By Jim Taylor
When Joan and I get tired of sex and violence on TV—something that happens increasingly often as TV channels vie for showing more and more blood and gore—we turn to the cable music channels.
On long holiday weekends, when the system runs on autopilot, the labels often fail to match the music. The album cover purports to be playing Michel LeGrand’s Windmills of your Mind. But the actual tune playing is Peewee Hunt’s Twelfth Street Rag. And the little blue line that indicates progress, second by second, seems to have no connection to either the audio or the video.
Setting up a music program can’t be terribly complex. How difficult can it be, to cross reference a visual image along with filename of the music itself?
Surely an algorithm—technically, the lines of code that run a program—should contain some auto-correction capabilities. If the song and the visuals don’t match, it should recognize that discrepancy. And fix it.
Algorithms need a built-in reality check.
I’m not thinking just about music programs.
Do I want a car’s self-driving system, for example, to set its cruise control for a highway’s Google-defined speed limit—say, 100 km/hr if it cannot recognize that the highway ahead has been ripped up for repaving?
Do I want a pharmacy to fill a prescription for a diuretic that eases heart congestion if it conflicts with another medication for malfunctioning kidneys?
Should I wear a raincoat because the forecasts call for showers, even though the skies are clear?
Reality check is, I think, the fundamental principle of science. As soon as someone comes up with a new theory, other scientists try to confirm or to disprove it. If they can’t replicate the results—remember the furore over “cold fusion” a few decades ago?—the theory goes into the garbage.
Scientists test for flaws in reasoning, errors in procedure, to find the exception that invalidates a general rule. A theory or process only becomes accepted wisdom when repeated attempts fail to find a flaw in it.
As an Australian chemistry professor reminded me, that’s where science differs from religion. Or from politics, for that matter. Science, in general terms, tries to disprove new theories; religion and politics try to defend old theories. Or dogmas, to use a more theological term. Challenges are attacked. Not welcomed.
Although, admittedly, even in science, many will still cling to an earlier “law” until a new generation of scientists take over.
Little wonder that for years, preaching classes in seminaries were called “apologetics”—creating an apologia, a defence, for historically accepted truths.
Politics doesn’t call speechifying apologetics, but the principle is the same—defend a party’s existing policies at all costs. Including one’s own personal integrity. Even when austerity budgets, tax cuts, or giveaway grants are no longer what the economy needs.
Sometimes when I listen to election debates, the handout melody and the hard-times label don’t seem to be connected.
And when I listen to a sermon or a homily, the message of unconditional love doesn’t match the promise of eternal punishment for sinners.
Sometimes, it seems to me, we need to step back from predigested programs and do a reality check.
Author Jim Taylor lives in Lake Country: rewrite@shaw.ca