Trade Offs

A listing of common and interesting trade-offs that are noticeable in a variety of fields.

Exploration vs Exploitation §

Bias Variance Trade Off §

Accuracy / Optimality vs Computational Performance §

• To have a more accurate or optimal solution requires more time and computational resources to solve.
• Conversely, some intractable problems require approximate solutions which can be evaluated at a faster speed at the cost of optimality.
• Applicable in: Machine Learning and Computer Science in general.

Social Reach Trade-Off §

• Probabilistically, the larger the social reach the more likely bad apples will be attracted. Also, the more people you know the harder it is to be intimately connected with one’s community.
• Conversely, the smaller the social reach the more tight-knit the community becomes at the cost of only having a niche reach.
• Applicable in: Social Media influencers / Communities

Freedom of Choice Trade-Off §

• The more degrees of freedom in a system, the more options one has at the cost of needing more effort to understand these options, which themselves might be confusing. Also, there are more avenues for errors.
• Conversely, the less options one has, the less flexible the system feels, but it is much easier to develop a cohesive structure for the system.

The Paradox of Choice §

• While we might believe that being presented with multiple options actually makes it easier to choose something we are happy with, having an abundance of options leaves us requiring more effort in decision making, and can leave us feeling unsatisfied.
• Hick’s Law applies here.

Abstraction Trade-Off §

• The more abstracted something is, the harder it is to understand the underlying internals of a component in the system.
• The less abstracted something is, the more prone it is to breaking. That is, it will end up with potential inconsistencies that need to be handled meticulously.
• Applicable in software development.

Humanization vs Numbers Trade-Off §

• It is easy to humanize someone or a group of people. However, the more people there are the harder it is to empathize with them. At a large scale, people are easily viewed as datapoints rather than individuals with their own stories.
• Then again, at a large scale it is hard to actually process each and every single person’s life.

Forest for the Trees Trade-Off §

• The cost of looking at the large scale is the tendency to forget what is right in front of you.
• The cost of looking at the small scale is the failure to see bigger picture. Perspective is lost within scale.

The Paradox of Technology §

• Technology with more capabilities at some point requires more complexity, which makes them harder to understand, use, and standardize.
• See The Psychopathology of Everyday Things.

Knowledge in the World vs Head Tradeoff §

• Knowledge in the World:
  • Information is readily and easily available whenever perceivable.
  • Requires no learning only interpretation, which is subject to what is being perceived.
  • Slow because it requires finding and interpreting the knowledge.
  • Ease of use at first encounter is high
  • Ugly and inelegant, potentially confounding due to clutter.
• Knowledge in the Head:
  • Material in working memory is readily available, otherwise considerable search is required. Search necessitates remembering there is something to be remembered.
  • Requires possibly considerable learning, made easier for structured or well-understood materials.
  • Efficient especially if well-learned to the point it is automated.
  • Ease of use at first encounter is low.
  • Requires nothing to be visible, leading to no clutter.
• See Knowledge in the Head and In the World

Boring vs Overstimulating Tradeoff §

• Simplicity is pleasing to the eyes, but too much makes the subject feel boring or bland.
• Dynamicity is pleasing to the eyes, but too much makes the subject feel confusing or jarring.
• Boredom is good for ideation. In fact, overstimulation can be considered a hazard

Social Surplus vs Personal Utility Tradeoff §

• An approach to achieving social surplus (i.e., an outcome that benefits society as a whole) may require sacrificing the personal utility of its individuals. In the end, Pareto optimality may leave some people unhappy.
• Conversely, following simple rational behavior may result in happy individuals but a society that is no-better off.

Risk vs Reward Tradeoff §

• The classic tradeoff of “High risk High reward’
• Can be subverted (i.e., high risk low reward) but more often than not prevalent in a lot of scenarios.

Space vs Time Tradeoff §

• This tradeoff applies to computation. On one hand, we can apply the computation as normal, which takes time.
• On the other time, we can reduce computation time if we memorize the answers at the cost of space.

Inventory Tradeoffs §