1
00:00:00,000 --> 00:00:06,080
no no no it's all on my side i am the producer i'm the producer josh so it's all my responsibility

2
00:00:06,080 --> 00:00:12,620
i hope i hope you found that all the brown m&ms were reduced from your bowl as as as your

3
00:00:12,620 --> 00:00:17,860
requirement yeah yeah trying to make everyone happy all right where did i see that i saw an

4
00:00:17,860 --> 00:00:22,120
episode i saw a tv show where that was like uh some character was established as being who

5
00:00:22,120 --> 00:00:29,980
who he or she says he was because those uh m&ms were well that that's that's a uh that's folklore

6
00:00:29,980 --> 00:00:35,420
which i think is real i have to look up on snopes around like van halen's writer that the way what

7
00:00:35,420 --> 00:00:42,120
they would do this is this is a uh a hack if you will is in order to make sure that people at the

8
00:00:42,120 --> 00:00:46,520
concert venues had the attention to detail that was required for them to be happy and assumably

9
00:00:46,520 --> 00:00:51,540
profitable they would have they would have this one little thing in the contract that was an absurd

10
00:00:51,540 --> 00:00:56,420
detail that they should have a bowl of m&ms with all the brown m&ms removed and that way when they

11
00:00:56,420 --> 00:01:01,280
showed up it was this quick heuristic where like well if they didn't read that part and actually do

12
00:01:01,280 --> 00:01:06,120
it then they've probably screwed up a bunch of other stuff so we need to go check on it so that's a

13
00:01:06,120 --> 00:01:12,260
little uh that's a little like uh preparatory sink bomb that you put in there i see like a

14
00:01:12,260 --> 00:01:20,340
a litmus test or something a veritable canary deploy in some 80s way i dig it okay well that was

15
00:01:20,340 --> 00:01:27,680
there's two references then uh both of which escaped me so nominally you you and i uh so i i thought to

16
00:01:27,680 --> 00:01:31,920
have you on because i got a tweet from someone the other day that they were listening to uh the maybe

17
00:01:31,920 --> 00:01:36,440
maybe not the last podcast because the last one we did was actually a little a little uh fun little

18
00:01:36,440 --> 00:01:41,260
video thing on my old my Coté show podcast which i like doing that was fun you know why it's fun

19
00:01:41,260 --> 00:01:45,480
it's because it's because we got to hang out which i you and i are in the same team people don't

20
00:01:45,480 --> 00:01:50,800
realize that but geez if i get to see you once or twice a year i'm lucky you know well we speaking of

21
00:01:50,800 --> 00:01:54,420
we're going to see each other next week in atlanta because because you're going to be you're going to

22
00:01:54,420 --> 00:02:00,740
be at devops days uh atlanta as as the listener might be asking and we were talking about this and

23
00:02:00,740 --> 00:02:05,160
i am and i was thinking like this would be a good topic for us to go over is is the relationship

24
00:02:05,160 --> 00:02:11,200
between uh microservices and devops because i think you know i don't want to speak on your behalf

25
00:02:11,200 --> 00:02:15,140
but i but i wouldn't say but i would love you i would love it if you if you would in fact

26
00:02:15,140 --> 00:02:21,020
i i've asked you if you wanted to join me in the talk and you seem to be reluctant to say yes but

27
00:02:21,020 --> 00:02:26,980
it'd be great if you were to give the talk to just prepare it i mean really just take the take over

28
00:02:26,980 --> 00:02:32,740
the whole thing just be a team player do it all for me you know i mean that'd be uh you know i'm

29
00:02:32,740 --> 00:02:39,360
sympathetic to that because that is basically uh also my proactive uh question to everyone that i meet

30
00:02:39,360 --> 00:02:44,580
in my life which is yeah if you would if you would like to do things i am supposed to do in place of

31
00:02:44,580 --> 00:02:51,640
me doing it go ahead that's fine and how's that stroked you so far because i'm starting to think

32
00:02:51,640 --> 00:02:57,420
that might be the uh well ingredient i find i find that in most walks of life well there's three

33
00:02:57,420 --> 00:03:03,520
instances where it doesn't work but in other walks of life in general it does oh good and in in in uh

34
00:03:03,520 --> 00:03:10,040
would it be uh uh ascending order of importance uh okay which is to say the last one is the most

35
00:03:10,040 --> 00:03:14,440
critical the first one is when you're with a group of people and you're trying to figure out where to

36
00:03:14,440 --> 00:03:19,940
go eat for lunch dinner occasionally breakfast and everyone knows that the i don't care where we go

37
00:03:19,940 --> 00:03:25,780
i'm out like that's that's the most annoying place so you do actually need to like be involved in the

38
00:03:25,780 --> 00:03:30,780
situation oh i don't know about that i'm just saying i don't care i know and i know there are people in

39
00:03:30,780 --> 00:03:36,160
the group who do care so i purposely say i punt whatever you want you know and and they would they

40
00:03:36,160 --> 00:03:42,520
would exert control anyway yeah you know you know i am no i am no camu or sartre but sometimes you

41
00:03:42,520 --> 00:03:47,980
just have to pretend that you care and just act in that charade right okay just just to get along

42
00:03:47,980 --> 00:03:52,280
with the other people uh so the giving the giving thing in that case is to take away the choice

43
00:03:52,280 --> 00:03:58,040
yeah yeah yeah or just or just get involved now the second situation is is whoever it is that is

