1
00:00:02,939 --> 00:00:05,819
Narrator: You're listening to
the humans of DevOps podcast, a

2
00:00:05,819 --> 00:00:09,449
podcast focused on advancing the
humans of DevOps through skills,

3
00:00:09,479 --> 00:00:13,799
knowledge, ideas, and learning,
or the skil framework.

4
00:00:17,699 --> 00:00:20,519
Max Howell: We see this all over
the place with open source. I

5
00:00:20,519 --> 00:00:23,369
think a lot of the time the
people who make successful open

6
00:00:23,369 --> 00:00:27,299
source, making a hobby project,
some kind, something they need,

7
00:00:27,659 --> 00:00:30,119
and then it becomes an essential
piece of internet

8
00:00:30,119 --> 00:00:30,959
infrastructure.

9
00:00:33,600 --> 00:00:37,800
Eveline Oehrlich: In December of
2004, a group of IIT students

10
00:00:37,800 --> 00:00:42,870
from DAP University in
Copenhagen has launched a open

11
00:00:42,870 --> 00:00:47,790
source beer, the one that a
version was described by those

12
00:00:47,790 --> 00:00:52,830
students as a great tasting,
energetic beer, and its world

13
00:00:52,980 --> 00:00:57,930
first open source beer. It's
based on a classic ale brewing

14
00:00:57,930 --> 00:01:01,980
tradition. But with added
Guarana for a natural energy

15
00:01:01,980 --> 00:01:06,390
boost. The aim was to see what
happens when an open source

16
00:01:06,390 --> 00:01:11,430
structure is applied to a
universally known product like

17
00:01:11,700 --> 00:01:17,430
beer. Hello there. This is
evolutionarily ish. We are not

18
00:01:17,430 --> 00:01:21,210
here to talk about beer. But
we're here to talk about open

19
00:01:21,210 --> 00:01:24,930
source. And I brought this
introduction because I think it

20
00:01:24,930 --> 00:01:28,980
fits in multiple ways you will
find out yourself. But I'm a

21
00:01:28,980 --> 00:01:33,780
beer Brewer by Hobby so that
caught my attention. Our podcast

22
00:01:33,780 --> 00:01:39,600
title today is open source,
Abreu and t. Now a few more

23
00:01:39,600 --> 00:01:45,570
details, which is very different
97% of commercial code contains

24
00:01:45,600 --> 00:01:49,260
open source and I actually
learned this from our guest.

25
00:01:49,680 --> 00:01:53,520
When looking at some of the tech
trends such as containerization,

26
00:01:53,820 --> 00:01:57,000
like Docker and Kubernetes,
originally developed at Google

27
00:01:57,000 --> 00:02:02,160
and released as open source in
2014. Big data like Apache Spark

28
00:02:02,400 --> 00:02:08,040
with Hadoop or Kafka, a lot of
open source is out there. Again,

29
00:02:08,040 --> 00:02:12,600
as I said, Welcome to humans of
DevOps podcast, I'm Avalon early

30
00:02:12,660 --> 00:02:17,220
Chief Research Officer at DevOps
Institute. Today, we have a very

31
00:02:17,220 --> 00:02:20,460
special guest, and I'm excited
to have him with us for multiple

32
00:02:20,460 --> 00:02:31,260
reasons. Max Howell, CEO of T
dot x, y, z. Hello, Max. Welcome

33
00:02:31,260 --> 00:02:32,280
to our podcast.

34
00:02:33,000 --> 00:02:34,800
Max Howell: Thank you. I'm
pleased to be here.

35
00:02:35,099 --> 00:02:39,839
Eveline Oehrlich: Yes, excited
for you to be with us. Tell us

36
00:02:39,839 --> 00:02:43,769
the first thing really I'm
curious about is your journey

37
00:02:43,799 --> 00:02:46,739
into an across open source.

38
00:02:48,210 --> 00:02:53,370
Max Howell: Yeah, well, it's a
fairly good story. So I did a

39
00:02:53,400 --> 00:02:55,590
chemistry degree because I
thought I wanted to be a

40
00:02:55,590 --> 00:03:00,750
scientist. And in the process of
doing that degree, and then

41
00:03:00,750 --> 00:03:05,610
doing a year in the industry, I
discovered that actually, it was

42
00:03:05,610 --> 00:03:11,790
really not for me, I was working
with this device that measured

43
00:03:11,790 --> 00:03:16,080
surface tension for like various
chemical solutions. And I

44
00:03:16,080 --> 00:03:19,620
realized that if I kept working
there, like in 10 years, I would

45
00:03:19,620 --> 00:03:23,100
still be using that single
machine and making very minor

46
00:03:23,100 --> 00:03:27,330
contributions to the world. So I
fell into this sort of

47
00:03:27,330 --> 00:03:30,900
depressing funk. And I
discovered open source.

48
00:03:31,590 --> 00:03:35,550
Programming was always something
that I had considered a hobby.

49
00:03:35,640 --> 00:03:39,480
My dad taught me when I was six,
which, you know, very young, but

50
00:03:40,410 --> 00:03:43,290
I only used it to like make
video games essentially, like

51
00:03:43,290 --> 00:03:46,500
most kids do when they learned
to program. And I never really

52
00:03:46,500 --> 00:03:51,120
considered it for a career until
my career was on the rocks. And

53
00:03:51,120 --> 00:03:54,630
I didn't know what I was doing
myself. So propelled into open

54
00:03:54,630 --> 00:03:59,670
source by installing Linux. And
I loved the community, I loved

55
00:03:59,670 --> 00:04:03,120
how everyone was super
passionate about what they were

56
00:04:03,120 --> 00:04:07,560
doing, and just doing it because
they wanted to do something that

57
00:04:07,590 --> 00:04:11,460
changed the world or improved
their workflow or just just

58
00:04:11,460 --> 00:04:16,260
making cool things. And I ended
up working on a few apps with

59
00:04:16,260 --> 00:04:18,570
people from all over the world
and it was just like nothing

60
00:04:18,570 --> 00:04:24,390
else. For me. Honestly, it's,
it's difficult to find it as

61
00:04:24,390 --> 00:04:28,620
good as it was when I first got
into it. And we were all on IRC

62
00:04:28,620 --> 00:04:33,060
together. And doing that led me
to getting a job in the

63
00:04:33,060 --> 00:04:36,510
industry, which you know, it's
it's kind of luck in many ways.

