Context-fu

Regardless of whether your in a team setting or at a conference, speaking to a senior or a junior, you can reliably assume: 1. you will be misunderstood by them; and 2. you will misunderstand them.

Truth is, everyone is misunderstood and everyone misunderstands.

A healthy habit is to

assume infinite ignorance, and infinite intelligence
(Max Delbrück; or Leó Szilárd; or Stanisław Jerzy Lec; purportedly)

and try to come up with an effective intro that will "collect" 1000 random people to begin to understand the talk you are about to deliver.

Tables¶

Think of your talk as "context-0", enclose it by "context-1", then "context-2" and "context-3".

Context level	Analogy	Num. of people in each context
0	Your talk	\(10^0=1\) (yourself)
1	Your niche topic	\(10^1=10\) (your closest cohort)
2	Room	\(10^2=100\) (people intrigued by the title of your talk)
3	Conference	\(10^3=1000\) (everyone attending the conference)

The goal of your intro is to collect 3 orders of magnitude outside of yourself. And you can do it in under 1 minute.

Diagrams¶

graph LR
subgraph conference["Conference (1000 people)"]
  subgraph room["Room (100 people)"]
    subgraph topic["Niche topic (10 people)"]
      topic-name["How many people will understand your _current_ talk?"]
    end
  end
end

Diagrams are phenomenal. A good diagram is a minimalist diagram. To produce one, you'd have to think hard what is relevant, and how much semantics is required to convey the point.

For example, consider a juju deployment diagram:

graph LR

subgraph application
mysql ---|scrape| grafana-agent
end

subgraph observability
prometheus ---|datasource| grafana
end

grafana-agent ---|remote-write| prometheus

(Edit a copy of this diagram)

In this diagram,

outer boxes are juju models (k8s namespaces)
inner boxes are juju applications (k8s pods)
connecting lines are juju integrations

Y-statements¶

One way to collect people quickly is by explaining a tradeoff. Y-statemets are short-form ADRs and look like this:

In the context of < use case/user story u >,
facing < concern c >,
we decided for < option o₁ >,
and rejected < options o₂, o₃, ..., oₙ >,
to achieve < quality q >,
accepting < downside d >.

Summary¶

Collect, into a Y-statement, providing a diagram, in ~1 minute.