes are mandatory - if Left blank taken to be a null
  • ALL terms are Resources

Attributes

Attributes

dc.source replaces BasisOfTerm?

References

Attributes will be drawn from Dublin Core

Primary source to be rfds definitions of DC terms

A second property with the same name as this property has been declared in the dcterms: namespace (http://purl.org/dc/terms/). See the Introduction to the document "DCMI Metadata Terms" (http://dublincore.org/documents/dcmi-terms/) for an explanation.

</SKOS:note>

</rdf:Property>

Attribute Intent and usege Example Content Source
rdfs:label What is the identifier of this resource Title http://dublincore.org/2008/01/14/dcelements.rdf#
rdfs:comment definition and explicatory text "A name given to the resource." http://dublincore.org/2008/01/14/dcelements.rdf#
rdfs:isDefinedBy URI rdf:resource="http://purl.org/dc/elements/1.1/" http://dublincore.org/2008/01/14/dcelements.rdf#
dc.source One to many references for why this term came to b Created on basis of entry in the Fllora of Australia profile http://dublincore.org/documents/dcmi-terms/
dcterms:issued   1999-07-02 http://dublincore.org/2008/01/14/dcelements.rdf#
dcterms:modified   2008-01-14 http://dublincore.org/2008/01/14/dcelements.rdf#
rdf:type rdf:resource   "http://www.w3.org/1999/02/22-rdf-syntax-ns#Property" http://dublincore.org/2008/01/14/dcelements.rdf#
dcterms:hasVersion   rdf:resource="http://dublincore.org/usage/terms/history/#title-006" http://dublincore.org/2008/01/14/dcelements.rdf#
SKOS:note   A second property with the same name as this property has been declared in the dcterms: namespace (http://purl.org/dc/terms/). See the Introduction to the document "DCMI Metadata Terms" (http://dublincore.org/documents/dcmi-terms/) for an explanation. http://dublincore.org/2008/01/14/dcelements.rdf#
Term Name: contributor     http://dublincore.org/documents/dcmi-terms/
Term Name: creator     http://dublincore.org/documents/dcmi-terms/
Term Name: coverage     http://dublincore.org/documents/dcmi-terms/
Title A name given to the resource. 1999-07-02 2008-01-14 − A second property with the same name as this property has been declared in the dcterms: namespace (http://purl.org/dc/terms/). See the Introduction to the document "DCMI Metadata Terms" (http://dublincore.org/documents/dcmi-terms/) for an explanation.

An example of original rdfs as follows


Title
A name given to the resource.

1999-07-02
2008-01-14



A second property with the same name as this property has been declared in the dcterms: namespace (http://purl.org/dc/terms/). See the Introduction to the document "DCMI Metadata Terms" (http://dublincore.org/documents/dcmi-terms/) for an explanation.

Discarded DCMI Terms

Term Name: date, Term Name: description, Term Name: format, Term Name: identifier, Term Name: language, Term Name: publisher, Term Name: relation, Term Name: rights, Term Name: source, Term Name: subject, Term Name: title, Term Name: type

Record of Taxon Profile Workshop 25/02/2010

Present: Jim Croft, Paul Alexander, Margaret Cawsey, Garry Jolley-Rogers, Paul Harvey