64
00:04:36,840 --> 00:04:41,070
But there was a company in
London that used one of the apps

65
00:04:41,070 --> 00:04:43,680
I was working on and invited me
to go and interview and it

66
00:04:43,680 --> 00:04:47,040
wasn't really as much of an
interview as just having a chat.

67
00:04:48,030 --> 00:04:51,090
So I actually got myself into
the industry. And a few years

68
00:04:51,090 --> 00:04:55,230
later after that, you know, I
kept doing open source and so I

69
00:04:55,230 --> 00:04:58,440
created homebrew, which is one
of the biggest open source

70
00:04:58,440 --> 00:05:03,600
projects of all time. At this
point, and I'm sure you'll have

71
00:05:03,600 --> 00:05:04,950
some more questions about that.

72
00:05:05,069 --> 00:05:07,919
Eveline Oehrlich: Yep. Yep, keep
going, because I will come,

73
00:05:07,919 --> 00:05:08,939
we'll come back to that.

74
00:05:10,500 --> 00:05:14,910
Max Howell: So I created
homebrew because at the time, we

75
00:05:14,910 --> 00:05:18,420
were making all these different
apps. I was at last firm in

76
00:05:18,420 --> 00:05:24,210
London, and we made six
different apps. One was for

77
00:05:24,210 --> 00:05:28,440
Linux and Mac and Windows, we
have an iPhone app, and an

78
00:05:28,440 --> 00:05:33,210
Android app, and even this
Blackberry app. And we we build

79
00:05:33,210 --> 00:05:38,940
them all on Mac, because Mac was
like the unified platform, time,

80
00:05:39,000 --> 00:05:42,060
and what it really has become
the platform for development

81
00:05:42,060 --> 00:05:49,350
now. But this was 2008. And at
that time, like, developers

82
00:05:49,350 --> 00:05:51,720
hadn't really decided what
platform they're using, it was

83
00:05:51,720 --> 00:05:55,140
all over the place, it was still
still a lot of people using

84
00:05:55,140 --> 00:05:59,790
Windows for development, which
is not as common nowadays. And

85
00:06:00,180 --> 00:06:02,940
in the our office was an awful
lot of people on Linux, and I

86
00:06:02,940 --> 00:06:07,620
was one of the few who converted
to Mac. Because Apple had a bad

87
00:06:07,620 --> 00:06:12,450
rep. With developers. It's hard
to believe now. But well, it's

88
00:06:12,450 --> 00:06:15,210
easier to believe over the last
few years. But there was

89
00:06:15,210 --> 00:06:18,570
certainly a period where
everyone in development used to

90
00:06:18,570 --> 00:06:22,860
back. But that was new. And we
were using it because it was the

91
00:06:22,860 --> 00:06:25,710
platform that you could do
Android Dev, you could do Linux

92
00:06:25,710 --> 00:06:29,010
Dev, you could have a Windows VM
running. So we could do all six

93
00:06:29,010 --> 00:06:32,490
platforms quite easily. And like
the package management solutions

94
00:06:32,490 --> 00:06:36,630
were the acceptable, they
weren't great. And they

95
00:06:36,630 --> 00:06:40,590
certainly weren't designed for
developers, their impact

96
00:06:40,590 --> 00:06:44,820
managers turned out with Linux.
And in many ways, they were the

97
00:06:44,820 --> 00:06:47,910
thing which defined the Linux
flavor. And it's still still the

98
00:06:47,910 --> 00:06:52,650
case. But I kind of felt that
they were designed for sis ops

99
00:06:52,650 --> 00:06:57,180
and DevOps, but not development.
So I wanted to build something

100
00:06:57,180 --> 00:07:03,900
that was more for that. And so I
got on with it. And after a few

101
00:07:03,900 --> 00:07:06,780
months, I realized that it was
kind of neat. So I open sourced

102
00:07:06,780 --> 00:07:10,020
it. And then it took a few
months before anyone noticed it.

103
00:07:10,260 --> 00:07:14,400
But then when he got noticed, it
took off, like amazingly

104
00:07:14,400 --> 00:07:17,760
quickly, huge amounts of
contribution, but a lot of

105
00:07:17,760 --> 00:07:21,090
excitement, because I've managed
to tap into something which

106
00:07:21,840 --> 00:07:25,500
people wanted, needed,
especially with this burgeoning

107
00:07:25,500 --> 00:07:29,640
developer platform, which was
what the Mac was becoming. And

108
00:07:29,730 --> 00:07:33,810
yeah, so here we are, like,
almost 14 years later. And yeah,

109
00:07:33,810 --> 00:07:36,900
it's it's an enormous project.
At this point. It's very hard to

110
00:07:36,900 --> 00:07:39,510
meet people that haven't heard
of it or used it.

111
00:07:40,680 --> 00:07:48,420
Eveline Oehrlich: You know, I
have watched your speak at Web

112
00:07:48,420 --> 00:07:54,840
Summit in 2022. And you talked
about something there called the

113
00:07:54,840 --> 00:07:58,620
Nebraska problem. And And
honestly, I've been in

114
00:07:58,620 --> 00:08:00,930
infrastructure operations for a
long time, but I'm not a

115
00:08:00,930 --> 00:08:06,540
developer. So maybe that's why
I've never really heard. I've

116
00:08:06,540 --> 00:08:09,960
understood, I've heard of it,
but I really don't understand or

117
00:08:09,960 --> 00:08:14,040
didn't understand what it
actually meant. So can you give

118
00:08:14,040 --> 00:08:16,980
us a quick explanation of what
is meant by the Nebraska

119
00:08:16,980 --> 00:08:20,100
problem? And how it relates to
the open source? And if

120
00:08:20,100 --> 00:08:24,420
possible, any examples you have
seen recently? Maybe?

