1
00:00:01,080 --> 00:00:05,679
How'd you like to listen to dot
NetRocks with no ads? Easy? Become

2
00:00:05,679 --> 00:00:09,839
a patron for just five dollars a
month. You get access to a private

3
00:00:10,000 --> 00:00:14,279
RSS feed where all the shows have
no ads. Twenty dollars a month,

4
00:00:14,279 --> 00:00:18,640
we'll get you that and a special
dot net Rocks patron mug. Sign up

5
00:00:18,640 --> 00:00:35,359
now at Patreon dot dot NetRocks dot
com. Hey, welcome back to dot

6
00:00:35,439 --> 00:00:40,000
net rocks. I'm Carl Franklin,
Amergard Campbell, and you probably know who

7
00:00:40,000 --> 00:00:42,560
we are if you've been listening to
the show as long as we've been making

8
00:00:42,560 --> 00:00:44,920
them. Hey, if it's your
first one, welcome, Yeah, where

9
00:00:44,960 --> 00:00:47,759
you've been? We've done eighteen hundred
and something of it? Where Were you?

10
00:00:50,399 --> 00:00:54,159
Episode eighteen hundred and eighty nine?
Can you believe it? There you

11
00:00:54,200 --> 00:00:58,039
go? And I love listening to
podcasts that started like in the two thousands

12
00:00:58,039 --> 00:01:00,359
and brag about how long they've been
doing podcast a. Yeah, we don't

13
00:01:00,359 --> 00:01:03,719
do that, right, we don't. We really don't. We just go

14
00:01:03,799 --> 00:01:07,120
to work and talk to people.
But we were talking the other day about

15
00:01:07,120 --> 00:01:10,680
it. I think we may be
the longest continuous running podcast in the world.

16
00:01:10,760 --> 00:01:12,599
Now we may be. Yeah,
yeah, it might be true.

17
00:01:12,959 --> 00:01:19,159
Well that's uh yeah, I know
we got that going for us. Or

18
00:01:19,200 --> 00:01:25,560
if you're in Mexico, it's fifty
five pisos. Don't rub it in anymore

19
00:01:25,680 --> 00:01:29,920
possibly than you possibly can. My
friend Richard's in Mexico. Yep, two

20
00:01:29,959 --> 00:01:33,879
weeks he's making us feel cold.
Yeah, you guys are having weather.

21
00:01:34,879 --> 00:01:38,480
Although weather in Vancouver. I don't
want to just talk about the weather.

22
00:01:38,560 --> 00:01:44,400
But man, yesterday was the first
day in Connecticut here where we could actually

23
00:01:44,480 --> 00:01:47,680
go out on the back deck at
night. It was sixty two degrees fahrenheit.

24
00:01:47,799 --> 00:01:49,480
Oh good, so broke. That's
good. I'm glad. Yeah,

25
00:01:49,680 --> 00:01:56,840
he had snow that turned the rain
in Vancouver and we're like, hah,

26
00:01:57,280 --> 00:01:59,840
well, Magnus Martinsen is here.
We're going to talk to him in a

27
00:01:59,840 --> 00:02:02,560
minit and Magnus, if you want
to jump in on this story that I

28
00:02:02,560 --> 00:02:13,800
have for better know framework feel free. Let's roll the crazy music. A

29
00:02:14,120 --> 00:02:15,639
man, what do you got?
Okay? So you know, I do

30
00:02:15,719 --> 00:02:21,840
this show called Security this week with
Patrick Hines and Duane Flotten, So I'm

31
00:02:21,879 --> 00:02:27,280
aware. I'm more aware now of
the stories that come across my desk about

32
00:02:27,599 --> 00:02:31,840
security, and AI is one of
them. Because everybody's worried about AI taking

33
00:02:31,840 --> 00:02:36,479
over the world, of course.
So this is an article from PC magazine

34
00:02:36,800 --> 00:02:42,919
and the headline is new malware worm
can poison chat, GPT and Gemini powered

35
00:02:42,960 --> 00:02:47,159
assistants. Generative AI models aren't immune
to viruses. A new worm can target

36
00:02:47,199 --> 00:02:53,360
AI powered assistance and break some of
Gemini and chat GPT security features. Research

37
00:02:53,400 --> 00:03:00,639
reveals, well, they're pointing to
an article in Wired and this this is

38
00:03:00,680 --> 00:03:06,599
a really irresponsible article. I don't
know what you thought, Richard when you're

39
00:03:06,599 --> 00:03:08,560
written, No, that's exactly what
I thought. And I was going to

40
00:03:08,560 --> 00:03:10,599
be annoyed with you for even bringing
it up. But if you want to

41
00:03:10,599 --> 00:03:14,639
come at the fact that there's some
irresponsible reporting coming on, I'm glad,

42
00:03:14,960 --> 00:03:17,879
yeah, because this is pretty irresponsible
at reporting. So the article from Wired

43
00:03:19,039 --> 00:03:22,919
is here, come the AI worms. Security researchers created an AI worm in

44
00:03:22,960 --> 00:03:29,319
a test environment and that can automatically
spread between generative AI agents, potentially stealing

45
00:03:29,360 --> 00:03:32,599
data and sending spam emails along the
way. And okay, this is a

46
00:03:32,680 --> 00:03:38,000
lab experiment right in a controlled environment. And PC magazine is like, hey,

47
00:03:38,199 --> 00:03:44,680
guess what you're don't use chut,
GPT and Gemini because you get this

48
00:03:44,759 --> 00:03:49,159
worm and you're screwed. Yeah it's
no worm, and yeah there is and

49
00:03:49,639 --> 00:03:53,240
Wired is a little more reasoned about
that. And actually the piece underneath this,

50
00:03:53,439 --> 00:03:55,639
which is you know, from a
set of researchers, is better still

51
00:03:55,680 --> 00:03:59,639
because right, really what they're talking
about is we do need to be prepared,

52
00:03:59,759 --> 00:04:01,759
Sure we do. Yeah, I
mean the most important part to me

53
00:04:02,039 --> 00:04:08,080
was the fact that it spanned open
Ai and Gemini and Lama. Yeah,

54
00:04:08,120 --> 00:04:13,520
because that's Facebook, because it shows
this is inherent to lllm's right, nothing

55
00:04:13,560 --> 00:04:16,879
to do with any particular model,
right, And basically, you know,

56
00:04:16,920 --> 00:04:21,399
they they gave it a what do
they do? They gave it a prompt

57
00:04:21,480 --> 00:04:27,720
that allowed it to they've been basically
jail broke it, right, And well,

58
00:04:27,800 --> 00:04:30,079
yeahs and jail break is one thing, but the bigger thing here was

59
00:04:30,120 --> 00:04:34,439
the recursion engine to have it generate
an output that then it then needs to

60
00:04:34,439 --> 00:04:38,879
immediately respond to. Yeah, you
know, the best answer I have to

61
00:04:38,920 --> 00:04:44,560
that is don't do that. Don't
do that exactly. So it's it's like

62
00:04:44,720 --> 00:04:47,480
fud right, it's a fun machine. So if you if you're on your

63
00:04:47,720 --> 00:04:53,079
chat rept channel and you've logged in
and you're authenticated and you're having a nice

64
00:04:53,079 --> 00:04:56,519
little conversation with chat chept. You
don't have to worry about any of this,

65
00:04:57,160 --> 00:05:00,399
right, Yeah, No, I
think they're thinking about a future where,

66
00:05:00,959 --> 00:05:03,040
you know, the thing they also
promise us with with the Amazon device

67
00:05:03,079 --> 00:05:06,040
which I won't see the name of, just enough ruin people's day, where

68
00:05:06,040 --> 00:05:10,399
you're supposed to be done with that
joke. Yes, automatically, you know,

69
00:05:10,600 --> 00:05:14,279
just order me more laundry soap.
Now you're imagining this agent doing that,

70
00:05:14,319 --> 00:05:18,560
and could that if it has full
authority to go make the order and

71
00:05:18,600 --> 00:05:24,000
complete the sale, you have a
risk of exploitation. That's true, whether

72
00:05:24,000 --> 00:05:28,040
it's an LM or in there or
not. Yeah, they And it comes

73
00:05:28,079 --> 00:05:31,600
back to our least privilege argument that
we've been having about computing ever since the

74
00:05:31,639 --> 00:05:35,920
dawn of computers. Well, and
having some gates around Okay, if you're

75
00:05:35,920 --> 00:05:39,839
going to write emails on my behalf, should I look at each one of

76
00:05:39,839 --> 00:05:43,279
them before they send? Or should
rules of what you can send? Like,

77
00:05:43,920 --> 00:05:46,920
you know, you could set some
parameters around that, but recursively generating

78
00:05:47,120 --> 00:05:53,360
code that then sends email. Yeah, all right, it's good that I

79
00:05:53,399 --> 00:05:57,360
can fix that. That's ten minutes. There we go. Yep, don't

80
00:05:57,360 --> 00:06:03,120
do that's all right. Well I
wanted to. I wanted to bring that

81
00:06:03,199 --> 00:06:06,240
up. Agreement on this one.
Yeah, we are in violent agreement and

82
00:06:06,279 --> 00:06:10,800
it's a good laugh. But you
know, sometimes we need to put these

83
00:06:10,839 --> 00:06:14,160
things in context. And you know
that's why you listen. And I want

84
00:06:14,160 --> 00:06:15,839
to believe the Cornell guys were working
in faith. It's like, hey,

85
00:06:16,120 --> 00:06:20,680
is there an exploit risk here just
to help lead folks into we should deal

86
00:06:20,720 --> 00:06:25,319
with this before it becomes a problem. But I'm on all the rest.

