1
00:00:01,080 --> 00:00:03,000
Speaker 1: How'd you like to listen to dot net rocks with

2
00:00:03,040 --> 00:00:07,879
no ads? Easy? Become a patron for just five dollars

3
00:00:07,919 --> 00:00:10,800
a month. You get access to a private RSS feed

4
00:00:10,839 --> 00:00:14,240
where all the shows have no ads. Twenty dollars a month.

5
00:00:14,279 --> 00:00:16,879
We'll get you that and a special dot net rocks

6
00:00:16,960 --> 00:00:21,000
patron mug. Sign up now at Patreon dot dot NetRocks

7
00:00:21,120 --> 00:00:37,719
dot com. Hey guess what might it's dot ad rocks

8
00:00:38,000 --> 00:00:40,600
episode nineteen seventy nine. And I said that with an

9
00:00:40,600 --> 00:00:45,320
Australian accent because I'm Carl Franklin here in Connecticut.

10
00:00:44,799 --> 00:00:48,359
Speaker 2: And I'm Richard Campbell, and I'm down in Queensland, Australia.

11
00:00:48,479 --> 00:00:51,200
Speaker 1: So and our guest is in Australia as well, So

12
00:00:51,240 --> 00:00:53,399
it's gonna be a down Under show today.

13
00:00:53,479 --> 00:00:55,240
Speaker 2: Say we got a Southern Hemisphere bias?

14
00:00:55,560 --> 00:01:00,320
Speaker 1: Yeah, all right, Uh so let's talk a bit out

15
00:01:00,359 --> 00:01:04,760
the show number. It's nineteen seventy nine, So what happened

16
00:01:04,799 --> 00:01:05,640
in that year?

17
00:01:05,959 --> 00:01:08,719
Speaker 2: Well, let me just oh a fair bit. Yeah, a

18
00:01:08,760 --> 00:01:11,200
lot of things. You perusing the list, I've got a

19
00:01:11,239 --> 00:01:13,400
lot of space and a lot, well a little bit

20
00:01:13,400 --> 00:01:16,159
of space, a lot of science at a huge amount

21
00:01:16,159 --> 00:01:17,120
of compute, Well.

22
00:01:16,959 --> 00:01:19,840
Speaker 1: A lot happened in Iran, China, and the Soviet Union

23
00:01:20,040 --> 00:01:26,359
in Afghanistan. Three Mile Island happened, the Iran hostage crisis,

24
00:01:27,519 --> 00:01:32,640
the Iranian Revolution, Soviet invasion of Afghanistan.

25
00:01:32,719 --> 00:01:35,680
Speaker 2: That didn't end well, Nope for them, for anyone.

26
00:01:36,359 --> 00:01:40,439
Speaker 1: US China relations. United States severed diplomatic ties with Taiwan,

27
00:01:40,599 --> 00:01:45,640
established full diplomatic relations with the People's Republic of China. Oh,

28
00:01:45,799 --> 00:01:50,239
let's see some The music for UNSEF concert was kind

29
00:01:50,280 --> 00:01:52,239
of important.

30
00:01:52,480 --> 00:01:56,159
Speaker 2: We're heading into the Live aid era, right, Yeah, that's coming.

31
00:01:56,359 --> 00:01:59,159
Speaker 1: The Dukes of Hazzard premiered on January twenty sixth I

32
00:01:59,200 --> 00:02:02,480
remember my brother and Nice to drive my mother crazy,

33
00:02:03,120 --> 00:02:05,560
all the car chases and stuff. She would like come

34
00:02:05,599 --> 00:02:07,840
in and turn the television off. It just drove her

35
00:02:07,920 --> 00:02:10,680
up the wall. I think that was the point of

36
00:02:10,680 --> 00:02:11,759
the Dukes, that hazard, wasn't it.

37
00:02:12,039 --> 00:02:12,240
Speaker 2: Driver.

38
00:02:12,520 --> 00:02:18,680
Speaker 1: Yeah, you'll talk about space. So an anti nuclear demonstration

39
00:02:18,879 --> 00:02:22,680
happened on September twenty third. Nearly two hundred thousand people

40
00:02:23,479 --> 00:02:27,960
protested nuclear power in New York City, or just nuclear

41
00:02:27,960 --> 00:02:34,599
in general. But tell us about space and compute what happened.

42
00:02:35,280 --> 00:02:38,960
Speaker 2: Ninety seventy nine is the year that the Columbia Shuttle

43
00:02:39,080 --> 00:02:41,759
is delivered to Kennedy Space Center. It's still two years

44
00:02:41,800 --> 00:02:44,639
away from flying, but they it has now been built.

45
00:02:44,800 --> 00:02:49,319
Enterprise is being retired as a museum article after finishing

46
00:02:49,400 --> 00:02:51,479
its testing. It was too expensive to refit it into

47
00:02:51,479 --> 00:02:54,159
being a spacecraft, although that had been the original attension.

48
00:02:54,159 --> 00:02:57,800
Instead they'll use a different test article, which will become Challenger. Yeah,

49
00:02:57,919 --> 00:03:01,759
but Columbia's now assembled, bunch of changes. It still has problems.

50
00:03:01,800 --> 00:03:04,520
It's the heaviest of the shuttles that will fly, and

51
00:03:04,560 --> 00:03:07,439
so it's limited by certain aspects. But it's still you know,

52
00:03:07,479 --> 00:03:09,639
we're making progress, although a couple of years away from flying.

53
00:03:10,919 --> 00:03:13,319
Voyager one and two make their fly bys of Jupiter

54
00:03:13,319 --> 00:03:16,159
in nineteen seventy nine, and Pioneer eleven does the very

55
00:03:16,159 --> 00:03:19,120
first flyby of Saturn. Wow. So those are all new things.

56
00:03:19,159 --> 00:03:26,120
And Skylab in July nineteen seventy nine now originally been

57
00:03:26,159 --> 00:03:29,199
launched in nineteen seventy three as part of the Apollo

58
00:03:29,520 --> 00:03:32,439
Missions projects. It was only intended for one hundred and

59
00:03:32,520 --> 00:03:36,400
forty days of use, okay, and so this is already

60
00:03:36,439 --> 00:03:39,120
six years later. They actually had one hundred and seventy

61
00:03:39,120 --> 00:03:42,080
two days of use over three missions they extended it.

62
00:03:42,080 --> 00:03:46,120
It had plenty of problems they and it had been

63
00:03:46,199 --> 00:03:48,879
left empty, although prepped for someone else to visit it.

64
00:03:48,919 --> 00:03:51,400
They had a welcome kit, the door been left unlocked,

65
00:03:51,400 --> 00:03:54,599
all that sort of thing. And they had thought that

66
00:03:54,639 --> 00:03:56,000
they would be able to get the shutle ready in

67
00:03:56,000 --> 00:03:57,560
time to keep an a do over. In fact, they

68
00:03:57,560 --> 00:03:59,120
figured it would stay it over till at least the

69
00:03:59,120 --> 00:04:02,199
early eighties, but there was a solar maximum going on

70
00:04:02,240 --> 00:04:05,560
then they and the additional solariation expands the atmosphere to

71
00:04:05,639 --> 00:04:09,400
the dragon Skylab was greater than expected. Also, there was

72
00:04:09,439 --> 00:04:13,800
a lot broken. You know, they used control moment gyros

73
00:04:13,840 --> 00:04:16,639
for directional stabilization on Skylab, and at that point one

74
00:04:16,639 --> 00:04:19,079
had already failed and another one was failing, and there

75
00:04:19,120 --> 00:04:21,360
had been no plan to make them serviceable, so there

76
00:04:21,439 --> 00:04:23,600
was really no way to fix them. They would literally

77
00:04:23,680 --> 00:04:26,360
have to deploy a new set of gyros on some

78
00:04:26,480 --> 00:04:28,360
kind of attachment to the space station. So there's going

79
00:04:28,399 --> 00:04:29,879
to get this point was like, you could just build

80
00:04:29,879 --> 00:04:30,879
another space station.

81
00:04:30,839 --> 00:04:32,720
Speaker 1: So Skylab was manned. At one point.

82
00:04:32,879 --> 00:04:35,639
Speaker 2: My Skylab had three mission sent to it, yes, okay, one, two,

83
00:04:35,680 --> 00:04:38,079
and three, and we talked about those you know, in

84
00:04:38,120 --> 00:04:42,839
the past few shows when when those happened. But now

85
00:04:42,879 --> 00:04:45,240
it had been empty for several years, the atmosphere had expanded,

86
00:04:45,240 --> 00:04:47,199
so it was reintercasterment and expected, and it came down

87
00:04:47,240 --> 00:04:50,759
over the Indian Ocean and Australia, and.

88
00:04:50,800 --> 00:04:54,319
Speaker 1: It lasted a lot longer than my and my brother's

89
00:04:54,600 --> 00:04:59,120
snow sculpture referring to the blizzard of seventy eight that

90
00:04:59,160 --> 00:05:01,000
we talked about last week's.

91
00:05:00,879 --> 00:05:03,959
Speaker 2: In the previous episode. One other science one before we

92
00:05:04,000 --> 00:05:06,399
get into the computing stuff, because the community stuff so extensive,

93
00:05:06,560 --> 00:05:09,439
is nineteen seventy nine is the year that we first

94
00:05:09,519 --> 00:05:15,199
found hydro thermal events. So this was the Riviera submersible

95
00:05:15,639 --> 00:05:19,600
experiments off the southern south of Baja California, about eighty

96
00:05:19,600 --> 00:05:22,120
five hundred feet of water with you remember the Alvin

97
00:05:22,240 --> 00:05:27,160
submersible woods whole. So they were looking around for underwater

98
00:05:27,800 --> 00:05:30,439
volcanic activity and came across these things we now know

99
00:05:30,519 --> 00:05:35,079
as black smokers. So these were jets of black material

100
00:05:35,160 --> 00:05:38,319
coming out of the ocean floor and the extremely hot

101
00:05:38,560 --> 00:05:41,319
three hundred eighty degrees celsius like seven hundred degrees fahrenheit

102
00:05:41,959 --> 00:05:46,079
in the deep dark abyssal parts of the ocean. And

103
00:05:46,120 --> 00:05:50,560
they're surrounded by life. There are tube worms and kinds

104
00:05:50,560 --> 00:05:52,879
of crabs and all sorts of things that are feeding

105
00:05:52,879 --> 00:05:55,800
off of the heat and the minerals that are pouring

106
00:05:55,839 --> 00:05:59,680
out of these black smokers. And it was a revolution

107
00:06:00,399 --> 00:06:03,000
in thinking around where life could emerge. It was this

108
00:06:03,079 --> 00:06:05,040
belief that you needed to be a certain distance away

109
00:06:05,040 --> 00:06:06,480
from the sun and have pad with water and all

110
00:06:06,480 --> 00:06:10,240
these things. And here was this completely dark place that

111
00:06:10,319 --> 00:06:13,360
had life emerging around it. And it speaks to the

112
00:06:13,480 --> 00:06:17,120
idea that we talk about these days about perhaps the

113
00:06:17,519 --> 00:06:22,839
Europa in orbit around Jupiter, with enough heat, could possibly

114
00:06:22,839 --> 00:06:25,639
have life under that ocean, underneath the ice cap that

115
00:06:26,000 --> 00:06:28,519
is Europa. So that all begins in seventy nine with

116
00:06:28,560 --> 00:06:30,759
its discovery. Nobody thought they were there. This was a

117
00:06:30,839 --> 00:06:32,600
fine So is the important moment.

118
00:06:32,600 --> 00:06:36,680
Speaker 1: You know what I remember fondly about this period National

119
00:06:36,720 --> 00:06:41,720
Geographic magazine. There was some great photos and especially from

120
00:06:41,800 --> 00:06:46,439
those space missions, and just amazing, amazing stuff. And as

121
00:06:46,439 --> 00:06:49,639
a kid, I was just wrapped by a National Geographic

122
00:06:50,160 --> 00:06:52,959
My mother recently passed and we're cleaning out her attic,

123
00:06:53,600 --> 00:06:57,240
and it turns out there they saved all their National

124
00:06:57,279 --> 00:07:00,199
Geographic magazines going back to the thirties.

125
00:07:01,040 --> 00:07:05,279
Speaker 2: Wow, that's crazy. Yes, So so here's a weird one

126
00:07:05,319 --> 00:07:08,600
about those black smokers. So you'll share this nationally graphic magazine.

127
00:07:08,720 --> 00:07:11,439
Those two birds are brightly colored. They're white and red.