44
00:03:58,040 --> 00:04:03,740
paying you occasionally they would like you to actually do the thing that uh that you're supposed

45
00:04:03,740 --> 00:04:09,240
to do instead of pawning it off on someone else now that's not always the case but i keep checking

46
00:04:09,240 --> 00:04:15,160
my spam mail to see if i get memos or anything from those folks i don't even know if they're if they

47
00:04:15,160 --> 00:04:21,020
know i work for them i i don't know man and then and then i think i think the third tier is uh what you

48
00:04:21,020 --> 00:04:25,900
know whoever you have some significant relationship with whether it's where it's a really good friend

49
00:04:25,900 --> 00:04:31,600
some significant other or whatever they tend to like you to uh actually do things uh around

50
00:04:31,600 --> 00:04:38,600
and in that regard i'm completely 100% agreed yeah yeah so so other than that i think i think yours and

51
00:04:38,600 --> 00:04:46,800
i's strategy about uh uh how how one one might pursue life happiness which is to delegate uh work to

52
00:04:46,800 --> 00:04:52,300
other people or professional hunting yeah i think it's solid i think that's a good approach that's a

53
00:04:52,300 --> 00:04:58,300
that's that's a hell of a hell of a way to run a railroad my friend so nominally yeah you've got you've

54
00:04:58,300 --> 00:05:03,520
got the the the microservices and the devops and i came up with this because i was reading the abstract

55
00:05:03,520 --> 00:05:08,360
for the uh for the talk that you have next week and that seemed to be the topic that it was on it was

56
00:05:08,360 --> 00:05:14,060
about it's about you know i spend well here's the problem i'm a little constrained we're both a little

57
00:05:14,060 --> 00:05:16,840
constrained actually isn't the talk supposed to be like 40 minutes or something like that it's

58
00:05:16,840 --> 00:05:21,300
30 minutes 30 minutes and i i don't know if you've ever had the dubious pleasure of

59
00:05:21,300 --> 00:05:29,560
seeing what am i uh what am i talked but they go on forever and and and though i do go fast uh you

60
00:05:29,560 --> 00:05:33,940
know i don't know if i'm going to be able to fit anything um i don't have anything i don't have

61
00:05:33,940 --> 00:05:39,800
anything much to say of wisdom uh uh anyway but 30 minutes my goodness if ever there was going to be

62
00:05:39,800 --> 00:05:44,520
if i'm going to say something worth listening to i'm not sure i can fit it in 30 minutes uh so

63
00:05:44,520 --> 00:05:51,000
so my constraint is how can i how can i bring the spring and do all the java stuff that i do

64
00:05:51,000 --> 00:05:55,840
in 30 minutes and and and and i would and that's normally for a developer audience so that's you

65
00:05:55,840 --> 00:06:00,380
know really that's not even the devops that's just it's just dev i care about i couch what i talk

66
00:06:00,380 --> 00:06:04,620
about in terms of the end result you know the outcome which is what i care about the most getting

67
00:06:04,620 --> 00:06:11,320
software into production right but but how do we how do we given the given that this is not dev days

68
00:06:11,320 --> 00:06:17,860
atlanta this is devops days um you know how do we really drive home that extra point and then how do

69
00:06:17,860 --> 00:06:26,700
we do so uh in uh in 30 minutes and by and so you may have noticed that i said we here i'm not

70
00:06:26,700 --> 00:06:32,260
i'm not schizophrenic well listen listen let me let me i i think i think what this episode short

71
00:06:32,260 --> 00:06:36,820
episode will be will be a little bit of consulting first of all first of all i want to say that you

72
00:06:36,820 --> 00:06:42,840
have just highlighted a key thing to the uh to the life choice of delegating and that is you're

73
00:06:42,840 --> 00:06:49,700
concerned in in this this delegation based lifestyle with outcomes you want outcomes you want you're

74
00:06:49,700 --> 00:06:56,260
you're concerned about what happens not necessarily how it happens now obviously you want it to be done

75
00:06:56,260 --> 00:07:03,860
in an ethical moral and legal method right these these are base assumptions you you part of having a

76
00:07:03,860 --> 00:07:10,200
positive outcome is ensuring that how that positive outcome was achieved does not in any way make it a

77
00:07:10,200 --> 00:07:15,920
negative outcome right like so that's that's one of the requirements is that uh the the outcome

78
00:07:15,920 --> 00:07:23,360
must be uh sustained by the purity of how it was achieved anyhow okay i'm saying if you had a good

79
00:07:23,360 --> 00:07:28,340
outcome right yeah and then it turns out that you basically crimed your way to that outcome

80
00:07:28,340 --> 00:07:36,360
that outcome would be less uh and perhaps even ruined uh is that so yeah i think so for example

81
00:07:36,360 --> 00:07:42,500
consider consider numerous instances of when journalists made things up right let's say you

82
00:07:42,500 --> 00:07:48,740
you're a journalist and you write a story and you fabricate 20 of the facts which were not facts in

83
00:07:48,740 --> 00:07:54,540
your story right so 20 of those are wrong but then how about the other 80 those are perfectly real

84
00:07:54,540 --> 00:08:00,080
but the end result is your entire story is garbage right yeah but that's okay that's the empirical

85
00:08:00,080 --> 00:08:05,060
sort of like you know common i don't know but what what is a crime in some countries is not a crime