87
00:06:26,240 --> 00:06:29,120
Indeed, Well, I don't know, magnety, have anything you want to

88
00:06:29,120 --> 00:06:30,839
add to that. I mean,
mister cloud guy. I was just listening

89
00:06:30,879 --> 00:06:36,399
to this and we've heard this time
and again that that it seems like sometimes

90
00:06:36,480 --> 00:06:44,399
the the the journalist or whatever we
should call this person is so desperate to

91
00:06:44,439 --> 00:06:47,360
do something, but they're they're going
to write just anything, whether it's you

92
00:06:47,399 --> 00:06:50,560
know, remotely true or not,
and and they, like you said,

93
00:06:50,560 --> 00:06:56,959
it's completely irresponsible and that I loathe
it. It's not it's not journalism.

94
00:06:57,000 --> 00:06:59,680
I don't know what that is.
Yeah, and if anybody would like cite

95
00:06:59,680 --> 00:07:02,639
this article on a podcast or something, and then nobody would do that,

96
00:07:02,759 --> 00:07:05,000
you know, put that in the
links. Who would do that? That's

97
00:07:05,000 --> 00:07:14,839
irresponsible, that's irresponsible clickbait. Just
it reprehensible. All right, Well that's

98
00:07:15,240 --> 00:07:17,800
the fun, fun way to start
the show today. Who's talking to us?

99
00:07:17,920 --> 00:07:20,959
Richard? Hey, I grabbed a
comment off of show eighteen eighty one,

100
00:07:21,000 --> 00:07:24,439
so not that long ago. That's
the one we did about dot net

101
00:07:24,439 --> 00:07:28,920
Aspire with David Fowler. Yeah,
and the comment comes from codeputier otherwise known

102
00:07:28,920 --> 00:07:31,480
as Richard Rukima, one of our
longtime listener and occasional guests. He's done,

103
00:07:31,560 --> 00:07:34,360
he lives in both worlds, and
he said, I've enjoyed dot net

104
00:07:34,439 --> 00:07:36,480
rocks for over a decade. Yeah, like it's longer than that. I

105
00:07:36,560 --> 00:07:40,680
found emails from you from the aughts, my friend. Oh yeah. I

106
00:07:40,720 --> 00:07:43,399
can't put my finger on it,
but this show really felt like this show

107
00:07:43,480 --> 00:07:46,160
is of old such great ton in
the conversation with an internal resize that we

108
00:07:46,199 --> 00:07:51,360
all need so desperately, Like David
hearing his perspective on Aspire was very inspirational

109
00:07:51,399 --> 00:07:54,720
as well as educational. And to
be clear, this is not a diss

110
00:07:54,759 --> 00:07:58,680
of previous content, but this episode
is a gold standard on wireless. Oh

111
00:07:58,720 --> 00:08:01,800
cool, And I was thinking about
it of a. I know, we're

112
00:08:01,839 --> 00:08:03,519
talking to the guy who invented the
technology. I mean, admittedly, David

113
00:08:03,560 --> 00:08:07,120
had a team around him and he's
still trying to figure out what he's made.

114
00:08:07,600 --> 00:08:11,399
You know. I saw him at
NBC Sydney and they were workshopping it

115
00:08:11,480 --> 00:08:15,439
and literally still trying to figure out
how to describe it. So, you

116
00:08:15,439 --> 00:08:18,399
know, one of the things I
like about that particular show is we were

117
00:08:18,439 --> 00:08:20,399
trying to understand what he was talking
about, and he was trying to understand

118
00:08:20,399 --> 00:08:24,279
what he was doing. So there's
this real wrestling match about I know you

119
00:08:24,439 --> 00:08:28,079
got something here, David, I
just don't know how to describe it.

120
00:08:28,120 --> 00:08:33,120
Yeah. I thought it was a
very fun show, definitely. And Richard,

121
00:08:33,159 --> 00:08:35,440
you know you already have a copy
of Music co b. I goodest

122
00:08:35,480 --> 00:08:39,080
noticed about every other thing, but
thank you so much for your comment,

123
00:08:39,279 --> 00:08:41,399
And if you'd like a copy of
Music Coobe, I read a comment on

124
00:08:41,440 --> 00:08:43,120
the website at dot na Rocks dot
com or on the facebooks. We publish

125
00:08:43,200 --> 00:08:46,159
every show there, and if you
comment there and I read on the show,

126
00:08:46,360 --> 00:08:48,559
we'll send you a copy of Music
COBA. And you can follow us

127
00:08:48,600 --> 00:08:50,799
on Twitter or x or whatever the
heck you want to call it, but

128
00:08:50,360 --> 00:08:54,080
all the cool kids are hanging out. I'm Masterdon, so follow us there.

129
00:08:54,240 --> 00:08:58,240
I'm at Carl Franklin at tech hub
dot social, and I'm Rich Campbell

130
00:08:58,279 --> 00:09:01,480
at McDon social. And you can
find all the other ways. You can

131
00:09:01,519 --> 00:09:05,840
reach me at Carl Franklin dot com. But yeah, send us a two

132
00:09:05,279 --> 00:09:09,399
Rudy too too. Okay, let's
bring on Magnus. I'm going to formally

133
00:09:09,440 --> 00:09:15,720
introduce him with his latest bio.
Magnus Martinsen is an entrepreneur, a Microsoft

134
00:09:15,840 --> 00:09:20,759
Most Valuable Professional for Azure since the
start of the cloud, and a Microsoft

135
00:09:20,799 --> 00:09:24,600
Regional Director. He's a consultant,
architect, product development lead, and he

136
00:09:24,720 --> 00:09:31,039
runs his own company, Loftysoft.
As an international speaker, Magnus travels the

137
00:09:31,039 --> 00:09:35,399
world to teach, network, learn
and experience. His passions include connecting with

138
00:09:35,480 --> 00:09:41,120
audiences and organizing conferences such as Cloudburst
and Global Azure. He is, of

139
00:09:41,159 --> 00:09:45,879
course, also very much into good
food, wine and great company. Except

140
00:09:45,919 --> 00:09:48,639
for that crap that he tried to
make us drink at Ted Neward's party,

141
00:09:48,639 --> 00:09:54,080
I would not recommend taking that tomping. That was my editorial. Tompy all

142
00:09:54,320 --> 00:09:56,240
is mind sharing. So when you
see him at a conference, come over

143
00:09:56,279 --> 00:10:01,039
and say hi, welcome back,
Magnus. Well, thank you, thank

144
00:10:01,039 --> 00:10:03,759
you very much. Good good of
you to have me. What the hell

145
00:10:03,879 --> 00:10:07,000
was that thing that you gave us
at Ted Newart's party back in the day.

146
00:10:07,480 --> 00:10:11,360
Uh, it's it's well, it's
based on the wormwood, so it's

147
00:10:11,399 --> 00:10:18,480
it's very very bitter and straight.
It was the most vile stuff I've ever

148
00:10:18,600 --> 00:10:24,120
It tasted like bile because I've tasted
that coming up out of my liver before

149
00:10:24,159 --> 00:10:28,879
when I was a drunken teenager say
some fairly horrible things. I remember a

150
00:10:28,919 --> 00:10:35,759
particular tin of fish, but that's
that's a different thing. Yeah, you

151
00:10:35,759 --> 00:10:41,279
don't inside, you'll clear a neighborhood. But here's the thing. It was

152
00:10:41,399 --> 00:10:45,840
just like, try this. All
the real men in Sweden drink this.

153
00:10:46,440 --> 00:10:48,440
It's good for you, absolutely,
and we do. And we do this

154
00:10:50,399 --> 00:10:52,879
when when they when they are hunting
in the in the lodge somewhere and they

155
00:10:54,000 --> 00:10:58,279
put this this twig of the wormwood
in the bottle and then leave it there.

156
00:10:58,399 --> 00:11:01,519
So like the first the first three
shots are pretty okay, but the

157
00:11:01,919 --> 00:11:11,000
bottom line is like whoa da the
states I think you got? I think

158
00:11:11,000 --> 00:11:13,399
you got from the top of a
bottle car and you're just so, You're

159
00:11:13,480 --> 00:11:18,720
just a week So I guess I
wasn't meant for the manliness of the Swedish

160
00:11:18,759 --> 00:11:24,000
liquor clouds. What can I tell
you? Yeah? So what are you

161
00:11:24,039 --> 00:11:26,200
thinking about these days? Magnus?
Oh? Wow, I'm thinking about a

162
00:11:26,200 --> 00:11:31,480
lot of things. Top of mind
in the community world is global Azure of

163
00:11:31,480 --> 00:11:35,039
course is coming up again. But
overall I'm working and I'm working with with

164
00:11:35,240 --> 00:11:43,440
enterprises of various various kinds. So
really helping large organizations fully adopt the cloud

165
00:11:43,879 --> 00:11:48,600
is a huge journey for for many
of them. The biggest challenge is that

166
00:11:48,639 --> 00:11:54,360
they don't have enough experienced people on
team on on on in their teams,

167
00:11:54,960 --> 00:12:01,039
so they're kind of struggling to really
really fully get you get all of those

168
00:12:01,080 --> 00:12:03,840
promises that we hear about the cloud. You know, it's pay per use,

169
00:12:03,919 --> 00:12:07,120
it's automation, it's agile, it's
quick, and it's all those things.

170
00:12:07,519 --> 00:12:11,879
But they can't really reach it because
it's hard when you don't have enough

171
00:12:11,000 --> 00:12:13,840
folks that can help you along the
way. So that's what I do.

