1
00:00:00,000 --> 00:00:03,100
The box is a style of working. 
It's about creating a 

2
00:00:03,100 --> 00:00:05,600
collaborative environment 
between the development team and

3
00:00:05,600 --> 00:00:09,900
the operations team and aligning
goals and incentives between 

4
00:00:09,900 --> 00:00:12,000
those two teams. 
Because when you think about it,

5
00:00:12,100 --> 00:00:15,800
so many of the problems that we 
encounter in life not just even 

6
00:00:15,800 --> 00:00:18,100
in technology is a misalignment 
of goals. 

7
00:00:22,700 --> 00:00:26,000
Hey everyone. 
My name is Henry Surya be Robin.

8
00:00:27,800 --> 00:00:31,400
And you're listening to the 
tekhelet Juno, the show will be 

9
00:00:31,400 --> 00:00:34,700
bringing you the greatest 
technical leaders practitioners 

10
00:00:34,900 --> 00:00:38,300
and thought leaders in the 
industry to discuss about their 

11
00:00:38,300 --> 00:00:43,100
Journey ideas and practices that
we all can learn and apply to 

12
00:00:43,100 --> 00:00:46,100
build a highly performing 
technical team and to make an 

13
00:00:46,100 --> 00:00:50,700
impact in your personal work. 
So let's dive into our Journal. 

14
00:00:55,900 --> 00:00:57,700
Hello, everyone. 
This is Henry. 201. 

15
00:00:58,700 --> 00:01:00,900
Welcome to another episode of 
the technology, you know, 

16
00:01:00,900 --> 00:01:03,200
podcast. 
Thank you for tuning in and 

17
00:01:03,200 --> 00:01:05,800
spending your time with me 
today, listening to this 

18
00:01:05,800 --> 00:01:08,200
episode. 
If you haven't, please follow 

19
00:01:08,200 --> 00:01:11,500
technology, you know, on your 
podcast app and social media on 

20
00:01:11,500 --> 00:01:14,900
LinkedIn, Twitter. 
And Instagram also consider 

21
00:01:14,900 --> 00:01:17,700
supporting the show by 
subscribing as a patron at 

22
00:01:17,700 --> 00:01:21,300
technology, you know, dot, f / 
Patron, and support me to 

23
00:01:21,300 --> 00:01:23,900
continue producing, great 
content every week. 

24
00:01:24,800 --> 00:01:28,000
There were UPS as a culture and 
practice is one of the most 

25
00:01:28,000 --> 00:01:30,900
widely talked about in the 
current technology landscape 

26
00:01:31,300 --> 00:01:34,900
according to the state of devops
report Elite and high devops 

27
00:01:34,900 --> 00:01:36,600
performing. 
Companies are leading the 

28
00:01:36,600 --> 00:01:40,100
industries in terms of 
organizational performance and 

29
00:01:40,100 --> 00:01:43,000
continue to outperform the 
companies that do not practice 

30
00:01:43,000 --> 00:01:45,800
devops optimally. 
And some of the reasons 

31
00:01:45,800 --> 00:01:49,300
companies do not practice devops
optimally are due to the 

32
00:01:49,300 --> 00:01:53,300
misconception of the whole Dev, 
Ops concept as a culture and 

33
00:01:53,300 --> 00:01:56,100
also some anti patterns. 
That may get adopted 

34
00:01:56,100 --> 00:01:59,700
unconsciously to shed more 
lights on devops culture and 

35
00:01:59,700 --> 00:02:01,700
practices. 
For today's episode. 

36
00:02:01,700 --> 00:02:05,000
I'm happy to share my 
conversation with Jeffrey Smith.

37
00:02:05,400 --> 00:02:08,800
Jeffrey is the author of a book 
titled operations and the 

38
00:02:08,800 --> 00:02:12,200
patterns, develop Solutions, and
he's also the director of 

39
00:02:12,200 --> 00:02:16,400
production operations at Centro.
In this episode Jeffrey, 

40
00:02:16,400 --> 00:02:20,000
describe devops Essentials and 
importantly a size. 

41
00:02:20,000 --> 00:02:23,600
What devops is not? 
He also broke down and explain 

42
00:02:23,600 --> 00:02:25,700
about cams. 
S-see AMS. 

43
00:02:26,100 --> 00:02:29,400
A framework that outlines the 
core components required for a 

44
00:02:29,400 --> 00:02:31,400
successful develops 
transformation. 

45
00:02:31,900 --> 00:02:35,100
We then discuss three anti 
patterns that are taken from his

46
00:02:35,100 --> 00:02:37,800
book. 
The paternity syndrome alert 

47
00:02:37,800 --> 00:02:41,100
fatigue and wasting a perfectly 
good incident. 

48
00:02:41,700 --> 00:02:44,900
Jeffrey explained, how to 
recognize those anti patterns 

49
00:02:45,200 --> 00:02:48,100
and how we can avoid them on our
devops journey. 

50
00:02:48,500 --> 00:02:52,000
And finally, Jeffrey, also 
talked about post-mortem and he 

51
00:02:52,000 --> 00:02:55,800
shared some tips on how we can 
cultivate a Post-mortem culture.

52
00:02:56,600 --> 00:02:58,600
This is such a fun conversation 
with Jeffrey. 

53
00:02:58,600 --> 00:03:01,800
And I really, really enjoyed it.
Especially discussing about 

54
00:03:01,800 --> 00:03:04,800
those three devops and the 
patterns that I could personally

55
00:03:04,800 --> 00:03:07,800
relate from my own experience. 
And I believe that you would 

56
00:03:07,800 --> 00:03:11,000
highly enjoy and learn a lot 
from this episode as well. 

57
00:03:11,400 --> 00:03:14,500
And if you do consider helping 
the show by giving it a rating 

58
00:03:14,500 --> 00:03:18,000
and review on your podcast app 
or share some comments on the 

59
00:03:18,008 --> 00:03:21,400
social media channels, those 
reviews and comments are one of 

60
00:03:21,400 --> 00:03:24,400
the best ways to help me get 
this podcast to reach more. 

61
00:03:24,400 --> 00:03:26,700
Listeners. 
And hopefully, they will also 

62
00:03:26,700 --> 00:03:29,500
benefit from all the contents in
this podcast. 

63
00:03:29,900 --> 00:03:33,600
So let's start our episode right
after our short sponsor message.

64
00:03:33,800 --> 00:03:36,900
Are you looking for a new cool 
swag taglit Journal. 

65
00:03:36,900 --> 00:03:40,000
Now offers you some swags that 
you can purchase online. 

66
00:03:40,400 --> 00:03:44,300
These wax are printed on demand 
based on your preference and 

67
00:03:44,300 --> 00:03:47,100
will be delivered safely to you 
all over the world where 

68
00:03:47,100 --> 00:03:50,200
shipping is available. 
Check out all the cool swag is 

69
00:03:50,200 --> 00:03:53,000
available by visiting 
technology, know the dev slash 

70
00:03:53,000 --> 00:03:55,900
shop, and don't forget. 
The bracket self once you 

71
00:03:55,900 --> 00:04:01,600
receive any of those tracks. 
Hey everyone, welcome back to 

72
00:04:01,608 --> 00:04:04,000
another 50 Journal podcast show.
Today. 

73
00:04:04,000 --> 00:04:06,100
I have with me an author named 
Jeffrey. 

74
00:04:06,100 --> 00:04:08,600
Smith. 
Jeffrey is an author of the book

75
00:04:08,600 --> 00:04:12,000
called operations and the 
patterns develop Solutions. 

76
00:04:12,100 --> 00:04:15,000
So I guess it's like using 
develop solutions to overcome 

77
00:04:15,000 --> 00:04:17,899
those anti patterns. 
So today we will be talking a 

78
00:04:17,907 --> 00:04:21,200
lot about what are some of the 
anti patterns, for operations, 

79
00:04:21,200 --> 00:04:24,300
people, or in the devops world. 
We're going to be talking about 

80
00:04:24,400 --> 00:04:27,100
Some of the things that maybe 
you should try to avoid in your 

81
00:04:27,100 --> 00:04:31,200
operations team and how we can 
actually do humor strategies 

82
00:04:31,200 --> 00:04:34,400
that actually can improve how 
you manage your systems, how you

83
00:04:34,400 --> 00:04:37,100
operate your systems? 
And so that using the devops 

84
00:04:37,100 --> 00:04:39,500
solutions we can overcome those 
things. 

85
00:04:39,800 --> 00:04:41,800
So thank you so much for 
spending your time with me 

86
00:04:41,800 --> 00:04:42,700
today. 
Jeffrey. 

87
00:04:43,000 --> 00:04:44,800
Hope you have a good 
conversation with you today. 

88
00:04:45,400 --> 00:04:46,800
Awesome here. 
Thanks for having me and I'm 

89
00:04:46,800 --> 00:04:49,400
looking forward to it. 
So Jeffrey in the beginning, 

90
00:04:49,400 --> 00:04:52,100
maybe if you can introduce 
yourself telling us more about 

91
00:04:52,100 --> 00:04:54,300
your career, any highlights or 
turning points. 

92
00:04:54,900 --> 00:04:56,500
Sure. 
Yeah, so as you mentioned, my 

93
00:04:56,500 --> 00:04:58,400
name is Jeff Smith. 
I'm currently the director of 

94
00:04:58,400 --> 00:05:01,000
production operations on a 
company called Centro in 

95
00:05:01,000 --> 00:05:03,100
Chicago, Illinois here, the 
United States. 

96
00:05:03,300 --> 00:05:06,100
We are A rad Tech platform, 
which didn't sound. 

97
00:05:06,100 --> 00:05:07,400
Like the most exciting thing to 
me. 

98
00:05:07,400 --> 00:05:10,500
When I originally took the job 
ad Tech is like, the bane of the

99
00:05:10,500 --> 00:05:13,200
internet, but in reality, it is,
what fuels the internet and keep

100
00:05:13,200 --> 00:05:15,300
so much of it, free. 
I'll talk a little bit about 

101
00:05:15,300 --> 00:05:17,600
more about that later. 
But prior to that, I was with a 

102
00:05:17,600 --> 00:05:19,600
company called GrubHub, and 
they're in the food. 

103
00:05:19,600 --> 00:05:22,900
Delivery space, my time to grow 
up, was really when I got first 

104
00:05:22,900 --> 00:05:25,600
introduced to devops and do It's
Concepts. 

105
00:05:25,900 --> 00:05:28,100
It was the first place. 
I could really sink my teeth 

106
00:05:28,100 --> 00:05:31,000
into it and experiment in my 
career, really started. 

107
00:05:31,000 --> 00:05:34,200
Twenty odd years ago back in 
Upstate New York where I grew 

108
00:05:34,200 --> 00:05:38,200
up, I had been working doing 
data entry at a local health 

109
00:05:38,200 --> 00:05:40,900
insurance provider in the area 
that I'd always been interested 

110
00:05:40,900 --> 00:05:43,800
in computers, but I was a 
terrible high school student 

111
00:05:43,900 --> 00:05:46,200
gentleman, who was the manager 
of operations? 

112
00:05:46,200 --> 00:05:49,000
Who would later become my mentor
walked by and saw me reading 

113
00:05:49,000 --> 00:05:51,500
Richard Stevens TCP IP 
illustrated book. 

114
00:05:51,500 --> 00:05:53,200
I don't know if you know that 
book, but it's like the 

115
00:05:53,200 --> 00:05:55,500
networking Bible, what? 
Didn't know at the time. 

116
00:05:55,500 --> 00:05:57,800
Was that book? 
Was way over my head. 

117
00:05:57,800 --> 00:06:00,100
I was reading it and I was 
getting some of it, but it was 

118
00:06:00,100 --> 00:06:01,800
pretty dense stuff. 
He saw. 

119
00:06:01,800 --> 00:06:03,800
And reached out was like, hey, 
what do you read in there? 

120
00:06:03,800 --> 00:06:06,500
We talked a little bit about it,
and we would have conversations 

121
00:06:06,500 --> 00:06:09,500
every now and then eventually an
operations position to opened up

122
00:06:09,500 --> 00:06:12,700
his team and he said, hey, would
you like to switch careers? 

123
00:06:12,700 --> 00:06:15,000
Stop doing data entry and do 
some computer stuff. 

124
00:06:15,000 --> 00:06:17,500
So I was like, yeah sure. 
So that's really what kicked off

125
00:06:17,500 --> 00:06:20,900
this 20-year romance with tech. 
I'll say so I was at that 

126
00:06:20,900 --> 00:06:23,500
company for about 10 years. 
I worked my way up to operations

127
00:06:23,500 --> 00:06:26,200
manager eventually. 
Seating him, you know, it's one 

128
00:06:26,200 --> 00:06:29,300
of those moments where you're 
like, wow is every place is 

129
00:06:29,300 --> 00:06:30,400
messed up. 
Is this one? 

130
00:06:30,600 --> 00:06:33,500
Maybe I need to Branch out and 
see what else is out there. 

131
00:06:33,700 --> 00:06:35,300
My wife, what? 
Girlfriend, at the time. 

132
00:06:35,300 --> 00:06:37,500
Now, wife said, why don't we 
open up a search? 

133
00:06:37,700 --> 00:06:40,400
Why don't we look anywhere 
instead of just looking in our 

134
00:06:40,400 --> 00:06:42,000
local area? 
So that's what brought us to 

135
00:06:42,000 --> 00:06:44,700
Chicago. 
I got a job in Chicago switch to

136
00:06:44,700 --> 00:06:47,000
a couple different companies. 
Had some success doing 

137
00:06:47,000 --> 00:06:49,800
performance tuning for a company
called Accenture and one of 

138
00:06:49,800 --> 00:06:53,000
their like lab environments, but
that was right during the start 

139
00:06:53,000 --> 00:06:55,700
of the financial downturn. 
I'd been brought out as a 

140
00:06:55,707 --> 00:06:58,600
contract to hire and then as 
soon as I got brought on the 

141
00:06:58,600 --> 00:07:01,800
financial crisis hit so they 
weren't really hiring and they 

142
00:07:01,800 --> 00:07:03,900
kept extending me. 
Extending the extended me. 

