Return to "How-to" Adoption Manual

Return to Case Studies

 

Lessons Learned

 

This page summarizes the lesson's learned that specifically pertain to the construction and implementation of the RWSL Nodes at North Island College in Courtenay, British Columbia and at the Colorado Community College System facility in Denver, Colorado and the NANSLO RDN.  It does not deal with the wider project management Lesson's learned although there is some over lap.  To see a page prepared by Catherine Weldon dealing with lessons learned from the wider project perspective, see: Lessons Learned-General .

 

Lessons Learned:

Purchasing and Delivery Issues:

Purchasing and delivery issues were experienced at both RWSL Nodes.  The time from when equipment was requested until it was delivered could vary dramatically.  Occasionally there were internal delays where a purchase requisition sat on an administrator's desk for an extra long time, but more likely the delays were due to back-ordering and slow service from vendors.  In some cases (particularly in Colorado) delays were caused by institutional procedures that required the research to be documented and approved before purchase requisitions could be signed and submitted to vendors. 

Suggested Solutions:

• 'Quick delivery' of equipment should not be assumed once decisions have been made to purchase.  Assume a 2 to 3 month delivery timeline in project planning.
• A thorough understanding of the greater local purchasing environment needs to be appreciated before equipment can be ordered.  Check with the purchasing people at your institution and leave time for these processes to run their course.

 

Equipment Quality:

Both NIC and CCCS experienced issues with the microscopes.  Because of funding constraints the North American vendor products could not be afforded so a Chinese vendor was chosen a a third of the cost.  This resulted in 3 of the 4 microscopes purchased by the 2 RWSL nodes having to be returned to China for repairs.  This resulted in significant project delays.  All microscopes were eventually repaired and put into service.

Suggested Solutions:

Budget for North American vendor products that carry North American based support policies.  This will reduce frustration and minimize project delays.

 

Physical Space:

Building an RWSL node without first securing space is problematic at best.  At NIC this was dealt with after the fact and has not been satisfactorily resolved yet.  Part of the problem was that no one had ever built an RWSL Node before so NIC was the first.  In Colorado space was secured before implementation began, with a much better result.

Suggested Solutions:

Secure physical space with adequate services ahead of time.

 

Connection Issues (Client connectivity):

A minimum 5 Mbs service for clients is recommended by the RWSL tech team.  This has to be the absolute minimum.  If a service does not maintain this speed dependably then the connection to the RWSL site will not be dependable and a client with a slow connection can cause the LabView server to crash.

Suggested Solutions:

The client Internet connection needs to be assessed by RWSL techs ahead of time.  The RWSL team at NIC is preparing a new architecture that will not be so susceptible to slow client connectivity issues.

 



Connection Issues (Overactive Security):

When the client is an institution it has been found that client security can interfere with or even prevent the connection to RWSL.

Suggested Solutions:

The client IT department needs to be included when access to RWSL is intended.  Various security protocols must be made aware of RWSL's requirements and work-arounds found before students and faculty can dependably access RWSL.

 

Connection Issues (Latency):

Latency is a time delay that occurs due to the distance from the RWSL site.  The further the client is from the RWSL unit the more delay will occur due to the limitations of the speed of light and the number of switches and routers the client must pass through to establish a connection.  Some issues that were originally contributed to latency turned out to be security issues (see above).  More testing is required to see just how much connection problems are due to latency and how much they are due to other problems is required.

Suggested Solutions:

Remote clients must test their connection with the RWSL techs before any lab exercise is attempted.  The new architecture will make this less of a problem will likely not eliminate this issue entirely.  Another solution will be the increase in overall bandwidth and increased speed that the Internet is likely to experience in during the foreseeable future, so simply waiting a year or a number of months may solve the latency issue for an individual client.  This is an area that the RWSL tech team needs to find better solutions for.

 

Transfer Credit/Articulation Issues:

There is a continued perception among many faculty and their institutions that on-line education is somehow inferior to face-to-face education.  This bias is particularly strong among many science faculty so when it is suggested that lab exercises can be delivered over the Internet from a remote location instead of in a face-to-face laboratory there are immediate knee-jerk reactions that question whether courses delivered on-line will retain their transfer credit.  This is in spite of the fact that real life science is more and more being conducted remotely (Mars rovers, autonomous deep sea vehicles, remote data collection networks, tele-surgery, computer controlled lab equipment, etc.).  This concern has significantly delayed the adoption of RWSL mediated lab exercises into college courses at least in BC.

Suggested Solutions:

• Education of faculty is the best solution for this issue.  Faculty need to see and be re-assured that RWSL
• is not-replacing faculty or jeopardizing their jobs, 
• is not another way to introduce simulated lab exercises, 
• will not replace ALL hands-on lab exercises,
• is intended to provide greater access to lab science programming and will in the long run result in increased science program enrollments because more students will have access to science programming,
• does not hurt (and may even enhance) student understanding,
• prepares students for the science jobs of the future where they will use remotely controlled science equipment regularly,
•  
• The NANSLO Faculty Sandbox that will give faculty and administrators ready access to explore some RWSL mediated lab exercises is a key component to faculty education.
• RWSL mediated lab exercises can be introduced into existing courses in small numbers initially without jeopardizing the courses transfer credit thus allowing time for faculty to become familiar with it and its capabilities 

 

Version Changes:

Changes in versions of hardware and software can be managed within a particular institution because the same group will be in charge of the RWSL node.  However, when the NANSLO RDN is implemented version changes can rapidly become a significant issue preventing the network from working smoothly.  Some changes occur internally, but some occur due to external forces (new operating systems, new Labview versions, etc)

Suggested Solutions:

It is critical that versions of RWSL labs be tightly controlled within the NANSLO RDN in a way that allows 'natural evolution' of RWSL labs to take advantage of new software and hardware and minimizes the financial strain on member institutions.  To this end the R&D and Technical Standards Committee needs to be implemented sooner rather than later.  This committee will represent participating institutions and oversee the evolution of RWSL technical standards.