172
00:12:13,919 --> 00:12:18,559
I help them make asure better.
How much of this is a dev problem

173
00:12:18,639 --> 00:12:24,159
versus an assisted problem? Wow,
dude, there is a dev problem in

174
00:12:24,200 --> 00:12:28,039
there, but there is a huge
governance problem in there because you just don't

175
00:12:28,080 --> 00:12:31,879
know what it is that you're supposed
to do. You can't even ask the

176
00:12:31,000 --> 00:12:35,120
questions. So that's a challenge,
right. So I work with teams of

177
00:12:35,159 --> 00:12:39,000
people, scores of people who have
twenty thirty years of experience. They're really

178
00:12:39,039 --> 00:12:43,039
good engineers of all kinds, and
they're really good at what they do,

179
00:12:43,440 --> 00:12:46,840
except they don't have mileage in the
cloud, so they don't know which questions

180
00:12:46,879 --> 00:12:50,039
to ask. I mean, if
somebody tells you have twenty years of experience

181
00:12:50,039 --> 00:12:54,320
in the cloud, they're lying.
Not in the cloud. No, not

182
00:12:54,440 --> 00:12:58,440
in the cloud, they are it
professionals. Yeah. To illustrate not knowing

183
00:12:58,480 --> 00:13:01,759
what to do, I'm going to
in a new acronym FO D A R

184
00:13:03,320 --> 00:13:09,559
FODD are fear of deleting Azure resources
because you know you have hundreds and hundreds

185
00:13:09,600 --> 00:13:13,000
of resources. You just create one
website you have You've created resources you may

186
00:13:13,039 --> 00:13:16,879
not even know about, and you
come back later and you're like, hmm,

187
00:13:16,039 --> 00:13:20,519
what's this? Should I delete it? That is an excellent, excellent

188
00:13:20,559 --> 00:13:24,039
acronym. It's costing me a lot
of money. We should you should?

189
00:13:24,120 --> 00:13:28,879
You should trademark, that acronym.
It's it's amazing, that's exactly. That's

190
00:13:28,960 --> 00:13:31,360
exactly a huge problem with many of
my customers is that they don't know.

191
00:13:31,679 --> 00:13:37,320
And so they a pretty recent study
indicated by you know, self engaging.

192
00:13:37,519 --> 00:13:43,480
They asked people who consume another cloud
roughly how much how much of your cloud

193
00:13:43,559 --> 00:13:48,559
spend would you say is waste?
And the self estimation was at about a

194
00:13:48,639 --> 00:13:50,399
third. So you got, you
gotta, you got, you got a

195
00:13:50,399 --> 00:13:54,919
million dollar you know. Ea with
Microsoft three hundred and thirty three thousand,

196
00:13:54,919 --> 00:13:58,799
it's just burning on the lawn.
And that's exactly it, Carl, that

197
00:14:00,039 --> 00:14:03,919
they don't know exactly what they have, and they do not use the automation,

198
00:14:03,159 --> 00:14:05,080
and they do not use the you
know, the things that you should

199
00:14:05,080 --> 00:14:09,519
the things that most most people know
that you should be doing, except they're

200
00:14:09,519 --> 00:14:11,559
not doing it. And when they're
not, yeah, I've found you know,

201
00:14:11,960 --> 00:14:16,480
oh look look at this. I
found this this cluster of vms here.

202
00:14:16,480 --> 00:14:18,360
What are you using that for?
Oh? Those five vms? Oh

203
00:14:18,399 --> 00:14:22,159
I don't know, Wow, do
we use them? I didn't. I

204
00:14:22,159 --> 00:14:24,600
didn't. I didn't even know.
I didn't even know three months ago,

205
00:14:24,679 --> 00:14:28,320
right, So that's like, you
know, ten months of compute just burning

206
00:14:28,440 --> 00:14:31,240
for not real Have you ever you
should? You could write a whole you

207
00:14:31,240 --> 00:14:35,679
could do a whole class on understanding
and itemized Azure bill. Have you ever

208
00:14:35,759 --> 00:14:41,000
looked at one of these things?
Yeah, by the way, this week

209
00:14:41,080 --> 00:14:45,399
because they keep changing the frickin thing
too. Yeah, fair enough. The

210
00:14:45,399 --> 00:14:48,559
point is that you never should get
to that position in the first place,

211
00:14:48,559 --> 00:14:52,799
so you should do it right.
And and and companies again, they're coming

212
00:14:52,840 --> 00:14:56,519
from a world of vms and and
let's order. Order a VM means by

213
00:14:56,600 --> 00:15:01,960
hardware, and that's where they're coming
from. And there's nothing wrong with that,

214
00:15:01,000 --> 00:15:03,559
except that's not what you're supposed to
be doing in the cloud. And

215
00:15:03,600 --> 00:15:07,679
so they transposed that thing to the
cloud and it just doesn't work. I

216
00:15:07,759 --> 00:15:11,759
had a nerving experience in Azure the
other day. I went to start a

217
00:15:11,840 --> 00:15:16,399
VM that I use for, you
know, video recording, and it said,

218
00:15:16,759 --> 00:15:22,440
can't start this VM because there aren't
enough resources in this region to support

219
00:15:22,480 --> 00:15:28,799
it. And I'm thinking, aren't
enough resources? You're Azure. You've got

220
00:15:28,039 --> 00:15:33,200
you know, underwater data centers all
over the world or whatever you have,

221
00:15:33,360 --> 00:15:37,559
you know, and you can't change
it. So I had to just wait.

222
00:15:37,960 --> 00:15:41,240
Well, the cloud is infinite for
infinite for you, but it's not

223
00:15:41,360 --> 00:15:48,000
infinite for Microsoft. Sure, I'm
throwing the sissemin hat on this thing because

224
00:15:48,000 --> 00:15:50,320
I do this conversation on run as
all the time. And one of the

225
00:15:50,440 --> 00:15:54,759
essential issues has been, and this
is you know, big in the pandemic,

226
00:15:54,879 --> 00:15:58,840
was we did this fast migration to
the cloud. And when you architect

227
00:15:58,879 --> 00:16:03,799
an app on pre or architect the
resources for an app on prem you architecture

228
00:16:03,840 --> 00:16:07,480
for peak. Right, if the
most we're ever going to consume, that's

229
00:16:07,519 --> 00:16:11,360
what I'll buy, even though it'll
represent one percent of the utilization. But

230
00:16:11,399 --> 00:16:15,399
that one percent is probably the most
profitable point. I don't want it out.

231
00:16:15,600 --> 00:16:18,799
And if you architect that way and
throw it into Azure, you're spending

232
00:16:18,840 --> 00:16:22,919
a ton of money on wasted resources. Yeah, absolutely right, and that's

233
00:16:22,960 --> 00:16:26,480
a terrible way to do it.
So the migration wizards that Microsoft have been

234
00:16:26,480 --> 00:16:30,519
working on for quite some time now, they're getting better and better, more

235
00:16:30,519 --> 00:16:33,360
and more sophisticated. They actually doing
a fair job. So long as you

236
00:16:33,759 --> 00:16:37,519
deploy a bunch of agents inside of
your network. First, you let them

237
00:16:37,759 --> 00:16:40,360
be there for a while and listen
to the traffic and see what's what,

238
00:16:41,200 --> 00:16:45,480
and then you process that data and
then you can write size as they call

239
00:16:45,519 --> 00:16:48,720
it when you migrate to the cloud, and sayf for you, I want

240
00:16:48,720 --> 00:16:52,039
to use this size of thing instead
in the cloud because that just makes more

241
00:16:52,080 --> 00:16:56,799
sense. And then you just right
sizing is a way of making sure that

242
00:16:56,840 --> 00:17:00,440
you just don't move it like one
to one on to the cloud and then

243
00:17:00,639 --> 00:17:03,599
you know, waste a lot of
money. And I love that idea that

244
00:17:03,640 --> 00:17:07,240
hey, you know, we don't
know any secrets here, we need to

245
00:17:07,279 --> 00:17:10,759
watch you. Yeah, so load
up these tools and let them run.

246
00:17:10,759 --> 00:17:15,000
And again, if you're a dev
probably listening to this show, you have

247
00:17:15,160 --> 00:17:18,319
to ask it, hey, let's
deploy these agents into the network, which

248
00:17:18,359 --> 00:17:22,559
is not going to make that person
happy, Like true, that's going to

249
00:17:22,599 --> 00:17:26,039
be a battle, but the alternative
is worse. Arguably, we had to

250
00:17:26,079 --> 00:17:33,279
do a security review and things like
that before that was allowed in the network

251
00:17:33,279 --> 00:17:37,039
of a large customer that a public
sector again that I'm working with right now.

252
00:17:37,440 --> 00:17:40,640
And so after having done all of
that, all the due diligence and

253
00:17:40,799 --> 00:17:44,440
figured out that this is not going
to steal our information or anything, then

254
00:17:44,599 --> 00:17:49,039
then that was allowed. And the
the we're migrating in the size of like

255
00:17:49,480 --> 00:17:55,039
eight nine hundred VMS is the first
migration step. Just moved move the things

256
00:17:55,039 --> 00:17:56,559
first, and that's not the biggest
thing we've ever seen, but it's it's

257
00:17:56,599 --> 00:18:02,000
sizable for sure, and move that
to the cloud to begin with. And

258
00:18:02,119 --> 00:18:07,039
that's kind of when the work starts
because the the Migration Wizard is basically automation.

259
00:18:07,400 --> 00:18:10,480
So now we have moved. Now, we have moved them. Now

260
00:18:10,480 --> 00:18:12,480
we have lots of service running in
the cloud. Now is where the work

261
00:18:12,519 --> 00:18:18,240
actually begins to to migrate some of
that those workloads off the machines that they're

262
00:18:18,319 --> 00:18:21,839
running on. They could be like
just hey, we're running on a we're

263
00:18:21,880 --> 00:18:25,640
running a sequel server on a VM. Well, dude, why you could

264
00:18:25,720 --> 00:18:30,440
use database database service same thing except
no VM that do you know? The

265
00:18:30,480 --> 00:18:34,680
problems aren't always our fault, right
one of the but you know, usually

266
00:18:34,759 --> 00:18:40,359
there's something that we can do to
remedy the situation. But one of the

267
00:18:40,400 --> 00:18:45,720
problems that I had recently was that
Microsoft was changing the IP address of a

268
00:18:45,759 --> 00:18:49,079
web app on me without telling me
they were going to do that. Had

269
00:18:49,119 --> 00:18:52,680
no notification except that, you know, somebody tweeted me and said, hey,

270
00:18:52,720 --> 00:18:57,279
this site's down. And when you
go and look oh that IP address,

271
00:18:57,319 --> 00:19:03,440
the A record is different, and
so yeah, yeah, I guess

272
00:19:03,480 --> 00:19:07,200
I have to fix it in DNS. I wrote a little I wrote a

273
00:19:07,200 --> 00:19:11,119
little PowerShell app. I know that
I can fix that in another way.

