Science of Collaboratories logo - link
 
 
 
An alliance to advance the understanding of collaboratories
Science of Collaboratories
   
   
 
Workshops : Comparative Investigation : Final Summary : EMSL

 
 
Enviornmental Molecular Science Laboratory (EMSL)

Presenter: Nathan Bos
EMSL participant: Jim Myers
Discussants: Ed Hackett, Peter Knoop

The EMSL collaboratory is primarily a 'shared instrument' collaboratory, with secondary classifications of product development and expert consultation. The EMSL makes Nuclear Magnetic Resonance (NMR) instruments at Pacific Northwest National Labs (PNNL) available to external users remotely. One of the reasons for EMSL's level of success is that the PNNL is a dedicated external-user facility. Its funder, the DOE, mandates that 50% of instrument time should go to external users (with the rest reserved for full-time PNNL researchers.) The EMSL collaboratory piggybacks on this, to provide remote access as an alternative to travel on-site for users of the NMR instruments.

Technology: There are two sets of technologies that were developed, or adapted, to support external users. The first was the CORE 2000 (Collaborative research environment) suite of synchronous collaboration tools. Built using NCSA's Habanero platform, the CORE 2000 includes:

  • Screen sharing,
  • Chat,
  • Whiteboard,
  • Video conferencing,
  • Remote-control camera on the instrument panel,
  • Molecular modeling tools, and
  • A voting tool.

This suite also now includes the VNC screen-sharing tool, which is distributed freely by AT&T labs in London. Because it is cross-platform and easy to set up VNC has become the most important remote access tool, (along with the telephone). The other tool developed to support remote access at EMSL is the Electronic Lab notebook. This is a web-based tool that provides access control, image capture, and a number of other features, and is freely distributed by EMSL.

Usage: Currently 25-30% of the external NMR users use the remote access capabilities. Interestingly, many users of the remote access tool do not use it as an alternative to travel, but instead use it as a way to involve more collaborators, such as non-traveling students or consulting researchers who want to be involved without traveling.

Diffusion of innovation: Since EMSL is a fairly well-established collaboratory, it is a good setting for studying diffusion of collaboratory-type innovation. Some of the factors that have promoted its adoption are:

  • User model: Since a large part of PNNL's mission is supporting external users, it was able to arrange institutional support fairly smoothly.
  • There are dedicated staff to help external users.
  • It is also fairly easy to enlist help from other on-site scientists.
  • On-site experts have a fund to bill hours to for external support, and they often get involved in projects as co-authors as well.

An interesting question is whether the current level of remote access (25-30%) is the peak, or should we expect continued growth?

Participant commentary

Jim Myers
An important use of the collaboratory has been for education. A paper on its use in undergrad class is available from EMSL website.

A study of initial users found that the nature of collaborations changed when using the CORE2000 tool. They had expected to collaborate the way they were accustomed to, which was checking in every few months. What they found was that their process changed as they got used to it, and they had more frequent contact. One real value of this was that they could check and edit each others' experimental setups early, and catch mistakes before they happened, thus saving time and frustration.

Discussants

Ed Hackett Instead of focusing on shared instruments for collaboratories, we should focus on shared instrumentalities. Collaboratories can help scientists share details of the research process; can make formerly hidden parts of the research process more apparent.

Perhaps Collaboratories could also become the middle part of the research process, between public and private spaces. Collaboratories may also help science scale up more effectively, such as in the example of National Center for Ecological Analysis and Synthesis (NCEAS).

Large, visible organizations such as collaboratories would be inviting targets for government wanting to direct science policy thru regulation.

Peter Knoop It's interesting that there is a combination of off-the-shelf and home-grown technology in EMSL. The challenge is always getting them to work together. We should learn more about the security policy of EMSL.

Discussion

Q: Why do scientists still choose to travel to PNNL to use instruments? A: (from Myers) the need to do complex sample preparation is often a driver of this.

Q: Why did you choose to develop your own tools rather than adapting others, such as those of Nestor Zaluzec? A: (from Myers) The two projects were working in parallel. There was communication, but Zaluzec's tools would not have been finished in time.

Q: What was the security policy for electronic lab notebooks? A: (from Myers) We reviewed more high-security measures, but these were not implemented because of other constraints. The notebook uses SSL. To get a higher level of security than the E-lab notebook would require use of Public Key Encryption. This was believed to be a barrier to adoption.

The EMSL lab notebooks have different security requirements than current 'grid' projects. Lab notebooks are for small groups to use; grids are more public. One of the most difficult aspects of security is transfer of access. E.g., if a scientist wants to temporarily grant access to secure data to a colleague, how do they do it?

Are there privacy concerns? Might collaboratory discussions be too permanent and public? It's similar to having dinner-napkin sketches made available to others.

How long should collaboratory records be kept? Would it be desirable to keep them indefinitely?

 
 
         
    
  Home | About SOC | Workshops | Resources | News & Events  

University of Michigan Logo

University of Michigan

School of Information Logo

School of Information University of Michigan