[al-gore-rhythm] A recursive computational procedure for solving a problem in a finite number of steps.
Amidst the recent web content explosion—blogs, user-generated content, social media, etc—finding relevant information on the web is truly a needle-and-haystack scenario. To make sense of the content chaos, as you know, Google employs a robust algorithm to deliver relevant results to web searchers. To manage the unwieldy masses of desperate singles, companies like eHarmony.com and Match.com leverage algorithms to determine relationship relevance and pair up compatible individuals. Even Amazon uses behavioral models to determine the right set of products to include in their product recommendations (you know, the "People who viewed this item ultimately purchased..." sections). While none of these services will ever be 100% accurate, the thought is that with enough information, we can build models to predict people's preferences and behaviors—or at least get to a degree of accuracy closer than a 50/50 coin flip.
In problem-solving, I'm always a fan of getting to the right information—faster, smarter—and customizing a fix, rather than applying a template solution that really may do more damage than good. Unfortunately, when it comes to divorce, the government isn't on the same page and tends to use context-irrelevant judgments in splitting up the spoils. Well, let me take a step back ... some states do put constraints on alimony payments—taking into account the length of the marriage and extenuating circumstances—but even still, the final determinations are highly-discretionary and hardly data-driven. The result: an unfair, inefficient distribution of wealth from one party to another.
Before we bring in algorithms to save the day, there are two basic principles that we need to ground ourselves on:
- In a divorce, a spouse needs to be compensated for the opportunity costs incurred from sacrifices to one's professional path. For example, if a wife forgoes her law practice to take a domestic role, she needs some repayment for that sacrifice.
- Give credit where credit is due. Who really worked harder—a athlete's x-wife who had no household obligation other than to wear out the stripe on her Black Card, or, the wife of a blue collar worker that struggled to feed three kids and maintain control of her household? And yet, by embracing the spouse is entitled to half methodology, we'll award the former $20 million and the latter $20,000. The point: just because you were along for the ride, does not mean that you contributed to the ultimate success and are entitled to a massive payout. What does determine that entitlement is the details of the situation—something which the legal system tends to gloss over.
Why can't we create an algorithm that internalizes a wide range of relevant inputs—spouse's income and career trajectory prior to marriage, industry, level of education, duration of marriage, level/type of contribution to the breadwinner's profession, etc—and combines it with historical data and trends to determine the fair amount of compensation? These could be guidelines for basing judgments rather than the arbitrary income divisions that we currently use. Sure, this is profiling which generally has negative connotations, but, with enough relevant inputs, the output becomes more reliable.
And no, this is not meant to be a misogynistic diatribe. Guy Ritchie shouldn't have gotten a penny from Madonna and instead of getting a payout, Nick Lachey should've been forced to compensate Jessica Simpson for putting him on the Hollywood radar (albeit the outer fringes). Furthermore, an algorithm could reveal that a supportive wife had a more material impact on her husband's earnings and therefore deserves more compensation. Whatever the case, we need to rely more on data than discretion.