274
00:19:11,160 --> 00:19:15,400
But in the meantime, I have
this PowerShell app that I wrote that runs

275
00:19:15,599 --> 00:19:22,960
against dot net rocks and other websites
that I have and tests for content that

276
00:19:22,799 --> 00:19:26,359
I know is going to be there. Yeah. So that's running once an

277
00:19:26,359 --> 00:19:30,839
hour and just in a script on
my machine. Yeah, well that works.

278
00:19:32,680 --> 00:19:37,039
It's emails the way if dot rocks
is down. Yeah, that's one

279
00:19:37,079 --> 00:19:40,640
way. I'm sure there are other
ways, but you know, you use

280
00:19:40,680 --> 00:19:44,279
the tools that you know before you
dive into the tools that you don't know.

281
00:19:44,799 --> 00:19:48,359
Yeah. Well, nowadays we have
enterprises and things that are ready to

282
00:19:48,400 --> 00:19:52,400
move to the cloud. Again.
It's it's weird for someone like me who

283
00:19:52,440 --> 00:19:56,000
has been in the cloud since the
very beginning. Means I have well,

284
00:19:56,200 --> 00:19:59,000
not the twenty years of experience that's
going to be hard, but at least

285
00:19:59,000 --> 00:20:03,599
the plug ten plus now right,
and now I'm talking to companies that have

286
00:20:03,839 --> 00:20:08,200
never touched it because you know,
reasons, because compliancy, because of you

287
00:20:08,240 --> 00:20:11,359
know, it's public sector and it's
been difficult and we haven't been able to

288
00:20:11,400 --> 00:20:18,480
move right and now they're getting there
and they're moving into this huge enterprise grade

289
00:20:18,640 --> 00:20:25,880
platform which has everything and everything extra. When we started using Windows Azure in

290
00:20:25,920 --> 00:20:27,920
the beginning as it was called Windows, it's called Windows in the beginning,

291
00:20:29,000 --> 00:20:32,079
right, it had like five services. That was that was it. That

292
00:20:32,160 --> 00:20:34,359
was that was all you could do. No VM, no VM. Even

293
00:20:34,519 --> 00:20:38,400
you only had a huge post of
service. Was the only thing you could

294
00:20:38,400 --> 00:20:44,880
do. So it's it's really interesting
to see these people who are coming into

295
00:20:45,039 --> 00:20:48,799
this now and haven't been part of
the journey. It's harder for them,

296
00:20:48,799 --> 00:20:51,279
I think. I mean, there's
an upside to being in early that you

297
00:20:51,359 --> 00:20:55,039
got to see the pieces built,
although you also went through all the convulsions

298
00:20:55,480 --> 00:20:59,440
of the sure getting because they made
new ones. Yeah, at the same

299
00:20:59,480 --> 00:21:03,079
time, like in New now with
stuff like those Migration wizard things seem to

300
00:21:03,119 --> 00:21:07,200
make life easier. But you ever
look at the portal, like that's a

301
00:21:07,200 --> 00:21:11,200
lot of stuff. There's a lot
of stuff, all the stuff. It

302
00:21:11,279 --> 00:21:17,279
is overwhelming and I'm so surprised at
you know, the questions that you get,

303
00:21:17,480 --> 00:21:21,960
the comments that you get, because
obviously I've lost that perspective. I

304
00:21:22,039 --> 00:21:25,000
know how the sausage was made,
how the cookies were baked. I know

305
00:21:25,079 --> 00:21:29,400
all of it. I know why
Azure Monitor is working the way it does.

306
00:21:30,000 --> 00:21:33,200
You know how the wormwood was mixed, my friend, I know exactly

307
00:21:33,200 --> 00:21:37,880
how the wormwood was mixed, and
it was strong. It was very very

308
00:21:37,920 --> 00:21:42,880
strong. Yeah. Yeah, it's
good to be able to lead them through

309
00:21:42,920 --> 00:21:47,680
that, but it definitely I feels
like it's a better time now. There

310
00:21:47,759 --> 00:21:53,039
is a whole conversation going on around
repatriation of I'm take people taking stuff back

311
00:21:53,079 --> 00:21:56,960
out of the class. Yeah,
I've heard a couple of cases of that,

312
00:21:57,039 --> 00:22:02,079
and some some kind of notable that
that somebody like wrote up a post

313
00:22:02,119 --> 00:22:03,599
about it then said hey, we
were going to use the cloud, and

314
00:22:03,640 --> 00:22:07,799
we started doing this and then we
just realized how expensive and crazy it was

315
00:22:07,839 --> 00:22:11,119
and we just moved away again.
Like Okay. When I look at that,

316
00:22:11,160 --> 00:22:14,599
I go like, okay, so
it's my specific expertise to help companies

317
00:22:14,680 --> 00:22:18,759
just make the cloud better. Okay, So I'm looking at what you're doing,

318
00:22:18,200 --> 00:22:22,799
and I just realized that you did
not approach this in a responsible way.

319
00:22:23,160 --> 00:22:27,000
You did not approach it in a
learning way. You didn't have the

320
00:22:27,559 --> 00:22:33,440
people to do this, and now
you're like finding that it's not pay per

321
00:22:33,599 --> 00:22:37,000
use, it's not you know,
optimized, and it's costing you too much,

322
00:22:37,039 --> 00:22:38,680
and it's not secure and all the
things that you're saying that it's not

323
00:22:40,200 --> 00:22:42,160
it's because you didn't do your homework. And now you're backing out. Now

324
00:22:42,160 --> 00:22:48,599
you're backing out and blaming the cloud. Hm, that's not fair. So,

325
00:22:48,680 --> 00:22:52,559
I mean, I was trying to
work through reasonable that seemed to me

326
00:22:52,599 --> 00:22:55,480
seems the most likely scenario. You
shipped a bunch of vms up in the

327
00:22:55,480 --> 00:23:00,119
cloud. Oddly enough, they cost
money, no surprise. Yeah, now

328
00:23:00,119 --> 00:23:03,240
you're going to ship a bunch of
vms back out of the cloud again where

329
00:23:03,400 --> 00:23:07,319
moving up to services and all those
other options. You know, you could

330
00:23:07,400 --> 00:23:11,039
have tuned this better, But I'm
trying to find better scenarios. I mean,

331
00:23:11,160 --> 00:23:14,039
one would argue, if you're running
a twenty four to seven knock,

332
00:23:14,079 --> 00:23:18,119
if you already have around the clock
SIS submits and so forth, your more

333
00:23:18,200 --> 00:23:21,680
or less could be operating your own
clouds, So why are you using somebody

334
00:23:21,680 --> 00:23:27,680
else's That the rest of the arguments
become so much harder, Like I'm just

335
00:23:27,680 --> 00:23:30,279
trying to figure out what the cloud
isn't good at that you could point it

336
00:23:30,319 --> 00:23:33,519
a service and go, yeah,
maybe that's better to keep on premp.

337
00:23:33,799 --> 00:23:37,039
I don't know. I mean,
if it's the only thing I come down

338
00:23:37,079 --> 00:23:40,920
to is if the service has,
you know, very very strict restrictions on

339
00:23:41,000 --> 00:23:45,039
data governance and where you can where
you can host it, and where you

340
00:23:45,039 --> 00:23:48,279
can run into But that's a feature, that's a yeah, And I feel

341
00:23:48,319 --> 00:23:52,599
like Microsoft is done a huge amount
of work on that, Like if all

342
00:23:52,640 --> 00:23:55,880
you got to do is search,
when it comes down to I need to

343
00:23:55,880 --> 00:24:00,039
comply with this set of regulations for
this country. They have ratifications in place

344
00:24:00,079 --> 00:24:04,359
for it. Yeah, yes,
and they support a lot of the services

345
00:24:04,400 --> 00:24:08,000
today support such things as bring your
own key and so forth. Right,

346
00:24:08,039 --> 00:24:12,519
so you have your own, your
very own key that that no one can

347
00:24:12,640 --> 00:24:18,279
decrypt, not even Microsoft. So
it's encrypted with your key except as your

348
00:24:18,279 --> 00:24:22,440
secretary because he left it on your
desktop. Oh yeah, post posted underneath

349
00:24:22,440 --> 00:24:26,079
the desk exactly. And there and
therein lies the bigger joke, which is

350
00:24:26,160 --> 00:24:32,200
arguably like I have seen a small
third party data center get rated by the

351
00:24:32,279 --> 00:24:37,519
FBI because of one of their customers, and they took and the FBI took

352
00:24:37,720 --> 00:24:41,680
all of the racks, and so
I had that customer that was in there

353
00:24:41,039 --> 00:24:45,960
and their machine. They would get
their machines back two years later, but

354
00:24:45,000 --> 00:24:48,680
in the meantime they're at the corridor. You know, the judge didn't understand

355
00:24:48,680 --> 00:24:52,720
the ratifications what they were doing.
They cleared that building. You can't do

356
00:24:52,759 --> 00:25:00,720
that with the cloud data center.
A nobody's got enough for cliffs like getting

357
00:25:00,759 --> 00:25:03,119
access to that data center. Are
those machines going to be even worth anything

358
00:25:03,119 --> 00:25:06,559
in two years? That's another question. That was the thing. It was

359
00:25:07,519 --> 00:25:10,440
largely destructed. It was with the
destruction of the business that we couldn't replace

360
00:25:10,440 --> 00:25:14,440
them. It's crazy because of one
customer. Wow, yeah, and yeah

361
00:25:15,160 --> 00:25:17,559
terrible. So you know that in
that sense, I said, you know,

362
00:25:17,599 --> 00:25:21,079
the upside to the cloud is go
ahead, knock yourself out, like

363
00:25:21,160 --> 00:25:25,559
it's just at that scale, that's
not possible. Great, agreed, But

364
00:25:25,599 --> 00:25:32,160
we were we were there before that, before such time where companies were legitimately

365
00:25:32,480 --> 00:25:34,799
legitimately they were saying that we can't
go to the cloud because it's not secure

366
00:25:34,920 --> 00:25:37,839
enough. We don't know that it's
secure, we haven't seen the evidence of

367
00:25:37,880 --> 00:25:41,680
that. We also don't know that
it's compliant. That was like a blocker,

368
00:25:41,720 --> 00:25:45,359
an adoption blocker in the early days. Nowadays people go to the cloud

369
00:25:45,400 --> 00:25:49,039
because it's more secure. Yeah,
and nothing you ever do is going to

370
00:25:49,119 --> 00:25:52,680
change that, you know, and
arguabily it always has been. It's just

371
00:25:52,720 --> 00:25:56,640
that now you can't deny it.
Yeah, yeah, true, fair enough.

