Ed is an EE, PE, and author in Poughkeepsie, New York. You can contact him at [email protected].
And on the pedestal, this legend clear:
My name is Ozymandias, King of Kings,
Look on my Works ye Mighty, and despair!
Nothing remains beside. Round the decay
Of that colossal Wreck, boundless and bare
The lone and level sands stretch far away.
—Percy Bysshe Shelley
In the early 1800s, Jessie Hawley got the entrepreneurial notion of planting wheat in the fertile plains of upstate New York—milling the grain, shipping the flour to New York City, and getting rich in the process. He went broke while discovering how shipping cost affects selling price and, Chapters 7 and 11 not yet having been written, wound up in the Canandaigua debtors' prison.
Having sufficient funds for paper and postage, he embarked on a letter-writing campaign extolling the virtue of spending federal money on a canal system across New York State from Lake Erie to the Hudson River, with the intent of slashing shipping costs and encouraging industry along the way. The Feds, including President Jefferson, thought he was nuts, but his proposal caught the fancy of soon-to-be-NY-Governor DeWitt Clinton (no relation) and, after many dramatic twists and turns, Hawley's idea became the first Erie Canal.
This past July, my family bicycled 400 miles from Buffalo to Albany along the canal as part of a group ride. The Erie Canal was one of the first full-throttle engineering projects in the then-new United States, which made the ride something of a pilgrimage for the engineering bears in the group.
Keep the telecom and computer biz in mind: These stones have a lesson.
Figure 1: Schoharie Aqueduct.
The Plan
Run a finger along a raised-relief map of the eastern United States to find the notch in upstate New York where the Mohawk River cuts eastward through the Appalachian Mountain Range. That gap separating the Catskills from the Adirondacks, the only such opening from Maine to Alabama, could hold a canal between the Hudson River and Lakes Ontario and Erie.
The Hudson, a navigable estuary with tidal flow extending north beyond Albany, could carry freight from the Atlantic Ocean to the Great Lakes. That proved a compelling incentive in the 1800s and, even today, the Port of Albany handles ocean-going freighters from around the world.
Canals are, by definition, level waterways, but the surface of Lake Erie is nearly 600 feet higher than the Hudson at Albany. State-of-the-art locks in 1800 could match about 12 feet of elevation difference and the canal would require 83 of them along its route.
A canal's cost depends strongly on the volume of earth moved as well as the number of locks. The final design called for a 363-mile-long rectangular prismatic channel 40-feet wide at the top, 28-feet wide at the bottom, and 4-feet deep. Thus, unless you were very short or stone-cold drunk, it was impossible to drown in the Erie Canal.
The channel accommodated two-way traffic and each lock could hold a single barge 61-feet long, 7-feet wide, and drawing 3.5 feet of water. Earth removed from the channel would form a single towpath on the downhill side.
Towpath? Mule teams at the end of long ropes hauled the barges both upstream and down, with an elaborate rope-handling ritual occurring when two barges passed in opposite directions. A quick-release hitch resolved protocol violations that would otherwise drag a team into the canal: Harnessed mules can drown in four feet of water.
Why not lay a railroad along the route? Recall that James Watt invented the separate-condenser steam engine in 1769 and the first steam locomotive hauled 10 tons at 4.5 MPH early in 1804. Heavy, bulky freight went by water in those days or it just didn't go at all.
By today's earth moving standards, this would be a moderately large project. Back then, it was incredible because it would be dug by hand. The Ages of Steam, Dynamite, and Internal Combustion were in the future.
The Team
In 1800, the United States boasted about 10 engineers, a situation so dire that six years earlier, George Washington had established an Engineering Corps at West Point to train future military engineers. By 1816, the number of civil engineers had nearly tripled, to 27. The Erie Canal was engineered by land surveyors.
Benjamin Wright, the canal's Chief Engineer, helped survey the terrain, lay out the canal's route, and design the channel and locks. Hydrodynamics hadn't yet been codified, which left some design decisions in the realm of guesswork.
At one point in the design phase, Governor Clinton suggested that they could skip most of the locks by simply building a channel from the Niagara Escarpment (elevation 570 feet) to Cayuga Lake (elevation 380 feet). Wright pointed out that, although the downstream trip would be exciting, hauling those barges upstream might pose a bit of challenge for standard mules. He also noted the need for 250-foot embankments at some points, which was well beyond the state of the art.
