It's a systemic bias

I’m reviewing the IRS guidelines for 501(c)(3) educational institutions, in preparation for registering my own organization for educational non-profit status. I think I’ve figured out why affirmative action is alive and well:

The IRS mandates something akin to this notice to be sent to a school’s community periodically:

“The M School admits students of any race, color, national and ethnic origin to all the rights, privileges, programs, and activities generally accorded or made available to students at the school. It does not discriminate on the basis of race, color, national and ethnic origin in administration of its educational policies, admissions policies, scholarship and loan programs, and athletic and other school-administered programs.” (IRS Publication 557)

This sounds great! Exactly the sorts of policies I hope to implement, except the IRS forgot to include gender in there (perhaps because athletics are already segregated by gender).

Then they say this:

“The IRS recognizes that the failure by a school drawing its students from local communities to enroll racial minority group students may not necessarily indicate the absence of a racially nondiscriminatory policy when there are relatively few or no such students in these communities. Actual enrollment is, however, a meaningful indication of a racially nondiscriminatory policy in a community in which a public school or schools became subject to a desegregation order of a federal court or are otherwise expressly obligated to implement a desegregation plan under the terms of any written contract or other commitment to which any federal agency was a party.”

And, even worse:

“A policy of a school that favors racial minority groups with respect to admissions, facilities and programs, and financial assistance is not discrimination on the basis of race when the purpose and effect of this policy is to promote establishing and maintaining the school’s nondiscriminatory policy.”

Now the “diversity committees” and differing admissions standards I’ve observed in existing universities start to make more sense. If minorities do not enroll in the school in approximately the same proportion that they exist in the community, the IRS immediately assumes something is wrong (rather than the more likely explanation that the university just doesn’t interest said students for some reason or another). Schools are then pushed into the position of being more lenient to minority students, because they basically have to fill a quota to avoid angering the IRS.

Now, this would be ok if not for some incorrect statistical assumptions. If the sampling of students from the general population were random (and taken exclusively from the local community), the proportions would be about the same as the local community and we’d be more or less fine.

Unfortunately, the sampling is not random. Certain majors will have certain racial compositions; others will have different ones. The majors that the school offers will thus affect the bias of the admissions process of the institution as a whole.

So there you go. Couched in the language of a “nondiscriminatory policy” lies a policy that very much discriminates. While it’s unjust, it’s unfortunately the law, so follow it I will.

"Best Practices"

Most people who use the phrase “Best Practices” probably don’t know what they are. I haven’t really found anyone using these terms who can explain why these practices are truly “best”, and in the realm of software engineering (not so much for web dev.), the code of people who endorse them tends to be awful. They can sometimes spit out why one model is better than another particular model, because they’ve memorized the dogma, but present them with a completely new paradigm and ask them to compare it and they cannot. It seems like they’re substituting process for talent.

The next type of fat that's going to be bad for you to eat.

Following (and indeed, likely due in no small part because of) the trans fat crusade, the next “bad fat” will likely be the interesterified fat resulting from total hydrogenation. Total hydrogenation, unlike partial hydrogenation, produces no trans fats. However, the resulting fatty acid is saturated and, worse, seems to cause a sharp rise in blood sugar. Given that the so-called “metabolic syndrome” consists of hypertension, obesity, and diabetes, it stands to reason that, should these fats become popular, the cause of this syndrome will eventually be laid in no small part at the feet of the interesterification process.

Milestone-based project fees as an alternative to hourly rates

A conversation with a client (who may be reading this now) prompted me to think over my payment structure.

My current rates for consulting are almost all hourly, even though I greatly dislike charging hourly rates: first, they’re predicated on the amount of time I’ve worked rather than what I’ve produced (and I’ve never held myself to a proficiency standard that said mere effort without results was ok). Second, they penalize me for working quickly. Charging by “lines of code” has similar problems, for conciseness instead of time: I can write one line of Perl that does what may take 10 in C++, for instance, but that one line may take as long as the 10 did to write.

I used to charge per-project rates. Unfortunately, aside from the difficulty of estimating an entire project in advance, my past experience has demonstrated why it is a truism never to accept a per-project rate that is less than four times what you think the project is worth: extra features.

It is common for features out of the initial requirements specification to crop up midway through a project. On a fixed budget, you are placed in the difficult position of refusing to implement the features (which annoys your clients), modifying your estimate (which might be difficult or impossible), or performing the extra work for free (which isn’t fair to you).

The model I’m now considering is a milestone-based model, under which milestones are set with particular deadlines and fees. From the time a milestone is agreed upon to the milestone’s completion or deadline, features leading up to it are frozen. Additional features can be requested, but these become part of the next milestone and factor into the cost estimate. This way, the client gets what he wants, you get compensated for the extra features, and your development time remains predictable. Everyone’s happy.

One Key to Leadership

I’ve never been particularly charismatic, but I have a knack for leadership based on ideas – I have the ability to motivate people on the strength of my vision. One important lesson I’ve learned while doing this is that criticism is a positive thing.

Ideas, particularly radical ones, always attract critics. If they attract too many negative criticisms relative to the positive ones, you might wish to rethink the idea, or at least figure out why, but even ideas that have overwhelmingly positive criticisms are bound to get negative ones as well from time to time.

The key is never to take criticism as a bad thing. Use it as an opportunity to improve, and always thank the critic for bringing something you need to address to light. You’ll be surprised how quickly your critics warm to the idea when they see how willing you are to consider and address their concerns.

In the end, you’re both working towards ideals, after all.

Some people change the world while they procrastinate on their dissertations

Both Google and Yahoo! were started by two Ph. D. students who decided they had more important things to complete than their dissertations. I find that funny as I try against all odds to change the very structure of modern academia rather than working on my dissertation 🙂

(Of course, my dissertation’s basically done already. Writing it was never the problem.)

Seriously, though, I think that there’s something about writing a dissertation that makes you realize there are more important things in life.