372
00:25:56,720 --> 00:26:00,480
Yeah, And we used to say
that do you think that you're security

373
00:26:00,480 --> 00:26:04,319
team who actually goes to bed at
night and works eight hours a day is

374
00:26:04,480 --> 00:26:10,920
more vigilant than you know, a
team of hundreds of people that are watching

375
00:26:11,279 --> 00:26:15,759
what's going on behind the scenes and
Azure twenty four to seven. It's just

376
00:26:15,799 --> 00:26:18,480
there's just no way. It's it's
not possible. There's no way. And

377
00:26:18,519 --> 00:26:23,880
somebody was saying, it's funny.
There was a networking company that was saying

378
00:26:23,920 --> 00:26:30,039
that, well, basically the slur
that that you know, networking for Microsoft,

379
00:26:30,079 --> 00:26:32,559
well it's not the best product,
it's not you know, I don't

380
00:26:32,599 --> 00:26:33,799
know that it's the most secure,
and so forth. I was like,

381
00:26:34,119 --> 00:26:37,000
are you shitting me? You just
pull that out of the air. You're

382
00:26:37,039 --> 00:26:41,880
talking about the company that probably has
the most of network of anyone on the

383
00:26:41,960 --> 00:26:47,640
planet, with with like the best
teams that do networking and networking security that

384
00:26:47,799 --> 00:26:52,319
even exists at all, and you're
just you're just blurting out there that you

385
00:26:52,400 --> 00:26:56,599
think that their product is not good
enough based on what dude? Yeah right,

386
00:26:56,160 --> 00:27:00,319
yeah, I can't wait to hear
your analysis in detail. I please

387
00:27:00,359 --> 00:27:04,039
have it at my desk by nine
o'clock tomorrow morning, exactly. Pass on

388
00:27:04,160 --> 00:27:10,880
stunt Kirk out. But it's a
it is, without a doubt, an

389
00:27:10,920 --> 00:27:14,480
extremely hard problem. Yes, and
gentlemen, we should stop at this moment

390
00:27:14,599 --> 00:27:21,319
for this very important message, and
we're back. It's done in rocks.

391
00:27:21,319 --> 00:27:25,039
I'm Richard Campbell, let's Carl Franklin, and we're talking to our friend Mangus

392
00:27:25,039 --> 00:27:29,559
Martins and about it his work in
the cloud, and we you mentioned it

393
00:27:29,599 --> 00:27:34,680
briefly at the top, but certainly
the Global Azure move event is a thing

394
00:27:34,759 --> 00:27:37,440
coming up in about another month,
right, it's in April, Yeah,

395
00:27:37,720 --> 00:27:44,920
April eighteen to through twenty. Is
what's happening this time, And it's been

396
00:27:44,960 --> 00:27:48,039
going on for a while now.
It's Global azer is a thing. Lot

397
00:27:48,200 --> 00:27:52,680
Lasher dot net is a thing that
has been around for quite some time.

398
00:27:53,359 --> 00:28:00,200
And what it really is more than
ten years, it's been twelve, so

399
00:28:00,880 --> 00:28:06,759
really what it is, it's I
guess it's instigation on learning, instigation on

400
00:28:07,000 --> 00:28:11,039
a global scale. If you will, we just hang out together and run

401
00:28:11,079 --> 00:28:15,400
this event everyone at once, and
everyone is. Any organizer of a community

402
00:28:15,400 --> 00:28:19,559
meet up anywhere who wants to learn
about Ashure, they simply just register their

403
00:28:19,599 --> 00:28:25,119
event as well on the same site. And the idea is that they can

404
00:28:25,119 --> 00:28:29,920
then benefit and their attendees can then
benefit from the shared sponsorships that we get,

405
00:28:29,960 --> 00:28:33,400
for example, from Microsoft. We
get ASHU passes from Microsoft, and

406
00:28:33,480 --> 00:28:36,599
if they need some, they can
get them through us. So we just

407
00:28:36,640 --> 00:28:40,799
do it all at the same time. Nice and it does speak too.

408
00:28:40,920 --> 00:28:45,720
I think a huge topic here,
which is learning to right size the utilization

409
00:28:45,759 --> 00:28:49,440
of Azure to get out of the
VM world and into the services world and

410
00:28:49,480 --> 00:28:55,000
into what it could be a platform
piece like the rethinking of the way you

411
00:28:55,039 --> 00:28:59,759
provide services to your company. Absolutely
right. We didn't even know what we

412
00:28:59,759 --> 00:29:03,359
were starting. When Global Azure did
start. We had no idea what was

413
00:29:03,400 --> 00:29:07,599
going to happen or what we what
we found. If you will, you

414
00:29:07,640 --> 00:29:11,839
know, they say that innovation is
just taking two pre existing ideas and combining

415
00:29:11,920 --> 00:29:15,400
them in a new way that no
one ever thought of yet. So I

416
00:29:15,400 --> 00:29:19,799
guess we did community and the Globe
because we just said, hey, how

417
00:29:19,799 --> 00:29:22,759
about you guys do an event at
the same time as we do, And

418
00:29:22,799 --> 00:29:26,319
then more and more people like,
hey, you also want to join,

419
00:29:26,400 --> 00:29:30,440
and you and you, and I
posted that thing internally at Microsoft for the

420
00:29:30,559 --> 00:29:33,680
MVPs, there is an alias called
the for Azure MVPs. I posted it

421
00:29:33,680 --> 00:29:37,160
there. I said, Hey,
a bunch of us are doing this community

422
00:29:37,200 --> 00:29:41,880
events on this one day, you
want to join, and ca boom.

423
00:29:41,000 --> 00:29:47,599
It was like an understatement. I
think it was like almost one hundred locations

424
00:29:47,680 --> 00:29:52,559
the first year we did it,
so apparently we found a potential that nobody

425
00:29:52,640 --> 00:29:56,200
knew would be there. Yeah,
that's kind of cool. There is noise.

426
00:29:56,240 --> 00:30:00,680
So how big is it now?
Well, it took a huge bump

427
00:30:02,079 --> 00:30:07,640
during the pandemic of this section.
Yeah, it's an important topic to cover.

428
00:30:07,839 --> 00:30:11,160
That community took a huge, you
know, whack on the head,

429
00:30:11,200 --> 00:30:18,160
and a lot of people just stopped
doing community and there weren't enough other people

430
00:30:18,200 --> 00:30:23,880
coming into doing community because you were
simply not allowed to have meetups TOLT.

431
00:30:25,200 --> 00:30:29,559
Yeah. Yeah, but just before
twenty nineteen was our largest year, and

432
00:30:29,960 --> 00:30:33,599
that year I think we were around
three hundred and thirty six locations at the

433
00:30:33,599 --> 00:30:37,759
same time. Wow, which is
huge. I don't know of any any

434
00:30:38,160 --> 00:30:44,160
larger community event than that. And
and what's one thing that I find is

435
00:30:44,440 --> 00:30:48,400
amusing is if you would aggregate all
the community events of the same kind that

436
00:30:48,480 --> 00:30:55,039
are called global something in you know, boot camp, there's there's AI,

437
00:30:55,200 --> 00:30:59,160
there's DevOps, there's plenty of them, even a couple coming out from Microsoft

438
00:30:59,160 --> 00:31:03,279
that are using the same naming standard. We were the first ones to ever

439
00:31:03,359 --> 00:31:08,200
do a global something boot Camp,
Global Azure, Azure boot Camp as it

440
00:31:08,240 --> 00:31:12,359
were. Nowadays there's tons of them
doing the same kind of thing. So

441
00:31:12,400 --> 00:31:18,200
I guess we showed to the world
that it is possible to do community activism

442
00:31:18,480 --> 00:31:22,359
on a global scale, and now
kind of everyone does that. Yeah,

443
00:31:22,359 --> 00:31:25,759
I'm thinking net comp which is very
much around the world kind of thing.

444
00:31:25,880 --> 00:31:30,960
But I do think it came after
the the Azure Global Yeah, well,

445
00:31:30,000 --> 00:31:33,119
it's it's fine a new idea though, I mean, no, you know,

446
00:31:33,319 --> 00:31:37,640
the Internet kind of encourages us to
do that, you know, it

447
00:31:37,680 --> 00:31:41,759
really does. Well, you think
about the original RDS in the nineties doing

448
00:31:41,759 --> 00:31:45,279
the simul past events right really before
the Internet was even up for that,

449
00:31:45,279 --> 00:31:51,319
there were going to say accept no
Internet. Yeah that was when we had

450
00:31:51,359 --> 00:31:56,400
regions and we directed something weird.
Yeah yeah, those funny titles of art.

