GLOBAL BIODIVERSITY

1
GLOBALBIODIVERSITY
WWW.GBIF.ORG
INFORMATIONFACILITY
The GBIF Data Repository Tool (New updated
version 3.0)
Hannu Saarenmaa EC CHM GBIF European Regional
Nodes Meeting Copenhagen, 2005-09-15/18

2
Outline

1. Objectives and background
2. Design and installation
3. Use
4. Demonstration

3
1. Objectives and background
4
Challenges in data sharing

• Eventually, all data sets become orphans
  Archiving services are a necessity.
• The concept share once, use many requires
  available data repositories.
• Data from archives must be available using
  standard mechanisms to portals such as GBIF.
• IPR, confidentiality, and benefit sharing must be
  respected at all times.

5
Goals of the GBIF Data Repository Tool

• Enable data custodians to manage their data and
  control its publishing.
• Provide mechanism such that spreadsheets, etc.,
  can directly be used for sharing data
• Hide the database complexities from users
• Make available a simple data warehouse tool for
  those who want to host datasets for the community
• I.e., lower the threshold of data sharing as low
  as possible.

6
2. Design
7
Functionalities

• Data must be formatted according to the Darwin
  Core standard and its extensions in flat
  spreadsheet format.
• In fact, any flat format will work (rows,
  columns)
• The system will check and parse the data into
  embedded MySQL database that becomes available to
  the public as a DiGIR/TAPIR resource.
• Owners can control the level of detail released
• Fuzzying of geographic coordinates is available
• Collector names and time periods can be hidden
• Approval of terms and conditions for data use can
  be required
• Owner can revoke release and update data.
• Metadata can be inherited to data to replace
  missing values as defined.
• Includes an embedded image server

8
Component architecture
9
Installation

• For Linux and Windows
• Based on Python, Zope 2.10 and MySQL
• Supports the DiGIR and TAPIR protocols of TDWG
• Turn-key installation
• Fits with directly into the EC CHM software
  package

10
(No Transcript)
11
(No Transcript)
12
3. Use
13
Steps for data owners

1. Prepare the data files
2. Create a nested folder structure on the
   Repository for the collection
3. Enter default metadata scope (to cover missing
   values in data, etc.)
4. Decide on access policies
5. Upload the files
6. Publish the data files

14
Create a collection and assign it to a data
custodian...
15
Enter metadata scope for the collection
(inherited)
16
Create the resources (databases) of the
collection and folders
17
Upload the files to folder, validate and release
them
18
Prepare the data file(s) in tab-separated format
19
(No Transcript)
20

• Access policy
• options
• Fully open
• Standard GBIF policy of acknow-ledgements
• No direct download and fuzzying for web service
  access

21
Data is now searchable locally and through the
DiGIR/TAPIR protocols
22
(No Transcript)
23
4. Demonstrationhttp//fmnh.eaudeweb.ro/