EURISCO Technical overview

1
EURISCO Technical overview

• Requirements
• Options
• Constraints
• Choices
• Import mechanism
• Design
• Live Demo
• Lessons learnt
• Next steps

2
Requirements

• Cost effectiveness
• Web based
• Automated system
• Easy to use
• Easy to implement
• Modular system(Can be exported/adapted to other
  communities)

3
Requirements Options

• Central Data Processing(all members would
  provide their datasets as is, its the
  responsibility of the Central platform to
  reformat and process the data)
• Decentralized Data Processing (all members would
  provide formatted datasets based on a common
  structure, the Central platform would be an
  automated process)
• Decentralized access(all members would have to
  provide an on-line Internet access to their
  national inventories, the Central platform will
  search on-line all national inventories)

4
Requirements Constrains

• Limited funding with implications on
• Software and design choices
• Developer staff time
• Project time frame (2001-2003)
• Focus on sustainability (2000-gt2010)
• Rapid changes in technology (i.e. Web, database)

5
Requirements Choices

• A Decentralized Replication model
• Open source tools
• Apache/PHP/MySQL, MapServ
• Source code free and documented
• All ancillary datasets free (i.e. GIS datafiles)
• Help Desk support
• Capacity building
• Sub-regional training workshops

6
Import mechanism - From National inventory
Reformat of data based on common standards
National Inventory
National Institutes
Phase 1
7
Import mechanism - File structure
NICODE
INSTCODE
ACCENUMB
GENUS
SAMPSTAT
FRA
FRA203
9342
Triticum
400
FRA203
9343
Triticum
400
FRA
National Inventory
FRA203
9344
Triticum
400
FRA
FRA203
9345
Triticum
200
FRA
FRA204
1123
Brassica
200
FRA
FRA204
1124
Brassica
301
FRA
FRA
FRA204
1125
Brassica
401
FRA
FRA204
1126
Brassica
402
FRA
FRA176
3457
Hordeum
100
FRA
FRA176
3458
Hordeum
100
8
Import mechanism - to EURISCO intranet
National Inventory
Central Infrastructure
Data Checking
EURISCO Web site
Phase 2
9
Live Demo
10
Lessons Learnt - Dataset errors

• File format Fatal errors
• ASCII character format (i.e. end line)
• Missing required descriptors (i.e. NICODE)
• Multiple declaration of same descriptor
• Process errors
• Multiple entry fields
• Geographical coordinates (latitude/longitude)
• Dates (wrong format)
• Ancillary data references (i.e. country ISO
  reference)

11
Lessons Learnt - Uploading

• Uploading process functionalities
• Slow upload gt Firewall ?
• Step by step instructions gt Tutorial ?
• Reporting
• Too complex reporting gt Brief reports ?
• Data quality checking gt more features !
• Help Desk
• Clear need to increase help desk support and
  training to ensure sustainability of EURISCO

12
Next steps

• Streamline the data uploading mechanism
• Improve the intranet front-end
• Improve data quality checking procedures
• Explore new technologies (XML, GBIF)
• Produce tutorial(s) and manual(s)
• Develop a generic infrastructure that can be used
  by other information networks
• Others?

13

• Thank you

Milko A. Skofic Database developer SINGERIPGRI/SG
RP Rome