128
00:07:11,519 --> 00:07:13,959
It's like it's dark. Why these things have colors? Yeah,

129
00:07:14,160 --> 00:07:19,319
I did do, all right. Let's talk about computers. Nineteen

130
00:07:19,360 --> 00:07:21,800
seventy nine the release of the Atari model four hundred

131
00:07:21,839 --> 00:07:26,360
and eight hundred, also the TI ninety nine that is

132
00:07:26,399 --> 00:07:29,120
a scientific calculator. No, no, t I nine nine was a

133
00:07:29,160 --> 00:07:32,319
computer made by talking instruments. Yeah no, I mean games

134
00:07:32,319 --> 00:07:35,439
for it back in the day. The Motor sixty eight

135
00:07:35,480 --> 00:07:40,120
thousand processor is released, and a couple of online things.

136
00:07:40,399 --> 00:07:45,079
The origin of CompuServe. Oh yeah, I love the CompuServe.

137
00:07:45,319 --> 00:07:50,000
So originally a company a subsidiary of Golden United Life

138
00:07:50,000 --> 00:07:53,120
Insurance out of Columbus, Ohio, which they actually started in

139
00:07:53,199 --> 00:07:56,759
nineteen sixty nine renting time on PDP tens, but in

140
00:07:56,839 --> 00:08:00,399
seventy nine they added dial up so you could island.

141
00:08:00,480 --> 00:08:03,240
It was acually internal customers primarily, but they started opening

142
00:08:03,319 --> 00:08:06,279
up to consumers and it didn't go particularly well. The

143
00:08:06,279 --> 00:08:08,319
company gets acquired in nineteen eighty by H and R.

144
00:08:08,360 --> 00:08:10,519
Block for twenty five million dollars. They only have they

145
00:08:10,560 --> 00:08:12,759
have less than a thousand users at that time, but

146
00:08:12,879 --> 00:08:15,279
by ninet eighty four it'll be one hundred and ten thousand.

147
00:08:15,360 --> 00:08:16,920
Speaker 1: Wasn't they owned by Sears at one point.

148
00:08:17,360 --> 00:08:19,720
Speaker 2: Yeah, it went through many hands before it end up

149
00:08:20,120 --> 00:08:20,759
at AOL.

150
00:08:20,879 --> 00:08:23,720
Speaker 1: But I just remember the snap packs which had you know,

151
00:08:23,759 --> 00:08:27,800
a code and it had the password explore plus world,

152
00:08:28,079 --> 00:08:28,959
right you remember that?

153
00:08:29,319 --> 00:08:32,039
Speaker 2: Yeah? Yeah, and so much fun. This is also just

154
00:08:32,159 --> 00:08:33,960
beginning of modes in general. So this is when I

155
00:08:34,039 --> 00:08:38,360
first got a motem as well. And for better or worse,

156
00:08:38,440 --> 00:08:40,919
like the Vancouver area had a ton of BBS's, it

157
00:08:40,960 --> 00:08:43,480
was very early on. It just ways for geeks to

158
00:08:43,519 --> 00:08:46,200
connect to each other. It's also the first computer worm

159
00:08:46,320 --> 00:08:50,279
was built by John Shock and John Hopp at Xerox Park,

160
00:08:50,480 --> 00:08:54,000
largely by accident of propagating piece of software across networks.

161
00:08:54,039 --> 00:08:56,960
And so there was your first computer worm. But by far,

162
00:08:57,240 --> 00:09:00,519
without a doubt, the most important thing for computing in

163
00:09:00,559 --> 00:09:04,639
nineteen seventy nine VisiCalc. Van Bricklin, the precursor to all

164
00:09:04,639 --> 00:09:07,840
the spreadsheets. Yeah, and the first real you know they

165
00:09:08,000 --> 00:09:11,559
at one point it was personal computers are described as

166
00:09:11,600 --> 00:09:15,440
an accessory for Visical. Right, this was the point of

167
00:09:15,519 --> 00:09:18,799
the personal computer first released in seventy nine on the

168
00:09:18,840 --> 00:09:21,759
Apple two and IBM sites in their history of the

169
00:09:21,799 --> 00:09:26,799
IBM PC that that product shipping made them accelerate the

170
00:09:26,840 --> 00:09:29,320
development of the IBM PC to get a version of

171
00:09:29,360 --> 00:09:32,200
Hysical Chronic. Now, the ibm PC made sense and they

172
00:09:32,200 --> 00:09:35,720
got it out in nineteen eighty. The story of VISICALCU

173
00:09:35,799 --> 00:09:39,480
is also one of mush drama too, because a former

174
00:09:39,879 --> 00:09:44,440
Visit Corp employee, Michiga Porl leaves and forms the Lotus

175
00:09:44,480 --> 00:09:47,639
Corporation Lotus one two three, create Lotus one two three

176
00:09:47,720 --> 00:09:51,320
and then the Electronic Frontier Foundation. Yeah, that was part

177
00:09:51,360 --> 00:09:52,960
of that, and that's nineteen eighty three, and what was

178
00:09:52,960 --> 00:09:54,519
the claim to fame for Lotus one two to three

179
00:09:54,639 --> 00:09:57,600
totally optimized for the IBM PC, and that of course

180
00:09:57,600 --> 00:10:01,000
will later lead to multi Plan by Microsoft, which will

181
00:10:01,000 --> 00:10:02,759
eventually become Excel. Yeah.

182
00:10:03,159 --> 00:10:05,440
Speaker 1: I remember my father bought a Tiers eight Model four

183
00:10:05,679 --> 00:10:08,480
and that's kind of my first foray into computers, and

184
00:10:08,639 --> 00:10:09,879
he bought it for Visical.

185
00:10:10,039 --> 00:10:11,919
Speaker 2: YEP. Visical was villefort because.

186
00:10:11,720 --> 00:10:14,720
Speaker 1: He did his taxes and bills and everything on Visical.

187
00:10:14,759 --> 00:10:16,759
Can I remember once a month and getting that thing

188
00:10:16,799 --> 00:10:19,200
out and putting it on the dining room table and

189
00:10:19,240 --> 00:10:20,919
we didn't see him for a few hours.

190
00:10:21,320 --> 00:10:26,279
Speaker 2: You know this is really up until now, personal computers

191
00:10:26,279 --> 00:10:28,840
have been toys. Yeah right, this is the product that

192
00:10:28,919 --> 00:10:33,879
made it actually a business product and important and changed everything. Yeah.

193
00:10:34,080 --> 00:10:36,559
Speaker 1: Still a critical tool in the arsenal today.

194
00:10:36,720 --> 00:10:40,360
Speaker 2: Oh yeah, spreadsheets, spreadsheets. Whole companies built around spreadsheets without

195
00:10:40,360 --> 00:10:40,639
a doubt.

196
00:10:40,759 --> 00:10:42,799
Speaker 1: You have a great joke, don't you, Richard? How about

197
00:10:42,840 --> 00:10:44,240
you and your wife in spreadsheet?

198
00:10:44,320 --> 00:10:47,440
Speaker 2: Oh but yeah, she's the industrial engineer. I'm a programmer.

199
00:10:47,440 --> 00:10:50,159
When we argue, it involves a spreadsheet. Yeah, and it's

200
00:10:50,200 --> 00:10:52,360
only because that happened, right. We were trying to we

201
00:10:53,200 --> 00:10:56,320
were arguing about how to redo a deck, and in

202
00:10:56,399 --> 00:10:58,600
the end resolved the argument with a spreadsheet. I'm like,

203
00:10:58,600 --> 00:11:01,600
this is who we are. It's great, this is reality

204
00:11:01,639 --> 00:11:04,600
for us. Awesome? So is that it? That's it? All right?

205
00:11:05,279 --> 00:11:07,320
Speaker 1: Well with that, let's roll the crazy music for better

206
00:11:07,360 --> 00:11:07,919
no framework?

207
00:11:15,720 --> 00:11:16,320
Speaker 2: Man, what do you got?

208
00:11:16,360 --> 00:11:16,480
Speaker 3: Well?

209
00:11:16,480 --> 00:11:19,639
Speaker 1: I realized when we were recording last week's show that

210
00:11:19,720 --> 00:11:22,080
this might have been a better, better no framework for that.

211
00:11:22,480 --> 00:11:30,159
But tech Nitium Software has a DNS server written in

212
00:11:30,360 --> 00:11:35,799
C sharp that's open source and their tagline is self

213
00:11:35,840 --> 00:11:40,240
host a DNS server for privacy and security, block ads

214
00:11:40,279 --> 00:11:43,679
in malware at the DNS level for your entire network,

215
00:11:44,679 --> 00:11:47,080
and technitium I don't know how you say it, but

216
00:11:47,759 --> 00:11:54,200
tech nit i um technitium probably technician who knows. DNS

217
00:11:54,240 --> 00:11:57,200
server is an open source, authoritative as well as recursive

218
00:11:57,279 --> 00:11:59,559
DNS server that can be used for self hosting a

219
00:11:59,639 --> 00:12:02,519
DNS server for privacy and security. It works out of

220
00:12:02,519 --> 00:12:05,480
the box with no or minimal configuration and provides a

221
00:12:05,559 --> 00:12:10,200
user friendly web console accessible using any modern web browser.

222
00:12:10,480 --> 00:12:11,399
Cool how about that?

223
00:12:11,639 --> 00:12:14,480
Speaker 2: Yeah? Yeah, interesting? And we always argument of like what

224
00:12:14,600 --> 00:12:17,519
DNA server should you run, because you know, there's plenty

225
00:12:17,519 --> 00:12:19,320
of free ones out there, but this is a free

226
00:12:19,320 --> 00:12:21,679
one too, guys. It's just open source, so good stuff.

227
00:12:21,720 --> 00:12:24,159
Speaker 1: But if you think about it, like your basic Windows

228
00:12:24,240 --> 00:12:28,960
land has everything except the DNS server, and back in

229
00:12:29,000 --> 00:12:31,360
the day we were editing host files and putting IP

230
00:12:31,480 --> 00:12:34,600
addresses and names in there, and then you had some

231
00:12:34,720 --> 00:12:38,879
other protocols on top of that that worked with anything

232
00:12:38,919 --> 00:12:42,159
but TCP IP, right, So you know, this is kind

233
00:12:42,200 --> 00:12:44,159
of a it's kind of an important thing. If you

234
00:12:44,200 --> 00:12:48,039
don't want to go outside of your network for DNS. Yeah,

235
00:12:48,080 --> 00:12:51,320
there you go. If you're gonna know if you need it. Yeah,

236
00:12:51,440 --> 00:12:52,720
we're not going to try to sell.

237
00:12:52,559 --> 00:12:55,840
Speaker 2: You on No. And without a doubt. It's like, yeah,

238
00:12:55,840 --> 00:12:58,840
it's one of those things. Just learned to configure it correctly.

239
00:12:58,879 --> 00:13:01,759
It's important, right, you know that whole running joke of

240
00:13:02,240 --> 00:13:08,039
it could be DNS, it's always DNS. Definitely, definitely DNS.

241
00:13:08,159 --> 00:13:11,120
It's impossible if it's DNS. It could be DNS. It's

242
00:13:11,240 --> 00:13:14,559
dns NS. Okay, Well that's what I got today. Richard,

243
00:13:14,559 --> 00:13:17,919
who's talking to us awesome, grabbed a comment off show

244
00:13:18,039 --> 00:13:21,039
nineteen seventy four, the one we did with Don Delamarski

245
00:13:21,080 --> 00:13:24,159
when we talked about Gethub spec kit. Yeah, and they've

246
00:13:24,159 --> 00:13:26,639
said a huge conversation, lots going down there. And I've

247
00:13:26,639 --> 00:13:28,360
read comments from this show before, and I'm gonna read

248
00:13:28,399 --> 00:13:30,200
oother one. This one's from Richard Cox who said, I

249
00:13:30,279 --> 00:13:32,320
just finished listening to this. I think there's one part

250
00:13:32,360 --> 00:13:35,519
that your guests got wrong. If m inference stopped, work

251
00:13:35,600 --> 00:13:38,759
stops thanks to the bubble bursting. Talking about the AI bubble,

252
00:13:39,000 --> 00:13:42,120
the models will start to degrade as new input like

253
00:13:42,200 --> 00:13:45,000
code using new versions of tools will not be included.

254
00:13:45,039 --> 00:13:48,639
Then the output will not use those new capabilities. IE.

