HISCOM 2008 AGM Fremantle

The 2008 HISCOM Annual General Meeting was held in Fremantle from 09:00 – 17:00 Saturday 25 October 2008. This document represents the minutes of the meeting.

Venue
Room ND 36.117, Notre Dame University Education Centre, Mouat St Fremantle. See TDWG Google Map.

Attendees

 * Alex Chapman, Western Australian Herbarium, Department of Environment and Conservation
 * Alison Vaughan, National Herbarium of Victoria, Royal Botanic Gardens
 * Ben Richardson, Western Australian Herbarium, Department of Environment and Conservation
 * Bill Barker, State Herbarium of South Australia, Department of Environment and Heritage
 * Donna Lewis, Northern Territory Herbarium, Department of Natural Resources, Environment the Arts and Sport
 * Greg Whitbread, Australian National Herbarium, Department of Environment, Water, Heritage and the Arts
 * Jim Croft, Australian National Herbarium, Department of Environment, Water, Heritage and the Arts
 * Karen Wilson, Royal Botanic Gardens, Department of Environment and Climate Change, New South Wales
 * Margaret Cawsey (invited, Fauna Collections Informatics Group, Australian National Wildlife Collection, Canberra)
 * Dave Martin (invited, Atlas of Living Australia)

Apologies

 * Peter Bostock, BRI
 * Aaron Wilton, CHR
 * Laurence Paine, HO
 * Paul Coddington, eResearchSA
 * Brett Summerell, NSW

FCIG
Margaret Cawsey, representing the Fauna Collections Informatics Group (FCIG). Margaret works at the National Wildlife Collection in Canberra.


 * Introduction to FCIG
 * It was agreed that HISCOM and FCIG would have reciprocal representation at each other's meetings (both face to face meetings and teleconferences)
 * Each will extend an invitation to the other
 * OZCAM as a product of CHAFC
 * FCIG have agreed that members will deliver their data only through OZCAM
 * Entomology collection may serve their own data as well
 * Maintain data in a common cache
 * Data will be available through both OZCAM and ALA
 * OZCAM will serve and resolve their LSIDs
 * Federated approach to sensitive data. FCIG is interested in HISCOM's approach to dealing with sensitive data. HISCOM to discuss this issue separately and report back to FCIG.
 * Management of images - follow TDWG standards
 * Margaret alerted HISCOM to a study where specimen labels were photographed and sent away to be triple keyed, then checked for correctness. The study showed that this method results in more accurate transcription of label data, for the same cost as having labels keyed once, locally. Several people pointed out that in-house databasing staff do a lot more than just transcribing specimen labels (value-adding, interpretation, error detection etc.), which needs to be factored in to the analysis.
 * FCIG have noted that herbaria are underrepresented in the Museums Australia network.
 * Piers Higgs is providing a short review of portal technology, and offered to pass this information on to HISCOM.
 * FCIG has made a recommendation for CHAH/CHAFC draft a memo to the Minister for the Environment, Heritage and the Arts about the importance of taxonomy, names etc.

Action: Bill to provide Margaret with CHAH AVH MoU (Bill)

Action: HISCOM to discuss treatment of sensitive data and report back to FCIG (All)

Recommendation: That CHAH consider raising the profile of herbaria in the Museums Australia network.

Action: FCIG to send Piers' review to HISCOM (Margaret Cawsey)

Recommendation: That CHAH provide the report from the November AVH meeting to FCIG and CHAFC

ALA Project Team
Dave Martin attended HISCOM, representing ALA.


 * Relationships. What is the relationship between AVH and ALA?
 * AVH and AVH data as a testbed for ALA infrastructure functionality
 * Agreed.
 * Principles. Can we define some simple, high-level principles by which the relationship should work?
 * ALA Tools. That any tools ALA develops should be designed for use as close as possible to the data provider, because we know that the best way to prevent errors is to perform validation upon data entry, rather than later.
 * Agreed, in principle. Examples include local implementations of validation tools, such as GeoLocate, BioGeomancer, Bioclim, GARP, etc.
 * AVH formats for ALA. That any formats chosen for ALA should be the same as those chosen for other high-profile sites such as EOL and GBIF to minimise the cost to each institution for being a part of the network.
 * Agreed. See the ALA data standards page.  Use existing formats.
 * One of the recommendations on this site is that botanical data sets should use HISPID5 and make the data accessible using the PyWrapper implementation of TAPIR. This recommendation caused some concern among HISCOM, as the Darwin meeting (2007) concluded that we should continue with the installation of BioCASE in those herbaria that do not already have it functional.
 * AVH Mapper. Future of the AVH Mapper in light of ALA.  There will be a continued business need for AVH to meet the needs of the herbarium and botanical communities

