You read her pull request
and felt slightly ill.
Clean commits.
Sensible names.
A test for the case
you would not have thought of
until the week after release.
Meanwhile yours is four days
of going the wrong way,
two abandoned approaches
and a branch you renamed twice
out of embarrassment.
You are comparing
two different kinds of thing.
What you saw was the edit.
She had the same four days.
She tried the approach that did not work.
She rewrote the history
before anybody looked.
She wrote that test
after the case bit her
in a project three years ago
and she has never mentioned that either.
Everyone ships the outcome
and keeps the process.
The talk that was rehearsed nine times.
The library whose first version
is not in the repository at all.
The colleague who just knew
where the problem was,
having spent a Sunday last spring
reading that subsystem
for reasons that had nothing to do
with being impressive.
So you calibrate yourself
against a world
that has been through an editor,
using material
that never was.
The conclusion is always the same,
and it is always wrong.
Everyone else finds this easy.
I am the slow one.
Do not let them see the middle of it.
And that last part
is where the real damage is,
because you then hide your own process,
which removes the only evidence
the next person had.
Two things fix it,
and both are small.
Ask.
Not how it works.
How long it took,
what you tried first,
what you threw away.
People answer that question honestly
far more often than you expect,
and the answers are a relief.
Then say yours out loud.
In review, in standup,
to the person three months in
who thinks you were born fluent.
I spent a day on this
going entirely the wrong way.
You lose nothing
that anybody sensible values,
and you hand somebody
the thing nobody handed you.
Polish is not a property of people.
It is a stage of work,
and you are only ever
watching other people's.
– Serguey Asael Shinder
