Ecosystem Approach Community of Practice: VRE planning
From IMarine Wiki
Virtual Research Environments (VREs) are web based working environments tailored to serve the needs of a specific eScience scenario.
The iMarine project is called to deploy and operate a number of VREs to satisfy the needs arising in selected scenarios identified by the Ecosystem Approach Community of Practice. The plan governing this VREs operation activity is documented in a dedicated wiki page.
Such a plan mainly results from the needs and desiderata with respect to VREs that the Ecosystem Approach CoP documented by this page.
VREs Specification Procedure
The specification of a VRE mainly consists in a characterisation of the expected service in terms of data/datasets to be made available, facilities to be offered, and qualitative aspects to be guaranteed. In addition to that the service characterization should include the target scenario it is conceived for.
Such a characterisation is expected to be produced by the CoP, however it requires an assessment before being submitted to the project Consortium for the actual implementation of the service. This assessment consists of a review procedure having three phases:
- iMarine Board Review: the members of this board are called to assess the technological and functional 'added value' that will be provided by any proposed VRE. This review procedure applies to all types of VRE's, services and resources identified in the project that will be exposed to the CoP;
- PEB Review: the Project Executive Board is called to evaluate the feasibility of the proposed VRE as well as to perform a cost-benefit analysis of the proposal;
- Consortium Review: the project members involved in the implementation of the VRE are called to evaluate the feasibility of the proposed VRE implementation and to agree on a deployment schedule;
If all the three review phases are passed, the project agrees to implement the identified VRE and start working on it.
Several VRE's are already used, while others are being implemented. Throughout the project, VRE's will be built, sometimes 'on top of' other VRE's. This VRE planning page is a living page, where the several stages of VRE development are tracked. These go from inception (1), proposal (2), review (3), implementation (4), to delivery (5).
After delivery, the functioning and use of the VRE will be subject to validation, for which a dedicated wiki page is available.
The delivered VRE's have been validated:
- FCPPS (5) The reporting tool;
- ICIS (5) The Timeseries management environment;
- AquaMaps (5) The species predictive modeling environment;
- VTI (5) The Vessel Transmitted Information management and modeling environment;
- Trendylyzer (5) Analytical tool, with a similar concept to Trendalyzer / gapminder for data manager to discover long-term trends in TimeSeries.
- SPREAD (5) Building on a subset of ICIS capabilities, SPREAD will add geospatial reallocation features ;
- FishFinderVRE (5) Reporting tool and work flow environment based on FCPPS to facilitate the collation and editing of Species Fact Sheets;
- FishFinderOffLine (5) Reporting tool based on FishFinder for authors to compile Species Fact Sheets.
- SpeciesNameFinder (5) Species name disambiguation tool. And nothing more than that;
- VME DB (5) The Vulnerable Marine Ecosystem database will contain VME Factsheets. These factsheets relate to services provided in FCPPS for document workflow and content management, and will be published using the FiMES schema. The use cases and requirements are completed (August2014);
- TaxonReconciliation (5) Using iMarine and FAO technologies, this service calculates distances between objects, often using lexical matchers. A first candidate is the FIN requirement for name reconciliation by taxonomists, delivered through BiOnym. Another application to integrate would be the FAO VRMF tool for vessel name reconciliation. Many other domains are being considered such as FAO-areas and a geonames locator;
In additon, several VRE similar gCubeApps have been released, most notably the
- BiodiversityResearchEnvironment (5) The environment to access and browse species occurrence data from multiple Darwin Core enabled repositories, and to view these in the GeoExplorer tool;
- EcologicalModeling (4) The combination of data, tools and services to deliver species predictive maps and datasets, conceived as a richer AquaMaps;
Already in an advanced state of implementation, either as independent VRE's, or as extensions to the above are:
- SmartFish (4) A toolset to offer a advanced search and data retrieval over 3 data repositories (XML, Excel and a database) focused on capture species data through the iMarine e-infrastructure.
