Group of Water Survey Canada hydrologists in the 1970s.

Is 49 Years Long Enough?

Photo: Lynne Campo, 2nd row, fourth from the left, shown in a group photo taken sometime in the 1970’s. The Water Survey of Canada has under-gone a lot of change (including gender distribution!) since that time. Lynne has been the pioneer of much of that change.

I have had the great pleasure of several decades of overlap of my career with that of Lynne Campo who retired on June 28th after 49 years of hydrometric data management for the Water Survey of Canada (WSC). I first met Lynne as a novice stream hydrographer in 1981 when I was sent to Vancouver for training in the art and science of hydrometric data processing.

In a few short days Lynne laid out the foundation for the rest of my career.

In 1981, hydrometric data processing was in the early stages of transition from the mainframe computer at the University of British Columbia (UBC) to the new minicomputer in the Vancouver office. My training was, perhaps, the last ever on the mainframe. Lynne guided me through the process of preparing an analog strip chart for digitization with the use of coding stamps, digitizing the chart using a digitizer that created computer punch cards, taking a large box of punch cards out to UBC in a taxi, and feeding the cards into the card reader. The output of the computer run were printouts on 24 inch wide paper to be taken back for quality review by spot checking against the original analog chart for either coding or chart interpretation errors.

I don’t remember exactly what the computer was that we used at UBC but this is from the manual:

The time required to process 20 stations of test data are as follows:  

Compile TimeExecution Time
IBM 37030.7 seconds42.7 seconds
CDC 640022.9 seconds18.6 seconds
UNIVAC71.7 seconds49.9 seconds

Computer digitization of strip chart records was still relatively new at the time.

Lynne also taught me how to manually digitize a recorder chart, using a specially designed clear plastic cursor, the same way she had been taught when she started her career in 1970. Thus she provided continuity in the business process of water level data handling from first principles through to the use of the latest technology. Lynne managed the transition from the mainframe to the new mini-computer which she kept operating for the next 15 years.

The STREAM, MANUAL and HOURLY programs that had originally been developed to run on the main frame computer and then adapted for the minicomputer were finally retired in the mid-1990’s when WSC transitioned to the Compumod system that ran on microcomputers. By this time the modernization of analog to digital water level recorders was nearing completion. However, a strategic cost-saving decision was made not to migrate all of the intermediate files from the old system to the new system. The paper records would be kept as the authoritative source of truth and only the results of data processing would be migrated into Compumod.

By this time Lynne had a quarter of a century of experience filling data requests from hydrologists, hydraulic engineers and fluvial geomorphologists.

She had always looked after them well and was not about to stop doing so because of a cost-saving decision. She knew that results aggregated to hourly or daily values are sufficient to solve the ‘easy’ problems of her clients but that the ‘hard’ problems required getting as close to the truth as possible, which means providing the point data digitized from the charts at irregular time intervals. She also did not trust that the bulky chart records would be preserved in perpetuity. Lynne, and her trusty colleague Lauren, quietly and almost surreptitiously, began what would become a 25 year long data rescue mission. In any lull in their busy days I would look over and notice that Lynne and Lauren would, rather than taking a well-deserved break, be busily salvaging the data that had been left behind in the data migration so that her clients could continue to have access to the point value data when needed to solve the ‘hard’ problems.

Lynne went on to manage the transition from Compumod to AQUARIUS in 2010 with a similar attention to the impact the change would have on the data, the field staff and–most importantly–to her clients. Needless to say, she is universally loved for her care and stewardship of the data and by extension everyone involved who either created, or used, data.

One of the many ways Lynne brought value to every conversation about data was the way she was uniquely qualified to connect how data are used with how data are produced.

This is a uniquely human task that comes from knowing not only what data are available and what data are being asked for (i.e., a good job for a web site) but what the problem is that data are needed to solve. This can only come from conversations with end-users of data (i.e., not a good job for a robot). 

Change is inevitable. And change can be disruptive. But with the wisdom borne of Lynne’s familiarity with change the integrity and continuity of WSC data has been sacrosanct.

Successful change comes from mavens like Lynne who deeply understand not only what the business processes are but why they exist. Some conventions are absolutely essential for data integrity and some only exist because of limitations of the previous generation of technology. It is with experience that covers at least 2 generations of data management systems that change can be managed with optimum benefit and minimum impact.  

A best practice for modern software engineering is for change to be continuous. Now quarterly software updates allow for continuous small improvements rather than long periods of stability followed by infrequent disruptive change. Continuous change must be measured against the continuously evolving needs of data users to ensure continuous improvement. Others may have faith in technologies like Google Analytics to provide this measure. My faith is with Lynne but I fear that she has retired too soon.

Lynne may be retiring but her role should not be retired. Not only should we be celebrating Lynne’s impactful career, but we should be honoring all of the hydrometric data mavens around the world for their stewardship of one of the most valuable resources of all, the story of freshwater.

No Comments

Post a Comment

Comment
Name
Email
Website