143
00:07:03,900 --> 00:07:05,100
And then finally, I said, you 
know what? 

144
00:07:05,100 --> 00:07:08,200
I need health insurance, right? 
Like I'm walking around here. 

145
00:07:08,200 --> 00:07:10,700
I bumped it away from Financial 
ruin. 

146
00:07:10,900 --> 00:07:13,800
So I started looking for a new 
opportunity in the day. 

147
00:07:13,800 --> 00:07:16,900
I got a new job was the day. 
They told me that they were 

148
00:07:16,900 --> 00:07:18,600
going to be able to extend my 
contract anymore. 

149
00:07:18,600 --> 00:07:19,900
And I was like, oh that's funny 
because I was gonna have the 

150
00:07:19,900 --> 00:07:21,300
city meeting with you to tell 
you. 

151
00:07:21,300 --> 00:07:23,700
I was leaving. 
So everyone was of the but, you 

152
00:07:23,700 --> 00:07:26,400
know, work the couple Jobs 
before getting to the pub. 

153
00:07:26,600 --> 00:07:29,400
I think GrubHub was a huge 
turning point one. 

154
00:07:29,400 --> 00:07:31,700
It was the first time I'd work 
for a company that actually made

155
00:07:31,700 --> 00:07:33,400
a product. 
I cared about and you never 

156
00:07:33,400 --> 00:07:35,200
really think about how important
that is. 

157
00:07:35,400 --> 00:07:37,700
So often you're just stuck in 
this field. 

158
00:07:37,700 --> 00:07:39,100
If you think about it 
technologist, a really 

159
00:07:39,100 --> 00:07:41,400
mercenaries, they don't really 
care about the field that 

160
00:07:41,400 --> 00:07:42,800
they're in. 
They're just like what language 

161
00:07:42,800 --> 00:07:44,600
you're my coding in the field or
whatever. 

162
00:07:44,600 --> 00:07:46,200
The company does is secondary 
to. 

163
00:07:46,200 --> 00:07:48,600
That was the first time I worked
at a company that actually made 

164
00:07:48,600 --> 00:07:51,400
a product that I've cared about 
the really fueled my interest in

165
00:07:51,400 --> 00:07:54,500
how things truly work, not just 
from a computer. 

166
00:07:54,600 --> 00:07:56,300
Active. 
But from a business perspective 

167
00:07:56,300 --> 00:07:58,100
as well. 
And that gave me all types of 

168
00:07:58,100 --> 00:08:00,100
insights that I was lying to 
before. 

169
00:08:00,300 --> 00:08:02,600
Because, you know, before that, 
it was like, I don't care about 

170
00:08:02,600 --> 00:08:04,800
health insurance, right? 
Your life, the inner workings of

171
00:08:04,800 --> 00:08:06,800
health insurance. 
I think care about legal tax 

172
00:08:06,800 --> 00:08:09,100
software, which I was 
responsible for running in a 

173
00:08:09,100 --> 00:08:11,500
company called Wolters kluwer. 
I just didn't care about those 

174
00:08:11,500 --> 00:08:12,500
things. 
So I didn't have a vested 

175
00:08:12,500 --> 00:08:14,300
interest in understanding how 
they work. 

176
00:08:14,500 --> 00:08:17,200
Being a grub hub and being 
curious about how things work. 

177
00:08:17,200 --> 00:08:20,800
Open me up to this world of 
possibility when it comes to 

178
00:08:20,800 --> 00:08:24,300
operations, so I carried that 
experience with the into Centro 

179
00:08:24,300 --> 00:08:27,200
and First, I didn't care about 
how I attack worked. 

180
00:08:27,300 --> 00:08:29,900
I knew that it was important and
could be career change and to 

181
00:08:29,900 --> 00:08:32,600
really understand the business. 
And I'm glad I did because that 

182
00:08:32,600 --> 00:08:36,100
took is way more fascinating as 
much as it is problematic from 

183
00:08:36,100 --> 00:08:38,299
the various different 
perspectives, but it is 

184
00:08:38,299 --> 00:08:41,200
definitely needed in the 
internet age because people 

185
00:08:41,200 --> 00:08:43,600
don't want to pay for Facebook. 
It's the trade-off. 

186
00:08:43,600 --> 00:08:47,000
So how do we do that and do that
in a respectful manner for our 

187
00:08:47,000 --> 00:08:51,800
customers clients and users. 
So that's a quick recap of 20 

188
00:08:51,800 --> 00:08:54,500
plus years of experience, but 
it's been a fun ride. 

189
00:08:55,100 --> 00:08:57,200
Thanks for sharing. 
Your story is very interesting. 

190
00:08:57,200 --> 00:09:01,000
How you started your journey? 
So through this book, TCP IP. 

191
00:09:01,100 --> 00:09:03,400
I have to agree that. 
Yeah, that book is also way over

192
00:09:03,400 --> 00:09:06,400
my head. 
It's a great book. 

193
00:09:06,400 --> 00:09:09,400
That is like absolutely Bible. 
But back then, at that time. 

194
00:09:09,400 --> 00:09:13,600
I was like a good prop was a 
really good problem. 

195
00:09:13,800 --> 00:09:16,900
So over these 20 years Journey. 
I'm sure you have seen a lot of 

196
00:09:16,900 --> 00:09:19,200
things right in the beginning. 
You also said that the messy 

197
00:09:19,200 --> 00:09:22,500
things of operations 
administration's, and now, in 

198
00:09:22,500 --> 00:09:25,400
this era of devops. 
Maybe you can If I first of all 

199
00:09:25,400 --> 00:09:28,400
the audience and listeners hear 
what is actually devops. 

200
00:09:29,100 --> 00:09:31,500
So the first thing I'm going to 
describe is what is not. 

201
00:09:31,500 --> 00:09:34,100
It is not a role. 
It is not a job title. 

202
00:09:34,300 --> 00:09:36,100
That's a major pet. 
Peeve of mine. 

203
00:09:36,200 --> 00:09:38,500
I think it might even be a 
little detrimental in my hiring 

204
00:09:38,500 --> 00:09:40,100
at first because people are 
searching for that. 

205
00:09:40,100 --> 00:09:43,100
Devops rolled pops is a style of
working. 

206
00:09:43,200 --> 00:09:45,500
It's about creating a 
collaborative environment 

207
00:09:45,500 --> 00:09:49,100
between the development team and
the operations team and aligning

208
00:09:49,100 --> 00:09:53,000
goals and incentives between 
those two teams, you know, we 

209
00:09:53,000 --> 00:09:54,500
say, Dev and Ops and focus on 
those team. 

210
00:09:54,600 --> 00:09:57,100
So it really can be any grouping
of Team. 

211
00:09:57,100 --> 00:10:00,200
It could be Devon Finance, 
option, finance, options, 

212
00:10:00,200 --> 00:10:03,000
security, and we keep coming up 
with all these acronyms, that's 

213
00:10:03,000 --> 00:10:04,700
a cop's Finn. 
It's just like, let's just call 

214
00:10:04,700 --> 00:10:06,900
the devops. 
We understand what the idea is. 

215
00:10:07,100 --> 00:10:10,200
I always say you would never 
post the position for an agile 

216
00:10:10,200 --> 00:10:11,900
engineer. 
That sounds insane. 

217
00:10:12,100 --> 00:10:14,200
You would never do that because 
actually the style of working, 

218
00:10:14,400 --> 00:10:17,600
same thing with that mops. 
How do we build cohesive joint 

219
00:10:17,600 --> 00:10:20,900
incentives? 
For teams to work towards a 

220
00:10:20,900 --> 00:10:24,500
common goal and putting that 
sort of front and center. 

221
00:10:24,900 --> 00:10:28,200
Shape a lot of your underlying 
decisions moving forward because

222
00:10:28,200 --> 00:10:31,100
when you think about it, so many
of the problems that we 

223
00:10:31,100 --> 00:10:34,100
encounter in life. 
Not just even in technology is a

224
00:10:34,100 --> 00:10:37,800
misalignment of goals. 
You set that it's not a role, 

225
00:10:37,800 --> 00:10:40,100
right, but I agree from this 
part of the world. 

226
00:10:40,100 --> 00:10:42,500
I keep seeing also devops 
engineer role. 

227
00:10:42,700 --> 00:10:45,100
What do you think should be the 
name of the title? 

228
00:10:45,100 --> 00:10:48,400
Is it like a sorry, as some 
people actually chose to prefer?

229
00:10:49,000 --> 00:10:54,200
I think that many system 
Engineers system administrators.

230
00:10:54,600 --> 00:10:56,500
Whatever you want to call. 
I think there are enough people 

231
00:10:56,500 --> 00:10:59,500
that have been in these proles 
previously better doing the same

232
00:10:59,500 --> 00:11:01,800
things that we're doing today. 
It's just with a different tool 

233
00:11:01,800 --> 00:11:03,700
set, right? 
When we made the switch from 

234
00:11:03,700 --> 00:11:05,800
Unix to Linux. 
We didn't really have to come up

235
00:11:05,800 --> 00:11:08,200
with a new job description. 
It was the same deal. 

236
00:11:08,200 --> 00:11:10,900
It was just, you know, now 
there's a new tool when we make 

237
00:11:10,900 --> 00:11:15,400
the switch from python to go. 
We don't come up with a new job.

238
00:11:15,400 --> 00:11:18,600
Title is just an asterisks under
the software developer title. 

239
00:11:18,800 --> 00:11:21,900
I think devops is the same. 
I think we have plenty of titles

240
00:11:21,900 --> 00:11:25,300
that could be used. 
I'm not naive to Could forces 

241
00:11:25,300 --> 00:11:28,700
though and I know that if you 
have a have Ops title, it's 

242
00:11:28,700 --> 00:11:32,900
probably an extra 10 or 15 
thousand dollars a year. 

243
00:11:33,200 --> 00:11:35,600
So I don't begrudge anyone using
that. 

244
00:11:35,600 --> 00:11:38,900
In fact, I tell my team that I 
manage if you want to call 

245
00:11:38,900 --> 00:11:41,300
yourself a devops engineer on 
LinkedIn, that's fine. 

246
00:11:41,300 --> 00:11:43,100
I'll pack you. 
But that's not what we're called

247
00:11:43,100 --> 00:11:44,200
here. 
Because the minute you give 

248
00:11:44,200 --> 00:11:46,500
someone a title like that. 
It becomes their job. 

249
00:11:46,600 --> 00:11:48,600
The minute. 
You have a QA team. 

250
00:11:48,900 --> 00:11:52,200
Guess whose job quality is 
Right? 

251
00:11:53,100 --> 00:11:57,100
Language is a very Finicky thing
because it does so much to the 

252
00:11:57,100 --> 00:12:00,400
way we perceive the world just 
by giving something a name or 

253
00:12:00,400 --> 00:12:02,300
title. 
That's why I try to avoid it so 

254
00:12:02,300 --> 00:12:04,700
that everyone knows is like this
devops things were talking 

255
00:12:04,700 --> 00:12:08,400
about, is everybody's job and we
need to think about that too. 

256
00:12:08,400 --> 00:12:11,500
From a times perspective, right?
Because we can quickly fall into

257
00:12:11,500 --> 00:12:14,900
a trap where we're like all 
we're building operations stuff.

258
00:12:14,900 --> 00:12:18,500
Nobody else needs to see this. 
Well that kind of runs counter 

259
00:12:18,500 --> 00:12:20,300
to this collaborative 
environment that we're talking 

260
00:12:20,300 --> 00:12:22,300
about. 
So maybe we do need to make sure

261
00:12:22,300 --> 00:12:25,100
that anyone and everyone can see
that maybe You to make sure 

262
00:12:25,100 --> 00:12:27,500
we're keeping the bodies out of 
the Ops Code so that we don't 

263
00:12:27,500 --> 00:12:30,400
have to lock down the repo. 
Some configuration file that 

264
00:12:30,400 --> 00:12:33,300
they're like, we can't share 
that with anyone because there's

265
00:12:33,300 --> 00:12:35,300
too much dirt in there. 
There's just way too many dead 

266
00:12:35,300 --> 00:12:37,100
bodies. 
We've got to keep that under 

267
00:12:37,100 --> 00:12:39,600
tight lock by putting that 
mindset forward. 

268
00:12:39,600 --> 00:12:41,800
Hopefully, to helps to prevent 
and eliminate those sorts of 

269
00:12:41,800 --> 00:12:43,700
traps. 
Thanks for giving that 

270
00:12:43,700 --> 00:12:45,900
clarification. 
So, in your view, I think you 

271
00:12:45,900 --> 00:12:47,500
mentioned is in the book as 
well. 

272
00:12:47,500 --> 00:12:49,300
There are few things that are 
very important. 

273
00:12:49,300 --> 00:12:51,800
When you want to adopt this 
devops culture, right? 

274
00:12:51,900 --> 00:12:54,400
This new style of working. 
So it's an acronym. 

275
00:12:54,500 --> 00:12:57,900
I'm called camps or some people 
actually call it calms. 

276
00:12:58,000 --> 00:13:02,200
See am SOC a LMS. 
So maybe you can give us a light

277
00:13:02,200 --> 00:13:04,000
here. 
What is cams? 

278
00:13:04,700 --> 00:13:09,000
So current is really dislike 
framework around devops as a 

279
00:13:09,000 --> 00:13:12,300
concept to think about the core 
components that you need in 

280
00:13:12,300 --> 00:13:15,100
order to make this 
transformation C is for culture.

281
00:13:15,400 --> 00:13:17,900
The A is for automation. 
The AL that you alluded to that.