86
00:08:05,060 --> 00:08:10,060
in others i'm just saying sure sure i mean i mean what you want to have is uh you got to have your

87
00:08:10,060 --> 00:08:15,600
bounded context for any any situation in life that you're trying to apply principles to so i'll give

88
00:08:15,600 --> 00:08:22,720
you that i'll give you that anyways you know maybe avoiding taxes in certain geographies is uh legal and

89
00:08:22,720 --> 00:08:28,080
ethical versus other geographies and so when you cross these boundaries the outcome being like good

90
00:08:28,080 --> 00:08:32,320
financial reports the way that you arrange your finances and your tax schemes and everything

91
00:08:32,320 --> 00:08:38,120
is all is all uh i don't know copacetic as my dad used to say depending on the geography you're in

92
00:08:38,120 --> 00:08:41,900
so there you go well we got that one cinched up so you know what you're going to be talking about

93
00:08:41,900 --> 00:08:48,740
next week now not actually i'm more confused than when i started wait so so you are you in uh i'll

94
00:08:48,740 --> 00:08:54,140
i'll help you josh okay let's do this so devops i you know i got that great quote in the beginning of

95
00:08:54,140 --> 00:08:58,680
the uh in the abstract right that was i thought that was nice i thought that was me kind of like uh

96
00:08:58,680 --> 00:09:05,760
you know i speak some of the language i walk among them i care about the same things i just want to

97
00:09:05,760 --> 00:09:10,740
really polish this off so that it's more i'm going to look up this abstract because unlike yourself i

98
00:09:10,740 --> 00:09:15,960
don't have uh i don't have it memorized i had the w edwards demings uh quote you know survival is not

99
00:09:15,960 --> 00:09:22,980
mandatory nobody says you need to to survive now see this is this is this is uh this is a good instance

100
00:09:22,980 --> 00:09:29,620
of uh synergies i think are coincidence is the the subtitle of the talk that i give nowadays is

101
00:09:29,620 --> 00:09:35,880
beyond survival is not mandatory oh my these might fold into each other very well josh wait am i am i

102
00:09:35,880 --> 00:09:42,060
preceding you i can i can you are preceding you're you're the first day and then i'm the the the opening

103
00:09:42,060 --> 00:09:46,940
talk for the second day so so here's here's my here's what i want to hear from you just thinking

104
00:09:46,940 --> 00:09:53,460
out loud about this is let let me let me just let me just soliloquy here a little bit uh if if i don't

105
00:09:53,460 --> 00:09:58,460
know what pentameter that would be in uh well first of all you've got itell reference in there so that's

106
00:09:58,460 --> 00:10:02,480
exciting i'm actually reading your abstract now but i would assume that the connection between

107
00:10:02,480 --> 00:10:09,200
microservices and devops is the following so first of all you've got devops and in my in my way of

108
00:10:09,200 --> 00:10:15,300
summarizing it the outcome of devops is that you can do something like release your software weekly if

109
00:10:15,300 --> 00:10:21,400
not daily and thus get this fast feedback loop that allows you to see if your idea of what the software

110
00:10:21,400 --> 00:10:26,920
should be doing is accurate and is solving your users problems and making them more productive now the

111
00:10:26,920 --> 00:10:32,060
ops part of that is that it would also be nice just as if it's nice for your outcome to not have been

112
00:10:32,060 --> 00:10:37,740
illegal it's also nice if your software actually stays up and running and can be used by people instead of

113
00:10:37,740 --> 00:10:43,600
crashes all the time or as we used to say fail whales so that's a lot of in my mind what the outcome of

114
00:10:43,600 --> 00:10:49,600
devops is right is supporting that kind of thing now how you achieve that uh is done through a whole bunch

115
00:10:49,600 --> 00:10:56,460
of like a a a tricky bag of interesting practices like having a unified team of people so that they

116
00:10:56,460 --> 00:11:00,840
take care of the thing all throughout and developers write their code to be more resilient and operations

117
00:11:00,840 --> 00:11:06,020
people know enough about writing code so that they write platforms and stuff so that it's resilient and

118
00:11:06,020 --> 00:11:11,560
you know they're working with each other and they all they all have skin in the game as as as our as that

119
00:11:11,560 --> 00:11:18,680
talib guy would say uh and and so another thing you have is you've got a a build pipeline so you can run

120
00:11:18,680 --> 00:11:22,960
tests and all this stuff and you automate as much as possible out of everything so you actually have

121
00:11:22,960 --> 00:11:28,000
the speed and then i would suspect i've never really seen this proven out but i would suspect that

122
00:11:28,000 --> 00:11:33,400
architecturally one of the better things you can do with how you arrange your code base and also mediate

123
00:11:33,400 --> 00:11:39,900
how some components talk with each other is to follow a microservices approach now without ditching

124
00:11:39,900 --> 00:11:46,440
into like how is microservices not just soa the idea of a microservices approach is that if we have

125
00:11:46,440 --> 00:11:54,020
a bunch of different services like i don't know what are some services josh oh there are some i don't

126
00:11:54,020 --> 00:11:58,420
know customer service order service fulfillment service product service you know yeah sure anything

127
00:11:58,420 --> 00:12:04,100
anything matching your bounded context in your there you go so you might you might have a a login service

128
00:12:04,100 --> 00:12:11,620
a service that tells you uh where a package is located in a warehouse or you might have a service that