451
00:31:56,559 --> 00:32:02,880
Yeah yeah. I do think the
interaction between the sissigmin and the DEV

452
00:32:04,079 --> 00:32:08,599
around right sizing resources here about you
know that because on one hand, I

453
00:32:08,680 --> 00:32:14,039
know, as the sissidmin, Hey
these vms are running all the time,

454
00:32:14,079 --> 00:32:16,359
but they're not actually busy, like
how does this break apart? But you've

455
00:32:16,359 --> 00:32:21,200
got to interact with the dev to
say, can you give me a measure

456
00:32:21,279 --> 00:32:24,079
to know when to dial something down
and to dial it up for what we

457
00:32:24,079 --> 00:32:29,359
can move to a service. And
so you're thinking, I think mostly about

458
00:32:29,440 --> 00:32:35,440
things such as you know, CPU
or utilization or those kinds of things.

459
00:32:35,440 --> 00:32:38,359
What I even I even include a
third role in this conversation normally, And

460
00:32:38,400 --> 00:32:42,279
I stated, it has to start
there. It has to start with what

461
00:32:42,319 --> 00:32:46,640
the business people wants to measure.
What actually is a measurable something for the

462
00:32:46,799 --> 00:32:51,079
for the industry for what we are
doing. Is it number of sales?

463
00:32:51,480 --> 00:32:53,359
Is it, you know, a
number of users? Is it a number

464
00:32:53,359 --> 00:32:58,920
of What is it that that makes
the day for the business owner in this

465
00:32:59,039 --> 00:33:04,200
context? And I always say that
you start by measuring that you don't put

466
00:33:04,240 --> 00:33:07,119
anything out there. You don't put
anything in the cloud unless you have a

467
00:33:07,160 --> 00:33:12,240
measurement that satisfies that what they are, what the suits, the being counters,

468
00:33:12,279 --> 00:33:15,000
they excel people what they are,
what they are asking for. That's

469
00:33:15,039 --> 00:33:17,880
what we really want to see.
And if we can have that on a

470
00:33:17,960 --> 00:33:24,200
dashboard in the beginning right arguably before
the workload moved, would be nice.

471
00:33:24,519 --> 00:33:28,160
Exactly, that would be the best. That would be the absolute best.

472
00:33:28,319 --> 00:33:31,680
And then we can actually see that
trend line over time and they can understand

473
00:33:32,720 --> 00:33:37,000
and that that I think is the
most important. And of course then then

474
00:33:37,079 --> 00:33:40,960
comes you know, CPU percentage and
memory usage and other things that are Yeah,

475
00:33:42,680 --> 00:33:49,240
I appreciate that those are very abstract
measures where we're yeah, cost of

476
00:33:49,319 --> 00:33:53,000
transaction, value of transaction, like
those kinds. I'm an email I did

477
00:33:53,000 --> 00:33:58,279
a lot of e commerce work.
Yeah, we made the dashboard that showed

478
00:33:58,279 --> 00:34:00,640
when the website was faster, we
made more money for customer. Yeah,

479
00:34:00,880 --> 00:34:06,839
which which needle is it that moves
the business? That's what we want to

480
00:34:06,880 --> 00:34:09,920
note that if I can get a
CFO excited, I can get more budget.

481
00:34:10,199 --> 00:34:15,960
That's correct. And also you won't
have you won't have that surprise conversation

482
00:34:15,159 --> 00:34:19,440
in the future where it's like,
why is our costing this much? We

483
00:34:19,440 --> 00:34:21,920
didn't think it was going to cost
this much. Oh, we're backing out

484
00:34:21,920 --> 00:34:23,480
of the cloud, We're moving away. This is not what we expected.

485
00:34:23,719 --> 00:34:28,320
Well, I also feel like we
did not measure stuff on prem Well,

486
00:34:29,559 --> 00:34:32,599
that's true. The side effective azure
is you are getting a monthly bill and

487
00:34:32,679 --> 00:34:39,199
that varies. It is and that
yes, And it's hard because we have

488
00:34:39,320 --> 00:34:45,719
been taught for decades that i T
is a budget item. It's a huge

489
00:34:45,199 --> 00:34:49,320
piece piece of Yes, cost center
on the budget. This like this,

490
00:34:49,320 --> 00:34:52,639
this many this many millions and and
that's it. And now there's no more

491
00:34:52,679 --> 00:34:57,360
conversation between the business and the i
T until next year. We're going to

492
00:34:57,440 --> 00:35:01,519
have the same conversation once. And
it's really weird. And that's exactly what

493
00:35:01,760 --> 00:35:06,719
how the cloud does not work.
Yeah, maybe that's the real issue bank

494
00:35:06,800 --> 00:35:12,159
This is that when I was directing
it, I had an annual conversation with

495
00:35:12,280 --> 00:35:15,360
finance about our spend for the year. That's right. And occasionally you go

496
00:35:15,400 --> 00:35:19,159
in for exceptions because hey, we're
making this big push and that I need

497
00:35:19,159 --> 00:35:21,320
to get a few more machines.
And it's got a two month leap time,

498
00:35:21,679 --> 00:35:25,199
but that was an exception. But
I had one year and typically four

499
00:35:25,320 --> 00:35:29,320
year horizons, like we're going to
renew the data set of contract, We're

500
00:35:29,320 --> 00:35:31,880
going to buy new hardware. And
these were two Comma spends right Like they

501
00:35:31,920 --> 00:35:36,679
were big spends. So you would
sit in the boardroom with the finance guys

502
00:35:36,719 --> 00:35:39,440
and hammer this stuff out and we'd
look at leasing deals and all these other

503
00:35:39,480 --> 00:35:44,599
things, and all of that goes
away with the cloud in exchange for a

504
00:35:44,719 --> 00:35:50,079
monthly bill that kind of like your
cell phone bill can surprise you. Yeah,

505
00:35:50,159 --> 00:35:52,519
well, when I turn on my
phone in bangal Or, it costs

506
00:35:52,559 --> 00:35:57,960
five bucks a minute, and good
luck trying to decipher and how well does

507
00:35:58,119 --> 00:36:01,119
az your advisor help with this kind
of thing? And do you use it?

508
00:36:01,199 --> 00:36:06,599
And what is it? So I
use azur Advisor to to look up

509
00:36:06,639 --> 00:36:08,760
the recommendations that are in there.
Is like saying there are things like,

510
00:36:08,800 --> 00:36:13,760
okay, you're you seem to be
running a lot of virtual machines, would

511
00:36:13,800 --> 00:36:19,480
you maybe consider using a reserved instance
plan instead? And makes like are you

512
00:36:19,519 --> 00:36:21,719
going to keep these machines for a
year? Are you going to keep them

513
00:36:21,760 --> 00:36:24,159
for three years? Does that look
like a thing you might do? Well?

514
00:36:24,559 --> 00:36:29,559
Then here's the recommendation that says you
should be buying a reserved instance plan

515
00:36:30,559 --> 00:36:35,840
to get to get a really really
big discount if you do. But that

516
00:36:36,079 --> 00:36:39,199
that really it sounds such a it
sounds like such a technical question, right,

517
00:36:39,599 --> 00:36:44,079
but it really is not. This
is this is a very very interesting

518
00:36:44,159 --> 00:36:47,320
question that I've been using for for
quite some time to illustrate what the problem

519
00:36:47,400 --> 00:36:52,800
really is. So you're saying,
it sounds like I can just say next,

520
00:36:52,800 --> 00:36:55,440
next, finish on this dialogue,
and and and all of a sudden,

521
00:36:55,440 --> 00:37:00,440
I have this great discount because I
now bought a reserved instance plan.

522
00:37:00,960 --> 00:37:04,559
Wonderful. Okay, So now I
just spent a couple one hundred thousand dollars

523
00:37:04,960 --> 00:37:07,000
doing that. Because if we had
a many we had many machines, maybe

524
00:37:07,000 --> 00:37:10,880
I should ask, you know,
the business people first. And now it

525
00:37:10,880 --> 00:37:16,880
becomes this question of are we going
to continue using vms as in infrastructure services

526
00:37:17,159 --> 00:37:21,760
or are we actually going to migrate
the applications away from those vms and put

527
00:37:21,760 --> 00:37:25,360
them on platform services instead. That
conversation, are you going to have that

528
00:37:25,400 --> 00:37:30,360
with a bunch of people who are
business people and IT people. Well,

529
00:37:30,400 --> 00:37:32,360
now you have to talk to the
devs. Are the debvs ready to actually

530
00:37:32,440 --> 00:37:37,920
move them out of the vms the
applications. That is right, and that

531
00:37:37,119 --> 00:37:44,239
strategy is much much larger conversation than
the question of should I activate reserve instances.

532
00:37:44,960 --> 00:37:47,119
You're right, but it also makes
more sense because I should have the

533
00:37:47,119 --> 00:37:50,239
finance guys in the room at that
point, I should have dev room.

534
00:37:50,280 --> 00:37:52,639
Now we're going to make a schedule. Hey, here are all the apps

535
00:37:53,199 --> 00:37:58,599
that are all running in vms right
now. We can do one a quarter,

536
00:37:59,480 --> 00:38:01,800
yes, these one, these ones
that are ten down the line,

537
00:38:01,840 --> 00:38:07,239
you can start buying years of reservedances
for them, lowering their for sure.

538
00:38:06,840 --> 00:38:09,840
For some of the other one absolutely
one hundred percent. For some of them.