282
00:13:17,900 --> 00:13:21,100
Some people call. 
It is lean metrics and sharing 

283
00:13:21,100 --> 00:13:24,400
or the lasted. 
So culture is like one soul. 

284
00:13:24,600 --> 00:13:28,400
Much of devops is cultural and 
you need to build a cultural 

285
00:13:28,400 --> 00:13:31,500
environment in your organization
where these sort of practices 

286
00:13:31,500 --> 00:13:36,000
and concepts are embraced, where
people are free to experiment, 

287
00:13:36,000 --> 00:13:39,100
without fear of Retribution. 
If they get something wrong, 

288
00:13:39,300 --> 00:13:43,100
there are all these small. 
Insidious things that happen in 

289
00:13:43,100 --> 00:13:45,700
our culture. 
That especially us as engineers 

290
00:13:45,700 --> 00:13:48,000
and technologists might gloss 
over. 

291
00:13:48,200 --> 00:13:51,800
Not realizing what an outsized 
impact that they have. 

292
00:13:52,000 --> 00:13:54,400
Have you ever worked for a 
company where the culture? 

293
00:13:54,500 --> 00:13:57,300
ER, is if you don't use your 
budget by the end of the year, 

294
00:13:57,300 --> 00:13:59,100
you lose it, and you won't get 
it next year. 

295
00:13:59,300 --> 00:14:01,800
That's a cultural thing. 
It doesn't have to be that way. 

296
00:14:01,800 --> 00:14:05,200
There is no golden Finance rule.
That's just the way people 

297
00:14:05,200 --> 00:14:06,700
operate. 
So what does that do? 

298
00:14:06,900 --> 00:14:09,500
That creates a culture? 
Where everyone is spending 

299
00:14:09,500 --> 00:14:11,900
aimlessly? 
They're spending a lessly at the

300
00:14:11,900 --> 00:14:14,600
end of the year to make sure 
that they use that budget and 

301
00:14:14,600 --> 00:14:16,700
I'm sure the people in finance 
are like sweet. 

302
00:14:16,700 --> 00:14:18,400
That's exactly what we wanted to
happen. 

303
00:14:18,600 --> 00:14:22,100
But you created that culture 
through a small rule change. 

304
00:14:22,100 --> 00:14:24,300
So culture is just something 
that we always have to keep 

305
00:14:24,300 --> 00:14:25,500
from. 
It's better when we're talking 

306
00:14:25,500 --> 00:14:29,100
about devops automation is 
really the thing that powers 

307
00:14:29,100 --> 00:14:30,800
this. 
One of the things is you're 

308
00:14:30,800 --> 00:14:32,400
always going to be asked to do 
more with less. 

309
00:14:32,700 --> 00:14:35,700
But the other thing that we need
to do is the more automation 

310
00:14:35,700 --> 00:14:38,400
that you have. 
The more empowerment. 

311
00:14:38,400 --> 00:14:40,800
You can give to other people in 
the organization. 

312
00:14:41,000 --> 00:14:42,300
It's this thing that in the 
book. 

313
00:14:42,300 --> 00:14:45,600
I talk about this idea of 
exporting expertise. 

314
00:14:45,900 --> 00:14:51,000
So it is a technical dance to 
fail over a database. 

315
00:14:51,300 --> 00:14:53,200
I don't know if you've ever had 
to fail over production 

316
00:14:53,200 --> 00:14:55,700
database, but like, everyone's 
been in that company where it's 

317
00:14:55,700 --> 00:14:57,900
all, we got to feel the database
over, get bob bobs. 

318
00:14:57,900 --> 00:14:59,400
The only one that can feel the 
database server. 

319
00:14:59,500 --> 00:15:02,500
The minute, Bob turns it into a 
script where someone just has to

320
00:15:02,500 --> 00:15:04,500
execute fail underscore 
database. 

321
00:15:04,700 --> 00:15:08,000
Bob has transferred, a large 
chunk of his expertise into this

322
00:15:08,000 --> 00:15:10,200
automation script and now he's 
empowered. 

323
00:15:10,200 --> 00:15:12,400
Tens of people that can do this 
thing. 

324
00:15:12,600 --> 00:15:15,200
So automation is key. 
I'll talk about lean briefly, 

325
00:15:15,200 --> 00:15:17,000
but I'm not a huge fan of adding
the elk. 

326
00:15:17,200 --> 00:15:20,300
Lean is just about operating 
dynamically and quickly. 

327
00:15:20,400 --> 00:15:22,700
I don't know that it falls in 
the category of devops because 

328
00:15:22,700 --> 00:15:26,300
typically, you're adopting a All
that is similar to the 

329
00:15:26,300 --> 00:15:28,500
organization. 
So like if you're forced to 

330
00:15:28,500 --> 00:15:30,900
adopt lean but during a 
waterfall shop that can be 

331
00:15:30,900 --> 00:15:35,100
problematic metrics, devops 
should be rooted in data as part

332
00:15:35,100 --> 00:15:37,100
of that empowerment. 
People need to know if the thing

333
00:15:37,100 --> 00:15:39,000
that they're doing is actually 
having an impact. 

334
00:15:39,200 --> 00:15:41,800
We always talk about it with 
systems and computers and 

335
00:15:41,800 --> 00:15:43,600
whatnot. 
But really extends everywhere. 

336
00:15:43,600 --> 00:15:45,700
Right? 
As we are running, our ticket 

337
00:15:45,700 --> 00:15:48,900
queue is actually performing the
way we want to perform and what 

338
00:15:48,900 --> 00:15:50,700
metrics are? 
We judging that success or 

339
00:15:50,700 --> 00:15:53,300
failure by? 
We should be able to objectively

340
00:15:53,300 --> 00:15:55,500
point to success. 
Or failure and we do that 

341
00:15:55,500 --> 00:15:57,300
through metrics and then 
sharing. 

342
00:15:57,500 --> 00:16:01,200
You know, how do we again back 
to that expertise exporting? 

343
00:16:01,300 --> 00:16:05,600
How do we share knowledge? 
How do we share access when 

344
00:16:05,600 --> 00:16:08,700
appropriate? 
How do we share responsibility? 

345
00:16:08,900 --> 00:16:11,000
How do we make sure that we're 
all sort of on the hook for the 

346
00:16:11,000 --> 00:16:13,600
same thing and we're all 
contributing to make that thing 

347
00:16:13,600 --> 00:16:15,500
better. 
Another thing that's sort of 

348
00:16:15,500 --> 00:16:17,400
front and center. 
So these things sort of collapse

349
00:16:17,400 --> 00:16:20,700
together to build this framework
and how we should think about 

350
00:16:20,800 --> 00:16:22,800
approaching these devops 
Transformations. 

351
00:16:23,000 --> 00:16:26,400
And when you look at, Nations 
that aren't doing well in their 

352
00:16:26,400 --> 00:16:29,500
transformation. 
You can usually find behaviors 

353
00:16:29,500 --> 00:16:32,400
or actions that tie to one of 
these five categories. 

354
00:16:32,600 --> 00:16:35,400
I keep leaving out lean. 
I'm sorry for all the lien fans.

355
00:16:35,500 --> 00:16:37,500
I'm not, hey, let me pay you in 
a little bit. 

356
00:16:39,200 --> 00:16:41,400
So thanks for sharing. 
I really like the concept 

357
00:16:41,400 --> 00:16:44,700
exporting the experts. 
So these things about automation

358
00:16:44,700 --> 00:16:46,800
sharing your knowledge. 
I think it's a great thing 

359
00:16:46,800 --> 00:16:48,600
especially in this devops 
culture. 

360
00:16:48,800 --> 00:16:52,200
Let's go into the anti patterns 
that you covered a lot in the 

361
00:16:52,208 --> 00:16:54,300
books. 
I think they are probably 10. 

362
00:16:54,500 --> 00:16:58,000
No, 12 kind of anti patterns so 
we can start with the favorite 

363
00:16:58,000 --> 00:17:00,300
one, probably. 
But in the beginning, let's 

364
00:17:00,300 --> 00:17:03,500
probably dive into why you wrote
this book. 

365
00:17:03,500 --> 00:17:05,599
Why are you covering 
anti-patterns? 

366
00:17:06,300 --> 00:17:07,800
Sure. 
So it's a funny story. 

367
00:17:08,000 --> 00:17:11,099
The whole journey book was 
actually a long winding road. 

368
00:17:11,400 --> 00:17:14,700
So Manning had reached out to me
years ago to write a book on 

369
00:17:14,700 --> 00:17:16,500
puppet. 
I was pretty active in the 

370
00:17:16,500 --> 00:17:19,099
public community at that time. 
At least from a user 

371
00:17:19,099 --> 00:17:21,099
perspective. 
You know, I thought about it was

372
00:17:21,099 --> 00:17:23,000
like, yeah, sure. 
But then before we got things 

373
00:17:23,000 --> 00:17:26,200
started, my first kid was born. 
So wasn't the ideal time to 

374
00:17:26,200 --> 00:17:28,600
write a book. 
So then years later they reached

375
00:17:28,600 --> 00:17:30,900
out and they were like, hey, 
curious, if you'd be interested 

376
00:17:30,900 --> 00:17:34,400
in writing a book on devops and 
I was like, I don't know, what a

377
00:17:34,400 --> 00:17:37,200
book on devops would be like, 
but then when I thought about 

378
00:17:37,200 --> 00:17:39,800
it, I was like, you know what, I
want a book that is practical 

379
00:17:39,800 --> 00:17:43,900
advice for people that aren't in
glossy startups. 

380
00:17:43,900 --> 00:17:47,700
That take so much for granted 
when you read a lot of these 

381
00:17:47,700 --> 00:17:50,900
books either their companies 
that are firmly rooted in 

382
00:17:50,900 --> 00:17:53,000
technology. 
And the fact that everyone is 

383
00:17:53,000 --> 00:17:56,100
sort of bought in is a Ian and 
that's not the reality that most

384
00:17:56,100 --> 00:17:58,500
people face, most people are in 
a company that's got a lot of 

385
00:17:58,500 --> 00:18:01,700
Legacy baggage. 
They've been around for 40, 50 

386
00:18:01,700 --> 00:18:04,100
years. 
They've got all this inertia 

387
00:18:04,100 --> 00:18:07,000
around bad practices that we 
don't really talk about in the 

388
00:18:07,000 --> 00:18:09,200
devops books. 
We just simply say stop doing 

389
00:18:09,200 --> 00:18:11,500
what you're doing and start 
doing this and it's like, okay, 

390
00:18:11,500 --> 00:18:14,000
easier said than done. 
The other thing was a lot of the

391
00:18:14,000 --> 00:18:17,200
books seem like they were 
written for ctOS, right? 

392
00:18:17,200 --> 00:18:20,400
Because you would need either a 
CTO or CT 0 by in to be able to 

393
00:18:20,400 --> 00:18:22,800
do a lot of the things that 
they're talking about and I was 

394
00:18:22,800 --> 00:18:24,800
like, well, there's got to be 
stuff that and Visual 

395
00:18:24,800 --> 00:18:27,800
contributor or line manager or 
something can do because I feel 

396
00:18:27,800 --> 00:18:30,600
like I've done it a couple times
and I'm not a CTO. 

397
00:18:30,900 --> 00:18:34,000
So that was really what sort of 
got me interested in writing 

398
00:18:34,000 --> 00:18:36,500
this book and it was originally 
titled devops for the rest of 

399
00:18:36,500 --> 00:18:40,000
us, but people felt that was an 
exclusionary Us Versus Them 

400
00:18:40,000 --> 00:18:42,400
thing and I hadn't thought about
it from that perspective and I 

401
00:18:42,408 --> 00:18:44,800
was like, yeah kind of hard to 
talk about a book about, you 

402
00:18:44,800 --> 00:18:47,600
know, bringing everyone together
and then immediately setting up 

403
00:18:47,600 --> 00:18:50,600
as an US versus them. 
So the first draft of the book 

404
00:18:50,600 --> 00:18:53,500
was not an anti patterns format.
Actually. 

405
00:18:53,800 --> 00:18:58,200
This is my first first book, I 
was very academic and formal and

406
00:18:58,200 --> 00:19:00,600
I got a piece of feedback in our
review process. 

407
00:19:00,600 --> 00:19:03,600
So what happens is with Manning,
you'll do a third of the book 

408
00:19:03,700 --> 00:19:06,300
and then you release it to a 
bunch of potential buyers that 

409
00:19:06,300 --> 00:19:07,800
will review it and give you 
feedback. 

410
00:19:07,900 --> 00:19:09,700
So we were doing that. 
With the first third of the 

411
00:19:09,700 --> 00:19:13,200
book, someone made a comment and
they said I've seen Jeff speak 

412
00:19:13,200 --> 00:19:15,900
and I was really excited to see 
this book because he's such a 

413
00:19:15,900 --> 00:19:19,600
fun energetic speaker and that 
personality does not translate 

414
00:19:19,600 --> 00:19:24,300
into this book at all. 
And I was like, wow, okay, so 

415
00:19:24,500 --> 00:19:27,100
All right, and my editor was 
like, I don't know you well 

416
00:19:27,100 --> 00:19:29,400
because I'd never really met 
miters who's like, but it sounds

417
00:19:29,400 --> 00:19:32,800
like this book, is you right 
now, so that's how we reworked 

418
00:19:32,800 --> 00:19:34,300
it into this anti-patterns 
thing. 

419
00:19:34,300 --> 00:19:36,900
So I could have a little fun 
with it, get more of my own 

420
00:19:36,900 --> 00:19:39,200
voice into it. 
That's how it everything sort of

421
00:19:39,200 --> 00:19:41,700
took shape, but it was 
definitely not intended to be an

422
00:19:41,700 --> 00:19:43,500
anti patterns book when it 
started. 

