WEBVTT

1
00:00:00.000 --> 00:00:03.280
When we share our work online, the default
posture is

2
00:00:03.280 --> 00:00:06.720
polish. We present the finished, gleaming
product.

3
00:00:06.720 --> 00:00:09.280
We engage in a performance of competence.

4
00:00:09.280 --> 00:00:11.600
Look at a standard architecture diagram.

5
00:00:11.600 --> 00:00:14.720
It builds systems from the top down,
implying the structure

6
00:00:14.720 --> 00:00:18.240
was flawless. But if you look behind that
diagram, the

7
00:00:18.240 --> 00:00:19.840
actual history is gone.

8
00:00:19.840 --> 00:00:22.960
The polished narrative deletes every wrong
turn.

9
00:00:22.960 --> 00:00:25.760
We are trained to wait until the dust
settles before

10
00:00:25.760 --> 00:00:28.640
we show anyone what we are doing, creating
the impression

11
00:00:28.640 --> 00:00:32.240
that success is a straight line from idea
to execution.

12
00:00:32.240 --> 00:00:35.840
Maintaining that highly curated posture
takes an immense amount of

13
00:00:35.840 --> 00:00:38.960
energy. It is exhausting to present a
version of yourself

14
00:00:38.960 --> 00:00:40.720
that never makes a mistake.

15
00:00:40.720 --> 00:00:43.920
The person reading a clean README learns
exactly what you

16
00:00:43.920 --> 00:00:47.120
built, but because you scrubbed away the
vulnerability, they do

17
00:00:47.120 --> 00:00:48.880
not learn how you built it.

18
00:00:48.880 --> 00:00:50.800
There is an alternative framework.

19
00:00:50.800 --> 00:00:52.880
It is called building in public.

20
00:00:52.880 --> 00:00:55.760
If we peel back the layers on the perfect
project

21
00:00:55.760 --> 00:00:59.600
up top, the actual reality at the bottom
looks incredibly

22
00:00:59.600 --> 00:01:00.000
chaotic.

23
00:01:00.840 --> 00:01:05.160
You stall into the solutions sideways,
hitting three false starts

24
00:01:05.160 --> 00:01:07.320
before making any real progress.

25
00:01:07.320 --> 00:01:10.840
You realize halfway through that your core
dependency is entirely

26
00:01:10.840 --> 00:01:14.680
unmaintained, forcing you to rip out the
foundation and reverse

27
00:01:14.680 --> 00:01:18.120
course entirely. This is why the code
commit log is

28
00:01:18.120 --> 00:01:22.200
so valuable. It provides a raw, unfiltered
history of what

29
00:01:22.200 --> 00:01:23.880
actually happened day to day.

30
00:01:23.880 --> 00:01:26.680
A standard change log gives you a curated
list of

31
00:01:26.680 --> 00:01:29.800
shiny features. The commit log gives you
the sequence of

32
00:01:29.800 --> 00:01:33.560
bad ideas, desperate fixes, and immediate
reversals.

33
00:01:33.560 --> 00:01:36.840
The friction and the chaos documented in
that timeline are

34
00:01:36.840 --> 00:01:38.600
not errors you need to hide.

35
00:01:38.600 --> 00:01:41.160
They are the actual mechanics of learning.

36
00:01:41.160 --> 00:01:44.680
This philosophy requires us to look at our
tools differently.

37
00:01:44.680 --> 00:01:48.040
Take open issues on a project repository,
for example.

38
00:01:48.040 --> 00:01:51.240
Instead of a customer support channel,
they become a public

39
00:01:51.240 --> 00:01:53.320
space for internal debate.

40
00:01:53.320 --> 00:01:54.840
You use that space to argue.

41
00:01:55.120 --> 00:01:58.160
with yourself about what to build next You
think out

42
00:01:58.160 --> 00:02:01.520
loud, right where others can observe the
thought process.

43
00:02:01.520 --> 00:02:03.280
Then there is the colophon.

44
00:02:03.280 --> 00:02:06.480
It is an exhaustive list detailing the
technology stack you

45
00:02:06.480 --> 00:02:10.000
chose, the specific design decisions you
made, and the credits

46
00:02:10.000 --> 00:02:11.520
for who helped you get there.

47
00:02:11.520 --> 00:02:14.640
And finally, there is the practice of
shipping drafts.

48
00:02:14.640 --> 00:02:17.680
This is the deliberate choice to keep your
dead-end experiments

49
00:02:17.680 --> 00:02:19.840
and half-finished projects visible.

50
00:02:19.840 --> 00:02:22.800
You do not sweep abandoned drafts under
the rug, because

51
00:02:22.800 --> 00:02:25.920
showing what fails provides vital context
for anyone trying to

52
00:02:25.920 --> 00:02:27.440
solve the same problem.

53
00:02:27.440 --> 00:02:30.880
Issues, colophons, and drafts provide the
context that a final

54
00:02:30.880 --> 00:02:32.480
product usually hides.

55
00:02:32.480 --> 00:02:35.120
Together, they form a map for any creator
trying to

56
00:02:35.120 --> 00:02:36.880
follow in your footsteps.

57
00:02:36.880 --> 00:02:39.840
Building in public carries a very real
cost.

58
00:02:39.840 --> 00:02:43.600
You expose your absolute worst decisions
before you correct them.

59
00:02:43.600 --> 00:02:46.000
There is a constant fear of losing control
of your

60
00:02:46.000 --> 00:02:49.280
narrative. Someone might look at your code
on the worst

61
00:02:49.280 --> 00:02:53.720
day of a a brutal refactor and form
permanent opinion,

62
00:02:53.720 --> 00:02:57.480
but that pressure fades when you realize
the curated version

63
00:02:57.480 --> 00:02:59.160
was always an invention.

64
00:02:59.160 --> 00:03:02.120
If the smooth launch is a fiction, you
don't have

65
00:03:02.120 --> 00:03:04.200
to spend energy maintaining it.

66
00:03:04.200 --> 00:03:06.680
You are not putting your work out there to
perform

67
00:03:06.680 --> 00:03:08.440
for a crowd of critics.

68
00:03:08.440 --> 00:03:10.920
You are doing it for the version of
yourself six

69
00:03:10.920 --> 00:03:14.520
months ago, the person who couldn't find a
single example

70
00:03:14.520 --> 00:03:17.480
of someone else doing it badly before they
did it

71
00:03:17.480 --> 00:03:21.560
well. Stop wasting your energy on the
illusion of polish.

72
00:03:21.560 --> 00:03:25.000
Redirect that effort into the actual work
and show the

73
00:03:25.000 --> 00:03:29.240
mess. The mess is the only honest map we
have.