Effectiveness vs. Efficiency

These two “E-F-F” words are what drives a lot of what happens in society today. If you want to run a successful business or a successful organization, you’ve gotta find a way to increase your efficiency, while maximizing your effectiveness. But which one is more important? As is often the case in life, it depends on the context.

For me, I’ve often defaulted to trying to be as efficient as I can (mostly). In the last few months, I’ve come to realize that my worshipping at the feet of efficiency might actually be costing me in effectiveness. Huh? Yeah, if you’re too efficient, it might mean that you’re lowering your effectiveness. In my head, I’m imagining one of those graphs where efficiency is plotted on the x-axis and effectiveness on the y-axis. As efficiency goes up, so does effectiveness. However, there’s a point towards the right of the graph where an increase in efficiency leads to a decrease in effectiveness. Becoming more efficient is no longer in yours (or the company’s best interest).

That’s a lot of words — how about an example. Artificial Intelligence and Machine Learning. A recent episode of The Next Big Idea illustrated how you can have a wholly efficient system in AI/ML, but this increase in efficiency is lowering the system’s effectiveness. For example, let’s say you’ve got an algorithm that’s screening out candidates for specific qualities in your hiring process. Using AI/ML can accelerate the efficiency — it can read 1,000’s of applications in the time it would take a human to read a handful of applications. However, the system probably won’t be well-calibrated for racial bias (as has often been showing on numerous occasions). Increase in efficiency, decrease in effectiveness.

I liked the way that Prof. Eberhardt framed her suggestion: “add more friction to the system.” This will lower the efficiency of it, but it will increase the effectiveness. By adding friction, you’ll slow down the processing speed, but the slow down in time will build-in an opportunity to correct misgivings before they become official.

This also reminds me of a Farnam Street article about the difference between speed and velocity:

Velocity and speed are different things. Speed is the distance traveled over time. I can run around in circles with a lot of speed and cover several miles that way, but I’m not getting anywhere. Velocity measures displacement. It’s direction-aware.

A lot of people think in terms of one dimension (speed). Almost all of those people are passed by people who think in multiple dimensions (velocity).

What is “Artificial Intelligence,” Anyway?

Sometimes, I wish I could go back to 1955 and prevent John McCarthy from calling it “artificial intelligence.” It’s a term that, depending upon where you work, you can’t go 5 minutes without hearing once or twice — which is great. It’s great that people are looking to the ‘future.’ It’s great that society is pushing forward with growth and expansion and all that warm and fuzzy stuff. Unfortunately, AI doesn’t really do justice to what it’s describing.

AI isn’t really “artificial” nor is it really “intelligent.” In fact, you could even argue that AI is really really dumb (wait, what?!). Yeah, dumb. Caveat: I’m speaking about the kind of AI that exists in this moment. If scientists can crack artificial general intelligence (i.e. Terminator, Hal, etc.), then, well, then that’s a whole new ballgame. But right, AI, as it exists right now can be thought of as a sort of ‘idiot savant.’ It can do the tasks that we tell it to do and do them extremely well.

Did you catch that? Let me say it again. It can do the tasks that we tell it to do and do the me extremely well. And that right there is the hitch. I can’t leave an AI at your doorstep an expect it to make you dinner. I need to give it some direction (NOTE: this is assuming that there isn’t some AGI out there that hasn’t be released). Maybe I give it a command like “make dinner” or “wash the dishes” and then it follows the rules/algorithms for navigating the space inside your hour or apartment to get to the kitchen, find the fridge (or the sink), and continue forward with its work.

When you think of it that way, that’s not really “intelligent,” is it? Nor is that really “artificial,” is it? And it’s certainly not artificial intelligence. Instead, it’s more like task automation. Granted, it’s a bit more sophisticated than that (any AI expert reading this is probably thinking I’ve lost my marbles), but that’s another thing that’s frustrating about nebulous terms like AI — they mean something very specific to the people that work in that field and to everyone else, it’s jargon. The problem with a term like AI is that the entertainment industry has given us plenty of images of what a fictitious AI might be able to do and so having a reasonable conversation with someone not versed in the particulars on the topic of AI can be daunting.