423
00:19:43,600 --> 00:19:47,100
That was the result of a pretty 
cold but honest piece of 

424
00:19:47,100 --> 00:19:51,400
feedback that I got. 
So I hope when we go through all

425
00:19:51,400 --> 00:19:54,200
these and Tibetans, you can also
use some of your font styles. 

426
00:19:54,400 --> 00:19:56,600
Going forward for that. 
So let's start with the first 

427
00:19:56,700 --> 00:19:59,800
anti patterns, which is called 
paternalist syndrome. 

428
00:20:00,000 --> 00:20:03,800
What is this actually, so I 
called paternalist syndrome. 

429
00:20:04,000 --> 00:20:08,400
It is when someone in a 
relationship assumes the role of

430
00:20:08,400 --> 00:20:11,800
the parent, even though no true 
hierarchy exists. 

431
00:20:11,900 --> 00:20:14,500
And you know who I'm talking 
about, Ops people, raise your 

432
00:20:14,500 --> 00:20:17,300
hand. 
We're all guilty of this the 

433
00:20:17,300 --> 00:20:20,300
bof, for those that don't know. 
Go ahead and Google it, but 

434
00:20:20,300 --> 00:20:22,900
there's a reputation of 
operation seems being the team 

435
00:20:22,900 --> 00:20:26,300
of know and I Instead, I spent 
my entire career not hop. 

436
00:20:26,300 --> 00:20:31,300
So I totally have all of that, 
baggage from years of devs, just

437
00:20:31,300 --> 00:20:34,500
throwing stuff into production, 
but I quickly was able to 

438
00:20:34,500 --> 00:20:37,400
identify that the reason again, 
was the fact that we had 

439
00:20:37,400 --> 00:20:40,500
misaligned incentives. 
So the paternalist syndrome, is 

440
00:20:40,500 --> 00:20:44,100
this behavior that operations 
people, anyone can do, if I'm 

441
00:20:44,100 --> 00:20:46,900
really talking to the office 
looks right now, this Behavior 

442
00:20:46,900 --> 00:20:50,900
where we assume everyone else is
out to destroy the system, and 

443
00:20:50,900 --> 00:20:54,000
we have to protect it. 
That means nothing can go to 

444
00:20:54,000 --> 00:20:55,500
production. 
Action without it, coming 

445
00:20:55,500 --> 00:20:58,400
through us first. 
That means no one can have 

446
00:20:58,400 --> 00:21:01,600
access to production. 
No matter how scene of a request

447
00:21:01,600 --> 00:21:05,000
it is you just can't do it 
because, you know, we can list a

448
00:21:05,000 --> 00:21:08,200
whole bunch of reasons and will 
invoke security as like a red 

449
00:21:08,200 --> 00:21:10,500
herring as to why you can't have
access. 

450
00:21:10,800 --> 00:21:14,100
Well invoke audit controls as a 
red herring as to why you can't 

451
00:21:14,100 --> 00:21:17,500
have access and sometimes those 
are true, but we're really 

452
00:21:17,500 --> 00:21:20,700
starting from the position of no
and defending that instead of 

453
00:21:20,700 --> 00:21:24,100
starting from the position of. 
Yes, but yes, but how do we 

454
00:21:24,100 --> 00:21:25,000
address? 
The audit. 

455
00:21:25,100 --> 00:21:28,600
Yes, but how do we ensure that 
access doesn't spread? 

456
00:21:28,600 --> 00:21:31,700
And it's really a mindset. 
So what the paternalist syndrome

457
00:21:31,700 --> 00:21:34,500
will do, is anytime that there's
a problem. 

458
00:21:34,700 --> 00:21:38,000
The first tool we reach in the 
toolbox, is another gate, 

459
00:21:38,000 --> 00:21:40,400
another checkpoint. 
Oh, this didn't go through 

460
00:21:40,400 --> 00:21:43,600
change control or this went 
through change control, but Ops 

461
00:21:43,600 --> 00:21:45,500
wasn't on the Change Control 
process. 

462
00:21:45,700 --> 00:21:47,700
So now Ops has to approve every 
change in. 

463
00:21:47,700 --> 00:21:51,200
Every time we do that. 
We're adding another layer of 

464
00:21:51,200 --> 00:21:54,200
slowdowns, very little value. 
Add work. 

465
00:21:54,300 --> 00:21:57,700
And waste of time for Ops folks 
to there was a book that I read 

466
00:21:57,700 --> 00:22:00,900
rework by chasing freed from 
37signals. 

467
00:22:01,200 --> 00:22:02,600
And it's one of my favorite 
quotes. 

468
00:22:02,600 --> 00:22:05,600
I use it all the time. 
Policies are organizational scar

469
00:22:05,600 --> 00:22:09,100
tissue and every time that 
there's a policy that's enacted.

470
00:22:09,100 --> 00:22:13,100
You could typically go back to 
some inciting incident where 

471
00:22:13,100 --> 00:22:16,400
that happened and the policy 
came out as a result of that, 

472
00:22:16,600 --> 00:22:19,100
almost to the point where it 
becomes lore and Legend, where 

473
00:22:19,100 --> 00:22:20,600
no one was even there for that 
anymore. 

474
00:22:20,600 --> 00:22:21,900
Right? 
It's like, oh, yeah. 

475
00:22:21,900 --> 00:22:24,200
There's some dude on life 
support in the lunchroom. 

476
00:22:24,400 --> 00:22:26,800
We keep him around just so he 
can explain to us. 

477
00:22:26,800 --> 00:22:29,200
Why we don't do stored 
procedures anymore. 

478
00:22:29,400 --> 00:22:30,900
There's always some story like 
that. 

479
00:22:31,100 --> 00:22:33,800
So the pattern listener on 
chapter talks about how we can 

480
00:22:33,800 --> 00:22:37,100
identify some of the common 
issues that we run into and 

481
00:22:37,100 --> 00:22:40,600
started the pattern of fear, 
honestly, because that's what it

482
00:22:40,600 --> 00:22:42,000
is. 
Fear around particular 

483
00:22:42,000 --> 00:22:45,000
situations and how we can go 
about breaking that down using 

484
00:22:45,000 --> 00:22:47,400
cams. 
How do we do that through 

485
00:22:47,400 --> 00:22:49,700
automation? 
So that we can Empower someone 

486
00:22:50,200 --> 00:22:54,200
perfect example as we don't let 
developers restart background. 

487
00:22:54,300 --> 00:22:56,500
Select Services, right? 
So why not? 

488
00:22:56,600 --> 00:22:59,800
They know way better than I do. 
How the system is going to 

489
00:22:59,808 --> 00:23:03,100
behave because they wrote it. 
So why wouldn't they be able to 

490
00:23:03,108 --> 00:23:05,100
reset it if they're seeing 
something weird? 

491
00:23:05,200 --> 00:23:07,700
Maybe we don't want to give them
SSH access to production. 

492
00:23:07,700 --> 00:23:09,500
That's fair. 
But there's a lot of different 

493
00:23:09,500 --> 00:23:12,100
ways that we can restart a 
service and then we can expose 

494
00:23:12,100 --> 00:23:15,200
those ways to the developer and 
Empower them to be able to do it

495
00:23:15,300 --> 00:23:18,400
all but then we don't know. 
Okay, and your program can send 

496
00:23:18,400 --> 00:23:20,100
an email, right? 
That's not crazy. 

497
00:23:20,300 --> 00:23:23,000
So if you really want to know, 
if you're not approving and you 

498
00:23:23,000 --> 00:23:25,800
just want to be notified. 
Have your script send an email 

499
00:23:25,800 --> 00:23:28,700
saying, hey so-and-so just 
restarted the service through 

500
00:23:28,700 --> 00:23:31,100
this access point. 
Boom, done. 

501
00:23:31,400 --> 00:23:34,100
And the other thing with the 
paternal syndrome list, we often

502
00:23:34,200 --> 00:23:38,100
insert ourselves as Gatekeepers 
when we're adding zero value to 

503
00:23:38,100 --> 00:23:40,000
the process. 
So I'll give you an example. 

504
00:23:40,000 --> 00:23:42,400
A real world example, from 
Central when I started at 

505
00:23:42,400 --> 00:23:45,400
Centro. 
There were a lot of requests for

506
00:23:45,500 --> 00:23:48,000
ad hoc script execution from 
development. 

507
00:23:48,000 --> 00:23:51,900
Hey, I need this Ruby script run
so we would have to get the 

508
00:23:51,900 --> 00:23:54,200
script copy it out to one of the
boxes. 

509
00:23:54,700 --> 00:23:58,800
SSH into the box, run the 
script, and no commit mode share

510
00:23:58,800 --> 00:24:01,200
that with the developer or he 
would look at it or she would 

511
00:24:01,200 --> 00:24:04,100
look at it and say, okay, it 
looks good and commit mode. 

512
00:24:04,100 --> 00:24:06,600
We'd run in commit mode and then
I'd send them the output. 

513
00:24:06,700 --> 00:24:09,500
What value do we add to this 
process? 

514
00:24:09,600 --> 00:24:12,700
Other than being a middleman? 
What value are we addict? 

515
00:24:12,800 --> 00:24:15,400
And the truth is 0? 
Because even if we wanted to be 

516
00:24:15,400 --> 00:24:17,700
part of the approval process, I 
don't know anything about the 

517
00:24:17,700 --> 00:24:19,800
script. 
I didn't write any of this code.

518
00:24:19,900 --> 00:24:23,400
Okay, you're updating flights to
make sure the campaign IDs 

519
00:24:23,400 --> 00:24:24,200
match. 
I did. 

520
00:24:24,300 --> 00:24:25,200
You know what? 
That means? 

521
00:24:25,200 --> 00:24:28,400
You just said a bunch of words 
that I'm going to take at face 

522
00:24:28,400 --> 00:24:29,700
value that. 
This is important. 

523
00:24:29,700 --> 00:24:33,000
So, we said, what if we were to 
set this up via gyro, maybe we 

524
00:24:33,000 --> 00:24:36,200
could create a change process 
where someone attaches a script 

525
00:24:36,200 --> 00:24:38,700
to the jira ticket. 
The Jura ticket, has to be 

526
00:24:38,700 --> 00:24:42,300
approved by another Dev who is 
way more qualified to approve it

527
00:24:42,300 --> 00:24:44,700
than I am. 
And once it approved, we have 

528
00:24:44,700 --> 00:24:47,300
some automation crab the script 
from the cheer, a ticket and 

529
00:24:47,300 --> 00:24:50,400
executed on the box and then 
attach the output to the ticket.

530
00:24:50,600 --> 00:24:53,800
Now, you don't even need us now.
There's not a Dev sitting around

531
00:24:53,800 --> 00:24:56,300
waiting like, My goodness is Ops
back from lunch yet. 

532
00:24:56,300 --> 00:24:58,500
I really need to run this script
but no one's there. 

533
00:24:58,800 --> 00:25:00,200
They can just do that on their 
own. 

534
00:25:00,300 --> 00:25:02,400
They self-police it, but he 
self-govern it. 

535
00:25:02,700 --> 00:25:04,400
Now everything is not without 
trade-offs. 

536
00:25:04,500 --> 00:25:07,600
So now you have the issue of 
like, oh, we don't have to fix 

537
00:25:07,600 --> 00:25:09,600
this problem because any time it
comes up with just run the 

538
00:25:09,600 --> 00:25:12,100
script and because there's no 
friction anymore. 

539
00:25:12,100 --> 00:25:15,500
A permanent fix is probably not 
as attractive to them. 

540
00:25:15,500 --> 00:25:18,300
Whereas before they had the pain
of going through Ops to be able 

541
00:25:18,300 --> 00:25:21,400
to entice them to do that. 
But still all in all it's a 

542
00:25:21,408 --> 00:25:22,900
boom. 
It's a win with some 

543
00:25:22,900 --> 00:25:26,000
standardized scripts with they 
read and able to push that 

544
00:25:26,000 --> 00:25:28,400
functionality down to customer 
service. 

545
00:25:28,600 --> 00:25:31,200
So, the customer service is 0. 
When you see this problem, you 

546
00:25:31,200 --> 00:25:34,300
have to submit this gyro ticket 
and then once it's approved, 

547
00:25:34,300 --> 00:25:37,000
you'll be able to execute this 
script to clean it up. 

548
00:25:37,300 --> 00:25:39,600
So the paternalist Ingram is 
really about changing your 

549
00:25:39,600 --> 00:25:43,000
mindset getting out of the habit
of just being no and shifting to

550
00:25:43,000 --> 00:25:45,600
a guess. 
But yes, but how do we solve for

551
00:25:45,600 --> 00:25:48,700
these problems? 
Because no one wants to feel 

552
00:25:48,700 --> 00:25:51,800
like they don't take their job. 
Seriously. 

553
00:25:52,000 --> 00:25:54,800
Everyone at work is doing their 
best with the I was at, they 

554
00:25:54,800 --> 00:25:58,600
have to do their job and this 
idea that there's a group that 

555
00:25:58,700 --> 00:26:01,100
instantly assumes you're an 
idiot is not a warm fuzzy 

556
00:26:01,100 --> 00:26:03,800
feeling for anyone. 
So, yeah, I assume mentioned all

557
00:26:03,800 --> 00:26:07,400
these stories like operations, 
people assuming the role of a 

558
00:26:07,400 --> 00:26:09,500
parent like it's what they 
always say. 