255
00:13:48,840 --> 00:13:50,960
Over time, as the rest of the world moves forwards,

256
00:13:51,039 --> 00:13:54,080
the lms will increasingly be stuck in the past. Yeah,

257
00:13:54,720 --> 00:13:57,440
I mean, I get the sentiment that, you know, when

258
00:13:57,440 --> 00:13:59,440
the bubble bursts, a certain amount of work's going to

259
00:13:59,440 --> 00:14:03,000
go away. Would argue there's too many models. Now, I'm

260
00:14:03,000 --> 00:14:05,000
sure there'll be a few models, so they'll continue to

261
00:14:05,039 --> 00:14:09,399
go forward. But we're seeing such incredible overspending at the moment, right,

262
00:14:10,000 --> 00:14:13,679
It'll be interesting to see what happens. And modelbility is

263
00:14:13,679 --> 00:14:17,559
getting easier, and I would also argue less important, like

264
00:14:17,600 --> 00:14:19,440
in that sense that it's just going to be part

265
00:14:19,519 --> 00:14:23,840
of the flow. It'll be interesting to see what models

266
00:14:23,960 --> 00:14:28,039
emerge post bubble burst, right, and boy, there's lots of

267
00:14:28,039 --> 00:14:30,320
noises about the bubble versus these days. It's be interesting

268
00:14:30,360 --> 00:14:33,960
to see what happens. Oh, I know it, but fair point,

269
00:14:34,600 --> 00:14:36,759
it's all we're you know. The interesting thing about this

270
00:14:36,879 --> 00:14:39,080
is this software in some ways does degrade because the

271
00:14:39,080 --> 00:14:41,399
world keeps moving on and these things need to keep

272
00:14:41,399 --> 00:14:44,720
being regenerated and optimized. So Richard, thank you so much

273
00:14:44,720 --> 00:14:46,200
for your comment and a copy of music co Buy is

274
00:14:46,240 --> 00:14:47,360
on its way to you, and if you'd like a

275
00:14:47,360 --> 00:14:48,799
copy of music, go buy. I write a comment on

276
00:14:48,799 --> 00:14:51,480
the website at dot netrogs dot com or on the facebooks.

277
00:14:51,480 --> 00:14:52,960
We publish every show there and if you comment there

278
00:14:52,960 --> 00:14:54,559
and ever reading the show, we'll send your copy us

279
00:14:54,639 --> 00:14:55,000
to go buy.

280
00:14:55,159 --> 00:14:57,279
Speaker 1: And of course needs to co by developed a while

281
00:14:57,360 --> 00:15:01,000
ago to provide music that's neither too boring or to

282
00:15:01,159 --> 00:15:06,360
distracting twenty five minutes track and there's now twenty three

283
00:15:06,480 --> 00:15:09,120
tracks and you can get it at Musicdoco by dot

284
00:15:09,200 --> 00:15:13,919
net and the entire collection in wave, flak and MP

285
00:15:14,000 --> 00:15:20,519
three formats. Okay, let's introduce Callum Simpson and he is

286
00:15:20,559 --> 00:15:24,679
a solution architect with fourteen years of software development experience

287
00:15:24,720 --> 00:15:30,600
as of this recording, and recently promoted VP of AI SSW,

288
00:15:31,320 --> 00:15:35,480
an enterprise consultancy based in Australia. Our friend Adam Cogan

289
00:15:36,039 --> 00:15:39,480
runs that he spends most of his time using AI

290
00:15:39,600 --> 00:15:43,120
to help deliver projects faster or building projects that use AI,

291
00:15:43,240 --> 00:15:46,360
and usually both. Despite being more productive than ever, he

292
00:15:46,399 --> 00:15:48,639
claims to have not written a complete line of code

293
00:15:48,679 --> 00:15:53,720
in over a year. He's also a product owner of

294
00:15:53,840 --> 00:15:59,480
SSW yak Shaver, an AI product. He'll tell us about today.

295
00:16:00,559 --> 00:16:03,320
Speaker 3: Callum, thank you very much, Carl, and great to be here.

296
00:16:03,440 --> 00:16:05,879
Speaker 2: Hey Richard, hey Man, great to have you. I was

297
00:16:05,919 --> 00:16:09,120
just down at SSW was part of their brainstorming day,

298
00:16:09,240 --> 00:16:10,600
so we had a chance to hang out.

299
00:16:10,679 --> 00:16:11,879
Speaker 3: Okay, what did you think about it?

300
00:16:11,960 --> 00:16:14,360
Speaker 2: Yeah, that's great. You know, it's not my first brainstorm.

301
00:16:14,399 --> 00:16:17,679
It's always fun. What I think you saw my clothes

302
00:16:17,720 --> 00:16:21,120
where I commented on you know, SSW is clearly all

303
00:16:21,120 --> 00:16:23,519
in on the AI space, and more than anything, what

304
00:16:23,639 --> 00:16:25,840
I saw was a bunch of different teams trying to

305
00:16:25,840 --> 00:16:29,480
find a way to rain the LMS in, to put

306
00:16:29,519 --> 00:16:33,360
parameters around them so that they focus on the things

307
00:16:33,360 --> 00:16:37,600
that are important productivity wise, whether that's around controlling architecture,

308
00:16:37,720 --> 00:16:40,679
deployment strategies you talked about. It was a group that

309
00:16:40,679 --> 00:16:43,559
we're talking about, I said, a UX frameworks that also

310
00:16:43,759 --> 00:16:46,240
the LMS would be pressed against. It's like this is

311
00:16:46,279 --> 00:16:49,759
how you build UI when you write, when you're building code,

312
00:16:49,759 --> 00:16:52,279
and I just thought it was really clear thinking from

313
00:16:52,360 --> 00:16:54,799
a group of really smart developers trying to get the

314
00:16:54,840 --> 00:16:56,000
most value from these tools.

315
00:16:56,279 --> 00:16:59,120
Speaker 3: That's awesome, that's right. Yeah, we're trying to use it.

316
00:16:59,399 --> 00:17:02,679
I guess in every way we possibly can to get

317
00:17:02,720 --> 00:17:05,440
as much value out of it. So anything that can

318
00:17:05,480 --> 00:17:07,880
be sold by AI, we are trying to do it.

319
00:17:08,000 --> 00:17:08,240
Speaker 2: Cool.

320
00:17:08,319 --> 00:17:13,359
Speaker 1: I am so curious about yak shaver. First of all,

321
00:17:13,400 --> 00:17:16,519
what a funny name. Wow, And I think.

322
00:17:16,400 --> 00:17:18,599
Speaker 2: Adam Cogan was involved. What do you think was going

323
00:17:18,680 --> 00:17:19,079
to happen?

324
00:17:19,240 --> 00:17:24,039
Speaker 1: I'm sure, yeah, But but there's probably just a very

325
00:17:24,079 --> 00:17:26,799
small handful of people in this world who have ever

326
00:17:26,920 --> 00:17:28,400
attempted to shave a yak.

327
00:17:28,680 --> 00:17:30,920
Speaker 2: Well, first you have to own a yak, don't you.

328
00:17:31,200 --> 00:17:33,599
Speaker 1: Well, you know, necessarily you could be a yak shaver

329
00:17:33,839 --> 00:17:36,559
professional and go around from yak to yak to yak,

330
00:17:37,559 --> 00:17:42,079
you know. I mean they have obviously yaq milkers who

331
00:17:42,119 --> 00:17:45,480
make butter from yak, you know, cream or whatever.

332
00:17:46,319 --> 00:17:48,680
Speaker 2: I've had yak milk in my coffee, in my tea.

333
00:17:48,759 --> 00:17:50,920
You know, it's the thing when you're in Nepal.

334
00:17:51,720 --> 00:17:53,920
Speaker 3: I was gonna say it's potentially an alternative name, but

335
00:17:54,119 --> 00:17:58,039
the name itself came from I think there was a

336
00:17:58,079 --> 00:18:01,759
guy called Colin Vieri in the nineties, was a PhD

337
00:18:01,880 --> 00:18:08,240
student at MIT, and he got the name from an

338
00:18:08,279 --> 00:18:13,000
episode of The Ren and Stimpy Show. In that show,

339
00:18:13,559 --> 00:18:17,279
there was a yak shaving day, and basically yac shaving

340
00:18:17,359 --> 00:18:20,440
Day was sort of like this thing where where these

341
00:18:20,480 --> 00:18:24,640
people do this series of ridiculous tasks. And so the

342
00:18:24,680 --> 00:18:28,200
concept of yak shaving is sort of like when you

343
00:18:28,359 --> 00:18:32,440
start doing your main mission and then you realize, all, right,

344
00:18:32,440 --> 00:18:34,960
in order to achieve this mission, I need to go

345
00:18:35,039 --> 00:18:37,279
down a side quest, and then in order to do

346
00:18:37,319 --> 00:18:39,400
this side quest, I need to go and do another

347
00:18:39,440 --> 00:18:41,759
side quest first before I can come back and finish

348
00:18:41,799 --> 00:18:44,240
my first side quest to finish the main goal. And

349
00:18:44,279 --> 00:18:47,519
then you end up going down this like ten ten

350
00:18:47,920 --> 00:18:52,039
layers of causality, and whatever you're doing has absolutely nothing

351
00:18:52,680 --> 00:18:55,200
apparently to do with the original goal, but you have

352
00:18:55,240 --> 00:18:57,599
to do it unblock all your other things to get

353
00:18:57,640 --> 00:18:59,039
back up to the original task.

354
00:18:59,240 --> 00:19:01,839
Speaker 1: And then every new layer you say to yourself, should

355
00:19:01,880 --> 00:19:04,000
I really be doing this exactly?

356
00:19:04,079 --> 00:19:05,640
Speaker 3: What am I doing with my wife?

357
00:19:09,720 --> 00:19:11,160
Speaker 2: Better use is of my time?

358
00:19:12,000 --> 00:19:14,640
Speaker 3: Exactly? And so that's sort of what we're what we're

359
00:19:14,640 --> 00:19:18,559
trying to do with with Yakshava is cut out as

360
00:19:18,680 --> 00:19:22,640
much of that sort of busy work as possible. Now,

361
00:19:22,640 --> 00:19:27,480
obviously we are software developers, so most of the busy

362
00:19:27,519 --> 00:19:31,119
work we do has to do with or at least

363
00:19:31,160 --> 00:19:32,799
a large part of it has to do with putting

364
00:19:32,839 --> 00:19:36,799
items in backlogs. So that was the sort of original concept.

365
00:19:36,920 --> 00:19:41,359
Is you know, when you've got an issue, when you

366
00:19:41,720 --> 00:19:44,599
see a bug on a website, what do you have

367
00:19:44,680 --> 00:19:48,920
to do to actually report that bug to the right team?

368
00:19:49,079 --> 00:19:51,559
You know, you need to figure out which backlog does

369
00:19:51,599 --> 00:19:56,880
this PBI belonging, who are the stakeholders of this project,

370
00:19:57,359 --> 00:19:59,720
all that sort of stuff, And if you're on a

371
00:19:59,720 --> 00:20:03,640
call with a bunch of important people, you can either

372
00:20:03,799 --> 00:20:05,920
sort of skip over the problem because you know you've

373
00:20:05,960 --> 00:20:08,880
got better things to talk about, or you can tell everyone,

374
00:20:08,920 --> 00:20:11,119
all right, wait for five minutes, I need to go

375
00:20:11,599 --> 00:20:14,000
figure out all these details, put in the right backlog

376
00:20:14,599 --> 00:20:17,480
all that stuff first. So it is a tough problem.

377
00:20:17,960 --> 00:20:20,519
And obviously we don't want to skip over issues when

378
00:20:20,519 --> 00:20:22,039
we see them, but we also don't want to waste

379
00:20:22,079 --> 00:20:25,240
everyone's time. So that's that's sort of the whole idea

380
00:20:25,279 --> 00:20:28,839
of what we're trying to achieve or the problem we're

381
00:20:28,839 --> 00:20:29,440
trying to solve.

382
00:20:29,559 --> 00:20:33,559
Speaker 1: So yak shaver does it? Is it sort of an

383
00:20:33,599 --> 00:20:36,519
agent kind of thing where you give it permission to

384
00:20:36,559 --> 00:20:41,880
do stuff, you know, like an MCP word and why

385
00:20:41,880 --> 00:20:43,759
wouldn't you just use an MCP.

386
00:20:44,200 --> 00:20:50,240
Speaker 3: Yeah, good question. So when Yakshava first was conceptualized, it

387
00:20:50,359 --> 00:20:54,000
was sort of like, I think it's almost three years

388
00:20:54,039 --> 00:20:58,440
ago now, so back then we didn't have mcps. We

389
00:20:58,480 --> 00:21:01,599
had only just got you know, of custom GPTs that

390
00:21:01,680 --> 00:21:05,400
can do tool calls and stuff like that. But yeah,

391
00:21:05,440 --> 00:21:08,640
the idea of an MCP was still very far off.