539
00:38:09,960 --> 00:38:14,519
It is not even worth it to
you know, that application, it's

540
00:38:14,559 --> 00:38:17,559
not so important, it's not that
expensive, and rebuilding it to move off

541
00:38:17,559 --> 00:38:22,320
of it. That fix it,
forget about it. That's not we'll just

542
00:38:22,400 --> 00:38:24,920
we'll just leave that one. But
the other one here is magnus. It's

543
00:38:24,960 --> 00:38:29,280
getting it's getting those guys in the
room. Like so often, we're doing

544
00:38:29,320 --> 00:38:32,800
our own job and not worry not
recognizing that this is a strategic conversation where

545
00:38:32,840 --> 00:38:36,880
all the players need to be in
the room and making a plan. Well,

546
00:38:37,000 --> 00:38:38,960
you know which person we haven't called
out yet, And I would argue

547
00:38:38,960 --> 00:38:42,800
that this is this is the most
important person to have in the room for

548
00:38:42,840 --> 00:38:49,599
this conversation. You want to know
the human resource director. Interesting, Yes,

549
00:38:50,039 --> 00:38:53,079
I only call it HR when I've
got problems. That's right. But

550
00:38:53,239 --> 00:38:57,000
you have a problem, but you
just don't know it yet. The problem.

551
00:38:57,280 --> 00:39:00,119
The problem is that you don't have
enough people with enough skills in azure.

552
00:39:00,519 --> 00:39:04,360
You have wonderful, wonderful people in
your business. They work there for

553
00:39:04,400 --> 00:39:06,679
a long time. They know their
stuff, they know it t they know

554
00:39:06,760 --> 00:39:08,559
dev, they know all that,
but they just don't know how to make

555
00:39:08,599 --> 00:39:14,159
an optimized, secure, compliant deployment
of your application in the cloud. The

556
00:39:15,199 --> 00:39:21,320
factory. That's a that's that's that's
exactly you have a factory pattern. Now

557
00:39:21,320 --> 00:39:25,519
you have two problems, right,
yeah, yes, So that means that

558
00:39:25,559 --> 00:39:30,880
you have to have a solid plan
for either acquiring skilled people. Good luck

559
00:39:30,880 --> 00:39:35,079
with that. Those the people who
are really skilled with cloud stuff, they

560
00:39:35,079 --> 00:39:37,800
already have a job. But you
also, yeah, they're busy. Uh.

561
00:39:37,840 --> 00:39:42,199
And then you have to also include
uh, you know, the training

562
00:39:42,280 --> 00:39:46,039
or retraining or upskilling or changing the
skill of your your existing people so that

563
00:39:46,079 --> 00:39:51,199
they will understand how to actually use
the cloud in a good way. I

564
00:39:51,239 --> 00:39:57,639
mean, I do debate the value
of training getting skilled in migration because that's

565
00:39:57,639 --> 00:40:01,360
going to end, yes in x
men years. It's operations that you want

566
00:40:01,400 --> 00:40:07,679
to be skilled. Operations, governance, the security and compliance pieces that are

567
00:40:07,679 --> 00:40:10,119
going to be so important. Maybe
you didn't know that to begin with that

568
00:40:10,320 --> 00:40:15,199
compliance was going to be important because
you didn't realize. And then much later

569
00:40:15,320 --> 00:40:17,719
you have an application running in the
cloud and your customer says, we want

570
00:40:17,760 --> 00:40:21,199
to be we need to be compliant
with so and so. Are you complied

571
00:40:21,280 --> 00:40:23,119
with so and so? Otherwise we
can't buy your service? And now compliance

572
00:40:23,199 --> 00:40:27,360
is very important, but it wasn't
before, So now we have to go

573
00:40:27,480 --> 00:40:30,519
fix compliance after the fact. Is
there any AI that Azure has, or

574
00:40:30,519 --> 00:40:37,119
maybe even chat GPT that can help
us find problems and correct them. It's

575
00:40:37,360 --> 00:40:42,559
getting to that point you can't You
can't release anything from Microsoft right now except

576
00:40:42,960 --> 00:40:45,320
going through like this AI wash first, right, is it AI enough?

577
00:40:45,480 --> 00:40:50,559
If it's not a enough, they
can they can get it. I mean,

578
00:40:50,599 --> 00:40:53,280
at the simplest level, you could
ask chat GPT here, tell me

579
00:40:53,320 --> 00:40:58,199
how to get a list of resources
that haven't been used in the last six

580
00:40:58,239 --> 00:41:04,039
months and help help me deactivate them. Yeah, but imagine having a complete

581
00:41:04,079 --> 00:41:07,559
AI engine inside of your governance in
your cloud use. I'm not so sure

582
00:41:07,559 --> 00:41:10,880
I want that. You'll ask it. You'll ask it. Hey, dude,

583
00:41:12,840 --> 00:41:15,079
what can I like? How can
I optimize my cost? And it's

584
00:41:15,119 --> 00:41:20,159
going to base that on your utilization, your usage, and it's going to

585
00:41:20,159 --> 00:41:23,199
base it on the experience of the
entire usage of everyone using Yeah, that

586
00:41:23,280 --> 00:41:27,760
I would like. I would like
something to tell me what to do when

587
00:41:27,760 --> 00:41:32,119
I ask a question. But as
we discovered in the better known framework story,

588
00:41:32,639 --> 00:41:37,639
you don't want something that just has
agency and it's going to go do

589
00:41:37,719 --> 00:41:40,320
things on your behalf without you being
involved. You know. No, No,

590
00:41:40,719 --> 00:41:44,400
that's not what we're talking about.
I guess no, No, I'm

591
00:41:44,400 --> 00:41:46,920
talking about just help me figure this
out. What am I doing wrong,

592
00:41:47,679 --> 00:41:52,320
you know, or why this application
seems to be down. And then this

593
00:41:52,480 --> 00:41:54,639
also comes back to this age old
question of what is an application in the

594
00:41:54,639 --> 00:41:59,159
cloud? If you have you know, my application consists of twenty five different

595
00:41:59,199 --> 00:42:02,920
resources, multiple different you know,
locations around my subscriptions and various things.

596
00:42:04,320 --> 00:42:07,920
I want to be able to see
that thing as one application because to me.

597
00:42:08,280 --> 00:42:12,679
It is one. So what is
this wonderful AI you're talking about?

598
00:42:13,800 --> 00:42:15,400
I was getting to it. I
mean Microsoft is getting to it. They're

599
00:42:15,440 --> 00:42:20,920
they're working all of these things into
their their existing services. They're taking all

600
00:42:20,960 --> 00:42:22,840
the services that they had and just
adding AI to them. If you yeah,

601
00:42:22,880 --> 00:42:25,760
we think it. Azure Advisor will
have an LM and it's sooner or

602
00:42:25,840 --> 00:42:30,440
later. Yeah, probably sooner.
Yeah. Yeah, and Azure monitor and

603
00:42:30,920 --> 00:42:35,719
you know this has been this has
been a while coming. Know that there

604
00:42:35,800 --> 00:42:39,559
was this Azure graph thing you could
have. Yeah, so Azure graph is

605
00:42:39,559 --> 00:42:44,679
is just a little graph that explains
that all the resources, all all the

606
00:42:44,719 --> 00:42:49,000
things you have. So that's one
step in that direction. But Microsoft is

607
00:42:49,000 --> 00:42:52,360
not taking multiple steps in that direction
over the I believe over the next six

608
00:42:52,400 --> 00:42:57,719
months is going to be an exciting
time for monitoring Azure. Okay, fair

609
00:42:57,840 --> 00:42:59,960
enough. Yeah, do you think
it's going to It's going to evolve.

610
00:43:00,039 --> 00:43:04,400
But I appreciate that probably they're not
going to introduce new tools. I mean,

611
00:43:04,639 --> 00:43:07,079
haven't helped me if they do,
but rather take the tools that we're

612
00:43:07,119 --> 00:43:14,000
already supposedly understand and add more LM
to them right here and there. Yeah,

613
00:43:14,039 --> 00:43:16,880
that would be wonderful. Yeah,
that's what we're going to get.

614
00:43:16,920 --> 00:43:22,559
I'm sure. I'm certain there's plenty
of evidence to show that they're doing deep

615
00:43:22,679 --> 00:43:25,280
out Like you look at some of
this other sentinels doing this day and its

616
00:43:25,320 --> 00:43:30,639
ability to look across multiple logs and
events and so forth and say, hey,

617
00:43:30,679 --> 00:43:34,320
there's some unusual behavior here. As
a composite of that view, that's

618
00:43:34,360 --> 00:43:37,199
a good summarizer engineer, Yeah,
it is. And that argument is what

619
00:43:37,239 --> 00:43:40,119
we have been using for a while
to talk about, for example, security

620
00:43:40,119 --> 00:43:45,079
in the cloud. Why is a
company why are the huge cloud giants so

621
00:43:45,239 --> 00:43:51,000
good at security? Well, it's
because they have a data pool of security

622
00:43:51,159 --> 00:43:54,719
data which is larger than anyone's.
Yeah, you might be on your business,

623
00:43:54,760 --> 00:43:58,840
your tiny little business, which you
think is huge, is very small

624
00:43:58,880 --> 00:44:01,920
in the cloud. It's being attacked
right now, except you can't see that

625
00:44:02,000 --> 00:44:07,360
based on the data you have.
But Microsoft can see the attack on you,

626
00:44:07,400 --> 00:44:09,599
an attack on other people as a
larger pattern, and they can tell

627
00:44:09,639 --> 00:44:14,480
you listen, this thing here that
you don't perceive as an attack is in

628
00:44:14,519 --> 00:44:19,320
fact an attack, right, And
that, you know, is why you

629
00:44:19,360 --> 00:44:22,920
want to have the power of that
information size, that data size, and

630
00:44:22,960 --> 00:44:27,639
if you put an NLM on that
then you have something. Yeah, you

631
00:44:27,679 --> 00:44:30,880
have some interesting things, So I
am looking. It will be interesting to

632
00:44:30,880 --> 00:44:34,840
see what that's going on. Of
course, we're recording this just before the

633
00:44:34,920 --> 00:44:37,760
MVP summit. You and I are
going to be there, I think magnets

634
00:44:37,880 --> 00:44:42,360
so exactly. You know, there's
some briefings along these lines we're not going

635
00:44:42,400 --> 00:44:45,880
to be allowed to talk about,
right, Yeah, clearly there's motion there.