129
00:12:11,620 --> 00:12:17,660
if you give it two images graphs them together so you can put a hat on a cat whatever uh and so

130
00:12:17,660 --> 00:12:23,660
the idea of my one of the organizational ideas of microservices by the way thanks for being here to

131
00:12:23,660 --> 00:12:27,700
listen to me just talk for this podcast hey it's it's what i've been looking forward to all week

132
00:12:27,700 --> 00:12:33,500
one of the one of the organizational ideas of microservices is that once you decompose your

133
00:12:33,500 --> 00:12:38,720
software and all these services uh you want them to be able to release on their own cadence and to

134
00:12:38,720 --> 00:12:44,860
some extent even innovate and evolve on their own cadence yeah but of course once you do that if you've

135
00:12:44,860 --> 00:12:50,020
got too strong of dependencies between all of them things start to break so if if i'm used to sending

136
00:12:50,020 --> 00:12:56,140
a jpeg two jpegs to the hat on cat service and then all of a sudden they don't support jpegs or they've

137
00:12:56,140 --> 00:13:02,500
had to support the jpeg 2000 or whatever uh and then they support pngs or pings instead and they

138
00:13:02,500 --> 00:13:07,300
don't tell me about this i keep sending jpegs to them and it breaks and then i lose millions of

139
00:13:07,300 --> 00:13:12,880
dollars a minute because i can't put a hat on a cat right and and so the idea of microservices is how

140
00:13:12,880 --> 00:13:20,020
might we redo things so that we can be releasing these services and updating them uh on their own cadence but

141
00:13:20,020 --> 00:13:26,220
also have the resiliency of one they they stay up and running but two as as different components start using

142
00:13:26,220 --> 00:13:33,160
them uh they they more gracefully fail or or you know they don't sort of just like um as soil the bed

143
00:13:33,160 --> 00:13:38,260
as people might say in a more saucy phrasing that's my very high level thing of how these two things

144
00:13:38,260 --> 00:13:44,100
might connect together what that actually means on a day-to-day basis as you know josh i am the person on our

145
00:13:44,100 --> 00:13:49,820
team who does slides so that's that's way beyond the slides not even in the notes section of powerpoint

146
00:13:49,820 --> 00:13:57,540
how this all actually works together well so in my in my in my talks uh i talk i focus on that one

147
00:13:57,540 --> 00:14:04,040
little bit which is uh you optimize for human or optimize your human organizations by uh uh factoring

148
00:14:04,040 --> 00:14:10,320
out your software into smaller chunks of work batches of work uh and then you optimize for each team

149
00:14:10,320 --> 00:14:16,300
to be able to release uh independently by moving these these separate sort of bound contexts into

150
00:14:16,300 --> 00:14:23,880
microservices uh and uh once you've done that you get in theory you get the ability to deliver software

151
00:14:23,880 --> 00:14:30,320
independently but the whole focus of my talk is uh going into practice which is that when you've got

152
00:14:30,320 --> 00:14:34,980
all these things communicating with each other over the network across network partitions uh you've got

153
00:14:34,980 --> 00:14:38,780
instability you've got things that could break and you need to build the architect accordingly and so

154
00:14:38,780 --> 00:14:42,560
that's really what i talk about my you know in this whole thing is how to build applications how to

155
00:14:42,560 --> 00:14:47,520
build java software using spring these days uh that that does the right thing in the face of service

156
00:14:47,520 --> 00:14:51,560
strategies and topology changes that benefits from the elasticity of a cloud environment you know that

157
00:14:51,560 --> 00:14:56,900
takes advantage of all these sort of uh characteristics so that you don't just invite more pain into your

158
00:14:56,900 --> 00:15:01,800
world now that said um all that other stuff you mentioned is good too so really i just i honed in

159
00:15:01,800 --> 00:15:05,700
on that one little thing which is that you which is what you mentioned now now you know you know

160
00:15:05,700 --> 00:15:10,460
knowing your talks and and unfortunately as we were talking about earlier your uh your your zoom

161
00:15:10,460 --> 00:15:15,480
function is broken so people will miss out on that but you know what i'm thinking what i'm thinking

162
00:15:15,480 --> 00:15:21,320
is good for a situation like this like a 30 minute talk to general audience is the wrong term but to a

163
00:15:21,320 --> 00:15:26,300
a non-developer audience right because normally what you would do is you would just roll in and be like

164
00:15:26,300 --> 00:15:32,160
i'm gonna demo stuff to you for the next uh 50 minutes right no better i'm gonna write software

165
00:15:32,160 --> 00:15:36,420
exactly exactly right i'm gonna dim it i'm gonna do the actual software that's the that's the magic

166
00:15:36,420 --> 00:15:43,700
trick of josh long right which which is great uh but maybe in this instance instead so you kind of go

167
00:15:43,700 --> 00:15:49,840
over that overview right and then you're basically like what are the principles that something like

168
00:15:49,840 --> 00:15:56,180
the spring framework various parts of spring framework embodies and follows such that it can deliver on

169
00:15:56,180 --> 00:16:01,480
this right and maybe there's like one or two slides that actually has some very simple code and some

170
00:16:01,480 --> 00:16:06,200
little boxes with lines going around but it's essentially like so in thinking through this

171
00:16:06,200 --> 00:16:12,460
what did we do with spring and then and then also how it integrates with with pivotal cloud foundry