559
00:26:09,500 --> 00:26:13,400
So the Ops job is to actually 
make the system safe, mix the 

560
00:26:13,408 --> 00:26:16,700
system secure, stable. 
Whatever that is while the other

561
00:26:16,700 --> 00:26:19,000
parts of the company or the 
other parts of the team, like 

562
00:26:19,000 --> 00:26:23,400
death is actually out there to 
introduce changes in stability, 

563
00:26:23,400 --> 00:26:25,300
like what you mentioned. 
It is like they are there to 

564
00:26:25,300 --> 00:26:27,900
destroy the company. 
So I think when you mention all 

565
00:26:27,900 --> 00:26:30,100
this anti-patterns, it's 
resonate a lot with the 

566
00:26:30,100 --> 00:26:32,000
traditional world. 
Yeah, there's something else 

567
00:26:32,000 --> 00:26:34,800
that when you said make the 
system save, you know, that's an

568
00:26:34,800 --> 00:26:37,400
interesting way to phrase it 
because I completely agree. 

569
00:26:37,600 --> 00:26:40,500
Think about the system that 
we're running this on as like a 

570
00:26:40,500 --> 00:26:43,500
huge tool box. 
Basically, what we're saying is 

571
00:26:43,500 --> 00:26:46,600
the only tool we have is a bunch
of sharp knives. 

572
00:26:46,900 --> 00:26:49,100
We don't want to give you access
to it because we're afraid 

573
00:26:49,100 --> 00:26:52,000
you're going to cut yourself and
it's like, well, okay, maybe we 

574
00:26:52,000 --> 00:26:53,500
could throw some different tools
and right. 

575
00:26:53,500 --> 00:26:56,400
And if you get through our Marin
maybe we could throw a spoon, 

576
00:26:56,400 --> 00:26:58,500
something. 
That's a little safer that still

577
00:26:58,500 --> 00:27:01,700
allows me to do the job. 
But yes, if you're asking me to 

578
00:27:01,700 --> 00:27:05,800
hammer a nail with a machete, 
this probably unsafe and that's 

579
00:27:05,800 --> 00:27:08,100
essentially what we're doing. 
Like, wow, because we've only 

580
00:27:08,100 --> 00:27:10,000
got machetes. 
I can't give anyone access. 

581
00:27:10,500 --> 00:27:12,700
So, yeah, for all the listeners,
who listen to this. 

582
00:27:12,700 --> 00:27:16,600
I hope you notice this patterns 
or empty patents in your team. 

583
00:27:16,600 --> 00:27:19,700
Make sure that you don't assume 
this parental role 

584
00:27:19,700 --> 00:27:22,400
unnecessarily. 
So let's move on to maybe the 

585
00:27:22,400 --> 00:27:26,600
next anti pattern which is quite
Amin for any administrators 

586
00:27:26,600 --> 00:27:29,200
operations people, which is 
alert fatigue. 

587
00:27:29,300 --> 00:27:31,900
So many times there are so many 
alerts are popping up. 

588
00:27:32,000 --> 00:27:34,300
Probably, we don't do actions on
most of them. 

589
00:27:34,600 --> 00:27:36,400
Can you explain a little bit 
more about alert? 

590
00:27:36,400 --> 00:27:38,800
Fatigue? 
Solar fatigue is actually a term

591
00:27:38,800 --> 00:27:40,300
borrowed from the medical 
industry. 

592
00:27:40,600 --> 00:27:45,100
It came about from nurses who 
would not respond to beeping 

593
00:27:45,100 --> 00:27:49,200
alarms and hospitals, because 
the alarms always go off, they 

594
00:27:49,200 --> 00:27:51,800
always go off. 
So, they became desensitized to 

595
00:27:51,800 --> 00:27:55,600
it to the point where even when 
there was Emergency, there was 

596
00:27:55,600 --> 00:28:00,400
no way to elevate that emergency
beep beyond the cacophony of 

597
00:28:00,400 --> 00:28:04,300
sounds that was always going off
from these machines. 

598
00:28:04,500 --> 00:28:07,900
So we borrow that term alert 
fatigue and Technology to say 

599
00:28:07,900 --> 00:28:12,400
like, what alert do we have that
are just constantly firing that 

600
00:28:12,400 --> 00:28:16,900
are drowning out more, critical 
useful actionable alerts. 

601
00:28:17,100 --> 00:28:20,100
I think the key word there is 
actionable, when we design 

602
00:28:20,100 --> 00:28:22,700
alarms. 
We design alarms from the 

603
00:28:22,700 --> 00:28:26,100
perspective of Things we think 
might be bad. 

604
00:28:26,100 --> 00:28:29,300
We said, oh well alert on high 
CPU utilization because that 

605
00:28:29,300 --> 00:28:31,300
sounds bad, but it isn't really 
bad. 

606
00:28:31,400 --> 00:28:33,600
We buy these machines to use 
them. 

607
00:28:33,800 --> 00:28:36,900
So this idea that we're worried 
that a machine is certainly 

608
00:28:36,900 --> 00:28:40,900
forty or fifty percent utilized.
That's not really a bad thing. 

609
00:28:41,100 --> 00:28:44,600
Let's say it is 90% utilize do 
we care? 

610
00:28:44,800 --> 00:28:49,600
What are the other factors CPU 
utilization on its own is not a 

611
00:28:49,600 --> 00:28:52,600
reason for concern, at least not
to wake someone up. 

612
00:28:52,800 --> 00:28:54,100
So I think we've all gotten 
those. 

613
00:28:54,300 --> 00:28:57,400
Wordsworth like database 
utilization is high and it's 

614
00:28:57,400 --> 00:28:59,600
3:00 a.m. 
We're running all of our batch 

615
00:28:59,600 --> 00:29:01,400
processing. 
Like that makes sense to me. 

616
00:29:01,400 --> 00:29:04,000
But why am I being woken up? 
And then the other thing to 

617
00:29:04,000 --> 00:29:08,700
think about is if you can't 
design an alert that leads up 

618
00:29:08,700 --> 00:29:12,400
engineer to take a next step or 
action, you need to seriously 

619
00:29:12,400 --> 00:29:14,300
question the value of that 
alert. 

620
00:29:14,700 --> 00:29:19,800
So when we have an alert that 
says replication, lag time is 

621
00:29:19,800 --> 00:29:21,700
high. 
You should be able to say in 

622
00:29:21,700 --> 00:29:24,000
that alert with this is actual 
alert that we have. 

623
00:29:24,200 --> 00:29:28,300
The alert will say replication 
slot, 1 has exceeded 

624
00:29:28,300 --> 00:29:30,800
replication. 
Time chances are, this is 

625
00:29:30,800 --> 00:29:34,500
related to the database 
replication service, being run 

626
00:29:34,500 --> 00:29:37,400
by the Bia team. 
You should investigate that 

627
00:29:37,400 --> 00:29:41,500
database and see if that's the 
cause of the replication lag. 

628
00:29:41,600 --> 00:29:45,800
If it is restart the division 
connector in order to catch it 

629
00:29:45,800 --> 00:29:48,000
up. 
That's a very specific set of 

630
00:29:48,000 --> 00:29:51,200
actions where I don't have to do
a bunch of things if that's not 

631
00:29:51,200 --> 00:29:53,100
the case, if that's not what's 
going on. 

632
00:29:53,100 --> 00:29:55,900
And it's like, well, clearly 
When we wrote this alert, we had

633
00:29:55,900 --> 00:29:58,400
a very specific set of scenarios
that we were worried about and 

634
00:29:58,400 --> 00:30:00,400
it's outside the boundary of 
that. 

635
00:30:00,600 --> 00:30:02,400
Maybe I should look deeper into 
this. 

636
00:30:02,700 --> 00:30:05,600
So this idea of alert fatigue is
this idea of making sure that 

637
00:30:05,600 --> 00:30:09,300
these alerts are actionable and 
if they're not actionable rid of

638
00:30:09,300 --> 00:30:12,500
it just get rid of it because 
it's not helpful. 

639
00:30:12,600 --> 00:30:16,900
Now, I'm assuming you've been on
call before and I'm assuming not

640
00:30:16,900 --> 00:30:20,100
casting any blame on anyone but 
I would imagine that you've 

641
00:30:20,100 --> 00:30:22,700
probably received an alert that 
you said. 

642
00:30:22,800 --> 00:30:25,600
Oh this alert usually clear. 
Has itself, let me snooze it for

643
00:30:25,600 --> 00:30:27,800
15 minutes. 
Everyone's done it. 

644
00:30:28,000 --> 00:30:31,100
Everyone's done it. 
So the question is why not just 

645
00:30:31,100 --> 00:30:34,100
increase the threshold of the 
alert and people like, oh, well 

646
00:30:34,100 --> 00:30:37,000
then if there's something wrong,
I won't know for an extra 15 

647
00:30:37,000 --> 00:30:39,100
minutes. 
Well, you don't know anyways, 

648
00:30:39,100 --> 00:30:41,700
because your first action is 
always the student for 15 

649
00:30:41,700 --> 00:30:42,900
minutes. 
You don't know? 

650
00:30:42,900 --> 00:30:46,300
Anyways, because that's the very
first thing you do is like, oh 

651
00:30:46,308 --> 00:30:49,800
man, that's stupid memory thing.
We always know that when the 

652
00:30:49,800 --> 00:30:52,400
code gets into this particular 
section against memory high and 

653
00:30:52,400 --> 00:30:56,000
it clears itself, so it's like 
Not really doing yourself any 

654
00:30:56,000 --> 00:30:59,900
service by having it. 
So push it 15 minutes when it 

655
00:30:59,900 --> 00:31:03,500
alerts you need to react quickly
because you're already behind 

656
00:31:03,500 --> 00:31:05,200
the eight ball. 
But guess what? 

657
00:31:05,500 --> 00:31:07,600
You're reacting and you know, 
it's real. 

658
00:31:07,900 --> 00:31:11,100
I would rather be a few minutes 
late to an alert. 

659
00:31:11,100 --> 00:31:15,400
But no, it's real then to be 
constantly alerted and not being

660
00:31:15,400 --> 00:31:18,400
sure if it's accurate or not and
having to figure that out and 

661
00:31:18,400 --> 00:31:22,100
decide because I know as human 
beings, we're going to err on 

662
00:31:22,100 --> 00:31:25,100
the side of the pattern and just
say, I'm going to do this. 

663
00:31:25,300 --> 00:31:27,600
If you're at a barbecue, you're 
having a friend. 

664
00:31:27,600 --> 00:31:30,200
You're out with your friends, 
having a cookout, you're eating 

665
00:31:30,200 --> 00:31:32,700
the sun's out. 
It's a beautiful day, your alarm

666
00:31:32,700 --> 00:31:34,800
goes off, and it's like all 
except memory alert again. 

667
00:31:35,000 --> 00:31:36,300
Never ever. 
Ever, ever again. 

668
00:31:36,300 --> 00:31:38,600
Be like, guys. 
I gotta go this thing that 

669
00:31:38,600 --> 00:31:41,800
alerts every 20 days or whatever
is alerting again, and I got to 

670
00:31:41,800 --> 00:31:43,300
look at it. 
No, you're going to snooze it. 

671
00:31:43,300 --> 00:31:44,900
You're going to continue 
chatting with your friends. 

672
00:31:45,000 --> 00:31:47,000
And then you crap your pants 
when you realize it's a real 

673
00:31:47,000 --> 00:31:48,200
alert and you've got to do 
something. 

674
00:31:48,600 --> 00:31:51,800
So alert fatigue has really 
focused on identifying those 

675
00:31:51,800 --> 00:31:54,100
patterns and trying to make them
better. 

676
00:31:54,500 --> 00:31:57,700
Or just simply eliminating them.
I think another big thing that 

677
00:31:57,700 --> 00:32:00,700
the chapter talks about it as 
well, is creating metrics that 

678
00:32:00,700 --> 00:32:03,700
reflect a business impact going 
back to the CPU. 

679
00:32:03,700 --> 00:32:07,300
Utilization example, if the 
database is at 90% CPU 

680
00:32:07,300 --> 00:32:10,600
utilization, but our 
transactions per second is 

681
00:32:10,600 --> 00:32:13,200
steady and isn't climbing, do I 
care? 

682
00:32:13,400 --> 00:32:17,000
I don't care work it. 
Yeah, sure. 90% utilization now,

683
00:32:17,000 --> 00:32:19,200
that's not to say that you don't
want the metric, right? 

684
00:32:19,200 --> 00:32:21,900
You just don't want the alert 
because the metric is good for 

685
00:32:21,900 --> 00:32:24,100
trending capacity planning, all 
of these. 

686
00:32:24,300 --> 00:32:26,000
Great things. 
I'm just saying I don't need to 

687
00:32:26,000 --> 00:32:28,500
know about a capacity planning 
alert at night at 3:30 in the 

688
00:32:28,508 --> 00:32:30,000
morning. 
That can be an email. 

689
00:32:30,400 --> 00:32:31,700
And that's another thing too. 
Right? 

690
00:32:31,700 --> 00:32:35,000
We always default to waking 
someone up as the default alert.

691
00:32:35,000 --> 00:32:37,500
There can be different types of 
alert, have a low priority alert

692
00:32:37,500 --> 00:32:39,400
that emails you so you wake up 
in the morning. 

693
00:32:39,400 --> 00:32:42,100
And you say, oh wow. 
We were at high CPU utilization 

694
00:32:42,100 --> 00:32:44,500
last night. 
Nothing else was impacted, but 

695
00:32:44,500 --> 00:32:46,000
that's a good data point to 
know. 