392
00:21:08,960 --> 00:21:14,039
So back then the idea was instead of just you know,

393
00:21:14,240 --> 00:21:17,559
letting the agent do whatever it wants because obviously models

394
00:21:17,640 --> 00:21:22,079
weren't as reliable, then we would force it through a

395
00:21:22,160 --> 00:21:25,440
pipeline where there's a couple of branching points where it

396
00:21:25,519 --> 00:21:28,599
can either do this or do that. Is, are you

397
00:21:28,720 --> 00:21:31,240
trying to report a PBI or send an email for example,

398
00:21:31,240 --> 00:21:34,640
as one of those branches, and we make each.

399
00:21:34,440 --> 00:21:37,000
Speaker 1: Of the all right, So it's not you're not giving

400
00:21:37,039 --> 00:21:39,799
a total agency, You're you're guiding it.

401
00:21:39,880 --> 00:21:43,079
Speaker 3: As you say, exactly, That's that's sort of the the

402
00:21:43,160 --> 00:21:49,359
Yakshav V one you have mentioned MCP. We are currently

403
00:21:49,519 --> 00:21:55,160
developing a V two that will indeed use MCP servers.

404
00:21:55,279 --> 00:21:57,960
Speaker 1: Okay, video context, what's that?

405
00:21:58,279 --> 00:21:59,039
Speaker 2: Yeah, exactly.

406
00:21:59,160 --> 00:22:04,319
Speaker 3: So the the main I think distinguishing factor between Yakshava

407
00:22:04,559 --> 00:22:09,039
and maybe firing up a you know, a MCP host

408
00:22:09,160 --> 00:22:12,079
like clawed Desktop and just saying hey, go create an

409
00:22:12,119 --> 00:22:14,960
item in the backlog is with Yakshava, we use a

410
00:22:15,079 --> 00:22:18,880
video as the input. So what you'll do is, you know,

411
00:22:18,920 --> 00:22:22,200
you'll share your screen and then you'll speak into your

412
00:22:22,200 --> 00:22:24,680
microphone and you'll say, hey, I'm just on this website.

413
00:22:25,920 --> 00:22:29,119
There's a URL, and here's the problem that I've got

414
00:22:29,319 --> 00:22:33,640
on the website, and then stop recording. And then basically

415
00:22:33,680 --> 00:22:38,440
the AI will obviously analyze the transcript, will analyze what

416
00:22:38,480 --> 00:22:41,839
it can from the screen that you shared, and then

417
00:22:42,039 --> 00:22:44,680
interesting go ahead and you figure out where to put

418
00:22:44,720 --> 00:22:46,640
it and how to format it all that stuff.

419
00:22:46,720 --> 00:22:49,480
Speaker 1: So it's a little more powerful than something like play rate,

420
00:22:49,599 --> 00:22:53,119
which can go and navigate a site and all that stuff.

421
00:22:53,160 --> 00:22:55,799
But what if you're not using a website, What if

422
00:22:55,799 --> 00:22:59,119
you're using a piece of software. Right, yeah, exactly, use

423
00:22:59,119 --> 00:23:02,039
a video screen. I'm sure it's brilliant.

424
00:23:01,640 --> 00:23:04,640
Speaker 3: That's right, And and some creative people have used it,

425
00:23:04,680 --> 00:23:10,400
even completely unrelated to software, which was an unexpected use case.

426
00:23:11,079 --> 00:23:15,559
You know, people like the building maintenance team when they find,

427
00:23:15,799 --> 00:23:19,440
you know, like an issue with the coffee machine, for example. Yeah,

428
00:23:21,000 --> 00:23:23,880
record a video of it and it gets filed off

429
00:23:23,920 --> 00:23:27,680
in the appropriate Obviously we're using back clothes for the

430
00:23:27,759 --> 00:23:29,599
office maintenance because Adam's the boss.

431
00:23:32,480 --> 00:23:34,119
Speaker 2: But in the same way that people use GitHub for

432
00:23:34,160 --> 00:23:37,599
recipes and things like, it's useful to have a coherent

433
00:23:37,680 --> 00:23:39,599
documentation chain for any of these things.

434
00:23:39,680 --> 00:23:42,720
Speaker 1: Well, chat GPT is good for that, and I find

435
00:23:42,759 --> 00:23:46,759
that the chat GPT is more consumer oriented that way,

436
00:23:46,920 --> 00:23:49,000
Like I can take a video when I could say, hey,

437
00:23:49,000 --> 00:23:52,039
what is what is this thing? You know, here's the

438
00:23:52,400 --> 00:23:54,960
here's a couple of pictures I snapped at my laptop,

439
00:23:55,000 --> 00:23:57,599
and it figures out what it is from that, here's

440
00:23:57,640 --> 00:23:59,759
the problem I'm having. I take a screen, you know,

441
00:24:00,200 --> 00:24:03,079
video of the screen, and it you know, can diagnose

442
00:24:03,119 --> 00:24:06,240
problems that way. So, but what I don't like about

443
00:24:06,319 --> 00:24:09,720
chat GPT is it doesn't understand the context of code

444
00:24:10,079 --> 00:24:15,039
right for example, and I don't want it to right Yeah.

445
00:24:14,759 --> 00:24:15,240
Speaker 2: Yeah, yeah.

446
00:24:15,319 --> 00:24:17,920
Speaker 3: So one of the I guess main things that we're

447
00:24:17,960 --> 00:24:21,240
doing with Yakshav, apart from the video input, is also

448
00:24:21,599 --> 00:24:26,720
the sort of organizational consistency. So what you'll do with

449
00:24:26,839 --> 00:24:29,759
Yakshav is set up you know, these are all my projects,

450
00:24:29,839 --> 00:24:33,960
These are all my people and who's associated with which project,

451
00:24:34,319 --> 00:24:36,240
and these are the formats that I like to have.

452
00:24:36,319 --> 00:24:40,359
My pbis created in all that sort of stuff, you'll

453
00:24:40,400 --> 00:24:42,640
sort of define it and that way everyone who uses

454
00:24:42,680 --> 00:24:47,160
it ends up with a consistent result, because obviously if

455
00:24:47,400 --> 00:24:50,279
everyone just used chat GPT and recorded a video of

456
00:24:51,119 --> 00:24:54,279
whatever with none of that consistency, then you're going to

457
00:24:54,359 --> 00:24:57,079
end up with a completely different thing.

458
00:24:57,160 --> 00:25:01,079
Speaker 2: Every time. The world is full of abandoned video that's

459
00:25:01,200 --> 00:25:03,359
just sort of a normal thing, right, Like this is

460
00:25:03,680 --> 00:25:05,839
I think the important part in all of these things.

461
00:25:05,839 --> 00:25:07,799
And I'm not going to point out they actuap per se,

462
00:25:07,920 --> 00:25:10,079
but it's like what do you do with it after?

463
00:25:10,319 --> 00:25:12,839
Like where does it go? How does anybody ever look

464
00:25:12,880 --> 00:25:16,799
at it? Although you've always got the quality problem right,

465
00:25:16,839 --> 00:25:19,680
like we've been you see this recurring theme with AI

466
00:25:19,799 --> 00:25:24,759
generated text in general. People are writing their corporate emails

467
00:25:24,839 --> 00:25:29,359
using these tools and they're over long and over complicated,

468
00:25:29,359 --> 00:25:32,640
and some ways you're pushing the problem down the line,

469
00:25:32,720 --> 00:25:36,200
like right, yeah, well, I guess this is the challenge

470
00:25:36,240 --> 00:25:38,440
with anything related to these AI tools is like how

471
00:25:38,440 --> 00:25:40,160
do you make sure the thing you're making is concise?

472
00:25:40,200 --> 00:25:42,839
That the next person down the line isn't being dumped

473
00:25:42,839 --> 00:25:44,200
with a lot of unnecessary work.

474
00:25:44,279 --> 00:25:46,599
Speaker 3: Yeah, well, I'm glad you mentioned that actually, because one

475
00:25:46,680 --> 00:25:49,799
of the other really cool things about yakshaber is not

476
00:25:49,880 --> 00:25:53,920
only does it generate the PBI, but it also puts

477
00:25:54,240 --> 00:25:57,799
the link to the video that you recorded on the PBI.

478
00:25:58,920 --> 00:26:01,599
So what it means is that the developer who picks

479
00:26:01,680 --> 00:26:05,440
up that issue and works on it, they, if they want,

480
00:26:05,480 --> 00:26:09,759
can just ignore all the AI generated PBI text and

481
00:26:09,799 --> 00:26:12,559
just watch the video of the user explaining the problem

482
00:26:13,000 --> 00:26:16,039
and what do you mean when you see PBI backlog items?

483
00:26:16,079 --> 00:26:19,880
So that's basically the of the bug. So if we've

484
00:26:19,920 --> 00:26:23,440
got a bug on a website, we've recorded a video

485
00:26:23,559 --> 00:26:27,720
showing that bug, and then the developer who picks up

486
00:26:27,799 --> 00:26:31,759
that that issue to fix can watch the video.

487
00:26:32,279 --> 00:26:37,079
Speaker 1: That's very Adam Cogan. He used to and maybe he

488
00:26:37,119 --> 00:26:41,599
still does. But when we had issues with our website,

489
00:26:41,599 --> 00:26:44,039
we would get emails from Adam that had screenshots that

490
00:26:44,079 --> 00:26:48,119
were annotated with you know, things circled and whatever, and

491
00:26:48,200 --> 00:26:50,720
you know this should be that, and that should be this,

492
00:26:50,880 --> 00:26:54,000
and yeah he's always been that way.

493
00:26:54,279 --> 00:26:58,160
Speaker 3: Yeah, yeah, yeah, But often there's heaps of details in

494
00:26:58,200 --> 00:27:02,079
the video that has an extra being transcribed in the text, right,

495
00:27:02,160 --> 00:27:05,039
and just being able to watch the video is it

496
00:27:05,079 --> 00:27:06,960
makes it so much easier because you don't have to

497
00:27:06,960 --> 00:27:09,160
go back to the original reporter and say, hey, what

498
00:27:09,200 --> 00:27:11,559
are your reproduction steps? So you know all that sort

499
00:27:11,599 --> 00:27:12,880
of stuff. You can just watch the video.

500
00:27:12,960 --> 00:27:15,240
Speaker 2: Yeah, anytime you can avoid having to go struggle with

501
00:27:15,279 --> 00:27:19,480
a reproduction is good. It's all about capturing that. But

502
00:27:19,519 --> 00:27:22,160
it's interesting to think about this from a workflow perspective

503
00:27:22,160 --> 00:27:24,519
of what's the next thing We're trying to avoid shave

504
00:27:24,559 --> 00:27:29,039
in the act here? Right? So are these just distraction

505
00:27:29,160 --> 00:27:31,640
items from a main thing? Because we also talked about

506
00:27:31,640 --> 00:27:33,240
the fact that there's many little things you need to

507
00:27:33,279 --> 00:27:35,319
do before you can get on with the main thing,

508
00:27:35,599 --> 00:27:38,599
right right, So I guess it's some of this just

509
00:27:38,599 --> 00:27:41,039
getting harnessing more people into the workflow so when they

510
00:27:41,039 --> 00:27:42,599
get those things done, you can move forward.

511
00:27:42,680 --> 00:27:46,319
Speaker 1: I'm curious about video transcription. I know that there are

512
00:27:46,359 --> 00:27:48,680
tools out there that do it, but they're kind of

513
00:27:48,920 --> 00:27:53,240
the transcription is usually integrated into those things, right, Like

514
00:27:53,680 --> 00:27:56,920
there's we use camtesa a little bit here, and I

515
00:27:57,000 --> 00:27:58,200
have maybe an older version.

516
00:27:58,319 --> 00:27:58,599
Speaker 2: I don't know.

517
00:27:58,640 --> 00:28:02,880
Speaker 1: If the newer versions do trans but transcription rather, but

518
00:28:03,720 --> 00:28:05,960
that is a huge thing for me, so if I

519
00:28:06,039 --> 00:28:11,440
have meetings, I know that Zoom can generate transcriptions, but

520
00:28:12,200 --> 00:28:15,680
just it would be really easy to to have a

521
00:28:15,720 --> 00:28:18,359
desktop application though. You could just drop a video into

522
00:28:18,400 --> 00:28:22,119
it and it could transcribe the video. Yeah, for sure,

523
00:28:22,599 --> 00:28:24,519
it seems like a simple thing. But then I could

524
00:28:24,519 --> 00:28:26,160
look it up. I could look up what we're we

525
00:28:26,240 --> 00:28:26,759
talking about.