Circling back to the task automation bit — to set the minds of AI experts at ease — I know, it’s not just task automation. It’s task automation that’s informed by reams of data (even that might get me into trouble with some folks who want to be more specific). That’s what makes it seem like there’s some kind of ‘magic’ at play. So, if the AI at your front door had reams of data about how you load your dishwasher or about how cities of people load their dishwasher or if it knew all the recipes that you might select from, how often you select and on what days, etc. Data. Data is the fuel that pushes the ‘task automation’ forward.

My point in discuss some of the finer points of AI today was not necessarily to get into the weeds of its definition, but more to illustrate that there are terms out there that have a very specific meaning to some folks, but when widely discussed by non-experts, could mean something very different. This reminds me of something I wrote a few years back about “the Economy.” It can mean something very different depending upon to whom you’re talking. For better or worse, AI seems to be one of those phrases and I’m sure it’s not the only one. I’m sure there are others out there. Can you think of any terms in your field that you’ve heard discussed in the popular press that seem to, rightly or wrongly, oversimplify its meaning?

What is Data Science?

There’s no question that “data science” is becoming more and more popular. In fact, Booz Allen Hamilton (a consultancy) found:

The term Data Science appeared in the computer science literature throughout the 1960s-1980s. It was not until the late 1990s, however, that the field as we describe it here, began to emerge from the statistics and data mining communities. Data Science was first introduced as an independent discipline in 2001. Since that time, there have been countless articles advancing the discipline, culminating with Data Scientist being declared the sexiest job of the 21st century.

Unsurprisingly, there are countless graduate and undergraduate programs in data science (Harvard, Berkeley, Waterloo, etc.), but what is data science, exactly?

Given that the field is still in its proverbial infancy, there are a number of different perspectives. Booz Allen offers the following in their Field Guide to Data Science from 2015: “Describing Data Science is like trying to describe a sunset — it should be easy, but somehow capturing the words is impossible.”

Pithiness aside, there does seem to be consensus around some of the pertinent themes contained within data science. For instance, a key component is usually “Big Data” (both unstructured and structured data). Dovetailing with Big Data, “statistics” is often cited as an important component. In particular, an understanding of the science of statistics (hypothesis-testing, etc.), including the ability to manipulate data and almost always — the ability to turn that data into something that non-data scientists can understand (i.e. charts, graphs, etc.). The other big component is “programming.” Given the size of the datasets, Excel often isn’t the best option for interacting with the data. As a result, most data scientists need to have their programming skills up to snuff (often times in more than one language).

What’s a Data Scientist?

Now that we know the three major components of data science are statistics, programming, and data visualization, do you think you could identify data scientists from statisticians, programmers, or data visualization experts? It’s a trick question — they’re all data scientists (broadly speaking).

A few years ago, O’Reilly Media conducted research on data scientists:

Why do people use the term “data scientist” to describe all of these professionals?

[…]

We think that terms like “data scientist,” “analytics,” and “big data” are the result of what one might call a “buzzword meat grinder.” The people doing this work used to come from more traditional and established fields: statistics, machine learning, databases, operations research, business intelligence, social or physical sciences, and more. All of those professions have clear expectations about what a practitioner is able to do (and not do), substantial communities, and well-defined educational and career paths, including specializations based on the intersection of available skill sets and market needs. This is not yet true of the new buzzwords. Instead, ambiguity reigns, leading to impaired communication (Grice, 1975) and failures to efficiently match talent to projects.

So… the ambiguity in understanding the meaning of data science stems from a failure to communicate? Classic movie references aside, the research from O’Reilly identified four main “clusters” of data scientists (and roles within said “clusters”):

Within these clusters fits some of the components described earlier, including two additional components: math/operations research (including things like algorithms and simulations) and business (including things like product development, management, and budgeting). The graphic below demonstrates the t-shaped-nature of data scientists — they have depth of expertise in one area and knowledge of other closely related areas. NOTE: ML is an acronym for machine learning.

NOTE: This post originally appeared on GCconnex.