696
00:32:46,100 --> 00:32:49,100
And I'm much more receptive to 
that data point this morning at 

697
00:32:49,100 --> 00:32:51,300
9:00 a.m. 
Well, I've got my coffee as 

698
00:32:51,300 --> 00:32:53,300
opposed to three in the morning 
when I don't know what I'm 

699
00:32:53,308 --> 00:32:56,100
really looking. 
We're trying to solve so get rid

700
00:32:56,100 --> 00:32:59,200
of alerts that don't mean 
anything to you tweak your alert

701
00:32:59,200 --> 00:33:02,000
notification settings so that 
you can do emails instead of 

702
00:33:02,000 --> 00:33:05,200
always paging out, try to tie 
your alerts to some sort of 

703
00:33:05,200 --> 00:33:07,500
business impact. 
So that you know, whether you 

704
00:33:07,500 --> 00:33:10,600
really need to wake someone up 
or not or if they'll or that's 

705
00:33:10,600 --> 00:33:12,900
firing is something that you 
actually care about, because 

706
00:33:12,900 --> 00:33:16,900
again, I like my databases busy 
as long as they are within their

707
00:33:16,900 --> 00:33:20,100
operating thresholds. 
So you mentioned something that 

708
00:33:20,100 --> 00:33:23,600
I pick interest which is it's 
okay to actually be alerted 

709
00:33:23,600 --> 00:33:25,900
late. 
Inside of always have fin first 

710
00:33:25,900 --> 00:33:27,800
minute, right? 
You got others popping up here 

711
00:33:27,800 --> 00:33:31,000
and there because some alerts do
actually recover by itself. 

712
00:33:31,200 --> 00:33:33,800
Because of this anti patterns 
for sure, people just put in 

713
00:33:33,800 --> 00:33:36,400
alerts, but actually they could 
recover over time in a short 

714
00:33:36,400 --> 00:33:38,800
period of time. 
But what you're saying here is 

715
00:33:38,800 --> 00:33:41,400
that it's okay to actually be 
alerted late. 

716
00:33:41,400 --> 00:33:44,200
As long as you can guarantee the
actually it's a real problem and

717
00:33:44,200 --> 00:33:46,100
you are supposed to take an 
action on it. 

718
00:33:46,100 --> 00:33:49,200
So I think it suggests that I 
think for everyone here who has 

719
00:33:49,200 --> 00:33:51,700
been in operation or still 
working in operations. 

720
00:33:51,800 --> 00:33:55,000
You should probably tweak your 
alerts in order to Behave much 

721
00:33:55,000 --> 00:33:57,200
more properly. 
Yeah, absolutely. 

722
00:33:57,300 --> 00:33:59,900
Absolutely, because very 
suddenly, those extra minutes 

723
00:33:59,900 --> 00:34:02,000
actually mean anything, right? 
And it's one of those 

724
00:34:02,000 --> 00:34:05,200
counterfactual is that we talked
about in these incident reviews 

725
00:34:05,200 --> 00:34:08,199
where it's like, oh man, if we 
had known about that Alert, five

726
00:34:08,199 --> 00:34:10,699
minutes earlier, we could have 
prevented the outage probably 

727
00:34:10,699 --> 00:34:12,900
not. 
You have no idea how quickly 5 

728
00:34:12,900 --> 00:34:15,100
minutes goes by. 
You're looking at something, 

729
00:34:15,100 --> 00:34:16,600
you're tracking down a red 
herring. 

730
00:34:16,600 --> 00:34:18,199
You're like, oh, yeah. 
It's probably this thing over 

731
00:34:18,199 --> 00:34:20,100
here when it's something 
completely unrelated. 

732
00:34:20,199 --> 00:34:22,199
A lot of times those five 
minutes are find you as much as 

733
00:34:22,199 --> 00:34:22,900
you think. 
They are. 

734
00:34:23,100 --> 00:34:24,100
Of course, your mileage. 
May. 

735
00:34:24,199 --> 00:34:25,400
Very take that with a grain of 
salt. 

736
00:34:25,400 --> 00:34:28,000
If you're in high frequency 
trading maybe it's a different 

737
00:34:28,000 --> 00:34:31,400
ball game, but for the most of 
us, but I target audience. 

738
00:34:31,400 --> 00:34:33,100
It's like, yeah, you'll be all 
right. 

739
00:34:33,600 --> 00:34:36,100
So let's move on to the next 
anti pattern which is about 

740
00:34:36,100 --> 00:34:38,800
wasting a perfectly good 
incident. 

741
00:34:39,000 --> 00:34:42,500
This is interesting because a 
perfectly good incident. 

742
00:34:42,600 --> 00:34:44,199
Maybe he can you explain about 
this. 

743
00:34:44,800 --> 00:34:48,100
Yeah, so actually stems from 
saying from local politician 

744
00:34:48,100 --> 00:34:50,699
here in Chicago that always 
would say never let a good 

745
00:34:50,699 --> 00:34:56,800
crisis go to waste it's the This
idea that there is so much to 

746
00:34:56,800 --> 00:35:01,200
learn from an incident because 
when you think about it, so we 

747
00:35:01,200 --> 00:35:03,500
have these mental models of our 
systems, right? 

748
00:35:03,500 --> 00:35:05,300
These systems are becoming so 
complicated. 

749
00:35:05,300 --> 00:35:08,300
So complex, so many different 
pieces. 

750
00:35:08,600 --> 00:35:11,300
So everyone in their head has 
this mental model of how they 

751
00:35:11,300 --> 00:35:14,200
think the system works. 
Then there's the reality of how 

752
00:35:14,200 --> 00:35:17,000
it actually works. 
And those two are seldom in full

753
00:35:17,000 --> 00:35:18,700
alignment because usually 
something off. 

754
00:35:18,800 --> 00:35:21,900
And the Delta, the time that you
find out that your model is 

755
00:35:21,900 --> 00:35:24,100
different than reality. 
There's an incident. 

756
00:35:24,700 --> 00:35:27,800
Other than that, you can spend 
the rest of your life and 

757
00:35:27,800 --> 00:35:30,400
complete the illusion thinking, 
you understand how the system 

758
00:35:30,400 --> 00:35:34,000
works, but then when there's an 
incident, suddenly the gap 

759
00:35:34,000 --> 00:35:37,300
between your understanding and 
reality is exposed in its raw 

760
00:35:37,300 --> 00:35:40,600
form. 
So often we sort of just close 

761
00:35:40,600 --> 00:35:43,600
it, incident ticket and move on,
but it's like hang on. 

762
00:35:43,700 --> 00:35:47,800
Let's dig into this incident and
try to make our mental models 

763
00:35:47,800 --> 00:35:50,200
better and learn from our 
mistakes. 

764
00:35:50,400 --> 00:35:53,100
It goes beyond just the tech 
side. 

765
00:35:53,100 --> 00:35:56,900
It's also the Human Side so 
simple things that you've 

766
00:35:56,900 --> 00:35:59,800
uncover like, okay. 
All right Henry, I noticed that 

767
00:35:59,800 --> 00:36:03,500
this alert fired. 
You got it in, snoozed it and 

768
00:36:03,500 --> 00:36:05,800
then it be alerted, 15 minutes 
later. 

769
00:36:05,900 --> 00:36:08,700
And then you engage what 
happened there? 

770
00:36:09,000 --> 00:36:12,800
Oh, well this system alerts all 
the time and it typically Auto 

771
00:36:12,800 --> 00:36:15,100
recovers. 
So when I got the alert, I 

772
00:36:15,100 --> 00:36:17,200
snoozed it. 
Thinking it was going to recover

773
00:36:17,200 --> 00:36:19,900
but then when it didn't recover,
I realize, okay, something's 

774
00:36:19,900 --> 00:36:22,400
really wrong. 
So, as a manager, be personally,

775
00:36:22,400 --> 00:36:24,000
right? 
If I'm not part of the on call. 

776
00:36:24,200 --> 00:36:26,200
Patient. 
I may not know that reality that

777
00:36:26,200 --> 00:36:29,000
Dynamic exist. 
So just by simply asking that 

778
00:36:29,000 --> 00:36:31,600
question in the incident review,
reveal something to me. 

779
00:36:31,600 --> 00:36:34,200
Like, whoa. 
Okay, we've got alerts that are 

780
00:36:34,200 --> 00:36:38,100
so bad people ignore them 
because they're so common and 

781
00:36:38,100 --> 00:36:40,600
I'm sure everyone on your team 
is going to be backing you up. 

782
00:36:40,600 --> 00:36:41,200
Yeah. 
Yeah. 

783
00:36:41,200 --> 00:36:42,800
I know that alert. 
I hate that alerts. 

784
00:36:43,000 --> 00:36:46,500
So now it's like, okay, so 
clearly I have poor learning 

785
00:36:46,700 --> 00:36:50,200
that poor alerting is impacting 
by on-call team, because every 

786
00:36:50,200 --> 00:36:53,200
time you're waking someone up, 
anytime you paid someone, you're

787
00:36:53,200 --> 00:36:55,600
interrupting their Life, they 
could be at dinner. 

788
00:36:56,000 --> 00:36:58,500
They could be in a movie. 
They could be taking care of 

789
00:36:58,500 --> 00:37:01,900
their sick, mother. 
You have to think when we page 

790
00:37:01,900 --> 00:37:03,900
out. 
What is this person doing? 

791
00:37:03,900 --> 00:37:06,000
In their life that I'm 
interrupting? 

792
00:37:06,000 --> 00:37:08,200
And is this worth it as a 
manager. 

793
00:37:08,300 --> 00:37:10,400
If I'm not part of the on-call 
rotation, right. 

794
00:37:10,400 --> 00:37:12,500
That is information that I can 
get on the incident review 

795
00:37:12,500 --> 00:37:14,200
process. 
Okay, cool. 

796
00:37:14,400 --> 00:37:18,000
Here's another example, a real 
world example, where two 

797
00:37:18,000 --> 00:37:22,600
Engineers were talking about the
same system using different 

798
00:37:22,600 --> 00:37:25,900
terminology and Didn't realize 
that they were talking about the

799
00:37:25,900 --> 00:37:29,500
same system because one team use
this term a and another team 

800
00:37:29,500 --> 00:37:32,400
uses term be. 
So my Ops guy was thinking that 

801
00:37:32,400 --> 00:37:35,000
there's some new system that he 
doesn't know anything about 

802
00:37:35,100 --> 00:37:36,900
that. 
He's tracking down and he's 

803
00:37:36,900 --> 00:37:38,500
pissed because there's no 
monitoring. 

804
00:37:38,500 --> 00:37:41,100
There's no metrics around it. 
But lo and behold. 

805
00:37:41,200 --> 00:37:43,400
Oh, no, we have a terminology 
difference. 

806
00:37:43,600 --> 00:37:47,400
What you're calling sidekick. 
We're calling consumer Damon and

807
00:37:47,400 --> 00:37:50,200
they're the actual thing. 
That's a huge disconnect. 

808
00:37:50,200 --> 00:37:53,300
But now that he understands 
we're talking about consumer 

809
00:37:53,300 --> 00:37:55,900
David. 
He instantly has a different 

810
00:37:55,900 --> 00:37:59,300
view of the entire scenario 
because he's like, oh now the 

811
00:37:59,300 --> 00:38:00,800
know we're talking about 
consumer Damon. 

812
00:38:00,900 --> 00:38:03,400
I understand that these aren't 
technical problem. 

813
00:38:03,400 --> 00:38:06,700
These for human meat, space 
problems that are coming out of 

814
00:38:06,700 --> 00:38:09,600
the incident review process, but
once you start to dig in and 

815
00:38:09,600 --> 00:38:14,100
peel back and get into people's 
head space, it's like, oh, okay.

816
00:38:14,300 --> 00:38:17,400
Alright, this is making sense. 
So I noticed Henry after that 

817
00:38:17,400 --> 00:38:20,000
was all over you restarted. 
The service. 

818
00:38:20,300 --> 00:38:22,200
What made you think to restart 
the service? 

819
00:38:22,500 --> 00:38:24,000
What information LED you to 
that? 

820
00:38:24,500 --> 00:38:28,300
Well, honestly, I was out of 
ideas and I thought maybe I 

821
00:38:28,300 --> 00:38:30,200
noticed the memory utilization 
was high. 

822
00:38:30,200 --> 00:38:32,200
So what made you look at the 
memory utilization? 

823
00:38:32,400 --> 00:38:35,400
Well, I normally don't look at 
that, but I happened to be 

824
00:38:35,400 --> 00:38:37,900
looking at a different screen 
and saw that it was high. 

825
00:38:38,100 --> 00:38:40,600
So, I just said, well, why not 
restart the service? 

826
00:38:40,800 --> 00:38:43,500
Okay, but when you restarted the
service, you didn't realize 

827
00:38:43,500 --> 00:38:46,900
billing was running and that 
interrupted the billing process.

828
00:38:47,100 --> 00:38:49,900
So, now, we're not getting bills
out on time, which was an 

829
00:38:49,900 --> 00:38:52,600
ancillary effect. 
Oh, I didn't realize the 

830
00:38:52,600 --> 00:38:54,900
service, communicated with them.
Billing process. 

831
00:38:55,000 --> 00:38:56,300
Oh, yeah. 
It's part of the billing 

832
00:38:56,300 --> 00:38:58,900
process, because they share a 
key or something like that. 

833
00:38:59,000 --> 00:39:02,200
Oh, well, it's not like I 
intended to impact filling. 

834
00:39:02,200 --> 00:39:04,500
I just had no idea. 
All of these sorts of 

835
00:39:04,500 --> 00:39:07,400
conversations happen because 
again, everyone has a slightly 

836
00:39:07,400 --> 00:39:09,200
different, mental model of the 
system. 

837
00:39:09,500 --> 00:39:11,200
So, wasting perfectly good 
incident. 

838
00:39:11,200 --> 00:39:14,600
Is this idea that you just say? 
Oh, yeah, system was low on 

839
00:39:14,600 --> 00:39:18,400
memory Henry, restart that 
service recovered service. 

840
00:39:18,400 --> 00:39:21,000
Didn't just recover. 
You impacted the billing team 

841
00:39:21,000 --> 00:39:24,000
who now has to rerun the billing
shop you're communicating with. 