636
00:44:45,920 --> 00:44:52,280
This last year seemed like the hype
cycle for especially from Microsoft, with

637
00:44:52,360 --> 00:44:55,519
the Copilot. Of all the things. I'm hopefully this is the year that

638
00:44:55,559 --> 00:45:00,599
we see a bunch of implemented stuff
that can make our lives better. Yeah,

639
00:45:00,599 --> 00:45:02,719
you get a copilot, you get
a call a copilot, everybody gets

640
00:45:02,719 --> 00:45:07,119
a co pilot. At one point
I was poking some friends inside of the

641
00:45:07,119 --> 00:45:09,000
company. There was like, there's
one hundred and eighty of these and most

642
00:45:09,000 --> 00:45:12,920
of them should not see the light
of day. Oh my god. No.

643
00:45:13,440 --> 00:45:17,840
Yeah, I can see why as
well. But that's to that point.

644
00:45:17,920 --> 00:45:21,599
It feels a little bit in the
beginning here, which is where we

645
00:45:21,639 --> 00:45:25,440
are in the beginning of the co
pilot era, is that some of these

646
00:45:25,480 --> 00:45:31,440
are so bland or they really they
really can't do anything yet that it's just

647
00:45:31,480 --> 00:45:34,960
like, if you're going to release
a new version of whatever you have,

648
00:45:35,400 --> 00:45:37,719
it's going to have to have AI
and a co pilot in it, or

649
00:45:37,719 --> 00:45:39,639
you're not shipping. So they are
like shipping this tiny little thing that they

650
00:45:39,679 --> 00:45:45,039
call a co pilot for whatever,
and it can't do anything yet. Yeah.

651
00:45:45,199 --> 00:45:47,440
But one of the reasons I mentioned
Aspire at the top of the show

652
00:45:47,719 --> 00:45:52,599
is it's it's about cloud native behavior
too, Like I'm hoping that our tools

653
00:45:52,760 --> 00:46:00,000
as devs get better at thinking helping
us fall into cloud thinking. Yeah.

654
00:46:00,039 --> 00:46:06,119
Yes, it's not just that I
identified that these are these are secure related

655
00:46:06,159 --> 00:46:08,320
strings or special tokens and so forth. Is that I'm also you know,

656
00:46:08,440 --> 00:46:13,840
pushing towards key vaults or where is
the secure stur for this that we can

657
00:46:14,239 --> 00:46:17,519
read against like that sort of cloud
native behavior so that we tend to do

658
00:46:17,559 --> 00:46:21,880
the right things that we're building software. Mm hmm, that's right, that's

659
00:46:21,960 --> 00:46:24,119
right. Well, I can't wait
to see what happens. I hope so,

660
00:46:24,159 --> 00:46:29,079
because I don't know that it's true
yet, we'll see, we'll see.

661
00:46:29,119 --> 00:46:30,480
I think it. I think it
will be true because it seems like

662
00:46:30,760 --> 00:46:36,119
an actual good use case of of
L MSM, so I saved the most

663
00:46:36,119 --> 00:46:40,920
important question for last. What flavors
of ice cream? Does lofty soft serve?

664
00:46:43,960 --> 00:46:50,119
Serves? That? Definitely observes it
serves chocolate, chocolate. Nice.

665
00:46:50,920 --> 00:46:53,440
I really can't pick on you for
a crazy name with a name like pwop

666
00:46:53,960 --> 00:46:55,920
Yeah, yeah, where are you
going to go with that? But I

667
00:46:55,960 --> 00:47:00,159
applaud your you know, being funny
there with lofty soft. It's kind of

668
00:47:00,159 --> 00:47:06,079
a funny sounding name, but you
won't forget it. Well, I just

669
00:47:06,079 --> 00:47:07,280
wanted to buy it a man that
had dot com at the end. I

670
00:47:07,400 --> 00:47:13,639
wasn't too expensive. It's not a
bad reasons the name you can get yeah,

671
00:47:13,719 --> 00:47:15,920
exactly. And people be like,
lofty does that mean that because you're

672
00:47:15,960 --> 00:47:19,119
doing things in the cloud? I'm
like, yeah, sure, yeah,

673
00:47:19,119 --> 00:47:22,880
that's exactly what that was. That
was what I was going for all along.

674
00:47:22,960 --> 00:47:28,519
Actually, yes, all the truth
is out Magnus now we know yep.

675
00:47:29,039 --> 00:47:34,079
Yeah. And Howard and meatballs at
home. It was the last time

676
00:47:34,159 --> 00:47:37,039
Jessica whipped up a batch of those, so they're good as ever. You

677
00:47:37,079 --> 00:47:40,480
know you should you should pop over
Yeah, be right there. It's been

678
00:47:40,519 --> 00:47:44,480
a long time, so we av
that's all started at an or dev right

679
00:47:44,519 --> 00:47:47,159
yeah, that's what we met.
Yeah, yeah, network keep seeing each

680
00:47:47,159 --> 00:47:52,480
other at every freaking conference around the
world. I guess I would go to

681
00:47:52,679 --> 00:47:55,760
ordev just to attend to go to
the bishop's arms and to come over for

682
00:47:55,840 --> 00:47:59,679
meatballs. I would do that.
There you go. Well, I'm happy

683
00:47:59,719 --> 00:48:06,159
to have It's just we can't do
it October November when it's freaking it's terrible.

684
00:48:06,320 --> 00:48:12,079
That's when you drink the wormwood my
mouse. Sorry. I don't know

685
00:48:12,079 --> 00:48:14,760
if they serve that at the Green
Lion, but I'd rather have scotch.

686
00:48:15,519 --> 00:48:19,519
Yeah, fair enough. Those are
those are good bars. So is there

687
00:48:19,519 --> 00:48:22,599
anything else you want to talk about
before we jump off? I mean,

688
00:48:22,079 --> 00:48:27,039
so, I'm really looking forward to
the next steps on the cloud, and

689
00:48:27,119 --> 00:48:30,719
I guess we kind of all we
already covered that. But I think that

690
00:48:30,199 --> 00:48:37,480
the new Horizons scenario, it's just
to continue to make it easier for people

691
00:48:37,519 --> 00:48:40,760
to get it. And that's hard
when you keep adding new things all the

692
00:48:40,800 --> 00:48:44,760
time. And I think, I
think though, I think though, that

693
00:48:44,880 --> 00:48:49,400
we are we are at a place
now where the cloud is It's hard to

694
00:48:49,440 --> 00:48:52,880
say it's stabilizing, but I think
it is, and it seems to be.

695
00:48:52,519 --> 00:48:57,119
It seems to be and a challenge
now is that now everyone is talking

696
00:48:57,159 --> 00:49:00,960
about AI. Okay, cool,
we love AI. AIR services are cool?

697
00:49:00,039 --> 00:49:04,159
AIR services are great? I get
it, I get it. Where

698
00:49:04,199 --> 00:49:07,599
are all the AI services running?
They're running in the cloud. So what

699
00:49:08,639 --> 00:49:13,000
a company, and this is kind
of a cautionary tail here, what a

700
00:49:13,039 --> 00:49:17,480
company really needs to understand is to
use AI services with you know, oh

701
00:49:17,559 --> 00:49:21,559
we have we're going to put you
know, all of our massive amounts of

702
00:49:21,639 --> 00:49:25,400
data and we're going to do our
own copilot and whatnot. We're going to

703
00:49:25,679 --> 00:49:29,239
run it on our own data and
so forth. Oh, that's good for

704
00:49:29,280 --> 00:49:32,559
you, except you now realize that
you have to understand how to use the

705
00:49:32,559 --> 00:49:37,159
cloud platform as well, because you're
moving petabytes of data into the cloud now,

706
00:49:37,400 --> 00:49:42,119
and they're security and governance and data
transport and all the things and costs.

707
00:49:42,920 --> 00:49:46,079
But the conversation right now seems to
be all about AI and how cool

708
00:49:46,119 --> 00:49:49,360
AI is and how we're going to
use that. Well, guess what,

709
00:49:49,480 --> 00:49:52,519
there's a platform underneath called Azure that
you may need to look into as well.

710
00:49:52,679 --> 00:49:57,159
So that's kind of a cautionary tale
here, and all that AI excitement

711
00:49:57,239 --> 00:50:01,000
is that you actually need to understand
the platform. Well, center enough sure,

712
00:50:01,320 --> 00:50:05,400
mangus. Thanks, it's been a
pleasure catching up with you again.

713
00:50:05,800 --> 00:50:07,239
Yeah, thank you very much.
All right, we'll talk to you next

714
00:50:07,239 --> 00:50:31,599
time on dot net rocks. Dot
net Rocks is brought to you by Franklin's

715
00:50:31,679 --> 00:50:37,639
Net and produced by Pop Studios,
a full service audio, video and post

716
00:50:37,639 --> 00:50:42,360
production facility located physically in New London, Connecticut, and of course in the

717
00:50:42,440 --> 00:50:47,320
cloud online at pwop dot com.
Visit our website at d O T N

718
00:50:47,360 --> 00:50:52,920
E t R O c k S
dot com for RSS feeds, downloads,

719
00:50:52,039 --> 00:50:57,320
mobile apps, comments, and access
to the full archives going back to show

720
00:50:57,400 --> 00:51:01,119
number one, recorded in September two
thousand into and make sure you check out

721
00:51:01,119 --> 00:51:05,800
our sponsors. They keep us in
business. Now go write some code.

722
00:51:06,360 --> 00:51:15,079
See you next time. You got
JAD metal vans by the Means Home,

723
00:51:15,599 --> 00:51:16,599
then my Texes