172
00:16:12,460 --> 00:16:17,620
so that you can you can um you can operate in this way and i don't again i don't really know what

173
00:16:17,620 --> 00:16:22,240
those answers are but having talked with you and other people about it in the past right like one of the things

174
00:16:22,240 --> 00:16:27,240
is uh and this is i don't know why i'm pulling this but like monitoring is built into things

175
00:16:27,240 --> 00:16:33,540
other stuff is that like you know we've got all the uh all those bulkheads and circuit breakers and stuff

176
00:16:33,540 --> 00:16:39,180
and and the way that which you deploy things gives you these uh you know little flags to turn things on

177
00:16:39,180 --> 00:16:44,560
and off and just like basically and there's a registry and i don't know just all the the natural way

178
00:16:44,560 --> 00:16:50,480
that you would want to do a microservices approach that gives you the the kind of like operational

179
00:16:50,480 --> 00:16:56,300
resiliency like these are the principles we found that allows that that uh that we embody to do it

180
00:16:56,300 --> 00:17:01,480
beats me again i don't even know what those are but it would be interesting to uh to cogitate on that

181
00:17:01,480 --> 00:17:05,860
a little bit i think i can do that but i think it's gonna be code and by the way you know the uh

182
00:17:05,860 --> 00:17:11,940
you know why i got invited to devops days no uh so the uh one of the organizers was in the audience at uh

183
00:17:11,940 --> 00:17:19,920
at um dev nexus in one of my talks ah yeah yeah and that was a thoroughly code-centric sort of

184
00:17:19,920 --> 00:17:26,560
sort of talk he came up to me afterwards and um said it wasn't the worst talk he's ever seen and

185
00:17:26,560 --> 00:17:31,820
you know yep here we are that's that's that's the review we can all hope for that wasn't the worst

186
00:17:31,820 --> 00:17:36,720
talk i've ever seen it didn't hurt as much as i thought it would that was a five-star review

187
00:17:36,720 --> 00:17:39,720
we call it we call that the worst plus one review

188
00:17:39,720 --> 00:17:47,820
yeah that's that's pretty good but yeah i i mean and then if you can figure out how to like show

189
00:17:47,820 --> 00:17:53,000
that in code then that's uh that's that's so much the better that's the that's what i'm thinking like

190
00:17:53,000 --> 00:17:59,700
if this guy was okay with my code i figure i figure uh i could probably try but but but the thing is i'm

191
00:17:59,700 --> 00:18:06,880
still curious about the 30 minute thing so either either i i cut down on i mean i i hate to say this

192
00:18:06,880 --> 00:18:10,640
you know i hesitate to say this but maybe i get rid of all you know i spend a good 20 minutes talking

193
00:18:10,640 --> 00:18:16,080
about myself sure sure in these talks and you know that's maybe maybe that uh well listen here's here's

194
00:18:16,080 --> 00:18:20,900
here's what i have found now yeah you sort of just answered your own question there don't talk about

195
00:18:20,900 --> 00:18:26,840
yourself for 20 minutes so there's that now now now the second thing i was just thinking here's here's

196
00:18:26,840 --> 00:18:33,100
it's always reducing things down to a single concept as i'm as i'm sure you know right like i think the

197
00:18:33,100 --> 00:18:38,820
style that both you and me have in our different ways is instead of following that principle the the

198
00:18:38,820 --> 00:18:43,680
and i'll speak on your behalf but the trick we like to do to impress ourselves is to cram as much as

199
00:18:43,680 --> 00:18:48,180
possible into a speaking slot right like i feel like i feel like i've spent my my professional

200
00:18:48,180 --> 00:18:53,520
rhetorical life honing the ability to jam as much as to essentially do the equivalent of putting seven

201
00:18:53,520 --> 00:18:59,900
pounds of content into a five pound bag that's sort of like what what i enjoy doing so yeah which you

202
00:18:59,900 --> 00:19:03,420
know there's actually a lot more extra space in there than you would think so maybe one thing but

203
00:19:03,420 --> 00:19:09,720
then to reduce it down maybe the intersection between these two is thinking about the the john

204
00:19:09,720 --> 00:19:15,020
willis and the extended john willis definition of devops which is uh cams some people would say

205
00:19:15,020 --> 00:19:22,580
calms right which is culture automation uh let's see if i can remember this what's the m stand for

206
00:19:22,580 --> 00:19:31,440
monitoring and uh s is uh it's probably security or something and then if if you if you add uh if you

207
00:19:31,440 --> 00:19:37,420
add i'm gonna look this up if you add in an l it's it's lean uh basically which which is fine and

208
00:19:37,420 --> 00:19:42,480
then you have calms uh you know whatever uh i don't mean whatever in a dismissive way it's just

209
00:19:42,480 --> 00:19:49,500
sometimes uh oh sorry the s is for sharing oh yeah which is a good sharing insight so okay so so that

210
00:19:49,500 --> 00:19:54,320
might be a good intersection right and it's basically like i can demonstrate this in code really briefly

211
00:19:54,320 --> 00:20:01,040
but how now how a microservices approach delivers on a lot of these and i would even reduce it down to

212
00:20:01,040 --> 00:20:08,100
another thing is that ultimately a lot of what devops is aiming for is resiliency on all levels right not

213
00:20:08,100 --> 00:20:14,080
not only resiliency on your your stuff running in production right your stuff's gonna go down but