842
00:39:24,100 --> 00:39:26,800
Engineer didn't realize you guys
were using the different 

843
00:39:26,800 --> 00:39:29,900
terminology had you known that. 
He might have made different 

844
00:39:29,900 --> 00:39:32,000
choices. 
If you knew that we were talking

845
00:39:32,000 --> 00:39:35,900
about this particular service. 
We've discovered that, oh, poor 

846
00:39:35,900 --> 00:39:39,700
alerting and because of that 
people are hesitant to do things

847
00:39:39,800 --> 00:39:41,800
right away. 
They're waiting until it really 

848
00:39:41,800 --> 00:39:44,200
hurts. 
There's all of this information 

849
00:39:44,300 --> 00:39:46,700
that could have been easily 
dismissed and wrapped up at a 

850
00:39:46,700 --> 00:39:49,300
memory utilization with high 
restarted Service close ticket. 

851
00:39:49,600 --> 00:39:51,400
So wasting a perfectly. 
Good instance, really? 

852
00:39:51,400 --> 00:39:53,900
This idea of like there's so 
much more to be done. 

853
00:39:54,100 --> 00:39:57,100
Information in a failure that we
can bring out. 

854
00:39:57,100 --> 00:40:00,200
If we just really want to put 
some energy towards thank you, 

855
00:40:00,207 --> 00:40:02,800
explain about all these 
different scenarios based on 

856
00:40:02,800 --> 00:40:05,400
anecdotes and all that. 
I realized this is like what 

857
00:40:05,400 --> 00:40:09,300
some people call it post, mortem
activity, but one thing that I 

858
00:40:09,400 --> 00:40:12,800
do find a challenge sometimes 
like for those people who are 

859
00:40:12,800 --> 00:40:15,200
involved in the crisis of the 
they solve the problem, of 

860
00:40:15,200 --> 00:40:17,200
course, they're like, okay. 
I don't want to deal with it 

861
00:40:17,200 --> 00:40:19,800
anymore. 
That culture of assessing this 

862
00:40:19,800 --> 00:40:22,500
incident assessing. 
What can we learn from it? 

863
00:40:22,700 --> 00:40:25,200
I think it's not there for Those
are the people, I would say. 

864
00:40:25,400 --> 00:40:27,600
So how do you actually inculcate
this culture? 

865
00:40:27,600 --> 00:40:30,100
So that it becomes a thing. 
It becomes a common thing that 

866
00:40:30,100 --> 00:40:33,200
people actually want to do it 
because they feel value out of 

867
00:40:33,200 --> 00:40:36,200
it. 
So the first thing that you can 

868
00:40:36,200 --> 00:40:38,400
do as an individual contributor,
I'm assuming you have the 

869
00:40:38,400 --> 00:40:39,800
willpower. 
So I'm going to say an 

870
00:40:39,800 --> 00:40:41,400
individual contributor in this 
scenario. 

871
00:40:41,700 --> 00:40:45,300
The first thing you need to do 
is, do it fast like immediately 

872
00:40:45,300 --> 00:40:47,100
after the incident within 24 
hours. 

873
00:40:47,100 --> 00:40:52,500
Why, because things have a sense
of permanence in our minds, but 

874
00:40:52,500 --> 00:40:54,500
only for a short period of time.
Time. 

875
00:40:54,600 --> 00:40:57,000
So it's like something 
permanent, but only, for a short

876
00:40:57,000 --> 00:41:00,000
period of time. 
That's how it works within that 

877
00:41:00,000 --> 00:41:02,800
24 hour period. 
This is the most important thing

878
00:41:02,800 --> 00:41:05,600
that has ever happened. 
But after that period, it 

879
00:41:05,600 --> 00:41:09,000
becomes just noise in the 
background while you've got all 

880
00:41:09,000 --> 00:41:12,800
of these other demands on you. 
So if you can get people in the 

881
00:41:12,800 --> 00:41:17,000
room as soon as possible to talk
about it, the actual incident is

882
00:41:17,000 --> 00:41:20,000
fresher in their mind, so they 
can recall more Vivid details 

883
00:41:20,200 --> 00:41:22,000
are much more energized around 
it. 

884
00:41:22,000 --> 00:41:23,900
You're really manipulating their
psyche. 

885
00:41:24,100 --> 00:41:26,900
Honestly, by doing it early to 
say like, hey, let's do it now 

886
00:41:26,900 --> 00:41:29,300
while you're super interested 
but our dealer never let you 

887
00:41:29,300 --> 00:41:31,200
lead the dealership. 
He doesn't want you to go home 

888
00:41:31,200 --> 00:41:33,400
and think about it. 
He wants you to remember what 

889
00:41:33,400 --> 00:41:37,200
that SUV felt like moment you're
in the store so he can sell it 

890
00:41:37,200 --> 00:41:38,900
to you. 
It's a basic human emotion. 

891
00:41:39,100 --> 00:41:41,600
So the biggest piece of advice I
can give is like I said schedule

892
00:41:41,600 --> 00:41:44,500
it as soon as possible 
definitely within 24 hours, but 

893
00:41:44,500 --> 00:41:49,100
once you do a few of these, I 
guarantee you even without any 

894
00:41:49,100 --> 00:41:52,300
real training in the process. 
You're going to discover things 

895
00:41:52,500 --> 00:41:55,700
and as you discover them people 
You will instantly see the value

896
00:41:55,700 --> 00:41:58,600
in it. 
The hard part, is the follow-up 

897
00:41:58,600 --> 00:42:02,900
action items? 
Because how do you influence 

898
00:42:02,900 --> 00:42:05,400
people who schedule and 
prioritization? 

899
00:42:05,400 --> 00:42:09,200
You have no control over to make
sure that some of these things 

900
00:42:09,200 --> 00:42:10,900
that were brought up, are 
addressed. 

901
00:42:11,100 --> 00:42:14,000
Sometimes, it's worth doing it, 
just to have the knowledge just 

902
00:42:14,000 --> 00:42:16,000
to be able to see like what we 
know these things now and even 

903
00:42:16,000 --> 00:42:18,300
though we're not going to 
correct it, the next incident, 

904
00:42:18,300 --> 00:42:21,200
we're aware of it, but for the 
things that you need to actually

905
00:42:21,200 --> 00:42:23,900
fix that, where you really need 
some leadership. 

906
00:42:24,100 --> 00:42:27,100
And could be up to like a, these
things were things, we discover 

907
00:42:27,100 --> 00:42:29,100
in the incident post-mortem 
process. 

908
00:42:29,300 --> 00:42:32,400
We really are going to need help
and commitment for your team to 

909
00:42:32,400 --> 00:42:36,200
address some of these putting 
dollar signs next to the actual 

910
00:42:36,200 --> 00:42:39,600
potential risk help sometimes. 
Yeah, we lost seventy five 

911
00:42:39,600 --> 00:42:41,600
thousand dollars for a 15-minute
outage. 

912
00:42:41,900 --> 00:42:44,300
Sometimes that can be 
persuasive, not all the time 

913
00:42:44,400 --> 00:42:46,600
because again, it's not coming 
out of anyone's paycheck. 

914
00:42:46,900 --> 00:42:49,600
So by Design, the only people 
that really care about that it's

915
00:42:49,600 --> 00:42:52,700
management. 
But yeah 24 hours proving the 

916
00:42:52,700 --> 00:42:56,000
use out of it and Documenting 
that use reading that 

917
00:42:56,000 --> 00:42:58,100
information is part of the 
sharing portion. 

918
00:42:58,200 --> 00:43:00,600
Have it somewhere where people 
can see and review it and 

919
00:43:00,600 --> 00:43:02,600
understand it. 
The other thing that really 

920
00:43:02,600 --> 00:43:05,100
makes it shine as when someone 
is looking at an incident and 

921
00:43:05,100 --> 00:43:07,900
they see this beautiful 
post-mortem incident review, and

922
00:43:07,900 --> 00:43:10,200
then they go to another incident
and it's just like restarted 

923
00:43:10,200 --> 00:43:11,000
service. 
We're good. 

924
00:43:11,100 --> 00:43:13,400
It's like, whoa. 
Whoa, where's this? 

925
00:43:13,700 --> 00:43:18,100
That sort of creates this 
pattern of like, hey Henry, I 

926
00:43:18,107 --> 00:43:21,000
noticed the last incident. 
You ran you didn't do a 

927
00:43:21,000 --> 00:43:23,600
post-mortem, like the other pin 
members of the team you should 

928
00:43:23,600 --> 00:43:25,500
be doing. 
To and it just becomes a 

929
00:43:25,500 --> 00:43:27,800
cultural thing. 
It just becomes this thing where

930
00:43:27,800 --> 00:43:30,400
like everyone rallies around it,
but it takes time but you know, 

931
00:43:30,400 --> 00:43:34,000
lead by example, do it yourself 
and when you do it, don't treat 

932
00:43:34,000 --> 00:43:35,700
it. 
Like it's some new fandangled 

933
00:43:35,700 --> 00:43:38,200
thing to it as if it's the most 
logical thing that you should be

934
00:43:38,200 --> 00:43:39,500
doing. 
Yes, of course. 

935
00:43:39,500 --> 00:43:41,200
We're going to do a post-mortem.
Why wouldn't we? 

936
00:43:41,200 --> 00:43:43,800
We just had an incident to hide 
the fact that this is the first 

937
00:43:43,800 --> 00:43:45,800
post mortem you've ever done 
before in your life. 

938
00:43:46,000 --> 00:43:48,500
Just tell it like it's the most 
natural Next Step. 

939
00:43:48,600 --> 00:43:49,800
Yeah. 
We're going to post more. 

940
00:43:49,800 --> 00:43:51,900
We got understand what happened.
Don't you think we need to 

941
00:43:51,900 --> 00:43:53,800
understand what happened? 
Say no to that. 

942
00:43:54,000 --> 00:43:56,300
That yeah, I guess we should 
probably have a better 

943
00:43:56,300 --> 00:43:57,700
understanding. 
Yeah, so that's why we're going 

944
00:43:57,700 --> 00:43:58,800
to the post-mortem that's 
enough. 

945
00:43:58,800 --> 00:44:00,500
Come on. 
So as you mentioned the 

946
00:44:00,500 --> 00:44:02,600
beginning, right? 
Probably all these also is part 

947
00:44:02,600 --> 00:44:04,700
of culture. 
So you can't just switch 

948
00:44:04,700 --> 00:44:07,700
everybody in a second like once 
an incident happened and yeah, 

949
00:44:07,700 --> 00:44:09,900
everybody will just do 
post-mortem for the next few 

950
00:44:09,900 --> 00:44:12,100
incidents. 
So sometimes I think we need the

951
00:44:12,100 --> 00:44:15,200
buy-in and also like I would 
say, probably policing what you 

952
00:44:15,200 --> 00:44:16,900
said is like leading by example,
right? 

953
00:44:16,900 --> 00:44:19,700
So the people at the top, 
probably also needs to spend the

954
00:44:19,700 --> 00:44:22,700
time or actually allocate the 
time to actually do this stuff. 

955
00:44:22,700 --> 00:44:25,800
Because it's important and High 
value for the company, not just 

956
00:44:25,800 --> 00:44:28,800
for that particular team. 
And like, with any cultural 

957
00:44:28,800 --> 00:44:30,700
change, you got three categories
people, right? 

958
00:44:30,700 --> 00:44:33,300
You've got supporters, 
detractors and fence sitters. 

959
00:44:33,600 --> 00:44:36,900
The majority of people are 
fencers the vast majority of 

960
00:44:36,900 --> 00:44:40,900
people and this is true of any 
context you look at politics. 

961
00:44:41,000 --> 00:44:43,700
Most people aren't on the 
extreme ends of either side. 

962
00:44:43,800 --> 00:44:46,600
Most people are in the middle, 
but the extreme ends of the loud

963
00:44:46,600 --> 00:44:47,900
one. 
So, those are the ones that we 

964
00:44:47,900 --> 00:44:51,800
focus on, it doesn't take a lot 
of people to change the culture 

965
00:44:51,800 --> 00:44:53,900
of an organization. 
It takes a few. 

966
00:44:54,000 --> 00:44:57,100
Supporters or even a few 
detractors, it works both ways. 

967
00:44:57,400 --> 00:45:00,200
So positive influence. 
It only takes a handful of 

968
00:45:00,200 --> 00:45:02,500
people to be able to change to 
have the fence sitters move 

969
00:45:02,500 --> 00:45:04,300
over. 
Same thing with the - though 

970
00:45:04,600 --> 00:45:07,100
with the negative behaviors, 
only takes a handful of people 

971
00:45:07,100 --> 00:45:09,600
to be like wall and also how 
we're going to do things and 

972
00:45:09,600 --> 00:45:11,400
then suddenly your cultures in 
the tank. 

973
00:45:11,700 --> 00:45:14,400
So think about that as you're 
recruiting people for these 

974
00:45:14,400 --> 00:45:16,600
different aspects, right? 
Who are the people that can 

975
00:45:16,600 --> 00:45:19,800
really be boosters for me? 
And I don't have to convert 

976
00:45:19,800 --> 00:45:23,100
everyone. 
I've got to convert a few really

977
00:45:23,300 --> 00:45:25,700
boisterous. 
Cheerleaders for this, and if I 

978
00:45:25,700 --> 00:45:27,700
convert them, people are going 
to follow. 

979
00:45:27,800 --> 00:45:30,300
And once people start following,
you have overwhelmed, the 

980
00:45:30,300 --> 00:45:34,800
detractors and they lose. 
So as I hear your insights about

981
00:45:34,800 --> 00:45:37,800
these and the patterns the 
audience here, actually can find

982
00:45:37,800 --> 00:45:40,900
more Independence in the book. 
They are some of the topics that

983
00:45:40,900 --> 00:45:44,000
I think like really relevant, 
but because of the time I'm sure

984
00:45:44,000 --> 00:45:46,200
we cannot cover all of them. 
So for people who are 

985
00:45:46,207 --> 00:45:49,300
interested, go by the book, or 
read the book, and you can learn

986
00:45:49,400 --> 00:45:52,300
all the fun styles and Tibetans 
from Jeffrey. 