121
00:08:25,200 --> 00:08:29,400
Max Howell: Yeah, sure. So what
we call the Nebraska problem is

122
00:08:29,400 --> 00:08:32,850
the open source funding problem.
And we call it that because

123
00:08:32,850 --> 00:08:36,630
there's this famous XKCD comic,
you'll see it trotted out

124
00:08:36,630 --> 00:08:40,800
whenever there's any discussion
about open source funding

125
00:08:40,980 --> 00:08:46,800
issues, and represent it shows
like this tower of blocks, and

126
00:08:48,240 --> 00:08:52,140
it represents all modern digital
infrastructure. And then near

127
00:08:52,140 --> 00:08:56,220
the bottom of the towers is
precarious, little pillar, and

128
00:08:56,670 --> 00:08:59,640
it's holding the whole thing up.
And there's an arrow pointing to

129
00:08:59,640 --> 00:09:04,230
it, which says, this open source
package maintained by some

130
00:09:04,380 --> 00:09:12,030
someone from Nebraska, since
2003, thanklessly. Are a

131
00:09:12,030 --> 00:09:16,350
suddenly experienced this with
homebrew, like, it was kind of

132
00:09:16,350 --> 00:09:18,540
the first time I'd really
experienced it, because even

133
00:09:18,540 --> 00:09:21,750
though I'd done some relatively
successful open source before

134
00:09:21,750 --> 00:09:26,220
that, well, you know, I did work
full time on a few projects

135
00:09:26,220 --> 00:09:30,870
before that, but I was I didn't
have a job and didn't know what

136
00:09:30,870 --> 00:09:33,030
I was going to do with myself. I
was still living with my

137
00:09:33,030 --> 00:09:36,990
parents. So it was to do it.
With homebrew, I had to have two

138
00:09:36,990 --> 00:09:41,520
full time jobs. One was paid and
one which didn't. And I took

139
00:09:41,520 --> 00:09:44,040
some time off here and there so
I could work on homebrew full

140
00:09:44,040 --> 00:09:47,610
time. But I never had enough
money saved up to work on it for

141
00:09:47,610 --> 00:09:50,550
a long period of time, but I
couldn't abandon it because it

142
00:09:50,550 --> 00:09:54,960
was well it was my easily at the
time like my greatest creation.

143
00:09:54,960 --> 00:09:59,850
I was extremely proud of it but
also people depended on it. And

144
00:09:59,850 --> 00:10:04,020
I Couldn't let them down, I just
didn't see how I could. So I

145
00:10:04,050 --> 00:10:08,130
work at, you know, my office
job, and often was doing some

146
00:10:08,130 --> 00:10:11,250
homebrew stuff to be honest. And
then in the evenings and the

147
00:10:11,250 --> 00:10:14,580
weekends, yeah, I didn't really
have a social life and just

148
00:10:14,610 --> 00:10:20,010
worked on homebrew. And that was
like, wanted or needed, but it

149
00:10:20,010 --> 00:10:22,740
wasn't sustainable. And in the
end, I did burn out on it. And I

150
00:10:22,740 --> 00:10:26,070
haven't really worked on
homebrew since 2016. I passed it

151
00:10:26,070 --> 00:10:30,030
to the community, which, you
know, I was, I'm lucky that I

152
00:10:30,030 --> 00:10:33,420
could, that suddenly other
people turned up. And this is

153
00:10:33,480 --> 00:10:36,270
how open source is and how it
should work. Of course,

154
00:10:36,480 --> 00:10:40,560
community is essential to open
source. And so it was good that

155
00:10:40,560 --> 00:10:43,770
I could hand it over, but I can
personally keep working on it.

156
00:10:44,670 --> 00:10:48,270
And we see this all over the
place with open source, I think

157
00:10:48,300 --> 00:10:52,020
a lot of the time the people who
make successful open source are

158
00:10:52,020 --> 00:10:54,870
just making a hobby project for
some kind, something they need.

159
00:10:55,200 --> 00:10:57,660
And then it becomes an essential
piece of internet

160
00:10:57,660 --> 00:11:02,220
infrastructure. And they're
stuck in a situation where they

161
00:11:02,220 --> 00:11:07,290
either like, abandon it. And
then that's not what they want

162
00:11:07,290 --> 00:11:11,280
to do. They can't afford to work
on it as much as they need to.

163
00:11:11,880 --> 00:11:15,750
While they try to find funding,
and let's face it most most

164
00:11:15,750 --> 00:11:20,670
funding doesn't doesn't work
very well for open source. A lot

165
00:11:20,670 --> 00:11:23,130
of the time then Nebraska
projects is what we call them,

166
00:11:23,130 --> 00:11:27,150
like famous one couple of years
ago was locked for J which is a

167
00:11:27,150 --> 00:11:31,950
logging piece of logging
software for Java applications.

168
00:11:32,490 --> 00:11:35,580
And there was a massive exploit
found in it where you could root

169
00:11:35,580 --> 00:11:39,600
servers just by typing stuff in
if the stuff you talked to him

170
00:11:39,600 --> 00:11:43,290
would go through the logging
software in some capacity. And

171
00:11:43,980 --> 00:11:47,940
no one had really heard of this
package, because it was deep in

172
00:11:47,940 --> 00:11:50,640
the stack. And the sounds
themselves often works, right.

173
00:11:50,640 --> 00:11:53,910
It's like these essential pieces
of software have been built. But

174
00:11:53,910 --> 00:11:56,130
over time, they've just been
buried on the things that are

175
00:11:56,130 --> 00:11:59,400
built on top of them. And you
got really mature pieces of

176
00:11:59,640 --> 00:12:02,100
software, which still need to be
maintained and everyone's

177
00:12:02,100 --> 00:12:05,610
forgotten exist. A lot of for
Jay was a great example of that.