Summary

Draft relationship set

DC:Resource every term component of a taxon profile is a DC:Resource
Internal, between WC terms (internal links within a concept scheme) Both External, between WC. and Other namespace terms (mapping links between concept schemes) Meaning
- - DC:source (Not WC:termBasis any more)
SKOS:related - SKOS:relatedMatch The property SKOS:relatedMatch is used to state an associative mapping link between two concepts.
SKOS:broader - SKOS:broadMatch The properties SKOS:broadMatch and SKOS:narrowMatch are used to state a hierarchical mapping link between two concepts.
SKOS:narrower - SKOS:narrowMatch The properties SKOS:broadMatch and SKOS:narrowMatch are used to state a hierarchical mapping link between two concepts.
- - SKOS:exactMatch The property SKOS:exactMatch is used to link two concepts, indicating a high degree of confidence that the concepts can be used interchangeably across a wide range of information retrieval applications. SKOS:exactMatch is a transitive property, and is a sub-property of SKOS:closeMatch.
SKOS:exactMatch is disjoint with each of the properties SKOS:broadMatch, SKOS:narrowMatch and SKOS:relatedMatch.
- - SKOS:closematch The property SKOS:closeMatch is used to link two concepts that are sufficiently similar that they can be used interchangeably in some information retrieval applications. In order to avoid the possibility of "compound errors" when combining mappings across more than two concept schemes, SKOS:closeMatch is not declared to be a transitive property.
- DC:isPartOf Contains DC:isPartOf -> A related resource in which the described resource is physically or logically included.
- DC:hasPart isContainedBy DC:hasPart -> A related resource that is included either physically or logically in the described resource.
- - OWL2:differentFrom (to distinguish confusing terms)
Actions

  • Paul to extract equivalent relationships for the RDFS namespace and see if they are better that the SKOS ones we have chosen (see above)
  • Paul to define the relationships between Flora of Australia and WWWattle
  • Paul to make a start on the list of attribute terms which will describe each WC.Term e.g.

Record of Taxon Profile Workshop 18/02/2010

Present: Jim Croft, Paul Alexander, Margaret Cawsey, Garry Jolley-Rogers, Paul Harvey

Summary

  1. The Wallace Core (WC) is our 'workbench' i.e. Reference Profile. All terms which end up in the reference profile, regardless of where we got them i.e. namespace of origin, will end up in the WC (ha, ha, joke).
  2. The WC is thus the equivalent of a data dictionary of Taxon Profile terms
  3. We should set out the descriptive framework for each WC term in a similar manner as has been done for the Darwin Core
  4. Leave the Type field blank for now, unless we really know what it is
  5. Basic Attributes are described on the photocopy of the whiteboard work, and that is listed here.

  • Name (Ours) *
  • Description, definition (Ours) *
  • Term basis (≥ 1)
  • Relationship(s) to other terms in this namespace (≥ 0)
  • Relationship(s) to other terms in other namespace(s) (≥ 0)
  • Domain/ Range/Values
  • Format
  • -
  • -
  • -
  • Comments (≥ 0) (? = notes)
  • Discussion
  • (Metadata for this record - Date, Contributions, etc.)

Actions

All by next Thursday 25th Feb 2010
  • All FoA? terms accepted in the first pass to be placed into the WC framework (see points 3-5 above) with the appropriate metadata e.g. name, definition, termBasis
  • The appropriate syntax for describing relationships to be determined e.g. is it WC.Identification OWL.sameAs FoA?.Identification or something else?
  • Relationships between these (now) WC terms and the World Wide Wattle profile terms to be placed into the framework; the namespace for World Wide Wattle is currently set at either 3W or W3, whichever Paul A prefers
  • The Pauls and Garry to brainstorm the kinds of data structures which will be required to manage this framework
  • Paul A. to nail down a starting list of preferred relationships in the appropriate syntax form e.g. they might be something like
    • SKOS.closeMatch
    • SKOS.broader
    • SKOS.narrower
    • SKOS.exactMatch or OWL.sameAs
    • RDF.seeAlso
    • RDF.comment
    • RDF.domain
    • etc.?