987
00:45:52,700 --> 00:45:53,800
Yeah, and you should buy the 
book. 

988
00:45:54,000 --> 00:45:59,800
I don't want to go over all 12. 
So Jeffrey, Before I Let You Go,

989
00:45:59,800 --> 00:46:02,800
normally I have this one last 
question that I always ask all 

990
00:46:02,800 --> 00:46:05,200
the guests which is called the 
tree technical leadership 

991
00:46:05,200 --> 00:46:07,400
wisdom. 
So this is just for people to 

992
00:46:07,500 --> 00:46:10,200
maybe learn from your journey. 
So what kind of wisdom that you 

993
00:46:10,200 --> 00:46:12,500
have in your career that 
probably you want to share with 

994
00:46:12,500 --> 00:46:14,200
everyone. 
Okay. 

995
00:46:14,400 --> 00:46:17,000
The first one is, of course, 
never let perfect be the enemy 

996
00:46:17,000 --> 00:46:21,200
of good or good enough. 
So there are a solution that is 

997
00:46:21,200 --> 00:46:25,800
70% effective is better than A 
perfect solution that has been 

998
00:46:25,800 --> 00:46:29,000
implemented. 
So, never let the fact that it's

999
00:46:29,000 --> 00:46:30,600
not perfect stop you, because 
guess what? 

1000
00:46:30,600 --> 00:46:32,200
It's never perfect. 
It's all perfect. 

1001
00:46:32,200 --> 00:46:34,900
In your mind. 
Keep pushing get it done. 

1002
00:46:35,200 --> 00:46:36,100
You're going to learn a bunch of
stuff. 

1003
00:46:36,100 --> 00:46:38,100
It's not going to be perfect. 
Something's going to be screwed 

1004
00:46:38,100 --> 00:46:40,500
up. 
Second piece of advice is 

1005
00:46:40,600 --> 00:46:42,200
belong. 
Same themes of this whole 

1006
00:46:42,200 --> 00:46:44,000
perfect. 
Is the enemy of good discussion 

1007
00:46:44,000 --> 00:46:49,200
is that there is no point in a 
project Journey that, you know, 

1008
00:46:49,200 --> 00:46:53,800
less about the requirements that
in the beginning, when you start

1009
00:46:54,300 --> 00:46:57,100
that is the most minimal amount 
of information. 

1010
00:46:57,100 --> 00:46:58,800
You're actually going to have 
about your requirements. 

1011
00:46:59,100 --> 00:47:02,000
So keep that in mind when you're
designing a solution for 

1012
00:47:02,000 --> 00:47:05,700
something because as you move in
the project, your requirements 

1013
00:47:05,700 --> 00:47:08,700
are going to become more and 
more concrete and understanding 

1014
00:47:09,000 --> 00:47:11,300
product people and project. 
People are going to come to you 

1015
00:47:11,300 --> 00:47:13,000
with all this list of 
requirements and it's going to 

1016
00:47:13,000 --> 00:47:17,300
look like they did a bunch of 
due diligence but know that they

1017
00:47:17,300 --> 00:47:21,000
only know so much at this point 
and cut him some slack as a 

1018
00:47:21,000 --> 00:47:23,700
result of that. 
So we have a problem now at 

1019
00:47:23,700 --> 00:47:26,500
work. 
We're Ops gets involved too late

1020
00:47:26,600 --> 00:47:29,700
in the life cycle of a project 
but a lot of it is because when 

1021
00:47:29,700 --> 00:47:32,800
they start they don't know if 
they're going to need op support

1022
00:47:32,800 --> 00:47:35,200
or not, because it's like, well,
if it's a feature in the 

1023
00:47:35,200 --> 00:47:37,400
monolith, we don't need Ops for 
anything. 

1024
00:47:37,500 --> 00:47:39,700
We can develop that on our own, 
they've given us all the 

1025
00:47:39,700 --> 00:47:42,300
automation tools. 
We need, we can, you know, do 

1026
00:47:42,300 --> 00:47:45,300
everything we need to do without
the but then if halfway through 

1027
00:47:45,300 --> 00:47:47,800
the project they pivot and they 
say, oh this needs to be a 

1028
00:47:47,808 --> 00:47:50,900
separate micro service will 
suddenly that's a whole new ball

1029
00:47:50,900 --> 00:47:54,400
game for Rob's, but we have to 
accept the fact that They did 

1030
00:47:54,400 --> 00:47:57,500
not know that when they started,
they made the best choice that 

1031
00:47:57,500 --> 00:48:00,400
they can make. 
So, always keep that in mind. 

1032
00:48:00,600 --> 00:48:03,200
I guess, the third thing is, as 
an engineer. 

1033
00:48:03,300 --> 00:48:07,300
You have an implicit bias and 
just about everything you do and

1034
00:48:07,300 --> 00:48:09,800
it's particularly bad in 
technology. 

1035
00:48:10,000 --> 00:48:13,800
There is a strong sense of this.
I didn't write it. 

1036
00:48:13,800 --> 00:48:15,100
Therefore. 
It's crap. 

1037
00:48:16,200 --> 00:48:20,800
Good engineer knows the 
difference between a preference 

1038
00:48:21,000 --> 00:48:23,500
and a problem. 
It's a pattern that I see all 

1039
00:48:23,500 --> 00:48:25,400
the time. 
An engineer comes in. 

1040
00:48:25,400 --> 00:48:28,000
They get hired and ready to get 
started. 

1041
00:48:28,000 --> 00:48:30,200
They look at some code or 
something in their life. 

1042
00:48:30,400 --> 00:48:32,600
This is all wrong. 
Can't do anything. 

1043
00:48:32,900 --> 00:48:35,500
It's funny because this thing's 
making this like 140 million a 

1044
00:48:35,500 --> 00:48:38,200
year. 
So tell me what is so broken 

1045
00:48:38,200 --> 00:48:39,500
about it. 
Is it perfect? 

1046
00:48:39,500 --> 00:48:42,400
No, of course, it's not perfect.
There's a bunch of problems with

1047
00:48:42,400 --> 00:48:44,400
it. 
See my previous two ideas and 

1048
00:48:44,400 --> 00:48:46,400
suggestions. 
You have to understand what to 

1049
00:48:46,400 --> 00:48:50,300
preference and what's a problem.
So that you know, where to focus

1050
00:48:50,300 --> 00:48:55,100
and put your energy because no 
one makes a We're out of 

1051
00:48:55,100 --> 00:48:56,800
rewriting. 
Something that was already 

1052
00:48:56,800 --> 00:49:00,000
working and just introducing 
different problems because you 

1053
00:49:00,000 --> 00:49:01,600
never actually fix it. 
Right. 

1054
00:49:01,700 --> 00:49:04,600
My shifted, you might change the
sort of problems, but there's 

1055
00:49:04,600 --> 00:49:07,100
always some new problem that 
you're going to be dealing with.

1056
00:49:07,200 --> 00:49:10,200
So accept that and know that 
everything you do is going to be

1057
00:49:10,200 --> 00:49:12,500
future. 
Use thing that they hate, you're

1058
00:49:12,500 --> 00:49:14,600
going to make a choice and 
someone's going to come in 5 

1059
00:49:14,600 --> 00:49:17,900
years later the who's dumb. 
Why don't you write it and go? 

1060
00:49:18,100 --> 00:49:20,300
Why didn't you write insert new 
language? 

1061
00:49:20,300 --> 00:49:23,000
That's hip and trendy and I'm 
going to add a fourth one to 

1062
00:49:23,000 --> 00:49:25,100
when you make it to try. 
Choice, document, your 

1063
00:49:25,100 --> 00:49:29,000
constraints and your context 
when you make any sort of 

1064
00:49:29,000 --> 00:49:31,800
technical decision. 
Why did you make that decision? 

1065
00:49:32,100 --> 00:49:34,300
What was the reality on the 
ground? 

1066
00:49:34,400 --> 00:49:37,800
I remember a story where a guy 
was telling me about all of this

1067
00:49:37,800 --> 00:49:41,600
handcrafted code, that was built
at his company and his Chi, 

1068
00:49:41,600 --> 00:49:44,700
don't understand why it is still
in use kubernetes to 

1069
00:49:44,700 --> 00:49:46,400
orchestrate. 
All this, it was, like, why did 

1070
00:49:46,400 --> 00:49:48,400
they write the stuff? 
He's like, oh, those probably 

1071
00:49:48,600 --> 00:49:51,300
eight ten years ago. 
That might be why they didn't 

1072
00:49:51,300 --> 00:49:55,700
use kubernetes right as I can. 
Yeah, I thought about that yet. 

1073
00:49:55,800 --> 00:49:58,800
So he got 10 years of energy 
built into this thing, just 

1074
00:49:58,800 --> 00:50:01,200
switching the kubernetes like 
that isn't an easy thing. 

1075
00:50:01,400 --> 00:50:03,800
So there's always context around
every technical decision that 

1076
00:50:03,800 --> 00:50:06,200
gets paid. 
If you can document that it'll 

1077
00:50:06,200 --> 00:50:09,000
save you some Hassle and future 
Engineers, some energy around 

1078
00:50:09,000 --> 00:50:11,600
understanding why particular 
decisions were made. 

1079
00:50:12,100 --> 00:50:13,300
Thanks for sharing these visit 
them. 

1080
00:50:13,300 --> 00:50:16,400
So I'm laughing, as you said all
this because I can see these 

1081
00:50:16,400 --> 00:50:19,800
patterns over and over again, in
any places that I went into. 

1082
00:50:19,800 --> 00:50:25,600
So this is just by Common, so, 
Thanks, Jeffrey for your time. 

1083
00:50:25,700 --> 00:50:28,500
So for people who wants to learn
more about you or connect with 

1084
00:50:28,500 --> 00:50:30,600
you or find the books. 
Where can they find you? 

1085
00:50:31,000 --> 00:50:32,900
Sure? 
Yeah, so I have a website that I

1086
00:50:32,908 --> 00:50:35,600
don't really update or maintain.
But if you feel nursing you can 

1087
00:50:35,600 --> 00:50:37,900
check that out at a noble 
devops.com. 

1088
00:50:38,200 --> 00:50:40,100
Most likely the best place to 
find me. 

1089
00:50:40,100 --> 00:50:43,200
Is that Twitter where I'm at 
dark and nerdy you can find the 

1090
00:50:43,200 --> 00:50:46,000
book operations into patterns 
develop solutions that 

1091
00:50:46,000 --> 00:50:49,300
Manning.com if you want to order
direct from them both physical 

1092
00:50:49,300 --> 00:50:52,100
and ebook copies, but it's also 
available on the Amazon book. 

1093
00:50:52,100 --> 00:50:54,600
Story inaudible I do. 
Read it though. 

1094
00:50:54,600 --> 00:50:57,100
Unfortunately, everyone's like, 
I bought this thing and you were

1095
00:50:57,100 --> 00:50:59,300
going to read it. 
Like I didn't even know they 

1096
00:50:59,300 --> 00:51:00,800
were turning it into an audio 
book. 

1097
00:51:00,800 --> 00:51:02,900
I got the email, the same time 
you guys did. 

1098
00:51:03,100 --> 00:51:05,800
So I'm going to make a Lobby for
the second edition that I read 

1099
00:51:05,800 --> 00:51:07,200
it though. 
So we'll see how that goes. 

1100
00:51:07,600 --> 00:51:08,700
Yeah. 
I mean like when you read it, 

1101
00:51:08,700 --> 00:51:10,600
probably you can impose this fun
style. 

1102
00:51:10,600 --> 00:51:13,800
So for people who are listening 
so that can also be entertaining

1103
00:51:13,800 --> 00:51:17,200
at the same time, which I think 
I'm doing it as you speak just 

1104
00:51:17,200 --> 00:51:18,500
now. 
So thanks again. 

1105
00:51:18,500 --> 00:51:21,000
Jeffrey for your time. 
I really learned a lot from this

1106
00:51:21,000 --> 00:51:24,200
conversation and I wish you good
luck for the things that You do.

1107
00:51:24,700 --> 00:51:25,600
All right. 
Thanks for having me. 

1108
00:51:25,600 --> 00:51:30,800
I had a really good time. 
Thank you for listening to this 

1109
00:51:30,800 --> 00:51:33,400
episode and for staying right 
till the end. 

1110
00:51:33,600 --> 00:51:36,500
If you highly enjoyed, please 
share it with your friends and 

1111
00:51:36,500 --> 00:51:39,900
colleagues who you think would 
also benefit from listening to 

1112
00:51:39,900 --> 00:51:42,100
this episode. 
And if you're new to the 

1113
00:51:42,100 --> 00:51:45,500
podcast, make sure to subscribe 
and leave me your valuable 

1114
00:51:45,500 --> 00:51:48,900
review and feedback. 
It really, really helps me a lot

1115
00:51:48,900 --> 00:51:51,400
in order to grow these podcasts 
better. 

1116
00:51:51,900 --> 00:51:55,200
You can also find the full show 
notes of this conversation on 

1117
00:51:55,200 --> 00:51:58,500
the episode page at technology. 
No, the death website. 

1118
00:51:58,700 --> 00:52:02,000
Including the full transcript 
interesting quotes, and links to

1119
00:52:02,000 --> 00:52:04,900
the resources and mentions from 
the conversation. 

1120
00:52:05,400 --> 00:52:08,200
And lastly make sure to 
subscribe to the show's mailing 

1121
00:52:08,200 --> 00:52:11,400
list on technology. 
No, the deaf to get notified for

1122
00:52:11,400 --> 00:52:14,200
any future episodes. 
Stay tuned for the next 

1123
00:52:14,200 --> 00:52:16,700
technique Journal episode. 
And until then. 

1124
00:52:16,900 --> 00:52:17,500
Goodbye.