The surveyor-engineer bears won that round and the side-by-side locks stepping through the Niagara Escarpment at Lockport remain a gorgeous example of practical artistry in stone.
Laborers drawn by relatively high wages ($8-12 per month) and an evening slug of rum came from everywhere. Wright imported German masons to cut the precise stone blocks required for each lock. Those masons, laid off at the end of the project, went on to build spectacular stone structures throughout New York rather than return to Germany.
The Problems
Construction began on July 4, 1817, at Rome, New York, with excavations extending in both directions. By starting in relatively flat land, they could demonstrate rapid progress, sign on early adopters, and work the bugs out of the system. In short, they could easily be 100-percent buzzword compliant.
Two years later, the 15-mile route between Rome and Utica opened. By simple extrapolation and ignoring the fact that this was the easiest section, the rest of the canal would be finished in about half a century. It was obviously time for some scheduled inventions.
In 1820, the middle of New York State was essentially a trackless wilderness, which meant the canal had few right-of-way issues. However, the workforce lacked enough axemen to cut down all those trees, had no good way to remove the stumps, and spent a lot of time carrying dirt. Nearly everything else became secondary to those problems.
Rather than cutting trees, the crews began attaching ropes to the treetops and simply winching them over. A massive vertical pulley arrangement then ripped the stumps directly out of the ground. They invented an ox-drawn earth scraper, designed larger wheelbarrows, and used mule-drawn carts to haul dirt faster than previously imagined. A three-man team (plus an ox and a mule or two) could now excavate one mile per season. Completing the canal on schedule became a simple matter of up-staffing.
And then the westward crews entered the Montezuma Swamp north of the Finger Lakes. In addition to saturated soil that required cofferdams to hold the waterway in place, they hit a different staffing challenge: In 1819, almost 1000 workers died of malaria. Without drugs or insecticides, progress on the canal halted until cold weather grounded the mosquitoes. The crews finished that section with a prodigious effort during the winter months.
Water admitted to each section as it was completed revealed another problem. No matter how closely those German masons fitted and grouted the stone blocks, the locks leaked. Wright appointed engineer Canvass White, who had earlier investigated the British canal system, to solve the problem. In 1820, White was awarded a patent for a hydraulic cement made from local materials. Nearly 400,000 bushels of the stuff waterproofed the stone locks.
The 6-inch clearance beneath loaded canal boats caused severe erosion of the puddled-clay bottom, which was partially solved by a 4-MPH speed limit. Repair boats hit 10 MPH on their way to spots where beavers, which considered the towpath bank an ideal residence, caused massive washouts.
The Payoff
DeWitt Clinton bailed Hawley out of the Canandaigua slammer in October 1825 for the first trip along the completed Erie Canal. The two journeyed from Buffalo to Albany, then along the Hudson to New York City, with a crowd of dignitaries and two barrels of Lake Erie water destined for the Atlantic Ocean. Ceremonial cannon fire preceded the Seneca Chief in a nonelectric "telegraph" that took 90 minutes to reach NYC. Much of the artillery had been captured from the British during the War of 1812, an event that had delayed the start of construction for a few years.
The most immediate effect of the canal was, as Hawley had predicted, to slash the cost of shipping to and from the Great Lakes by an order of magnitude. Disruptive technology, indeed!
People accompanied all that freight as workers, passengers, opportunists, and vagrants. NYC was a bustling city of 150,000 that would balloon by 1 million people in the next 30 years. Buffalo, a sleepy village of 2000, gained another 40,000 people by 1850. Rochester went from the middle of nowhere to the middle of the canal and saw its population jump from 1500 to 56,000. The canal powered rapid development of the Midwest frontier.
The original canal design anticipated 1.5 million tons of cargo a year, which proved to be entirely inadequate. The channel and locks were enlarged almost immediately, again in the late 1800s, and were replaced by the adjacent NY Barge Canal in 1918. The canal was so successful that tolls were abolished (!) in 1883 after $121 million had been collected.
The Aftermath
In 1853, the New York Central Railroad forged 10 smaller railroad companies into a system that connected Buffalo with Albany. Because railroads and canals both require smooth, flat terrain, the two transportation systems ran roughly in parallel across New York.
The advent of motorized transport in the early 1900s wove a third strand, often overlaying the long-disused original Erie Canal. Nearly every town along the way has a wide, flat, bar-straight, east-west Erie Avenue or Canal Street.
