Lawyers vs. Programmers

[selenite0]

There's a joke I heard in the SCA:

The SCA runs the way it does because its laws are written by computer programmers. The USA runs the way it does because its computer programs are written by lawyers.

Which is not strictly true, but lawyers have a huge impact in how we do things, from every decision made to avoid a potential lawsuit, to the increasing complexity and obscurity of laws, to the one-winner one-loser assumption in all political contests. It's become so pervasive that some people get law degrees without ever intended to practice law, just so they can function in their preferred profession. Now there's a change in society--lots of people are learning programming as part of their jobs and former programmers are carrying their skills on to new careers. Jeff Jarvis thinks this will push us toward a more transparent system instead of the deliberate obscurity lawyers like to produce to protect their guild. Another blogger points out that while lawyers want to maintain the status quo programmers are producing constant change.

I'm more interested in another difference between the professions. To lawyers "truth" is ultimately what an individual or group says it is--a judge, a committee, Congress, the Supreme Court. That's who has the last word, and everyone has to live with it. Programmers--and other kinds of engineers--test their theories against the real world. If it works, great. If it doesn't, no appeals to any court or amendment to a law will fix it--you have to change the design/theory/approach. I think our society would be a lot healthier if we routinely tested ideas against Reality instead of Authority.

Comments

[patgund.livejournal.com]

Problem being is that those in authority *think* their desires trump reality.

[joyeuse13.livejournal.com]

Problem is, with programming problems it's very easy to define "it works." Does this line of code produce the correct output? Yes/No. Law-type problems are not so concrete, and may take a lot longer to test. In addition, the answers may change over time.

[selenite.livejournal.com]

What I'm hoping will change on the law/government side is that they'll start measuring the results of a government program and comparing them to the results of other programs. Right now there's a built-in assumption that supporting X is good and providing more funding for it is better. So DARE gets more money despite studies showing it's not reducing drug use, and more useful educational programs have their budgets cut.

Hmmm--software shops have full-time testing people trying to find ways to figure out if the code is working right. I wonder if we'll ever have a Texas Department of Testing and Measurement doing research to figure out how well each of those expensive government programs works. Or if funding could be found to support something like that as an NGO. That'd be an interesting job. The hard part would be trying to get professionals who don't already have a vested interest in the outcome of each evaluation.

Lawyers

[p-o-u-n-c-e-r.livejournal.com]

Rare is the opportunity for a legislator to create new law upon a blank sheet of paper. More often, the problem is that conflicting INTERPRETATIONS of written law, applied to situations the earlier law-writers never envisioned or intended, must somehow be reconciled. The interpretations have become the "common law" or "case law" ...

Case law is less analogous to computer code than it is to the procedural hacks users work out. Import the report text to a spreadsheet and massage, export the results to a delimited flat file, import THAT to a data-base wizard to assign keys and break into 4 related tables, update one table to a keyed standard table, (rejecting any duplicated records) in order to pick up recent changes ... In an ideal world the procedure could be coded (a macro?) to react to the incoming text by updating the records. In practice, it's a lucky office that has such procedures written down and not passed from departing user to incoming replacement user via folklore, oral tradition, and one-time on-the-job-training.

SCA is getting lawyer-y

[selenesue.livejournal.com]

They are rewriting their rules about how and when to expel members:

http://www.sca.org/BOD/announcements/commentary.html?id=12

[kd5mdk.livejournal.com]

I'd have more trust in this idea if I believe most programmers were good at providing for the exceptions that every rule needs, or dealing with people in general.

[selenite.livejournal.com]

Programmers don't have to be "good" at that to be better than lawyers.