178
00:12:05,610 --> 00:12:10,440
And they fixed the bug. And they
asked that maybe they could get

179
00:12:10,440 --> 00:12:13,170
some funding in future so that
they could afford to spend more

180
00:12:13,170 --> 00:12:15,870
time actually maintaining the
software and maybe making sure

181
00:12:15,870 --> 00:12:18,900
it doesn't have these security
holes. And I don't think they

182
00:12:18,960 --> 00:12:21,690
ever received anything, because
once the bug was fixed, everyone

183
00:12:21,690 --> 00:12:26,610
forgot about them again. And
then more recently, there was

184
00:12:26,610 --> 00:12:34,260
core J S, nine D in downloads in
is since it was released, the

185
00:12:34,260 --> 00:12:39,480
core of the base of every node
app that exists, every you know,

186
00:12:39,540 --> 00:12:43,590
everything was built on Node or
uses Node, and an essential

187
00:12:43,590 --> 00:12:47,580
piece of infrastructure really,
but the person who maintains it

188
00:12:47,640 --> 00:12:50,820
is giving up because they're
just fed up with the fact that

189
00:12:50,820 --> 00:12:53,370
they don't get any funding.
Everyone uses this thing.

190
00:12:53,370 --> 00:12:57,750
Everyone is extremely entitled
about how they treat open

191
00:12:57,750 --> 00:13:03,240
source. I'm afraid to say it's
often true. And he's just like,

192
00:13:03,240 --> 00:13:05,880
well, I can't afford it I'm
giving up. And what's going to

193
00:13:05,880 --> 00:13:09,540
happen, I don't know, it's not a
good thing to use at the

194
00:13:09,540 --> 00:13:14,700
infrastructure of and, you know,
I, I've done loads of open

195
00:13:14,700 --> 00:13:18,960
source. And the truth is, I
really should just been doing

196
00:13:18,960 --> 00:13:23,190
open source all these years like
I created one homebrew with the

197
00:13:23,460 --> 00:13:28,380
small amount of free time I had.
People I myself shouldn't have

198
00:13:28,380 --> 00:13:33,030
to make a choice between working
on software that improves the

199
00:13:33,030 --> 00:13:36,150
world and is beneficial to like
all layers of the software

200
00:13:36,150 --> 00:13:39,960
stack. Or some of the some of
the jobs I did were not

201
00:13:40,110 --> 00:13:45,870
particularly beneficial to the
world, really. So that's the

202
00:13:45,870 --> 00:13:46,860
Nebraska problem.

203
00:13:48,149 --> 00:13:50,879
Narrator: Skill up it learning
is the perfect online

204
00:13:50,879 --> 00:13:53,489
destination to learn about
DevOps and digital

205
00:13:53,489 --> 00:13:57,389
transformation, anytime,
anywhere. Our digital learning

206
00:13:57,389 --> 00:14:00,239
platform provides you with all
the resources you need to

207
00:14:00,239 --> 00:14:03,779
upskill and learn about these
important topics, including

208
00:14:03,809 --> 00:14:07,949
expert led course videos, access
to certification, prep courses,

209
00:14:08,189 --> 00:14:11,759
and a community of supportive
peers. Our subscription based

210
00:14:11,759 --> 00:14:14,459
model makes it easy for you to
gain the skills needed for

211
00:14:14,459 --> 00:14:18,569
success in the modern workplace.
Visit DevOps institute.com now

212
00:14:18,599 --> 00:14:20,819
and explore our different
learning plans.

213
00:14:21,929 --> 00:14:25,739
Eveline Oehrlich: Wow, wow,
quite quite involved. quite

214
00:14:25,799 --> 00:14:31,379
complicated. And when you
started out just said, you know,

215
00:14:31,379 --> 00:14:34,679
this noble cause of contributing
to the world really with what

216
00:14:34,679 --> 00:14:38,429
you have done with homebrew.
We'll get to T XYZ in a minute.

217
00:14:38,459 --> 00:14:43,499
But I wanted to ask a couple
other questions. Because of the

218
00:14:43,499 --> 00:14:48,209
topic of open source, there's
still some people seem to not

219
00:14:48,209 --> 00:14:52,439
understand that perspective. So
that's why I was very, very

220
00:14:52,439 --> 00:14:56,729
interested in your thoughts and
your journey there. Let's switch

221
00:14:56,729 --> 00:15:01,229
a little bit to the benefits in
terms of of leveraging open

222
00:15:01,229 --> 00:15:06,329
source. You know, take it from
either the benefits into from a

223
00:15:06,329 --> 00:15:10,379
developer perspective
collaborating and into the open

224
00:15:10,379 --> 00:15:15,479
source community, or for the
benefits of the community over

225
00:15:16,349 --> 00:15:20,459
in its large meaning the world
right. Talk a little bit to us

226
00:15:20,459 --> 00:15:24,629
about what you see the benefits
are of open for open, not open

227
00:15:24,629 --> 00:15:26,219
force, but open source.

228
00:15:27,240 --> 00:15:29,880
Max Howell: Yeah, open source is
really interesting. I don't

229
00:15:29,880 --> 00:15:33,840
think that human history is
really got any other equivalent

230
00:15:33,840 --> 00:15:40,230
examples. Or markets usually up
here, because there's an

231
00:15:40,230 --> 00:15:44,550
economic incentive there. Like I
don't know if anything else like

232
00:15:44,550 --> 00:15:48,330
this. But I've been thinking
about this quite a lot recently.

