Welcome to EtherPad!

This pad text is synchronized as you type, so that everyone viewing this page sees the same text.  This allows you to collaborate seamlessly on documents!

Agenda
    Member node metrics (per Bruce)
    Diagrams
    Action items (next steps)

DataONE Metrics

What we log (data users)
    IP address
    User ID    
    When (date)
    Which Member Node
    Unauthenticated read = "nobody" (perhps "guest" or "anonymous")
    Which browser reportedly used
    What they did (event was - read/update)
    
Member Node API
    MN logs information immediately 
    Periodic log back to Coordinating Node
    
Availability
    to anyone with rights permission
    to object (owner and anyone he grants)
    e.g., DAAC object owner (ORNL DAAC)
    
Permissions
    permissions to read
    permissions to write
    
    
    
Governed by who has rights permission for particular access

Unauthenticated searches
    still provide date, time, IP address
    can see this information if you have permission
    
The "user agent" field - provides information about searches that are built

Careful to explain
    for a lot of things (for the user) there is limited privacy
    log on - we know who you are
    not logged in - limited use 
    
    
    
Static Measures   
1. Cost-benefit analysis [static measure]
    right software in place
    adequate staff (metrics from existing MN, CN would be useful here)
 
Dynamic Measures    
2. Services provided by DataONE, MN, CN (how much and how often used)
    software
    
3. Where metrics collected, aggregated
    Metrics can be collected either at CN or MN; MN metrics would have to be sent to CN for aggregation across DataONE
    
4. Supports - helpdesk for the user
  
    how often helpdesk contacted, response time
    
5. Helpdest for the MN, CN

6.   interacting support between MN and CN (making sure monitored to maintain standard)
        interoperability (in the face of changes at either). Software interoperability is set up at the beginning but this needs to be maintained. This is important for CI to understand (new 'improvements' in software are not neccessarily well received)).
        
What are metrics supposed to be doing?
Sustain system
Adds to individual researcher prestige by being provider (DOI)
Prestige for MN 
MN gains reputation because the more used datasets are highly valued and provide reflective glory to MN: demonstrated ROI; attracts funding


    


How often used?
e.g., DOI increasingly used as citation
How much data volume sent to a particular user? (transmission information)
Which particular datasets accessed by the user?
    Important for credit to originator (almost as important for citations in publications)
    Important for managing digital repository 
    
Cost-benefit analysis
    right software in place
    

thanks everyone!

not quite sure how I log out! Lovely to chat.