526
00:28:26,880 --> 00:28:30,960
Speaker 3: You know that's true, And I mean that sounds like

527
00:28:31,000 --> 00:28:34,279
an interesting future feature that weak we can backlog. Maybe

528
00:28:34,440 --> 00:28:40,480
we could actuate it right now and you your details

529
00:28:40,480 --> 00:28:42,640
about it will form the video, and then the developer

530
00:28:42,640 --> 00:28:44,480
who picks it up, we'll have all the context of

531
00:28:44,480 --> 00:28:45,400
why you wanted it, so that.

532
00:28:45,559 --> 00:28:47,559
Speaker 2: Yeah, I like it. All right, take a break?

533
00:28:47,640 --> 00:28:49,200
Speaker 1: Yeah, all right, Well this seems like a good place

534
00:28:49,240 --> 00:28:51,039
to take a break, so we'll be right back after

535
00:28:51,079 --> 00:28:56,279
these very important messages. Did you know you can easily

536
00:28:56,319 --> 00:29:00,559
migrate asp net web apps to Windows containers on Aws.

537
00:29:01,240 --> 00:29:05,279
Use the app to Container tool to containerize your iis

538
00:29:05,279 --> 00:29:10,119
websites and deploy to AWS managed container services with or

539
00:29:10,160 --> 00:29:14,200
without Kubernetes. Find out more about app to Container at

540
00:29:14,240 --> 00:29:19,240
aws dot Amazon dot com. Slash dot Net, slash Modernize,

541
00:29:23,240 --> 00:29:25,599
and we're back at starting at Rocks. I'm Carl Franklin.

542
00:29:25,599 --> 00:29:28,119
It's my friend Richard Campbell. Hey, and this is Callum

543
00:29:28,160 --> 00:29:32,319
Simpson from SSW and he's down in Australia with Richard.

544
00:29:32,480 --> 00:29:35,799
Right now, we're talking about Yakshaver and is this going

545
00:29:35,880 --> 00:29:38,119
to be an open source product? I know it's a

546
00:29:38,160 --> 00:29:39,240
desktop application.

547
00:29:39,519 --> 00:29:42,559
Speaker 3: Yeah, So I mentioned earlier we've got sort of like

548
00:29:42,799 --> 00:29:46,920
a V one Yakshaver and a V two Yakshaver. So

549
00:29:47,000 --> 00:29:50,319
the V one Yakshaver as it currently is is not

550
00:29:50,640 --> 00:29:56,599
open source. It's basically a cloud based pipeline subscriber before

551
00:29:56,640 --> 00:29:59,400
where we sort of force everything through a pipeline with

552
00:29:59,480 --> 00:30:03,519
a few brand options. But what we really wanted to

553
00:30:03,559 --> 00:30:09,559
do with V two is to make it open source

554
00:30:09,920 --> 00:30:13,839
desktop application. The reason why we wanted to make it

555
00:30:13,880 --> 00:30:16,480
open source is because it's a desktop application that we

556
00:30:16,640 --> 00:30:23,319
expect users to install on their machine. So obviously it's

557
00:30:23,400 --> 00:30:27,359
not doing anything crazy. All it's doing is orchestrating MTP

558
00:30:27,519 --> 00:30:30,680
servers from the local machine. So why not make an

559
00:30:30,720 --> 00:30:34,079
open source so anyone can open the lead and see

560
00:30:34,079 --> 00:30:34,920
what's happening inside.

561
00:30:35,200 --> 00:30:36,200
Speaker 1: It's really, really cool.

562
00:30:36,440 --> 00:30:39,160
Speaker 2: Now what are the key AI parts here? Is it

563
00:30:39,279 --> 00:30:42,079
just a transcriber? Like, what do you use an LM

564
00:30:42,240 --> 00:30:44,960
for their other generative ail? What's going on inside this out?

565
00:30:45,559 --> 00:30:47,359
Speaker 3: Are we talking about the V two up with the

566
00:30:47,480 --> 00:30:48,319
MCP service?

567
00:30:48,519 --> 00:30:50,880
Speaker 2: Yeah? Yeah, So basically.

568
00:30:52,279 --> 00:30:55,960
Speaker 3: You'll record your video well, as you said, transcribe the

569
00:30:56,440 --> 00:31:01,039
text of the audio of that video, and then we

570
00:31:01,240 --> 00:31:05,359
throw it over to the MCP orchestrator. And basically what

571
00:31:05,559 --> 00:31:10,119
that is doing is saying, all right, here's the transcript

572
00:31:10,319 --> 00:31:13,599
that the user has submitted. Here's the system prompt. So

573
00:31:14,079 --> 00:31:16,720
inside yak Shava you can sort of define how you

574
00:31:16,799 --> 00:31:19,920
want it to work. We leave that sort of up

575
00:31:20,000 --> 00:31:22,880
to the user, but you would have I guess an

576
00:31:23,160 --> 00:31:27,839
organizational default that you can then customize if you so wish,

577
00:31:28,720 --> 00:31:32,440
and then Yakshava will basically take that and then start

578
00:31:32,599 --> 00:31:35,680
using the MCP servers that you made available to it

579
00:31:36,640 --> 00:31:39,559
to get through whatever it is you're trying to do

580
00:31:39,680 --> 00:31:43,000
with that transcript. So, for example, you could give yak

581
00:31:43,079 --> 00:31:47,640
Shava the GitHub MCP server and you could give it

582
00:31:47,759 --> 00:31:54,000
instructions to say, you know, figure out which backlog or

583
00:31:54,119 --> 00:31:57,960
project I'm talking about in my transcript, put this issue

584
00:31:58,359 --> 00:32:02,200
in that backlog. Surely you include the link to the

585
00:32:02,240 --> 00:32:08,039
YouTube video in that issue and use whatever whatever PBI

586
00:32:08,440 --> 00:32:11,400
template is available in that repo as well. Uh, and

587
00:32:12,319 --> 00:32:14,640
when when approaches it is it will then use you know,

588
00:32:14,680 --> 00:32:18,400
the GitHub MCP to search all the projects you've got available,

589
00:32:18,680 --> 00:32:21,039
figure out all right based on that transcription, I think

590
00:32:21,079 --> 00:32:24,680
he's talking about that one, and then you know, search

591
00:32:24,960 --> 00:32:30,000
the that repo for the templates, find the right template,

592
00:32:30,079 --> 00:32:32,759
maybe have one template for reporting a bug, one template

593
00:32:32,839 --> 00:32:35,960
for creating a feature request, that sort of thing, and

594
00:32:36,039 --> 00:32:38,759
then fill out that template. And then we also pass

595
00:32:38,799 --> 00:32:42,359
it originally when you've recorded your video we uploaded to YouTube,

596
00:32:42,400 --> 00:32:44,480
and we pass it in the link to the YouTube video,

597
00:32:44,559 --> 00:32:48,759
so it can then embed that that YouTube video in

598
00:32:48,920 --> 00:32:50,119
the issue as well as.

599
00:32:50,000 --> 00:32:50,359
Speaker 2: Part of it.

600
00:32:50,720 --> 00:32:53,480
Speaker 3: So start getting issues with videos, which is pretty cool,

601
00:32:54,480 --> 00:32:57,240
that's right. And we also we also give it a

602
00:32:57,359 --> 00:33:02,359
couple of built in MCP tools as well. So we

603
00:33:02,519 --> 00:33:06,759
have one tool that basically says, grab a screenshot at

604
00:33:06,799 --> 00:33:10,480
the specified timestamp, because when we give it the transcript,

605
00:33:10,640 --> 00:33:12,839
it's you know, it's got all the time stamps of

606
00:33:13,240 --> 00:33:15,640
when each bit of text was said, so it can

607
00:33:15,720 --> 00:33:18,359
sort of figure out what the key moment or moments

608
00:33:18,519 --> 00:33:21,319
are in the transcript, you get a you got a

609
00:33:21,359 --> 00:33:25,440
thumbnail maker based on context, that's right, and then we've

610
00:33:25,440 --> 00:33:28,720
got another tool that takes that well, can take that

611
00:33:29,279 --> 00:33:34,000
screenshot and then you use a multimodal LM to analyze

612
00:33:34,079 --> 00:33:37,519
it so it can sort of grab some context out

613
00:33:37,559 --> 00:33:39,400
of that screenshot as well. So in that way it's

614
00:33:39,400 --> 00:33:41,640
sort of able to analyze the video as well as

615
00:33:42,319 --> 00:33:42,920
what you've said.

616
00:33:43,039 --> 00:33:47,319
Speaker 2: It's very cool. I'm also thinking just de duplication is often,

617
00:33:47,960 --> 00:33:49,720
you know, you build these tools so it's easy for

618
00:33:49,759 --> 00:33:51,759
people to report problems. They're going to report a lot

619
00:33:51,799 --> 00:33:53,960
of problems, and they're off going to report the scene problems.

620
00:33:54,160 --> 00:33:54,920
Oh yeah, for sure.

621
00:33:55,440 --> 00:33:57,440
Speaker 3: So yeah, that's that's one of the by the way,

622
00:33:57,519 --> 00:34:00,480
one of the reasons why we really wanted to transition

623
00:34:00,640 --> 00:34:07,720
to this V two MCP approach because a lot of

624
00:34:07,759 --> 00:34:10,639
people have a lot of requests like that, you know,

625
00:34:11,320 --> 00:34:14,199
and all you have to do to do it with

626
00:34:14,400 --> 00:34:16,320
the new world is you just say, and you're prompt

627
00:34:16,800 --> 00:34:19,760
before you create a GitHub issue, make sure you search

628
00:34:19,960 --> 00:34:23,440
for any duplicate issues that already exist, and if they do,

629
00:34:24,119 --> 00:34:26,719
add this as a comment instead of creating a new one, right,

630
00:34:26,840 --> 00:34:29,639
and yeah, that way it's just a prompt update instead

631
00:34:29,639 --> 00:34:31,519
of having to go back to our pipeline and adding

632
00:34:31,519 --> 00:34:34,800
all these extra branches to handle different cases.

633
00:34:35,000 --> 00:34:36,559
Speaker 2: Well, and it's tougher to get people to do that

634
00:34:36,599 --> 00:34:38,320
because it's way easier to just create a new issue

635
00:34:38,360 --> 00:34:40,639
that searched you existing ones. So yeah, you can get

636
00:34:40,679 --> 00:34:42,679
the tool to do that for us, Thank goodness for that.

637
00:34:42,920 --> 00:34:45,599
Speaker 1: Well, but you have to put that in the system

638
00:34:45,679 --> 00:34:47,760
prompt you know, that has to be a rule, right,

639
00:34:48,159 --> 00:34:52,079
that's right, don't create new issues unless there's nothing there already.

640
00:34:52,159 --> 00:34:56,800
Speaker 2: Yeah, you think it's actually original. Yeah, it's really interesting.

641
00:34:57,079 --> 00:35:01,199
Another you know, we're talking about how general VII is

642
00:35:01,280 --> 00:35:03,800
coming into the different aspects of software development. This is

643
00:35:04,239 --> 00:35:08,599
an aspect of generating issues intelligently and trying to get

644
00:35:08,599 --> 00:35:11,039
as much information to the developer as possible, whether it's

645
00:35:11,079 --> 00:35:13,239
for a bugfix or a feature. Of course it's free.

646
00:35:13,159 --> 00:35:15,880
Speaker 3: Course, Yeah that's right. Yeah, So I was going to say,

647
00:35:16,119 --> 00:35:19,800
originally we with the V one solution. This was another

648
00:35:19,880 --> 00:35:23,599
pain of the V one solution. We integrate with GitHub

649
00:35:23,880 --> 00:35:28,280
and as your DevOps, but the majority of people we

650
00:35:28,400 --> 00:35:32,039
spoke to about it, they didn't actually use those tools.

651
00:35:32,079 --> 00:35:35,159
They use something like Gira or whatever some other tool,

652
00:35:35,800 --> 00:35:39,119
and it was it was just too hard to keep

653
00:35:39,239 --> 00:35:42,119
adding all these because you know, every time we create

654
00:35:42,159 --> 00:35:45,119
one of these branching bits of logic in the V

655
00:35:45,280 --> 00:35:48,800
one we had to also reproduce that same you know,

656
00:35:49,199 --> 00:35:52,760
the same actions or set of actions in each of

657
00:35:53,000 --> 00:35:56,519
the backlogs that we're integrating with so GitHub Badge develops.