233
00:15:48,330 --> 00:15:53,100
And I remember how in the 90s,
the entire software stack was

234
00:15:53,100 --> 00:15:57,390
basically owned by Microsoft.
And, you know, there was

235
00:15:57,390 --> 00:16:00,840
numerous issues with this. But
certainly, as the Internet

236
00:16:00,840 --> 00:16:04,590
became a thing, it became more
and more clear that Microsoft

237
00:16:04,590 --> 00:16:08,940
was holding the Internet back,
because Internet Explorer was

238
00:16:08,970 --> 00:16:12,780
the only browser that anyone
used, and they didn't want to

239
00:16:12,840 --> 00:16:16,650
advance it. They didn't care,
too, there are no real

240
00:16:16,650 --> 00:16:20,550
incentive. And it was this kind
of attitude, which led to open

241
00:16:20,550 --> 00:16:25,080
source taking more and more of a
hold. In the end, like now, it's

242
00:16:25,080 --> 00:16:27,870
difficult to imagine a world
where open source isn't the base

243
00:16:27,870 --> 00:16:31,260
of every stack, like we say, 97%
of all commercial software has

244
00:16:31,260 --> 00:16:35,940
open source involved in it in
some capacity, and like, this is

245
00:16:35,940 --> 00:16:39,330
a really free percent that
doesn't, I'd like to see what

246
00:16:39,630 --> 00:16:42,990
those source bases and
understand how they get away

247
00:16:42,990 --> 00:16:49,560
with not having any open source
at all. It Over time we, we

248
00:16:49,560 --> 00:16:53,880
built the open source app, and
we swapped off the Microsoft

249
00:16:53,880 --> 00:16:59,730
stack. And it became like the
point where developers realized

250
00:16:59,730 --> 00:17:03,480
that they not only gained like
all this functionality for free,

251
00:17:03,480 --> 00:17:06,570
but also they were getting these
step ups in productivity. And

252
00:17:06,570 --> 00:17:09,870
proficiency is open source is
usually designed to be like

253
00:17:09,870 --> 00:17:13,440
these tiny little self contained
pieces that you slop together,

254
00:17:13,590 --> 00:17:15,930
and you build on top of each
other. And then you've got all

255
00:17:15,930 --> 00:17:18,690
this slight advanced
functionality. But in the

256
00:17:18,690 --> 00:17:21,750
process of replacing Microsoft
with open source, we never

257
00:17:21,750 --> 00:17:24,750
transferred all that money that
makes over making while they

258
00:17:24,750 --> 00:17:28,530
were trillions of dollars in.
And instead, we just sort of

259
00:17:28,710 --> 00:17:33,540
ended up with Freebase without
real any maintenance on it,

260
00:17:33,990 --> 00:17:38,670
which, you know, we're trying to
solve AT T realized the the

261
00:17:38,670 --> 00:17:46,350
value of open sources. People
throw out ideas and the good

262
00:17:46,350 --> 00:17:50,580
ones stick, there's no marketing
behind it. There's no like

263
00:17:50,760 --> 00:17:53,700
someone's trying to sell you
something that actually they

264
00:17:53,700 --> 00:17:57,480
know isn't a good fit for you.
You put some open source out

265
00:17:57,480 --> 00:18:02,190
there. And if it works for the
developers of this world, they

266
00:18:02,190 --> 00:18:06,150
adopt it? Well, this is sort of
people's world, they adopt it.

267
00:18:06,450 --> 00:18:11,550
And then over time, if it
continues to fit, it builds

268
00:18:11,550 --> 00:18:14,250
until it becomes like a mature
block in the stack.

269
00:18:17,760 --> 00:18:22,230
Eveline Oehrlich: So with that
there are challenges as well.

270
00:18:22,380 --> 00:18:27,120
What are some challenges you
would say? Again, from multiple

271
00:18:27,120 --> 00:18:30,480
perspectives, take it from the
developer, take it from the

272
00:18:30,480 --> 00:18:37,230
community or take it from the
software world of vendors out

273
00:18:37,230 --> 00:18:37,620
there?

274
00:18:39,570 --> 00:18:42,450
Max Howell: Well, I think
vendors have the highest risk

275
00:18:42,510 --> 00:18:47,580
right now, right? But if you're
building an app, and it has 6000

276
00:18:47,580 --> 00:18:51,870
dependencies, either, there's no
way you can vet that. And be

277
00:18:51,870 --> 00:18:56,010
sure that all those dependencies
are secure. First off, also

278
00:18:56,010 --> 00:19:00,420
without any malicious insertions
of any kind. It's one thing

279
00:19:00,420 --> 00:19:05,130
we're going to try and fix with
what we're doing at T. But I do

280
00:19:05,130 --> 00:19:10,140
wonder how, you know, if I had
to report to someone and say

281
00:19:10,170 --> 00:19:13,920
that our app is secure, we we
can guarantee that I don't know

282
00:19:13,920 --> 00:19:20,040
how I'd say that with a straight
face 6000 depths in there. So

283
00:19:20,040 --> 00:19:25,440
there's a lot of risk there for
vendors. I think developers,

284
00:19:25,470 --> 00:19:29,220
most of them just enjoy open
source existing because it

285
00:19:29,220 --> 00:19:35,400
really does make computing more
enjoyable in general, especially

286
00:19:35,400 --> 00:19:39,480
now. Like 14 years ago when I
made homebrew, like the amount

287
00:19:39,480 --> 00:19:43,440
of people doing open sources so
much smaller. And now you can't

288
00:19:43,440 --> 00:19:46,170
go a day without dozens of new
things being released

289
00:19:46,170 --> 00:19:50,550
continuously. And some of those
are great. But in most

290
00:19:50,550 --> 00:19:53,220
developers feel bad that they
just consume this open source

291
00:19:53,220 --> 00:19:56,580
and don't give back but I don't
see how they can effectively

292
00:19:56,580 --> 00:20:00,690
give back right because you can
try and spot through a few open

293
00:20:00,690 --> 00:20:04,530
source steps here and there, but
we're all using 1000s and 1000s

294
00:20:04,530 --> 00:20:07,980
of open source projects. And
it's just infeasible without

295
00:20:07,980 --> 00:20:12,450
some layer of automation to
contribute to those properly.