Action: HISCOM to invite ALA to participate in HISCOM meetings and teleconferences (HISCOM Chair)

eFlora

 * Projects (all)
 * Flora of Australia on-line
 * Marine Benthic Flora of South Australia
 * e-Flora of South Australia
 * Flora of New South Wales
 * Flora of the Sydney Region
 * FloraBase -- the Western Australian Flora
 * etc.
 * Schemas, standards (all)
 * ABRS (i.e. Helen) offered to assess and map of existing ontology to those agreed at the Adelaide e-Flora's workshop, working with Greg and Jim.
 * Discussion on how this relates to the AVH
 * An implementation (lite?) of SPM, working with TDWG
 * Working with GBIF, TDWG, EoL, GISN, Plazi, etc.
 * LSIDs and (APNI ID in particular) seen as a critical linking element
 * Use the HISPID development process as an operational model
 * (Aside: is SPM a use case of SDD?)

Action: HISCOM members to check draft schema against their own flora products to test what components be can accommodated. Members to report back to HISCOM by next teleconference (to be scheduled after the November meeting). (All)

Action: ABRS to evaluate draft schema against various TDWG and other comparable ontologies (Helen Thompson)

LSIDs

 * urn:lsid: is a new protocol and thus is not automatically available in most domains. DNS specialists are required to make them work as described in the LSID specification, and this has resulted in many in the biodiversity informatics community using a HTTP equivalent (a so called "proxy") instead.
 * better resolved as http://.../urn:lsid:..
 * will not have to create a new protocol
 * TDWG has created a resolver
 * e.g. http://lsid.tdwg.org/urn:lsid:authority:namespace:identifier, see a working example displaying data for Euphrasia collina
 * Proxy for lsid GET METADATA request
 * Will return an RDF fragment
 * Not implementing GET DATA request
 * See extensive discussion on TDWG wiki
 * can be applied to: specimens, names, taxa, descriptions, Images
 * see Canberra workshop report

Action: HISCOM members who understand the issues with LSID implementation to put some practical examples on the HISCOM wiki (HISCOM people who know stuff)

Action: Refer DNS resolver issues to TDWG working group and the ALA (HISCOM people who know stuff)

Action: Approach ALA to present a simplified explanation and use case of LSIDs in the AVH situation (Who wants to do this?)

AVH3 release and launch programmes
AVH Update from Paul Coddington (delivered by Bill): Paul believes that the latest version of AVH has moved on to the QA phase. The synchronisation of user registration on multiple sites remains to be done, but it is not urgent given that we have one working site. The code has been released on SourceForge. The user guide is built in, but there is no documentation (still to be written). Shunde was making the help pages dynamic (this should be done by now); they will be updated dynamically.

Version 3.1.0 is the most recent version at time of writing and has been released on SourceForge.

AVH Help Pages
The help text used in AVH 3 is being developed in the HISCOM wiki, see AVH Public Help Page and AVH Restricted Help Page. The public help text will be a subset of the restricted set. The reasoning for this is that it ensures that all HISCOM members have equal access to modifying the help text. The text from these two wiki pages will be pulled from the wiki into AVH 3 each time a formal release of the software is made, and will thus only change when an AVH node updates their AVH software to a new version (is this correct?).

Text for some form fields still be to be updated, for example, the following items. Any other item in the two help text wiki pages highlighted in  red  are also in need of work.


 * Country
 * State
 * Botanical Region

Some questions remain with this approach:


 * How to deliver data in a canonical form?
 * How to manage unambiguous delivery of the canonical form?
 * Option for State plus 'variants' - and be explicit in the help pages
 * Cultivated field - not all herbaria delivering cultivated data records

Action: HISCOM members to review the information on the help pages and to update information relating to their data as needed. (All)

Action: Bill to contact Paul Coddington about guidelines/instructions to manage/edit wiki help page content. (Bill)

AVH Data Inconsistencies
A long discussion ensued about how or if to handle data inconsistencies so that users see a better product without the need for too much churn in the data being provided from participating herbaria. Jim felt strongly that the AVH should make no attempt to change the underlying data for the user, which was generally agreed, but all other members agreed that the user interface should make some allowance for inconsistency in the underlying data. A use case was put forward for the location of Norfolk Island. There is some disagreement over whether it is a country, or a state or territory or region of Australia.