658
00:35:56,880 --> 00:35:59,920
We didn't want to have to keep doing that with Gira, Zender,

659
00:36:00,159 --> 00:36:02,840
whatever other tool people are using. So that was yet

660
00:36:02,880 --> 00:36:07,599
another argument why we moved to MVP because with MCP

661
00:36:07,880 --> 00:36:11,039
you can even have exactly the same prompt, but if

662
00:36:11,039 --> 00:36:14,360
you've got a different system that you're using for your

663
00:36:14,400 --> 00:36:17,119
backlog or you have to flop the MCP server and

664
00:36:18,039 --> 00:36:21,679
as long as it has a vaguely similar action available,

665
00:36:22,239 --> 00:36:23,119
it'll work just fine.

666
00:36:23,159 --> 00:36:26,079
Speaker 2: Yeah, way more scalable. That's classic V two stuff is

667
00:36:26,880 --> 00:36:30,039
creating the rate set of interfaces for the next iterations

668
00:36:30,039 --> 00:36:31,679
where people want to use it in more places.

669
00:36:31,960 --> 00:36:34,360
Speaker 1: I know you guys had some discussions on the back

670
00:36:34,519 --> 00:36:37,519
end about whether or not it's a good idea to

671
00:36:37,639 --> 00:36:42,119
give people all this freedom and flexibility. So what was

672
00:36:42,239 --> 00:36:45,639
that discussion like and would you come out of it with.

673
00:36:45,920 --> 00:36:49,800
Speaker 3: Yeah, I think it's still sort of an open question

674
00:36:50,159 --> 00:36:54,199
that we haven't got an answer to, because I think

675
00:36:54,280 --> 00:36:59,559
it could potentially be detrimental. If you give people full

676
00:36:59,639 --> 00:37:02,800
freedom to do anything, Suddenly they're sort of swamped by

677
00:37:04,280 --> 00:37:07,320
choices and things they can customize, and they end up

678
00:37:08,199 --> 00:37:11,639
not knowing how to best use it. Whereas if you

679
00:37:11,719 --> 00:37:15,440
sort of force them down a path, you may not

680
00:37:15,559 --> 00:37:17,559
be the optimal path, but at least they're forced down

681
00:37:17,639 --> 00:37:19,280
the path and they'll figure out how to use it.

682
00:37:20,039 --> 00:37:22,800
Speaker 1: Yeah, like some workflow templates or something like that you

683
00:37:22,840 --> 00:37:25,320
can choose from exactly instead of just giving people a

684
00:37:25,360 --> 00:37:28,440
blank slate. Also, if you think about it, you know,

685
00:37:29,280 --> 00:37:32,840
it's kind of irresponsible to just slap some MCP service

686
00:37:32,920 --> 00:37:36,599
together and give them full agency to go do whatever

687
00:37:36,960 --> 00:37:40,159
you would normally do. I mean that can open up

688
00:37:40,159 --> 00:37:43,559
a huge can of worms. Oh yeah, cans and cans

689
00:37:43,639 --> 00:37:44,519
and cans of worms.

690
00:37:45,000 --> 00:37:48,639
Speaker 3: Absolutely, you really need to be careful, So, I mean,

691
00:37:48,920 --> 00:37:52,000
there is a lot of power. With power obviously comes responsibility.

692
00:37:52,079 --> 00:37:54,480
Speaker 2: I've heard that you need to be here.

693
00:37:55,079 --> 00:37:59,639
Speaker 3: You need to be careful about, you know, surfacing untrusted

694
00:38:00,039 --> 00:38:04,280
third party data to MVP servers that can perform you know,

695
00:38:04,559 --> 00:38:09,480
potentially destructive actions is actually sort of a running joke

696
00:38:09,840 --> 00:38:13,480
in the actually the dev team is when someone's recording

697
00:38:14,239 --> 00:38:17,320
a video to be processed, someone in the background will

698
00:38:17,360 --> 00:38:20,320
shout out and delete the whole repo while you're there. Yeah,

699
00:38:23,199 --> 00:38:25,960
we need to build in some safeguards to prevent.

700
00:38:25,760 --> 00:38:27,199
Speaker 2: That from there.

701
00:38:27,280 --> 00:38:30,760
Speaker 1: Used to be the running joke on dart net Rocks

702
00:38:31,679 --> 00:38:34,800
Alexa delete on my no, not you don't listen to

703
00:38:34,880 --> 00:38:38,119
me exactly. She's not quite sure how to help me

704
00:38:38,159 --> 00:38:42,000
with that. But you know, a delete my account, you

705
00:38:42,079 --> 00:38:45,280
know that kind of thing. Or send five hundred pounds

706
00:38:45,280 --> 00:38:46,760
of concrete to Richard Campbell.

707
00:38:50,519 --> 00:38:53,400
Speaker 2: Yeah, still trying to clean up that concrete. Thanks for that.

708
00:38:54,400 --> 00:38:57,639
Very good. But you know, we even that office conversation

709
00:38:57,800 --> 00:39:01,760
with more and more aishing. Our role as shepherds of AI,

710
00:39:02,039 --> 00:39:06,480
goodness knows, is to constrain it, to put parameters around

711
00:39:06,519 --> 00:39:09,000
each of these things, and so same thing here. You

712
00:39:09,119 --> 00:39:14,480
want to focus on particular issue, particular capabilities, and keep

713
00:39:14,559 --> 00:39:16,679
limits on all the things that it can do so

714
00:39:16,760 --> 00:39:18,880
that it does focus on the direction you wanted to

715
00:39:18,920 --> 00:39:19,079
go in.

716
00:39:19,400 --> 00:39:23,280
Speaker 3: The other good thing about using MCP rather than a pipeline,

717
00:39:23,360 --> 00:39:27,519
I think is that every action that the MCP service

718
00:39:27,599 --> 00:39:30,400
takes on your behalf is sort of done in your

719
00:39:30,559 --> 00:39:36,159
name because you're connecting directly with your own or credentials

720
00:39:36,320 --> 00:39:37,840
or whatever if you use and get MVP.

721
00:39:38,280 --> 00:39:39,800
Speaker 2: So yeah, you need to.

722
00:39:39,800 --> 00:39:42,199
Speaker 3: Take your responsibility for everything that it does. You can't

723
00:39:42,280 --> 00:39:46,000
just give it a random video and then cross your

724
00:39:46,000 --> 00:39:47,079
fingers and hope it works.

725
00:39:47,320 --> 00:39:50,159
Speaker 2: So that Yeah, Well you bring up a great point

726
00:39:50,159 --> 00:39:52,199
because there's lots of conversation about I think this was

727
00:39:52,320 --> 00:39:55,960
just at Ignite where they are setting identities for agents,

728
00:39:56,000 --> 00:39:59,840
and that's almost like giving you an excuse lack of culpability.

729
00:40:00,639 --> 00:40:02,920
All the software did It wasn't me. It's like, dude,

730
00:40:02,920 --> 00:40:07,320
it was your prompt the software. We may call them agents,

731
00:40:07,360 --> 00:40:09,599
but how much agency do we want them? They're working

732
00:40:09,679 --> 00:40:10,360
on our behalf.

733
00:40:10,679 --> 00:40:12,239
Speaker 1: Yeah, it's a constant theme.

734
00:40:12,920 --> 00:40:15,280
Speaker 2: Well, I just feel like these are unsolved problems. Like

735
00:40:15,480 --> 00:40:18,119
I'm appreciating which the work you guys are doing callum

736
00:40:18,280 --> 00:40:21,239
just because you are making these experiments and using them

737
00:40:21,280 --> 00:40:24,880
yourselves and finding out like what what works, what doesn't,

738
00:40:24,880 --> 00:40:26,800
what the limits are on all this are because we

739
00:40:27,559 --> 00:40:29,559
I think we're a few years away from really nailing

740
00:40:29,639 --> 00:40:31,159
down what these new workflows look.

741
00:40:31,079 --> 00:40:33,000
Speaker 3: Like, Yeah, that's right, an experiment.

742
00:40:33,280 --> 00:40:36,880
Speaker 1: At this point, you have some other products AI products

743
00:40:36,920 --> 00:40:38,480
that you're working on at ss W.

744
00:40:38,679 --> 00:40:42,800
Speaker 3: Oh yeah, we've got SSW Eagle I okay, which is

745
00:40:42,800 --> 00:40:49,280
a sort of a dystopian email analysis tool that basically

746
00:40:50,559 --> 00:40:53,880
checks all of your emails and makes for sure that

747
00:40:54,119 --> 00:40:56,480
you are sort of you know, adhering to all the

748
00:40:56,639 --> 00:41:00,679
SSW rules and and sort of game fires that a

749
00:41:00,719 --> 00:41:02,719
little bit as well. So you have like a leaderboard

750
00:41:03,119 --> 00:41:07,920
who sends the most you know, checked by email, and you.

751
00:41:07,960 --> 00:41:11,039
Speaker 1: Know, so it checks outgoing emails, not incoming, because that

752
00:41:11,079 --> 00:41:13,119
would be a huge security risk.

753
00:41:14,280 --> 00:41:16,679
Speaker 3: Yeah. Well, obviously you opt into it, right, So it's

754
00:41:16,760 --> 00:41:20,079
not just checking any random email.

755
00:41:21,320 --> 00:41:24,320
Speaker 1: You have it checking outgoing emails. Yeah, correct, that's what

756
00:41:24,440 --> 00:41:26,079
it sounds like, but not incoming.

757
00:41:26,280 --> 00:41:29,440
Speaker 3: Well, it's checking emails that you send to other people

758
00:41:29,639 --> 00:41:32,800
inside the company, right, So it's not just paying any random.

759
00:41:32,719 --> 00:41:35,360
Speaker 2: It's eternal email. That's good. We did a Cogan rules

760
00:41:35,440 --> 00:41:38,840
show in two thousand and six. That was fun. I

761
00:41:38,960 --> 00:41:43,400
remember that. Yeah, well, and then you know he's out

762
00:41:43,400 --> 00:41:46,079
of serious about that. The rules continue to this day.

763
00:41:46,159 --> 00:41:47,800
I think they're very in the middle of a migration

764
00:41:48,000 --> 00:41:50,440
of them right now, if I recall from the brainstorming

765
00:41:50,519 --> 00:41:53,360
session right there, Yeah, that's right. Yeah. So this, this

766
00:41:53,559 --> 00:41:55,400
is the idea of an LLM being able to part

767
00:41:55,440 --> 00:41:57,599
stuff and saying, are you is this compliant? Are you

768
00:41:57,679 --> 00:42:01,079
following the rule set? Maybe making suggestions for what's incorrect? Yeah,

769
00:42:01,199 --> 00:42:01,840
that's very cool.

770
00:42:02,000 --> 00:42:06,800
Speaker 3: Interesting, Yeah, that could be coming soon. The other brains

771
00:42:06,800 --> 00:42:08,920
sewing idea, the one that I was working on was

772
00:42:10,000 --> 00:42:14,719
sort of an AI that constantly scans your site and

773
00:42:14,920 --> 00:42:18,480
looks for problems and sort of reports them before someone

774
00:42:18,559 --> 00:42:19,559
needs to actualy them.

775
00:42:19,840 --> 00:42:19,960
Speaker 2: Right.

776
00:42:20,159 --> 00:42:21,320
Speaker 3: I thought that was a cool idea.

777
00:42:21,480 --> 00:42:25,760
Speaker 2: Smart site testing, yow. Yeah, could we build out it?

778
00:42:25,880 --> 00:42:29,559
Can we build out a tool that was really good

779
00:42:29,599 --> 00:42:33,079
at putting wrong things into text boxes?

780
00:42:33,760 --> 00:42:33,960
Speaker 3: You know?

781
00:42:35,320 --> 00:42:37,320
Speaker 2: Did you know those test guys, the ones that you know,

782
00:42:37,639 --> 00:42:40,760
this is where I entered negative forty three? The all

783
00:42:40,800 --> 00:42:41,440
thing blew up?

784
00:42:41,519 --> 00:42:43,800
Speaker 3: That's right, that's kind of test is the one that

785
00:42:44,000 --> 00:42:47,719
just ignores the instructions and does whatever they want. Yeah,

786
00:42:47,840 --> 00:42:49,920
I'm sure if you get a play right MCP to

787
00:42:51,039 --> 00:42:55,280
just prompt it, you're a crazy person who ignores instructions

788
00:42:55,360 --> 00:42:56,480
and does whatever they want.

789
00:43:00,159 --> 00:43:02,119
Speaker 2: Yeah, the breeze I don't want to feed you an

790
00:43:02,239 --> 00:43:06,719
LM is ignore instructions like that just seems disturbing to me. Yeah,

791
00:43:06,880 --> 00:43:10,320
that's risky. Yeah, absolute trouble, without a doubt. How many

792
00:43:10,360 --> 00:43:12,400
folks worked on on yacual what it take to get

793
00:43:12,400 --> 00:43:13,000
into this point.