214
00:20:14,080 --> 00:20:19,880
what's important is that it recovers quickly and then it it uh fails gracefully right not that it just um

215
00:20:19,880 --> 00:20:26,120
soils the bed no and that's important by the way for my code uh because my my code in particular

216
00:20:26,120 --> 00:20:31,200
failure fails a lot i mean just a staggering amount so anything i can do to protect myself

217
00:20:31,200 --> 00:20:36,320
against those intellectual failures i yeah and and then you also think you think about the other

218
00:20:36,320 --> 00:20:41,960
stuff right so culturally we talked about the idea of uh and this is the microservices thing that the

219
00:20:41,960 --> 00:20:46,600
the cheap cheap shot which is accurate is like conway's law and then you're done with that right like

220
00:20:46,600 --> 00:20:52,320
that's like microservices is basically like how do we how do we judo our way through positive use of

221
00:20:52,320 --> 00:20:58,640
of conway's law uh and then and then the automation part is equally easy it's right which which is the

222
00:20:58,640 --> 00:21:03,160
the main sort of point that that you and i and other people make is like here's a bunch of concepts

223
00:21:03,160 --> 00:21:09,600
i know what you're thinking is like i'm gonna i'm gonna open up my intellij or my sublime edit or

224
00:21:09,600 --> 00:21:14,400
whatever it is the kids are what are the what are the kids use the get get editor adam yeah that's

225
00:21:14,400 --> 00:21:20,020
pretty legit and i'm just gonna start code i'm gonna start coding from scratch and the idea is like

226
00:21:20,020 --> 00:21:24,160
not only should you automate stuff in production but you should probably just like you shouldn't

227
00:21:24,160 --> 00:21:28,360
write a string tokenizer it's probably a good idea not to write a new framework unless you're doing

228
00:21:28,360 --> 00:21:34,120
javascript in which case the answer is to always start writing a framework immediately but please don't

229
00:21:34,120 --> 00:21:41,340
and actually that that wow we have uh right now you know as you know on the spring team we've got this uh

230
00:21:41,340 --> 00:21:47,400
uh this reactive initiative we're trying to to make to make work and boy oh boy is that a

231
00:21:47,400 --> 00:21:54,220
you know the things that it touches and every every uh sort of aspect of the ecosystem every aspect of

232
00:21:54,220 --> 00:22:01,260
the stack oh wow i just can't imagine what somebody's you know the road that somebody has to walk to to

233
00:22:01,260 --> 00:22:06,720
be able to build or build all that stuff themselves you know yeah that exactly and then so then going to

234
00:22:06,720 --> 00:22:11,820
the next thing you've got measurement we alluded to that earlier right so servability yeah yeah a key

235
00:22:11,820 --> 00:22:17,880
part of any microservices implementation is not only throwing off a bunch of metrics and information

236
00:22:17,880 --> 00:22:23,840
right otherwise you can't like manage it and do do what's the netflix called hysterical or hysterics or

237
00:22:23,840 --> 00:22:28,480
whatever yeah so you need all this information to know how to manage it and to like trip your circuit

238
00:22:28,480 --> 00:22:34,000
breakers and all this other stuff right but also i think i think there's a key insight that you learn

239
00:22:34,000 --> 00:22:38,920
from uh like reading the google sre book you know andrew and i andrew schafer and i talked about this

240
00:22:38,920 --> 00:22:44,820
a few episodes ago is it'd be really nice if it was easy to instrument your own applications right

241
00:22:44,820 --> 00:22:49,640
and if you can release daily you can start putting in new measurements and instrumentations into your

242
00:22:49,640 --> 00:22:53,980
application so you have more metrics and that's something that microservices would support right

243
00:22:53,980 --> 00:22:56,940
we support we talk about that with the spring boot actuator all the time

244
00:22:56,940 --> 00:23:04,100
yes actuator point and and then uh and then finally i mean lean again i think i think uh i think lean

245
00:23:04,100 --> 00:23:08,540
comes you could explain that but we don't have a lot of time so i'm going to skip over that so then so

246
00:23:08,540 --> 00:23:13,440
then you go to sharing and that's the whole point of microservices right is like and turn it into

247
00:23:13,440 --> 00:23:19,340
knowledge people never really make reference to this but like long ago we had this idea of mashups

248
00:23:19,340 --> 00:23:24,380
and they were going to be wonderful and save the world and all of that stuff but just like just

249
00:23:24,380 --> 00:23:31,740
like those uh those bandana face festooned people in the late 2000s the wt or 1990s the wto protesters

250
00:23:31,740 --> 00:23:36,060
the idea of mashups seems to have gone away but the notion is that like if we have all these little

251
00:23:36,060 --> 00:23:40,640
services we can compose them together and mash them together into new things and thus we can share

252
00:23:40,640 --> 00:23:44,720
in a more effective way but it's not just sharing in kind of an uncontrolled way you have a very

253
00:23:44,720 --> 00:23:50,460
structured way of sharing that uh adds resiliency and protects people pretty well distributed lego blocks

254
00:23:50,460 --> 00:23:56,480
yeah there you go standardization is sharing i'm a fan and and uh okay all that sounds very good

255
00:23:56,480 --> 00:24:02,380
so josh yes sir 30 minutes now now the unspoken rule is that you always go over five minutes

256
00:24:02,380 --> 00:24:07,340
i always do they don't need their time to get to the next hallway it's next room it's just down the