Record of Taxon Profile Workshop 11/02/2010

Present: Jim Croft, Paul Alexander, Margaret Cawsey, Garry Jolley-Rogers

Summary

  1. We have our own Namespace: the Wallace Core, or WC for abbreviation. This will provide the basis for a generic Reference Profile.
  2. We will build the Reference Profile by selecting elements/terms from other namespaces, choosing each term on the basis of our own decision that we like it and its definition e.g. WC.Habitat may well be selected from the Flora of Australia list of terms
  3. We would then define the relationships between the WC Reference Profile element WC.Habitat and other related terms in other namespaces e.g. SameAs FoA.Habitat, SimilarTo TDWGSPM.Distribution or whatever
  4. We continue to adopt terms from other namespaces as they become relevant to the taxon profiles our collaborators are interested in building, mapping the relationships between each term in the Wallace Core Reference Profile to related terms in other namespaces.
  5. If we can't find a term we like in any namespace, we make one up and map its relationships as per point 4. above.
  6. Thus, the Wallace Core will provide a reference point for all other taxon profile namespaces
  7. Because no other namespace has everything we want in it.

Goals of the Taxon Profile Project

  1. To represent (all) existing Species Profile Models
  2. To provide the capacity to represent any new SPM
  3. Improve searchability across taxon profile models

Actions

By next Thursday 18/02/2010, Paul Alexander to mock up a particular WC reference profile element
  • Put a definition on it
  • Cherry-pick the relationships we think will be the most informative, regardless of whether they are regarded, sensu stricto, as 'relationships' within the originating namespace
  • Map the term to terms in other common profiles
  • Make up our own relationships
  • Find out whether we've invented any relationships; if one turns out to be the same as something else, in some other namespace, make ours the same as theirs
  • Don't have any confusing redundancy.

Record of Taxon Profile Workshop 16/07/2009

Present: Jim Croft, Paul Alexander, Margaret Cawsey, Garry Jolley-Rogers

taxonprofilemodel.jpg

Definitions

Taxon Profile definitions

A taxon profile is the structure of a description of distinguishing traits of a defined taxon (e.g. Family, Genus, Species), which includes information as to the who, what and where of this taxon.

Resource

A resource is the source of a particular profile, or an instance of a taxon profile class, e.g. a species page providing a profile of a particular taxon e.g. the Shovel-nosed Snipe. A resource can be anything that might be identified as such, e.g. a book, a species page, an electronic document, the Flora of Australia, a website, a service (DCMI).

Element

An element is a property of a specific aspect, characteristic or relation used to describe resources.

Class

A class is a category of elements things e.g. Locality is a class of Locality elements in the Darwin Core set.

Design Rule #1

We will choose our elements such that each belongs to only one class. We will revisit this if we hit the wall somewhere as the design of the Taxon Profile structure proceeds.

Attribute

Elements are defined by attributes , which include Dublin Core elements such as: dc.Name dc.Label dc.URI dc.Definition dc.TypeOfTerm dc.Comments dc.NarrowerThan etc.

Representations

Elements may be described through a variety of representations in the way the information for that element may be presented. There may be more than one representation of information for any given element. \x{2003}

Example

Class: Spatial Distribution

Definition:

Where we find it in space and time. dc:Coverage The spatial or temporal topic of the resource, the spatial applicability of the resource, or the jurisdiction under which the resource is relevant. Note: Spatial topic and spatial applicability may be a named place or a location specified by its geographic coordinates. Temporal topic may be a named period, date, or date range. A jurisdiction may be a named administrative entity or a geographic place to which the resource applies. Recommended best practice is to use a controlled vocabulary such as the Thesaurus of Geographic Names [TGN]. Where appropriate, named places or time periods can be used in preference to numeric identifiers such as sets of coordinates or date ranges.

See also:

i.e. does NOT include: biophysical, ecological, climatic, seasonal distribution

Elements

Geographic Extent

Definition:

Where we find the taxon in horizontal space: An equivalent schema would be Darwin Core Geospatial Extension schema>
Representations:

GeographicalExtent?.Text; Definition: dcTermsDescription? An account of the resource. Note: Description may include but is not limited to: an abstract, a table of contents, a graphical representation, or a free-text account of the resource.

GeographicalExtent?.Box; Definition: dcTermsBox The set of regions in space defined by their geographic coordinates according to the DCMI Box Encoding Scheme.