296
00:20:14,670 --> 00:20:19,230
But open source developers are
the ones that I feel for but you

297
00:20:19,230 --> 00:20:21,990
know, most of them are just do
it, they start off by doing side

298
00:20:21,990 --> 00:20:24,150
projects, it really is, it's
like, it's one of the beautiful

299
00:20:24,150 --> 00:20:26,640
things about software
engineering is that it's so

300
00:20:26,640 --> 00:20:29,820
cheap and easy. All you need is
time, that if you're gonna lay

301
00:20:29,820 --> 00:20:31,920
there, and you got some time,
you can, you can create some

302
00:20:31,920 --> 00:20:36,450
open source. But what happens
with all the open source devs,

303
00:20:36,450 --> 00:20:42,480
I've talked to who are, have got
successful projects is that

304
00:20:43,140 --> 00:20:49,170
after a while, it's a huge
amount of work. And you just

305
00:20:49,230 --> 00:20:54,870
can't abandon that. You have to
somehow find the time to do it.

306
00:20:56,010 --> 00:21:00,870
People on myself, we've, the way
I approached, it was always to

307
00:21:00,870 --> 00:21:05,850
try and make it so that as much
as possible, it was robust. And

308
00:21:05,850 --> 00:21:08,520
then but if you did break, it
would tell the user what they

309
00:21:08,520 --> 00:21:10,650
can do to try and fix it
themselves. And they'll make

310
00:21:10,650 --> 00:21:13,650
sure the documentation was as
good as I can get it. And that

311
00:21:13,650 --> 00:21:17,070
would reduce the burden on me as
much as I could, as much as

312
00:21:17,070 --> 00:21:20,100
possible. And then, you know, if
you if you're lucky, a community

313
00:21:20,100 --> 00:21:23,160
turns up, mostly open source,
I've done communities that turn

314
00:21:23,160 --> 00:21:24,180
up, homebrew was

315
00:21:24,690 --> 00:21:27,840
Eveline Oehrlich: really quite
an exception. Hmm, fascinating.

316
00:21:28,170 --> 00:21:31,680
Absolutely fascinating. So
again, at the Web Summit, you

317
00:21:31,680 --> 00:21:35,400
said, I love this vision you
painted, you said, you want to

318
00:21:35,400 --> 00:21:38,760
support those who are passionate
on open source development, but

319
00:21:38,760 --> 00:21:42,150
make sure that they get
rewarded. And you actually call

320
00:21:42,150 --> 00:21:46,890
it I love this to digital
revolution for developers. What

321
00:21:46,950 --> 00:21:51,840
do you mean by that? And then
we'll get to T XYZ, because I

322
00:21:51,840 --> 00:21:55,980
think there is a relationship.
So tell us about the digital

323
00:21:55,980 --> 00:21:57,450
revolution for the developers?

324
00:21:58,440 --> 00:22:03,030
Max Howell: Well, let's face it,
in the sense of web two, as we

325
00:22:03,030 --> 00:22:07,110
call it, which was, you know,
mid 2000s, till, kind of

326
00:22:07,110 --> 00:22:12,360
recently, I'd say, a huge amount
of money was made by people

327
00:22:12,360 --> 00:22:16,620
building on top of open source.
And it was gradual, you know,

328
00:22:16,620 --> 00:22:20,100
like the way open source became
so essential to what we're

329
00:22:20,100 --> 00:22:23,910
building and software was
gradual. And I think a lot of

330
00:22:23,910 --> 00:22:30,270
these essential pieces, if they
were built by hackers who were

331
00:22:30,270 --> 00:22:33,480
building them for other hackers,
and then when web two turned up

332
00:22:33,480 --> 00:22:36,180
and figured out how to monetize
all that, there wasn't any

333
00:22:36,210 --> 00:22:39,000
transition of wealth. And so
that's why it's a digital

334
00:22:39,000 --> 00:22:44,910
revolution. These huge, mega
billion dollar companies that

335
00:22:45,510 --> 00:22:49,080
have never really given any
substantial amount to the open

336
00:22:49,080 --> 00:22:55,320
source that they depend on. It's
not right, like we were okay

337
00:22:55,320 --> 00:22:58,980
with it as it happened, because
it was gradual. But now we like

338
00:22:58,980 --> 00:23:04,260
hanging on. You know, why am I
working two full time jobs when

339
00:23:04,320 --> 00:23:07,440
Microsoft has billions and
billions of dollars coming in

340
00:23:07,530 --> 00:23:12,780
every month? Doesn't make sense.
So what we're trying to build a

341
00:23:12,780 --> 00:23:19,710
T is a way to successfully
remunerate how open source is

342
00:23:19,710 --> 00:23:25,170
funded. There's been lots of
attempts, sponsorship, bounties,

343
00:23:25,200 --> 00:23:30,000
things like this, but we see
that as being reward favorites.

344
00:23:30,030 --> 00:23:32,550
And some people make a lot of
money with sponsorship. But the

345
00:23:32,550 --> 00:23:36,180
truth is, there's millions of
open source packages. And most

346
00:23:36,180 --> 00:23:41,940
of them are Nebraska projects
that nobody knows about. React

347
00:23:42,030 --> 00:23:45,180
doesn't need to be funded,
because Facebook, pay for it

348
00:23:45,180 --> 00:23:48,210
essentially, which is another
thing we want to fix instantly.

349
00:23:48,210 --> 00:23:51,420
I don't believe that these huge
corporations should have so much

350
00:23:51,420 --> 00:23:56,760
agenda over how these open
source projects that the

351
00:23:56,760 --> 00:23:59,700
community needs to figure out
what makes sense for open

352
00:23:59,700 --> 00:24:05,010
source. Facebook could just like
close down react tomorrow. I'd

353
00:24:05,010 --> 00:24:09,420
want with what we're building a
tee is for these developers at

354
00:24:09,420 --> 00:24:13,290
Facebook and Microsoft who do
work on open source or who are

355
00:24:13,290 --> 00:24:16,380
just extremely talented people
to be able to quit and work on

356
00:24:16,380 --> 00:24:19,710
open source full time because
the pay is exactly the same or

357
00:24:19,710 --> 00:24:24,060
better. Even, frankly, we're
building an economy on code. And

358
00:24:24,270 --> 00:24:27,360
if you make a bunch of
successful open source packages,

359
00:24:27,630 --> 00:24:31,110
providing you're using the T
protocol, you could be much

360
00:24:31,110 --> 00:24:33,690
wealthier than you are currently
as a software developer.

