Mosquitoes, Trains, and Software

by breeve 23. January 2011 15:26

Ferdinand de Lesseps could not be doing better. He had just built the Suez canal and no sooner had the grand opening ceremony finished on November 17th 1869—in typical French party fashion—did Ferdinand begin to stir thoughts of a canal in central America and Panama in particular. Ferdinand, who had no engineer training, may have mastered the art of persuasion but Central America and Panama in particular was no sandy sea level desert.

While the environment of the Suez may be hot, Panama was a death trap. If the severe rains, dense forest, and staggering humidity didn’t kill you off then the diseases of Yellow Fever and Malaria would. During construction of the railroad, a windy stretch of rails cutting through the thick jungle from the Atlantic to the Pacific, 6000 lives were lost with some estimates being double that. So many died in Panama that the survivors swore it was haunted.

If Ferdinand was aware of the difficult conditions on the other side of the sea, he didn't share it. Soon he started what he had at the Suez: raising money and spreading dreams. He purchased the Panama railroad and by 1881, 2000 men were at work. Actual digging commenced on January 1882.

The plan from the beginning was an ambitious one. The canal, like the Suez, was to be a sea level canal which was no small feat considering the highest point along the route was Culebra point a full 300 feet above sea level. As actual work on the Culebra Cut began, it became apparent just how much dirt was involved. The news back in France, however, was misleadingly upbeat reporting 2 million cubic meters a month when in reality only 146,000 in May 1883 and 156,000 in June 1883.

And when the railroad wasn’t carrying alive bodies it was carrying dead ones. Disease was taking its toll on the workers. Yellow Fever and Malaria—still believed to come from the vapors of trash and the rain forest soil despite investigations at the time pointing to mosquitoes as spreading agents—was killing, by some workers estimates, two often three for every four coming from France. So bad were the conditions that the head engineer Jules Dingler lost his entire family to disease just one year after bringing them all to Panama.

By February 1889, not even Ferdinand’s upbeat propaganda could stop bad news from spreading. The investors had caught on and the French Panama effort quickly ended. The United States Panama story would be different largely because of two figures: John Stevens and William Gorgas.

John Stevens became the 2nd chief engineer of the US effort after John Wallace—who didn’t do anything differently than the French and consequently had all the same problems—quit. Upon arriving in Panama Stevens, a distinguished railroad engineer, immediately saw what the French and Wallace had missed. Building the canal was less of an engineering problem and more of an infrastructure problem.

The main problem the French had, it turns out, was not the digging but rather removing the large amounts of earth away from the digging sites. The French, in their eagerness to dig, had neglected to build the infrastructure needed to remove earth fast enough to keep the excavators working full time.

But few were prepared for Stevens' next move. He put an end to all work on the canal and instead put everyone to work building communities with houses, paved streets, mess halls, barracks, hospitals, laundries, sewage systems, running water, churches, cold-storage facilities, clubhouses, schools, reservoirs, and hotels. The railroad was upgraded to handle bigger cars and more track added.

By the time he was done if you kidnapped someone from the United States and dropped them in one of the communities he built, it would take them days to figure out they weren't at home. If the top brass back home were worried about progress, the workers were not. Ice cream made regular appearances on the work lines. Visitors from the US commented on what a nice place Panama was.

Even more impressive, when work continued, Stevens' ingenuous design of rail roads carried the dirt away from Culebra Cut to 60 different dump sites as fast as the excavators could dig.

But if Stevens was the engineering wizard, Gorgas was the mad scientist. Gorgas, who believed in the Yellow Fever spreading mosquito theory, spent much of his initial time in Panama studying the mosquitoes. By the time he was done with his study, no one on earth would know more about the mosquitoes of Panama than Gorgas.

But before Gorgas could squash his first mosquito, he would need Stevens' help. Stevens, who viewed the whole mosquito theory as speculative at best, found the strength to trust Gorgas and gave him the resources he so badly needed. With the resources of the entire engineering staff behind him, Gorgas set immediately to putting windows and nets in buildings, getting rid of open water containers, setting up garbage cans, and fumigating every house.

By December 1906, a year and a half after Gorgas program started, Yellow Fever—the disease which plagued the world for centuries—was eradicated from the Isthmus. So great was his work that many visitors to Panama would have a hard time deciding which work was greater: the canal or the complete absence of insect life from the work zone. In the final year of construction, the death rate among canal employees was 7.92 per thousand which was better than the death rate in the United States at that time. Turnover—a big problem previously—was low and moral was never better.

What Stevens and Gorgas did in Panama was extraordinary. By taking the time to build infrastructure, they succeeded where the French failed by turning a sick and inefficient operation into a healthy efficient one.

Unfortunately, the lessons of Stevens and Gorgas is lost in most software shops. Like the French, who wanted to sling dirt rather than build rails, slinging code is given the highest priority. Like moving dirt away from the excavators, how long does it take to make a build and deploy the bits? Like the mosquitoes spreading deadly disease with each bite, are code bugs fixed before writing new code or is disease spreading causing high turnover and low moral? Like the watchful canal employee eliminating standing water, is QA given proper time and resources to find and eliminate future mosquitoes?

Software projects can get away with neglecting infrastructure for a time. But only temporarily. Sooner or later the software foundations will start to crack, work progress will slow, and everyone will be standing around staring at piles of dirt wondering how they are going to be moved.



Powered by BlogEngine.NET
Theme by Mads Kristensen

About Me

I am a Principal Engineer with 16 years experience developing and releasing software products. I started developing in C/C++ then moved into .NET and C#. Currently working in Python/Flask and Docker. Have tech lead multiple projects. I have developed products in Windows Forms, ASP.NET/MVC, Silverlight, WPF, and Python. I currently reside in Austin, Texas.

Own Projects

Pickaxe - An easy to use web page scraper. If you know a little SQL and how basic CSS selectors work, no web scraping product will be easier to use.


FligthQuery - Query FlightAware FlightXml API with SQL


Created ASP.NET MVC forum originally targeting home owner associations but now in use by an investor group.

Currently Reading