GeographicalExtent?.Image; Definition:

GeographicalExtent?.GISfile Definition:

Vertical Extent

Definition:

Where we find the taxon in vertical space
Representations:

VerticalExtent?.Text; Definition:

VerticalExtent?.Range; Definition:

Temporal Extent

Definition:

Where we find the taxon in time
Representations:

TemporalExtent?.Text; Definition:

DateRange?; Definition: dcTermsPeriod? The set of time intervals defined by their limits according to the DCMI Period Encoding Scheme.

Record of Taxon Profile Workshop 4/08/2009

Present: Jim Croft, Paul Alexander, Margaret Cawsey, Garry Jolley-Rogers

Framework

Taxon Profile Framework

What does Jim perceive to be a community need (Paul A's fundamental niche)
Agile approach

Testing Process

All the following points feed into one another

  • Recommend vocab for taxon Profile
  • Terminology mapped across taxa (i.e. accommodating alternative vocabularies)
  • I.e. suitable at a broad level for all taxa
  • Templates for using the recommended vocab to produce cute species pages
  • Suite of taxon forms based on the vocab

Community needs

  1. A way to document existing profiles
  2. Record new ones
  3. Place to store and render
  4. So need standards - structure and recommendations

Principles

  • Standards
  • Reuse (not reinvent)
  • Explicit (defined well)
  • Framework is DCMI
  • Dealing with broad blocks of text rather than codified, atomised pot points
Focus of domain is:
  1. Fact sheets
  2. Floras
  3. Interactive keys
  4. Species pages
Focus of domain is not:
  1. atomised descriptive data
  2. the key matrix


I Attachment Action Size Date Who Comment
03082010.jpgjpg 03082010.jpg manage 191 K 2010-08-18 - 14:30 GarryJolleyRogers  
17082010001.jpgjpg 17082010001.jpg manage 152 K 2010-08-18 - 14:29 GarryJolleyRogers  
Profile1_Geocode_Relationship.gifgif Profile1_Geocode_Relationship.gif manage 12 K 2010-10-05 - 18:15 GarryJolleyRogers  
Profile2_location_Text_Relationship.gifgif Profile2_location_Text_Relationship.gif manage 11 K 2010-10-06 - 09:39 GarryJolleyRogers  
Profile3_Distribution.gifgif Profile3_Distribution.gif manage 12 K 2010-10-06 - 09:41 GarryJolleyRogers  
Profile3_Distribution_.gifgif Profile3_Distribution_.gif manage 12 K 2010-10-06 - 09:42 GarryJolleyRogers  
Relationship_axes.gifgif Relationship_axes.gif manage 13 K 2010-10-05 - 17:58 GarryJolleyRogers  
Relationship_definition_mapping_tool.gifgif Relationship_definition_mapping_tool.gif manage 32 K 2010-07-21 - 09:33 GarryJolleyRogers  
TaxonProfileModel.gifgif TaxonProfileModel.gif manage 52 K 2009-10-14 - 14:17 GarryJolleyRogers  
TaxonProfileModel.jpgjpg TaxonProfileModel.jpg manage 84 K 2009-07-17 - 13:03 GarryJolleyRogers taxon profile representation
directedgraphplugin-1.pngpng directedgraphplugin-1.png manage 71 K 2010-12-09 - 10:57 GarryJolleyRogers  
directedgraphplugin-1.svgsvg directedgraphplugin-1.svg manage 17 K 2010-12-09 - 10:50 GarryJolleyRogers  
igp_250_TaxonProfileModel.gifgif igp_250_TaxonProfileModel.gif manage 13 K 2009-08-22 - 15:42 UnknownUser  
wc_def_and_mappingn_workflow.gifgif wc_def_and_mappingn_workflow.gif manage 19 K 2010-06-25 - 17:32 GarryJolleyRogers