361
00:24:35,790 --> 00:24:39,180
Eveline Oehrlich: Interesting.
So how can developers get

362
00:24:39,180 --> 00:24:43,740
involved if I am sure we have a
few of those who are going ooh,

363
00:24:43,740 --> 00:24:48,060
I love this. I love hearing
this. How can they get involved?

364
00:24:49,349 --> 00:24:52,829
Max Howell: Well, four months
ago I released the tea package

365
00:24:52,829 --> 00:24:58,769
manager so tea is the successor
to brew you know I've stopped

366
00:24:58,769 --> 00:25:04,469
working on brew in mid to 2010s.
And I didn't stop thinking about

367
00:25:04,499 --> 00:25:09,599
what could be better about
broom. But I didn't really have

368
00:25:09,989 --> 00:25:14,039
a good incentive I feel to make
another brutal I'd already done

369
00:25:14,039 --> 00:25:20,459
it. And I remember the tireless
hours that I spent working on it

370
00:25:20,489 --> 00:25:24,179
robust, defying it, filling out
the package graph, answering

371
00:25:24,179 --> 00:25:26,789
people's support questions, and
just like trying to make sure it

372
00:25:26,789 --> 00:25:29,999
was a successful and useful
piece of software. But I didn't

373
00:25:29,999 --> 00:25:36,089
really want to go and do that
again. I think I like doing

374
00:25:36,089 --> 00:25:39,719
things once usually, but But 18
months ago, while I was looking

375
00:25:39,719 --> 00:25:43,349
into web three, and crypto,
which basically was the first

376
00:25:43,349 --> 00:25:45,239
time I've looked into this
stuff, because I've never been

377
00:25:45,239 --> 00:25:49,709
particularly interested in it. I
thought Bitcoin was pretty

378
00:25:49,709 --> 00:25:55,049
impressive. As a wonderful
mystery story to it also pretty

379
00:25:55,079 --> 00:26:01,109
ingenious, how it works, and how
it has successfully become so

380
00:26:01,109 --> 00:26:04,949
large, even though it's not
controlled. But I never really

381
00:26:04,949 --> 00:26:07,889
thought that there was much more
that was interesting there for

382
00:26:07,889 --> 00:26:11,009
me. But as I was diving into web
three stuff, I started to see

383
00:26:11,009 --> 00:26:14,309
the different contracts were
actually pretty interesting, I

384
00:26:14,309 --> 00:26:18,629
had this moment of inspiration
while messing around an open sea

385
00:26:18,629 --> 00:26:23,729
with NF Ts. Because NF T's
allowed you to well view the

386
00:26:23,729 --> 00:26:28,529
open sea digital contract anyway
forced any repurchases to put

387
00:26:28,529 --> 00:26:33,179
10% back to the original
creator. I was like, Oh, so you

388
00:26:33,179 --> 00:26:35,939
can write a digital contract
that funnels money automatically

389
00:26:35,939 --> 00:26:38,789
to different entities. And I
realized suddenly that the

390
00:26:38,789 --> 00:26:43,469
package graph of all open source
was that kind of, if you put a

391
00:26:43,469 --> 00:26:46,859
digital contract on it anyway,
you could funnel money to all

392
00:26:46,859 --> 00:26:50,159
the dependencies, all the
packages. So all these Nebraska

393
00:26:50,159 --> 00:26:54,629
projects could get like little
bits of token from any insertion

394
00:26:54,629 --> 00:26:57,389
of token near the top. And
that's how it works, right? Like

395
00:26:57,389 --> 00:27:00,179
people sponsor these big
projects of near the top. And

396
00:27:00,179 --> 00:27:04,289
it's all the ones underneath
that that are failing, have are

397
00:27:04,289 --> 00:27:09,359
maintained by people without any
things. And they're essential

398
00:27:09,359 --> 00:27:14,009
pieces of the digital
infrastructure. Without them,

399
00:27:14,039 --> 00:27:16,769
everything would collapse. I
like everything being that

400
00:27:16,769 --> 00:27:20,249
fragile is scary, like people
remember that for a long time?

401
00:27:20,519 --> 00:27:24,689
How it could all collapse any
moment, but somehow it doesn't.

402
00:27:24,809 --> 00:27:29,639
And we're just waiting for a
bigger claps really. But also,

403
00:27:29,639 --> 00:27:32,369
like, think about what would
happen if it was all funded

404
00:27:32,369 --> 00:27:36,839
correctly. People myself, like
make, like occasionally big and

405
00:27:36,839 --> 00:27:40,289
important projects. But we have
to do it in our spare time. What

406
00:27:40,289 --> 00:27:44,099
if I was working on that full
time? And then, you know, we

407
00:27:44,189 --> 00:27:48,689
find security holes and open SSL
every two months? Well, if it

408
00:27:48,689 --> 00:27:52,949
was funded correctly, so that
well, either of us so could

409
00:27:52,949 --> 00:27:57,059
really afford to hire people who
could, you know, like kind of

410
00:27:57,059 --> 00:28:01,889
like a corporation is what we're
thinking things light, each of

411
00:28:01,889 --> 00:28:05,429
these big projects will have
their own dowels on the T

412
00:28:05,429 --> 00:28:11,099
protocol. And they can operate
them. Similar to a kind of

413
00:28:11,099 --> 00:28:15,119
corporation, they can hire
people using the token, they can

414
00:28:15,149 --> 00:28:19,589
split the rewards that they're
getting from the tee protocol

415
00:28:19,829 --> 00:28:24,599
across everyone is contributing
to open SSL. So if you open a

416
00:28:24,599 --> 00:28:27,689
pull request, you could end up
getting paid for that pull

417
00:28:27,689 --> 00:28:32,009
request. Because the DAO is a
set of smart contracts that

418
00:28:32,039 --> 00:28:35,129
figure out how to channel the
token based on the governance

419
00:28:35,129 --> 00:28:41,099
structure that those projects
have chosen. CL I came up with

420
00:28:41,099 --> 00:28:44,699
the idea, like build a package
manager and build it on top of

421
00:28:44,699 --> 00:28:48,839
blockchain. So that open source
can be remunerated, then we

422
00:28:48,839 --> 00:28:53,369
managed to raise 80 million, and
we've been building it since. So

423
00:28:53,399 --> 00:28:56,669
I released the package manager
itself four months ago, got six

424
00:28:56,669 --> 00:29:00,029
sales and styles now got some
really passionate users. I think

425
00:29:00,029 --> 00:29:03,899
it's special. And if anyone is
interested in getting involved

426
00:29:03,899 --> 00:29:06,389
with what we're doing, that's
the that's the place to start.