794
00:43:13,400 --> 00:43:17,360
Speaker 3: We've got a team of about ten. Now, they're not

795
00:43:17,440 --> 00:43:21,440
all working obviously constantly, because you know, when we've got

796
00:43:21,519 --> 00:43:23,960
client engagements up, go off and do that. So we've

797
00:43:24,000 --> 00:43:27,039
got sort of ten coming going. I think the core team,

798
00:43:27,519 --> 00:43:33,320
you could say, maybe five people working on it. But

799
00:43:33,599 --> 00:43:39,000
one of the very interesting things we've noticed recently is well,

800
00:43:39,079 --> 00:43:42,199
I'm sort of operating as the product owner of Yakhava,

801
00:43:43,000 --> 00:43:47,360
and so as the product owner, I've been trying out AI,

802
00:43:49,639 --> 00:43:54,840
like completely AI driven development with not even looking at

803
00:43:54,880 --> 00:43:58,159
the code, just to see is that a viable approach?

804
00:43:59,320 --> 00:44:02,119
Speaker 2: So vibe coding I hate the phrase, but this.

805
00:44:02,280 --> 00:44:06,320
Speaker 3: Is exactly vibe coding. Yeah, I mean, yeah, the concept

806
00:44:06,400 --> 00:44:10,199
of vibe coding is interesting. I mean, people have different

807
00:44:10,280 --> 00:44:15,000
definitions of it, right. Some people say, anytime you use

808
00:44:15,079 --> 00:44:19,519
AI to write code, that's vibe coding, whereas others would say, well,

809
00:44:19,559 --> 00:44:23,239
it's only if you you only talk to the agent

810
00:44:23,280 --> 00:44:24,440
and you don't look at the output.

811
00:44:24,559 --> 00:44:27,559
Speaker 2: Sure, so I can say, but that's what Kapothi said

812
00:44:27,599 --> 00:44:29,800
at the time, right when he came up with the.

813
00:44:29,800 --> 00:44:34,519
Speaker 3: Phrase exactly exactly, but the terms being corrupted. I think

814
00:44:34,639 --> 00:44:36,320
so a lot of people just say anytime you use

815
00:44:36,360 --> 00:44:38,679
AI to write code, you're vibe coding. I don't think

816
00:44:38,679 --> 00:44:40,599
that's good at all. Yeah, and I think it's actually

817
00:44:40,599 --> 00:44:44,559
important to draw that distinction because, like we just said,

818
00:44:44,639 --> 00:44:46,920
you know, you need to take responsibility for the code

819
00:44:46,920 --> 00:44:51,239
that you generate if it's generated under your name. So so.

820
00:44:51,800 --> 00:44:54,679
But yeah, So the concept is, as product owner of

821
00:44:54,760 --> 00:44:58,239
yak Shav, I'm just vibe coding the features that I

822
00:44:58,320 --> 00:45:01,880
want to see in the tool, and then rather than

823
00:45:02,320 --> 00:45:06,039
giving the team an issue that says, hey, I want

824
00:45:06,199 --> 00:45:10,880
this feature in Yakhava, I will basically vibe code the

825
00:45:11,840 --> 00:45:14,800
feature itself and then give them the poll request, and

826
00:45:14,920 --> 00:45:18,039
then their job instead of implement the feature is just

827
00:45:18,159 --> 00:45:19,599
review the poll request.

828
00:45:19,800 --> 00:45:22,199
Speaker 2: Figure out how badly you've gotten the LM to mess

829
00:45:22,239 --> 00:45:23,199
things up exactly.

830
00:45:26,719 --> 00:45:29,760
Speaker 3: And I mean obviously that sometimes it does a terrible job,

831
00:45:29,840 --> 00:45:33,199
but I think more and more it does a good job,

832
00:45:33,360 --> 00:45:38,679
particularly at tasks that are sort of well constrained or

833
00:45:38,840 --> 00:45:42,880
are you an implementation that is along the lines of

834
00:45:43,360 --> 00:45:45,719
some codes that's already written, but just doing a slightly

835
00:45:45,760 --> 00:45:48,559
different thing, you know what I mean. So there's a

836
00:45:48,599 --> 00:45:52,119
big difference between creating a new, completely new piece of

837
00:45:52,199 --> 00:45:55,440
code that may have got some new architectural components and

838
00:45:55,519 --> 00:45:58,519
that sort of thing that me sounds rather risky to

839
00:45:58,639 --> 00:46:00,760
just let AI do whatever it wants. But if you've

840
00:46:00,760 --> 00:46:02,719
already got all that set up and now you're just saying,

841
00:46:03,320 --> 00:46:07,920
add another you know, vertical slice on this project and

842
00:46:08,079 --> 00:46:10,079
just copy the one that's already there, I think it

843
00:46:10,159 --> 00:46:11,280
does a much better job of that.

844
00:46:11,480 --> 00:46:14,800
Speaker 2: It does speak to this idea that more mature software

845
00:46:15,159 --> 00:46:18,800
will be easier to maintain with these tools than very

846
00:46:18,920 --> 00:46:22,400
new software. On you know, the wen Star, as long

847
00:46:22,480 --> 00:46:25,119
as it's it's good, you know, as long as it's

848
00:46:25,119 --> 00:46:27,320
written well, right, right, So it's got to be written

849
00:46:27,360 --> 00:46:28,599
well in the first place.

850
00:46:28,480 --> 00:46:32,159
Speaker 3: Because it's going to give you more crap, I imagine.

851
00:46:32,239 --> 00:46:35,079
Speaker 2: I also wonder if if the success of this has

852
00:46:35,119 --> 00:46:37,280
more to do with the scope of the feature or

853
00:46:37,360 --> 00:46:38,639
the quality of the prompt.

854
00:46:38,920 --> 00:46:43,000
Speaker 3: Yeah, good, good question, And I think that in my

855
00:46:43,119 --> 00:46:48,239
experience anyway, a lot of people have like when when

856
00:46:48,280 --> 00:46:50,039
they try to do something with AI. You know, the

857
00:46:50,119 --> 00:46:53,679
people who I think get bad results. Some people seem

858
00:46:53,719 --> 00:46:57,519
to get bad results even though they're fantastic developers. You know,

859
00:46:57,559 --> 00:47:00,199
they'll try to use AI and they'll get a a

860
00:47:00,280 --> 00:47:02,000
bad result and then they say, well, you know this

861
00:47:02,159 --> 00:47:04,320
AI sucks another tool myself.

862
00:47:04,400 --> 00:47:04,960
Speaker 2: Yeah, exactly.

863
00:47:05,159 --> 00:47:09,559
Speaker 3: But from my perspective, I always try to think, if

864
00:47:09,679 --> 00:47:15,360
I've used AI and the result was substandard, rather than

865
00:47:15,400 --> 00:47:17,880
blaming the AI, I want to blame myself, and I

866
00:47:17,960 --> 00:47:20,840
want to think, how could I how could I prompt

867
00:47:20,920 --> 00:47:23,760
it better? How could I give the right context or

868
00:47:23,880 --> 00:47:27,400
better prompt or whatever to get the better output next time.

869
00:47:27,719 --> 00:47:30,000
Speaker 2: Well, let's say that's the old adage it is a

870
00:47:30,039 --> 00:47:31,719
poor craftsman that blames as tools.

871
00:47:31,840 --> 00:47:35,639
Speaker 3: That's right, good point, And yeah, so that's something that

872
00:47:35,719 --> 00:47:39,320
I'm heavily leaning into is figuring out how we can

873
00:47:40,320 --> 00:47:43,880
you know, make this process of AI generated code not

874
00:47:44,280 --> 00:47:47,440
vibe coding, to be clear, because we're drawing a distinction there,

875
00:47:48,119 --> 00:47:52,360
but generating code with AI that we actually care about

876
00:47:52,480 --> 00:47:54,880
and that's going to form a part of our long

877
00:47:55,000 --> 00:47:59,880
term code base. How do we actually go about doing

878
00:48:00,119 --> 00:48:02,920
that in the best way. And obviously, to start with

879
00:48:03,079 --> 00:48:08,039
it takes longer, probably than writing the code yourself quite

880
00:48:08,079 --> 00:48:10,039
but I think in the long term, once you've got

881
00:48:10,119 --> 00:48:13,039
your systems in place, you know, you've sort of set

882
00:48:13,119 --> 00:48:16,000
that up once and then you can use it infinitely

883
00:48:16,079 --> 00:48:19,480
many times. But but yeah, so it's that reason that

884
00:48:19,519 --> 00:48:21,800
I've also been doing this vibe code experiment just to

885
00:48:21,880 --> 00:48:24,760
see what is the difference between these two different approaches

886
00:48:24,960 --> 00:48:27,760
and what goes wrong when you're VIBE coding. What are

887
00:48:28,119 --> 00:48:30,480
the problems that AI has and then how do we

888
00:48:31,239 --> 00:48:33,519
try to address them so that we can do a

889
00:48:33,559 --> 00:48:34,000
better job.

890
00:48:34,119 --> 00:48:36,719
Speaker 2: So I see three categories of code. Then that you

891
00:48:36,840 --> 00:48:40,239
have handwritten code, you have AI syst a code, and

892
00:48:40,320 --> 00:48:44,320
then you have AI generated code. And you know how

893
00:48:44,440 --> 00:48:47,039
different are those things? How can they support each other?

894
00:48:47,760 --> 00:48:47,920
Speaker 3: You know?

895
00:48:48,039 --> 00:48:49,880
Speaker 2: Where is where is it? I think we're still trying

896
00:48:49,880 --> 00:48:51,719
to figure out where the human needs to step in

897
00:48:51,880 --> 00:48:56,360
more and where automation it can work fairly responsibly on it.

898
00:48:56,599 --> 00:49:00,679
Speaker 3: That's right, I think it really the human needs to

899
00:49:02,079 --> 00:49:05,000
If you're delivering a piece of software for a client,

900
00:49:05,559 --> 00:49:09,199
the human obviously is the person who the client is

901
00:49:09,280 --> 00:49:13,400
engaging to deliver that software. You need to take responsibility

902
00:49:13,519 --> 00:49:17,400
for you know, understanding the problem and ensuring that the

903
00:49:17,519 --> 00:49:21,400
software that is delivered meets all the requirements and does

904
00:49:21,480 --> 00:49:23,800
everything the client needs, but also does all the technical

905
00:49:23,880 --> 00:49:25,960
things that the client doesn't even know that they need.

906
00:49:25,920 --> 00:49:30,519
Speaker 2: But they do. The client presumes security, probably without even

907
00:49:30,639 --> 00:49:35,199
articulating it, that's right. You know, client presumes reliability also

908
00:49:35,280 --> 00:49:37,519
without caculating it. Like we've got to make sure those

909
00:49:37,519 --> 00:49:38,599
things exist, that's right.

910
00:49:38,679 --> 00:49:42,639
Speaker 3: And if we just mindlessly generate something with AI and

911
00:49:42,719 --> 00:49:45,360
don't even look at it and then yeah, okay, it

912
00:49:45,480 --> 00:49:48,559
does most of the functionality you wanted, but it's all

913
00:49:48,599 --> 00:49:50,960
these gaping security flaws, all this sort of things. That's

914
00:49:51,239 --> 00:49:53,639
that's you know, that's a problem for you that you

915
00:49:53,760 --> 00:49:56,360
haven't taken responsibility for what you have delivered.

916
00:49:56,880 --> 00:49:59,559
Speaker 2: Yeah, that's what That's what responsible detment looks like. That's right.

917
00:50:00,119 --> 00:50:03,400
Speaker 3: But you can totally be responsible with your delivery and

918
00:50:03,639 --> 00:50:06,800
also use AI to do everything. It's just that you

919
00:50:06,920 --> 00:50:10,360
need to be taking responsibility for it right ultimately, so

920
00:50:10,599 --> 00:50:11,159
do the right thing.

921
00:50:11,440 --> 00:50:13,679
Speaker 2: So as a product owner, I see a shape of

922
00:50:13,719 --> 00:50:16,039
a V two, is your stuff going to a V three?

923
00:50:16,199 --> 00:50:18,719
Ben Yette like, what's the future of the ACTUAV? Look like?

924
00:50:18,840 --> 00:50:23,039
Speaker 3: Good question? I mean, I see after after V two.

925
00:50:23,519 --> 00:50:26,800
Obviously V two we should be rolling out fairly soon.