257
00:24:07,340 --> 00:24:11,520
room down the hallway yeah yeah but but what do you what do we got going on here how's how's your feel

258
00:24:11,520 --> 00:24:17,360
for the content well i i mean i'm kind of back to where i i was gonna touch on a lot of that stuff

259
00:24:17,360 --> 00:24:21,200
anyway but what i'm really one of what i'm struggling with here friend of friends is how do i get you

260
00:24:21,200 --> 00:24:27,600
uh involved you know i feel like i feel like it's gonna be okay i feel like worst case it won't be

261
00:24:27,600 --> 00:24:32,820
the worst talk that these people have ever seen uh statistically but it may not be the best certainly

262
00:24:32,820 --> 00:24:40,420
not the best i very likely not the best um yeah to extend our metaphor you know basically in our in

263
00:24:40,420 --> 00:24:46,100
our business any talk you want to give should fall into the range of worst plus one and best minus one

264
00:24:46,100 --> 00:24:51,060
that's pretty much where you want to be right okay with that yeah you don't want to this is another

265
00:24:51,060 --> 00:24:56,660
learning from the google sre book is you don't want to over deliver for too long because then people

266
00:24:56,660 --> 00:25:02,100
suspect that's how you normally operate and then when you revert back to the mean just problems right

267
00:25:02,100 --> 00:25:06,980
so you never really want to do your best unless you really really have to and it's understood that

268
00:25:06,980 --> 00:25:12,100
this is an out of the ordinary situation it's not sustainable otherwise um did you like that book

269
00:25:12,100 --> 00:25:17,700
uh yeah i i like that with the following footnotes one i haven't finished it yet and and two uh

270
00:25:17,700 --> 00:25:22,340
i feel free skipping over a lot of things like i don't really need to know the intricacies of

271
00:25:22,340 --> 00:25:27,940
distributed storage i understand that it's hard and um i'm i'm happy as we talked about earlier i'm

272
00:25:27,940 --> 00:25:33,220
happy to delegate uh my success or creating my success to other people that's perfect because

273
00:25:33,220 --> 00:25:40,260
why should we be involved no uh yes i i read that book i like the book but i was a lot of it felt a

274
00:25:40,260 --> 00:25:47,700
little um it's sort of like they hadn't talked to anybody outside of google you know sure sure yeah i

275
00:25:47,700 --> 00:25:52,500
i think i think that's fine and and it's it's it's um you know it's sort of like peering into another

276
00:25:52,500 --> 00:25:58,500
culture that you vaguely know about which which which is fine that's all it purports to be right

277
00:25:58,500 --> 00:26:07,220
as as they say works as designed yeah so now now what's more of a curiosity is is is how o'reilly

278
00:26:07,220 --> 00:26:12,820
came saw fit to publish it but i would suspect for two reasons one it probably is commercially viable

279
00:26:12,820 --> 00:26:18,260
right certainly everyone wants to read that kind of thing and two like uh i think it's a big enough

280
00:26:18,260 --> 00:26:24,020
deal that you would want to document it yeah and i'm also number three if i may speak for you uh

281
00:26:24,020 --> 00:26:29,140
they'll they'll publish anything at ariley my book is almost done my book cloud native java also at

282
00:26:29,140 --> 00:26:34,020
ariley uh that's what i'm as i as i'm talking to you right now i'm staring at in adam actually

283
00:26:34,020 --> 00:26:38,580
the aforementioned adam editor this is what i'm staring at right now my friend what do you uh

284
00:26:38,580 --> 00:26:44,580
what what do you write that in some sort of markdown latex are you a latex person oh yeah you like that

285
00:26:44,580 --> 00:26:52,340
ascii doc or whatever doctor doctor there's a doctor of ascii doc uh-huh it's called ascii doctor i

286
00:26:52,340 --> 00:26:58,260
know it it gets me i think i think uh i think ascii doctor and the condiment king are two of the

287
00:26:58,260 --> 00:27:08,740
lesser-known super villains in the dc universe oh that'd be so bad well we we should have a maybe a

288
00:27:08,740 --> 00:27:12,980
brief well maybe we'll try to do some live recordings maybe even a brief follow-up where we we see what

289
00:27:12,980 --> 00:27:17,460
josh actually talked about but i think i feel like i don't know about you josh but i i enjoyed listening

290
00:27:17,460 --> 00:27:22,020
to myself i enjoyed listening to yourself as well buddy and i think i think i think we got some good

291
00:27:22,020 --> 00:27:26,900
input from you about uh about about how how you think about these two things uh collide together

292
00:27:26,900 --> 00:27:32,820
some marginalia yeah and i and i also think that you know having me as a guest on your show and i

293
00:27:32,820 --> 00:27:42,020
delegated to you to uh to talk it embodies a lifestyle i think we i think you have trained me very well

294
00:27:42,020 --> 00:27:48,340
now uh so my last question before we wrap up josh so last we met uh this is last week we were in paris

295
00:27:48,340 --> 00:27:54,180
and i was just strolling in to the uh the le meridian you know you speak french right yeah yeah

296
00:27:54,180 --> 00:28:01,940
it's well yeah now but my question was yeah so i i thought the lounge at the the the le meridian

297
00:28:01,940 --> 00:28:07,460
etol or whatever that was a very nice starwood lounge now i agree yeah domestically you go to the