To grossly oversimplify events, the railroads killed the canal with faster and cheaper cargo transportation, whereupon trucks and automobiles killed the trains with faster and more convenient access. In each case, designers could not foresee the next disruptive technology or its effect on their planned system.
The New York Barge Canal, now the NYS Canal System, handles only a few pleasure craft. Cargo ships ply the St. Lawrence Seaway between the Great Lakes and the Atlantic Ocean to eliminate cargo transshipping. The remains of the original canal serve as a biking/ jogging trail and a rather skinny tourist attraction.
The New York Central Railroad tracks, now part of Conrail and Amtrak, still carry freight and passenger traffic. Amtrak's Maple Leaf took us from Albany to Buffalo to retrieve our van, passing miles of idle automobile carriers and freight cars along the way. Many corroded brick buildings display faded rail-side ads: "National Biscuit Company. Biscuit Box 5 cents."
The Future
Although we like to think we're better at technology than our forebears, the evidence that we're approaching a discontinuity seems clear. Some data points will illustrate why—a few decades from now, our works may seem as transient as the Erie Canal.
The cost of long-distance telephone traffic is dropping asymptotically toward zero, as evidenced by the collapse of the telecom industry. Based on regulations and physics, telecoms strung thousands of fiberoptic strands across the country. The regulations changed and wavelength-division multiplexing boosted fiber bandwidth by a factor of 1K, so that about 95 percent of all those long-haul fibers remain dark. Voice-over-IP is poised to blowtorch the local access market and few rational business models generate any revenue.
Wireless providers plunked down big money to acquire bandwidth that will likely never be used, as improved coding and changing application models rendered those frequencies useless. Perhaps paying the military to move out of other frequencies will generate some return on investment; certainly nothing else has.
Hardware design continues to be blindsided by Moore's Law, with CAD tools lagging far behind the available transistor count. Producing high-end chips now requires iterating through the architecture, layout, and routing while simulating the final circuit floorplan until the timings converge. There's no way to determine how a given system will perform until it's actually laid out, an end-to-end task not well handled by existing tools.
Software design, such as it is, has foundered upon the increasing complexity of network-enabled systems. Building large projects from small, well-tested components has demonstrated both the lack of small, well-tested components and that Newton's Laws don't apply to code. At least we now understand that we don't understand the inherent complexity of large-scale projects.
Embedded developers, only recently converted to off-the-shelf componentry, now struggle with multiple CPUs and operating systems in devices networked to the world. Unfortunately, the Internet model of open access has conspicuously failed, as simple programming errors in mission-critical systems permit remote access for morally stunted crackers.
All else has rotted away from the cut stones of those locks and aqueducts. Can we design long-lasting structures that our great-to-the-nth-grandchildren will admire and use?
Reentry Checklist
More on Jessie Hawley and the Erie Canal is at http://www.wxxi.org/canaltowns/transcripts/tgrasso.htm. A canal timeline is at http://www.home.eznet.net/~dminor/UCanal.html, and a timeline of major inventions is at http://www.wikipedia.com/wiki/Invention_timeline. The history of the New York Central Railroad is at http://www.crisny.org/not-for-profit/railroad/nyc_hist.htm.
A 1925 pamphlet describing the Erie Canal is at http://www.canals.state.ny.us/cculture/history/finch/finch_history_print.pdf. The homepage shows the canal's current state of affairs.
Syracuse University Professor Sam Clemence's presentation on Erie Canal engineering is available at http://www-fcms.syr.edu/showcase/SPClemence/ErieCnl/. I was in the front row of his talk during our bike ride. Syracuse the city has more canal stories, maps, and links at http://www.syracuse.com/features/eriecanal/. An open-channel flow calculator Wright would have died for can be found at http://www.lmnoeng.com/manning.htm.
Information about the New York Parks and Conservation Association's annual Cycling the Erie Canal ride is at http://www.nypca.org/canaltour/index.shtml. Go for it!
You can find several authoritative versions of Shelley's sonnet. I used the handwritten copy at http://www.nla.gov.au/worldtreasures/html/theme-literature-2-ozymandias.html, with further background at http://www.rc.umd.edu/rchs/waith.htm.
Vernor Vinge discussed the coming technological discontinuity at length in his books, most notably Across Realtime (now out of print), and in the article found at http://www.ugcs.caltech.edu/~phoenix/vinge/vinge-sing.html.
DDJ