427
00:29:06,389 --> 00:29:12,929
Currently, what we got is
github.com/t. XYZ and you can

428
00:29:12,959 --> 00:29:17,129
check it out, install it, and I
think it's a better package

429
00:29:17,129 --> 00:29:21,239
manager the brew it's, it's more
than a package manager, really,

430
00:29:21,929 --> 00:29:25,079
the ideas I had over the years
for how the package manager

431
00:29:25,079 --> 00:29:28,439
could be improved. I realized
that it's a unique piece of

432
00:29:28,439 --> 00:29:31,139
software, right? It sits
underneath all other tooling.

433
00:29:31,499 --> 00:29:35,309
There's so many things that you
can do, which nobody's really

434
00:29:35,309 --> 00:29:37,709
tried before. I don't really get
it I think perhaps package

435
00:29:37,709 --> 00:29:42,149
managers just aren't very sexy.
Is the truth and don't many

436
00:29:42,149 --> 00:29:44,159
people that are interested in
working on them but for some

437
00:29:44,159 --> 00:29:48,419
reason I've always enjoyed it. I
love getting into that sort of

438
00:29:48,419 --> 00:29:49,349
area of it.

439
00:29:50,520 --> 00:29:52,740
Eveline Oehrlich: It is a
brilliant idea. You are

440
00:29:52,740 --> 00:29:57,930
brilliant. I am honored to talk
to you. If I come up with a name

441
00:29:57,930 --> 00:30:01,350
for this that I will share it
with You but it is. It's

442
00:30:01,350 --> 00:30:06,840
certainly a super cool and
absolutely honorable, really

443
00:30:06,840 --> 00:30:10,350
honorable thing you're doing. I
love the idea. I'm, I'm very

444
00:30:10,350 --> 00:30:17,250
grateful. Now I have one closing
question. Hmm, maybe it involves

445
00:30:17,250 --> 00:30:20,250
beer drinking, but what do you
do for fun? What do you do for

446
00:30:20,250 --> 00:30:24,570
fun if you don't do these
brilliant things and manage

447
00:30:24,570 --> 00:30:28,800
these fantastic things? You are
You were you were talking about?

448
00:30:30,030 --> 00:30:34,590
Max Howell: Yeah, well, so like,
programming was always a

449
00:30:34,590 --> 00:30:39,690
hobbies. Switch when I switch to
doing it for career did become

450
00:30:39,690 --> 00:30:42,420
kind of what I did one as well.
But certainly, you know, I used

451
00:30:42,420 --> 00:30:46,290
to live in the UK. I live in the
States now. And I used to love

452
00:30:46,290 --> 00:30:50,040
going to pubs friends. So there
was some beer drinking in my

453
00:30:50,040 --> 00:30:55,260
palace for sure. But I got a son
now he's 10 months. And so

454
00:30:55,290 --> 00:30:57,720
there's certainly less beer in
my life at the moment.

455
00:30:59,040 --> 00:31:03,150
Otherwise, for fun, so Blimey, I
do like video games here and

456
00:31:03,150 --> 00:31:10,530
there on Sunday not playing them
as much as I used to. And I like

457
00:31:10,530 --> 00:31:15,030
hiking or going out or camping
isn't some of the things I

458
00:31:15,030 --> 00:31:15,690
really enjoy.

459
00:31:16,200 --> 00:31:18,600
Eveline Oehrlich: Excellent.
Well, if you make it to my part

460
00:31:18,600 --> 00:31:21,360
of the world, I'm happy to take
you out for beer. We got some

461
00:31:21,390 --> 00:31:27,120
pretty good beer in Germany. So
yes, yes. Well, this has been

462
00:31:27,120 --> 00:31:31,230
fantastic. Thank you so much
Max. We have been talking to max

463
00:31:31,230 --> 00:31:37,440
Howell CEO of T dot XYZ and
brilliant thought leader in the

464
00:31:37,440 --> 00:31:40,830
topic of open source Mexican,
thank you so much for joining me

465
00:31:40,830 --> 00:31:45,840
today on humans of DevOps
podcast. Thank you very much.

466
00:31:47,190 --> 00:31:51,030
Humans of DevOps podcast is
produced by DevOps Institute.

467
00:31:51,510 --> 00:31:54,870
Our audio production team
includes Julia pape, Daniel

468
00:31:54,870 --> 00:31:58,800
Newman, Schultz and Brendan Lee,
shout out to my colleagues who

469
00:31:58,800 --> 00:32:03,030
do a fantastic job. I'm humans
of DevOps podcast, Executive

470
00:32:03,030 --> 00:32:06,660
Producer, evolutionarily, if you
would like to join us on a

471
00:32:06,660 --> 00:32:12,480
podcast, please contact us at
humans of DevOps podcast at

472
00:32:12,510 --> 00:32:17,370
DevOps institute.com. I'm
evolutionarily talk to you soon.

473
00:32:19,830 --> 00:32:21,930
Narrator: Thanks for listening
to this episode of the humans of

474
00:32:21,930 --> 00:32:25,500
DevOps podcast. Don't forget to
join our global community to get

475
00:32:25,500 --> 00:32:28,860
access to even more great
resources like this. Until next

476
00:32:28,860 --> 00:32:32,280
time, remember, you are part of
something bigger than yourself.

477
00:32:32,610 --> 00:32:33,360
You belong