298
00:28:07,460 --> 00:28:12,980
sheratons they don't even have a lot of lounges here in the us which is frustrating now i also stayed

299
00:28:12,980 --> 00:28:19,220
at a sheraton at the paris airport the uh charles de gaulle their lounge though small was magnifique

300
00:28:19,220 --> 00:28:24,260
really right like like it had a nice selection of wine it had some warm foods a cheese selection they

301
00:28:24,260 --> 00:28:29,940
were really they were doing a good job there now anyhow known to self you travel quite a lot what would

302
00:28:29,940 --> 00:28:35,620
you say the best starwood lounge is what's the one that like you're like oh i'm going so and so

303
00:28:35,620 --> 00:28:40,420
lucky me i'm looking forward to strolling into the lounge and getting something i don't know okay

304
00:28:40,420 --> 00:28:44,020
let me think about that one but i can tell you the best airport lounge i've ever been in that's

305
00:28:44,020 --> 00:28:51,700
good too hands down are you a starwood uh sorry are you are you a star lines uh no no i'm i'm just

306
00:28:51,700 --> 00:28:59,620
on the one world oh oh well that's awkward well okay so i there's a believe it or not the airline

307
00:29:00,260 --> 00:29:09,540
the star lines uh turkish airline in in the in istanbul was it is so far beyond any other lounge

308
00:29:09,540 --> 00:29:16,500
i've ever seen uh i feel you go in and by the way the airport is okay but you go into this lounge and

309
00:29:16,500 --> 00:29:21,780
and it's just it's multiple levels there's whole floors dedicated to like you know you get massages

310
00:29:21,780 --> 00:29:27,220
and there's showers and spas and whatever there's there's i there's imax people are just there's

311
00:29:27,220 --> 00:29:32,500
banks and banks full of imax where people can like log in check their email do whatever there's you know

312
00:29:32,500 --> 00:29:38,740
every kind of food you can imagine drinks teas coffees i mean just the most it's yeah it's it's it

313
00:29:38,740 --> 00:29:45,140
sounds it sounds like the uh japan airlines lounge in tokyo in narita there oh which i've only been

314
00:29:45,140 --> 00:29:50,260
to one or two times which is also nice i would that's a good tip i'll i'll uh yeah i'll have to

315
00:29:50,260 --> 00:29:55,620
look that up well thanks for being on josh hey my pleasure are we done did that happen already sure

316
00:29:56,340 --> 00:30:01,940
i'm i'm wrapping up here i see are you ready are you ready for the wrap-up speech yeah yeah go hit

317
00:30:01,940 --> 00:30:06,180
it well first first of all first of all if people want to uh catch up with you on the internet like

318
00:30:06,180 --> 00:30:10,980
what's your what's your twitter handle i'm at starbucks man s-t-r-b-u-x-m-a-n and uh

319
00:30:11,700 --> 00:30:16,100
they'll uh they'll probably regret it but i'd love to love to and you usually write like a uh like a

320
00:30:16,100 --> 00:30:20,900
weekly what's happened in the spring world blog over like what spring.io like yeah spring.io

321
00:30:22,340 --> 00:30:28,500
yeah and and there is a one to two percent chance that wherever you live josh might be at a conference

322
00:30:28,500 --> 00:30:34,980
in in uh your geography today that's tomorrow so that's so eventually eventually you will uh you'll

323
00:30:34,980 --> 00:30:40,340
come across him so with that thanks for listening this has been another episode of Pivotal Conversations

324
00:30:40,340 --> 00:30:44,660
if you want to get new episodes as quickly as possible what you should do is subscribe to

325
00:30:44,660 --> 00:30:49,540
the podcast feed in your itunes your overcast whatever a stitcher is maybe you want to write

326
00:30:49,540 --> 00:30:55,460
your own your own script to download these files from the rss the mp3s whatever it is just subscribe

327
00:30:55,460 --> 00:31:01,300
to it you can go to soundcloud.com slash Pivotal Conversations all one word and you can find the

328
00:31:01,300 --> 00:31:05,060
rss feed there usually if you search in the directories of these various things you can find

329
00:31:05,060 --> 00:31:10,260
it it's also nice if you leave us a review or a star rating and it's even better the uh what's

330
00:31:10,260 --> 00:31:15,460
the french word for just like the ultimate thing the mayor i mean the best that's where you go there

331
00:31:15,460 --> 00:31:20,340
you go they may or as it were that sounds like spanish but if you just write something in twitter

332
00:31:20,340 --> 00:31:25,220
and tell your friends about it and help us spread the word word of mouth as seth godin used to teach us

333
00:31:25,220 --> 00:31:30,020
is the most effective marketing there is it's always a little complicated his advice for marketers yeah

334
00:31:30,020 --> 00:31:35,140
because because his advice was basically the equivalent of become independently wealthy at a young age

335
00:31:35,140 --> 00:31:41,380
which was the following if you have a great product marketing is not a problem so you know there's a

336
00:31:41,380 --> 00:31:47,940
searing insight for you uh so also if you want to read the full show notes which uh put a link to some

337
00:31:47,940 --> 00:31:54,020
to josh and maybe the presentation and things like that uh and and to the columns and columns uh if

338
00:31:54,020 --> 00:31:58,980
you go to pivotal.io slash podcast you can find the show notes there and with that we'll see everyone

339
00:31:58,980 --> 00:32:01,140
next time bye bye thanks again