A second use case was put forward for country names that have changed over time. The terms Papua New Guinea, New Guinea, and a number of other historical names all exist in the databases of participating herbaria. A user of AVH might reasonably expect that if they request a set of records from Papua New Guinea, AVH will retrieve the records containing older, synonymous, names for the country also, without changing the underlying data &mdash; a critical point, because the underlying data is not in error &mdash; specimen data collected in 1950 in what we now know as Papua New Guinea will not necessarily contain the term "Papua New Guinea".

This requires a table of country name synonyms to be implemented in AVH 3. Other fields with a similar requirement will need to be treated in the same manner. The agreed outcome was to create the table of synonyms.

Action: A table of country name synonyms needs to be created and handed to eResearch SA, who is doing this?

AVH 3 Interface Design

 * Code to be made more modular - separate programming from design
 * Ben to modify interface layout and operations
 * Ben not up to speed with AVH code and will liaise with Gerson for some changes (eResearch SA)
 * Basic principle to separate the design from the programming
 * Code to be updated to allow a new AVH provider to be added
 * AVH design consideration can be found on AVH Interface Design

Action: HISCOM to decide what needs to be changed about the design. Comments to be made on AVH web pages - issues, suggested actions (All)

Action: Establish wiki page and populate with design requirements (2 weeks) (Ben, Jim)

Data harvest and cache model

 * General principle of replicating the AVH cache to return to the distributed architecture
 * Work required to get all providers working
 * Work required to achieve compliance
 * to avoid repeated data handling and fixing with each load
 * need a measure of the class and scale of the problem
 * What is a reasonable refresh time?
 * Aspire to two weeks? Reality might be closer to 3 months
 * Might have to flatten out the exchange format?

Action: Bill to request that Paul Coddington and Rex Croft report on scale of load problem and provide a summary of the non-compliance issues that need to be fixed at each herbarium (Paul, Rex)

Action: Participants of HISPID Mapping to ABCD to look at HISPID 5 and update documentation (Ben, Bill, Rex, Greg)

Early Warning System and Weed Tracker

 * QA version by end December
 * Wiki page established at Early Warning System

AVH November meeting

 * One day joint CHAH/HISCOM meeting in Melbourne
 * Evaluation of AVH to be considered
 * What will be the direction for AVH 4.0?
 * APNI/APC as a backbone
 * How to fix non-compliance issues?
 * TDWG ideas
 * Extension of current functionality
 * Need to bring CHAH up to speed - briefing paper?

Action: HISCOM to prepare a briefing paper for CHAH for November meeting AVH Briefing Paper November 2008 (All)

APNI/APC
(Greg, Jim)


 * Integration with AVH
 * 'Union concept' fundamentally flawed
 * Crude approach
 * Need to document ambiguity and allow user to follow desired route
 * Use APNI/APC for disambiguation
 * Algorithms to 'fall back' through the literature
 * TAPIR provider being divided
 * Effort needed on APNI/APC content
 * Interface needs to be redeveloped
 * Need for a central (accepted) names server for all Australian organisms
 * EDIT WP5 has a very similar data model
 * Common Data Model (CDM) simple domain model based on TDWG ontologies
 * See http://dev.e-taxonomy.eu/trac/wiki/CommonDataModel
 * ALA needs a name service from APNI/APC

Action: Establish a nomenclatural disambiguation discussion wiki: Nomenclatural Disambiguation (Done)

Pragmatism is the new Standard
Over the last week at TDWG we've been hearing a lot about the trouble people are having, either in transferring the amount of data required to build our networks or in the usage of the tools and standards, particularly BioCASE and ABCD. A proposal has been put forward by GBIF to use delimited files to transfer data because this reduces the amount of data that traverses the network to the bare minimum. It makes network transfer (in place of postal hard disk transfer) a viable option. What, if anything, do we want to do in light of this? Does it suggest a better way forward for AVH?

Action: Participants to update this section with their impressions of the TDWG meeting (All who attended TDWG)

Other Business

 * Electronic house journals (Bill)
 * The potential for taking a joint approach to the production of electronic house journals was noted but not discussed
 * CHAH website and wiki
 * CHAH already has its own domain http://www.chah.gov.au/
 * Will be costs involved in moving, reconfiguring
 * Already under CHAH/HISCOM control
 * Preferable solution will be to have links from http://www.taxonomy.org.au to CHAH
 * HISCOM Chairperson
 * HISCOM Members welcomed Brett as the new chair of HISCOM, in absentia
 * Meetings for 2008/9
 * Supported quarterly teleconferences
 * Supported attempting HISCOM meeting following CHAH/HISCOM AVH meeting

Action: Inform Brett and HISCOM members about proposed meeting (Ben)

Action: Check availability of facilities at MEL (Alison)

Action: HISCOM to advise CHAH on domain name issues(?Ben)'

Close: 17:00
Well, sort of...