(1/23) Lausanne, April 15-16, 2004 EWME workshop Sharing online laboratories and their components -...

(1/23)

Lausanne, April 15-16, 2004EWME workshop

Sharing online laboratories and their components

- a new learning experience

K. Jeppson, P. Lundgren, J. del Alamo, J. Hardison, D. Zych

Chalmers University of Technology Massachussetts Institute of TechnologySolid State Electronics Laboratory Cambridge, MA, USAGöteborg, Sweden

(2/23)

Lausanne, April 15-16, 2004EWME workshop

Sharing online laboratories

Experiences from using MIT WebLab in large classes (~350 students)

• What are critical issues for successful sharing?

• How do students perceive, relate to and use this tool?

• How do we design courses to benefit from the resources of shared online laboratories?

(3/23)

Lausanne, April 15-16, 2004EWME workshop

So what is WebLab?A remote laboratory is ...

• a cost effective way of opening up the laboratory hall for measurements 24 hours a day

• a cost effective way of making state-of-the-art devices available to students

• a measurement tool organized to simplify data aquisition and to minimize time spent on practical details

• more specific, WebLab is an online remote laboratory setup for I-V characterisation of MOSFETs

(4/23)

Lausanne, April 15-16, 2004EWME workshop

WebLab graphical interface

(5/23)

Lausanne, April 15-16, 2004EWME workshop

Critical issues for sharing

• Lecturer must have opportunity to set course focus so that resources offered by the online laboratory can be fully utilized and appreciated

• Good and fast communication between host and user site when system goes down or device is broken

• Simple and self-instructive graphical interface• Easy access – no waiting time• Dependable – the system must be accessible

whenever students have planned their measurement session

(6/23)

Lausanne, April 15-16, 2004EWME workshop

Critical issues for sharing

• adding online laboratory exercises to courses previously without hands-on exercises is one thing…

• ...but successfully replacing traditional on-campus laboratory exercises with remote ones online is quite another

• The on-campus and the remote laboratories are two qualitatively quite different learning tools - just as listening to a lecturer is something else than reading a book

(7/23)

Lausanne, April 15-16, 2004EWME workshop

Can shared online laboratories help create a competitive learning

environment?

• We choose to change a traditional closed-task laboratory assignment to an open task where students were expected to plan measurements themselves and to find important device parameters to study

• Exploring device properties and how to model them must be an integrated part of the course

(8/23)

Lausanne, April 15-16, 2004EWME workshop

Measurement task

The objective of using WebLab became an issue of moving student focus • from handling instruments for collecting data• to analysing (readily available) data by comparing

experimental data to models

Measurement results expected to be presented orally

(9/23)

Lausanne, April 15-16, 2004EWME workshop

An important measure of success is the impact on student learning

(in relation to the teaching costs)

• Student time spent on subject• Student attitude towards subject• Student focus within subject• Student learning outcome

(10/23)

Lausanne, April 15-16, 2004EWME workshop

When did students login to WebLab?

2-22

2-24

2-26

2-28

3-2

3-4

3-6

3-8

3-10

3-12

3-14

2-26 2-28 3-2 3-4 3-6 3-8 3-10 3-12 3-14 3-16

Presentation date

(11/23)

Lausanne, April 15-16, 2004EWME workshop

Positive comments on WebLab• Access (43%):

– ”You can decide for yourself when to do the laboratory exercise!”

– ”means less stress!”– ”gives opportunity to see how different settings affect results”– ”offers flexibility – you can work from home at your own pace”

• Interface (19%): – ”Clear graphs!”

• Real devices (16%): – ”You get a feeling for realistic values”

• Repeated use (15%): – ”Measure one day – think a bit – then measure again!”

• Methodology (9%): – ”focus is on assignment, not on instrumentation or wiring”– ”avoids many practical problems”

(12/23)

Lausanne, April 15-16, 2004EWME workshop

What students thought about WebLab

0

5

10

15

20

25

30

35

40

45

Miserable Poor OK High Outstanding

Accessibility and stability User friendliness Educational value

“Accessibility” and “Educational value” ratings not correlated

(13/23)

Lausanne, April 15-16, 2004EWME workshop

Technical Problems

• The use of WebLab in undergraduate courses at Chalmers was the largest and most ambitious deployment of WebLab to date

• This was bound to result in identification of new bugs and problems not seen before

(14/23)

Lausanne, April 15-16, 2004EWME workshop

Peak performance

• On February 25, 2003 between noon and 1 PM EST WebLab performed 134 characterization experiments in one hour – on average that means one experiment every 27 seconds

• This was a 35% increase over previous WebLab record

(15/23)

Lausanne, April 15-16, 2004EWME workshop

Two types of technical problems

• A handful of system blackouts during which WebLab was unavailable for measurements – due to improper resource allocation setting in web server

• System returned error message in response to valid experimental request – problem was eliminated by increasing time-out settings of the device driver

(16/23)

Lausanne, April 15-16, 2004EWME workshop

Negative comments on WebLab

• System instabilities (35%):– ”WebLab performs poorly – at first web site is not

available, then it keeps on crashing”

• Supervision (30%): – ”WebLab is difficult to handle – much to learn and no

instructors”

• Time consumption (17%): – ”It takes TOO MUCH TIME!”

• Reliability (16%):– ”Quite a few bugs in new graphical interface”

• Speed: – ”WebLab is slow”

(17/23)

Lausanne, April 15-16, 2004EWME workshop

Negative impacts on students

• At Chalmers students worked in groups while at MIT assignments were of individual nature – more difficult for group to re-schedule when WebLab was down >> frustration & project delay

• Time zone differences – even trivial problems with WebLab took long time to correct since MIT staff was off duty

(18/23)

Lausanne, April 15-16, 2004EWME workshop

Improvements needed

• More powerful server to handle heavy traffic• I/V range limits to prevent device breakdown• Measurement examples available online• Options to save graphs directly• Simplified data export to Excel and/or Matlab

(according to student opinion)

(19/23)

Lausanne, April 15-16, 2004EWME workshop

WebLab vs hands-on

• Lucidity and tangibility are important• Provides practical device experience...• ...but gives no instrumentation experience• However, avoids hazzle with ”boring”

instruments• A real lab is better – because supervisor is

available and topic is ”easier” to grasp

(20/23)

Lausanne, April 15-16, 2004EWME workshop

Video evaluation

(21/23)

Lausanne, April 15-16, 2004EWME workshop

Conclusions

• It is not trivial to design a course that fully benefits from the resources of shared online laboratories

• Online laboratories are not simply replacements for traditional hands on laboratories

• The most important concern from the students´ point of view is that of accessibility

• On the negative side, lack of supervision when stuck on trivial matters is very frustrating for many students

(22/23)

Lausanne, April 15-16, 2004EWME workshop

Conclusions

• A successful implementation should carefully address the topics of how assignment is organized, where supervision is available, and maybe supply a list of FAQs and some measurement setup examples

• Nevertheless, WebLab motivated students to undertake more advanced data analysis than before

• Over all, WebLab was received positively in the introductory (second year) microelectronic device course