- Codelistmanager (4) The management of Codelists requires the further development of the facility in ICIS, and the development of Cotrix;
- CodelistMapper (3) The generation of persistent mappings between codes, code lists, and data using those codes requires that similar functionality developed in ICIS be made available as a separate VRE. In addition, it will have to be enabled to extract data using Cotrix components; a code lists and vocabulary management environment being developed by FAO and others;
- FishFrame (4) The support for the IRD / EU FishFrame format and COST model. The specifications are provided by ICES, and implemented in an experiment;
- NEAFC2ICES (4) Support to NEAFC to 1.) anonymize confidential data in their infrastructure, 2) publish these anonymized datasets in a controlled repository, 3) ensure that these data-sets can be analyzed in a variety of other VRE's for e.g. statistical trend analysis, plotting, graphing and tabulation.
- MixeR (4) Support to read from / write to an integrated, parallel R script. This component will build on the existing R functions in ICIS, and is developed as an iMarine Experiment with FAO Stock Assessment software. The expected reduction in processing time is from 12 days to 12 hours.
- EA-Linked Open Data (4) A toolset to produce Linked Open Data in the iMarine e-infrastructure.
- SpeciesModeller (4) The AquaMaps VRE with added algorithms, enriched with additional algorithms supported by e.g. OpenModeler. This can be achieved through the Statistical Manager facility;
- SpeciesChart (4) Interactive display and editing of OBIS and other occurrence data, related to TrendyLyzer;
- TBTI (2) The Too Big To Ignore VRE will serve the EA-CoP interested in the socio-economic data specific to small scale fisheries, their geospatial context, and document retrieval.
In addition, a range of new VRE's that answer to specific data manipulation needs has been identified to support other VRE's; they provide functionality that other VREs may need to increase their uptake, or that open new data sources to existing VRE's
- Semantic Annotator (3) The FAO services to annotate a document. Existing FAO services that might be re-purposed here are FLOD, the AgroTagger and VocBench;
- KB Browser (2) This service will prepare request to FLOD, and manage the results by e.g. integrating these in the a FCPPS report. Also IRD will have to describe their expectations.
Additionally, the CoP offers a host of independent services that will require to be interoperable with the VRE's. Examples are OpenSDMX, Cotrix, and Spread. these can be described in separate wiki's:
- OpenSDMX. OpenSDMX is the FAO Open Source project in support to SDMX data acquisition and dissemination.
- CotrixCotrix. Cotrix is the development of a shared code list manager. It is currently identical to the code list manager above. It can become a Open Source development tool, once FAO departments decide that it is worth investing in.
- FLOD is the Semantic data-set combining strategic FAO Assets in a knowledge base. FAO uses iMarine resources to enrich this dataset, and position it as a knowledge base for other resources to refer to concepts rather than string literals. The FLOD content management is currently being bolstered through the development of a Semantic data work-flow.
Other tools extract data from the iMarine e-infrastructure, but do not rely on specific services or resources. The data may be prepared in a VRE, but the are consumed e.g. through a web service only.
- iMarine Species App; (3) The updated version of the AppliFish mobile app;
- AppliFish2; (2) A redesigned AppliFish with better integration with e-infra services, such as search and country-statistics.
VRE Specification Template
The many opportunities provided by Virtual Research Environment operated through a gCube-based infrastructure at times may be overwhelming and difficult to grasp by representatives from e.g. the Communities of Practice behind the Business Cases.
In order to facilitate the collection of and negotiations on specific desiderata, a template that captures first the CoP 'wishlist', and then discusses these with Board representatives, WP3 members, and PEB delegates.
The current template contains:
- Describe the expected service in maximum 3 sentences
- Priority to CoP
- List proposed solution priority following the iMarine Board priority setting criteria:
- Potential target community;
- Potential for co-funding;
- Structural allocation of resources;
- Referred in DoW;
- Business Cases;
- How does the proposed action generally support sustainability aspects;
- How consistent it is with EC regulations/strategies (eg INSPIRE);
- Re-usability – benefits – compatibility;
- Relation to CoP Software
- Relation to D4S technologies
- Are the proposed measures effective?
- Does it reduce a known workload?
- How must the component be delivered to users? (UI Design / on-line help / training material / support)
- Are there any policies available that describe data access and sharing?
- Have the Copyright / attribution / metadata / legal aspects been addressed from a user and technology perspective?