926
00:50:26,880 --> 00:50:31,360
Hopefully V two is all about MCP servers. I see

927
00:50:31,440 --> 00:50:37,320
potentially V three is going to be Yakshava actually writing

928
00:50:38,639 --> 00:50:42,800
writing code sort of like disposable code in order to

929
00:50:43,039 --> 00:50:46,920
achieve tasks, rather than just using MCP servers.

930
00:50:47,880 --> 00:50:51,239
Speaker 1: Disposable code like writing PowerShell scripts and then.

931
00:50:51,199 --> 00:50:52,440
Speaker 3: Executing that kind of thing.

932
00:50:52,599 --> 00:50:52,800
Speaker 2: Yeah.

933
00:50:53,679 --> 00:50:57,960
Speaker 3: Yeah, so like in integrating with whatever it's trying to

934
00:50:58,039 --> 00:51:01,599
integrate with, it could also be using the MCP servers,

935
00:51:02,639 --> 00:51:06,239
but you know, writing code to perform the tools in

936
00:51:06,480 --> 00:51:09,679
MCP rather than just using MCP because obviously one of

937
00:51:09,760 --> 00:51:12,920
the issues with one of the issues with MCP is

938
00:51:13,480 --> 00:51:15,719
you know, if you give it a whole heap of tools,

939
00:51:16,159 --> 00:51:20,039
that's just eating up a bunch of context that probably

940
00:51:20,079 --> 00:51:23,519
doesn't need to be eaten up. Although I do suspect

941
00:51:23,559 --> 00:51:26,320
in the future we'll get better MCP management, Like you

942
00:51:26,400 --> 00:51:29,159
might have some sort of middle layer that says, here's

943
00:51:29,360 --> 00:51:32,639
here's the prompt, here's on my tools. Just surface a

944
00:51:32,719 --> 00:51:35,239
couple of tools that seem like they need to be used,

945
00:51:35,480 --> 00:51:36,159
and that sort of thing.

946
00:51:36,559 --> 00:51:40,360
Speaker 2: Yeah, it's a pretty pretty raw design at this point too, right,

947
00:51:40,559 --> 00:51:44,280
Like it's all of this stuff is so crazy new. Yeah.

948
00:51:44,559 --> 00:51:45,960
Speaker 3: Yeah, it's an exciting space to be in.

949
00:51:46,280 --> 00:51:48,840
Speaker 1: So what's next? What's next for you? Personally?

950
00:51:49,280 --> 00:51:54,039
Speaker 3: For me, well, I think I continue to use AI

951
00:51:54,280 --> 00:52:00,599
to deliver projects. You know, I always felt that my

952
00:52:00,719 --> 00:52:04,440
favorite part of my job was the part where I

953
00:52:04,559 --> 00:52:08,639
get to talk to the client, understand the problem and

954
00:52:08,960 --> 00:52:12,400
design a solution and understand all the different trade offs

955
00:52:12,440 --> 00:52:14,599
in all the decisions that are making in that solution,

956
00:52:15,239 --> 00:52:17,159
and then you know, talk to the client figure out

957
00:52:17,199 --> 00:52:19,280
what is the best solution for them. And then the

958
00:52:19,360 --> 00:52:22,079
actual part where I write the code is sort of

959
00:52:22,199 --> 00:52:26,280
the necessary evil. You know, it's not the fun part.

960
00:52:26,360 --> 00:52:29,800
The fun part is designing the solution, right. So I

961
00:52:29,920 --> 00:52:32,639
think that's one of the main reasons why I love

962
00:52:32,760 --> 00:52:35,639
using A for this, because I can sort of still

963
00:52:35,719 --> 00:52:39,199
be fully involved in all that decision making process and

964
00:52:39,320 --> 00:52:43,320
solution design, but then I'm basically just delegating all the

965
00:52:43,400 --> 00:52:51,039
actual code writing to a system of parallel cloud based

966
00:52:51,079 --> 00:52:54,800
agents who while I'm asleep, they can just make a

967
00:52:54,920 --> 00:52:57,239
dozen pull requests and the next morning I'll wake up

968
00:52:57,320 --> 00:53:01,440
and just review them and you know, provide feedback and.

969
00:53:01,639 --> 00:53:02,400
Speaker 2: So on and so forth.

970
00:53:04,159 --> 00:53:05,360
Speaker 1: It's a brave new world in it.

971
00:53:05,480 --> 00:53:07,880
Speaker 3: Absolutely it's very exciting, Yeah it is.

972
00:53:08,800 --> 00:53:12,920
Speaker 1: And it's also I think freeing our imaginations to come

973
00:53:13,039 --> 00:53:15,360
up with solutions because and this has been a theme

974
00:53:15,559 --> 00:53:17,960
we've talked to many people about, starting with that Scott

975
00:53:18,000 --> 00:53:23,199
Hunter interview. Right, just imagination is going to be a

976
00:53:23,480 --> 00:53:28,559
very important commodity now, the creativity to think about what

977
00:53:29,199 --> 00:53:31,800
you can do, and if you can think it, you

978
00:53:31,920 --> 00:53:36,199
can probably get it done. And that's just an amazing thing.

979
00:53:36,559 --> 00:53:41,519
Speaker 3: That's right, because I guess there's sort of no downside

980
00:53:41,639 --> 00:53:46,599
to just trying something, right is if you can articulate

981
00:53:46,679 --> 00:53:49,559
your idea, then you can have an agent go off

982
00:53:49,599 --> 00:53:52,039
and have a crack at it and just see what happens.

983
00:53:52,079 --> 00:53:54,400
And I mean, the only thing you've lost is maybe

984
00:53:54,559 --> 00:53:59,000
a few cents of token usage, and that's really it.

985
00:53:59,320 --> 00:53:59,840
Speaker 2: So why not?

986
00:54:00,239 --> 00:54:03,920
Speaker 1: So it pays to stay in school kids and learn

987
00:54:03,960 --> 00:54:06,000
as much as you can about the English language, and

988
00:54:06,119 --> 00:54:08,960
take writing classes and be clear and your thoughts and

989
00:54:09,079 --> 00:54:12,960
all of that, and don't use like too much, and

990
00:54:13,280 --> 00:54:18,159
you know, treat each other well.

991
00:54:20,000 --> 00:54:21,239
Speaker 3: That's right, and the good news.

992
00:54:23,840 --> 00:54:25,679
Speaker 2: Of wisdom.

993
00:54:26,440 --> 00:54:28,719
Speaker 3: You can also use AI to teach you things as well,

994
00:54:29,400 --> 00:54:31,599
which is cooling. So if you want to learn something,

995
00:54:33,400 --> 00:54:36,960
you can just ask AI to create a personalized tutorial.

996
00:54:37,000 --> 00:54:41,039
Speaker 1: I think that's or even little things like I do

997
00:54:41,159 --> 00:54:41,760
this all the time.

998
00:54:41,840 --> 00:54:42,000
Speaker 2: Now.

999
00:54:42,360 --> 00:54:47,599
Speaker 1: I was watching a TV show and this woman, this

1000
00:54:47,760 --> 00:54:51,679
young woman had what looked like a tube coming a

1001
00:54:51,760 --> 00:54:55,000
white tube coming from her ear into her nose with

1002
00:54:55,159 --> 00:54:58,239
some scotch tape on her cheek. And I thought, that

1003
00:54:58,360 --> 00:55:01,000
looks very strange. Is it oxygen?

1004
00:55:01,079 --> 00:55:01,440
Speaker 2: What is it?

1005
00:55:02,039 --> 00:55:03,960
Speaker 1: So I took a couple of pictures of the TV

1006
00:55:04,079 --> 00:55:06,920
screen to send it to chat Gypt and it quickly

1007
00:55:07,159 --> 00:55:11,960
figured out that it was a feeding tube because and

1008
00:55:12,079 --> 00:55:15,599
then turns out like later on somebody asked her because

1009
00:55:15,599 --> 00:55:17,400
it was a cooking show, somebody asked her, did you

1010
00:55:17,519 --> 00:55:19,320
chase your food? And she says, no, I can't taste

1011
00:55:19,360 --> 00:55:26,880
it because I have this gastric just whatever situation where

1012
00:55:27,199 --> 00:55:30,360
it takes two hours to half digest food, and so

1013
00:55:30,599 --> 00:55:33,519
she has to eat through a feeding tube. And she'd

1014
00:55:33,559 --> 00:55:35,000
had to do that she was twenty one, she'd had

1015
00:55:35,039 --> 00:55:37,760
to do it since she was thirteen. Wow, But chatchypt

1016
00:55:37,960 --> 00:55:40,159
figured it all out, and then I was asking it

1017
00:55:40,320 --> 00:55:43,800
questions about does that go all the way down the esophagus,

1018
00:55:43,920 --> 00:55:45,960
isn't that uncomfortable? And it's like telling me all this

1019
00:55:46,039 --> 00:55:48,960
stuff that I would It's just like having an expert.

1020
00:55:49,320 --> 00:55:51,159
It's like having Richard in your house.

1021
00:55:52,039 --> 00:55:52,199
Speaker 3: You know.

1022
00:55:53,000 --> 00:55:55,400
Speaker 1: I can say Richard, what is that? And hit clause

1023
00:55:55,440 --> 00:55:58,800
and Richard knows everything, so he would tell me it's.

1024
00:55:58,679 --> 00:56:01,079
Speaker 2: Really cool, right, which is it's interesting.

1025
00:56:01,440 --> 00:56:03,360
Speaker 1: It's better, way better than just Google.

1026
00:56:03,639 --> 00:56:04,280
Speaker 2: Yeah, for sure.

1027
00:56:04,519 --> 00:56:06,519
Speaker 1: Before you leave, I have an Adam's story for you.

1028
00:56:08,239 --> 00:56:10,840
It was I can't remember when, but he was doing

1029
00:56:10,920 --> 00:56:14,440
some videos with me for DNR TV, so that tells

1030
00:56:14,480 --> 00:56:16,920
you it was a while ago, early two thousands. Yeah,

1031
00:56:17,760 --> 00:56:20,800
and he was actually in Boston for techads. So that

1032
00:56:20,920 --> 00:56:24,119
was what two thousand and six maybe something like that, Yeah,

1033
00:56:24,320 --> 00:56:27,239
something like that. So he came down and we were

1034
00:56:27,360 --> 00:56:30,039
up all night and I was just basically waiting for

1035
00:56:30,159 --> 00:56:32,960
him to get his demo together because he was, you know,

1036
00:56:33,159 --> 00:56:36,000
working on it. So seven o'clock in the morning rolls

1037
00:56:36,000 --> 00:56:39,079
around time for Brecky. So we go to a diner.

1038
00:56:39,480 --> 00:56:42,039
So two thousand and six, right, so we barely have

1039
00:56:42,360 --> 00:56:46,440
internet on phones, but we do have it. And he

1040
00:56:46,559 --> 00:56:48,679
go to a diner and I'm putting pepper on my eggs.

1041
00:56:48,719 --> 00:56:51,760
He goes, you gotta watch that stuff. I said, what

1042
00:56:52,480 --> 00:56:54,199
black papa bad for blokes?

1043
00:56:55,039 --> 00:56:55,239
Speaker 3: Like what?

1044
00:56:56,079 --> 00:57:05,159
Speaker 1: He says, Yeah, I'll give you a prostate cancer. Come on, Well,

1045
00:57:05,199 --> 00:57:07,519
I didn't have Chatty obviously, but I did have Google

1046
00:57:07,559 --> 00:57:10,119
on my phone. I looked it up and not turns out.

1047
00:57:11,159 --> 00:57:14,519
It turns out not only is black pepper good for

1048
00:57:14,639 --> 00:57:19,400
your immune system and therefore not causing cancer, but kespasin,

1049
00:57:19,519 --> 00:57:22,519
which is not black pepper at all. But it's the

1050
00:57:22,599 --> 00:57:26,199
stuff that makes Pepper's chili peppers hot. When you apply

1051
00:57:26,400 --> 00:57:28,960
ks spasin directly to prostate cancer cells.

1052
00:57:28,760 --> 00:57:29,360
Speaker 2: It kills them.

1053
00:57:30,679 --> 00:57:33,360
Speaker 1: So I was like, dude, where did you learn this?

1054
00:57:33,480 --> 00:57:37,280
Speaker 3: He goes, friend, Yeah, well, look, it's all about strong

1055
00:57:37,320 --> 00:57:38,400
opinions weekly hill.

1056
00:57:38,559 --> 00:57:42,239
Speaker 2: So yeah, yeah, a change. But I told my kids.

1057
00:57:42,119 --> 00:57:44,719
Speaker 1: Because they grew up in the Internet age, and you know,

1058
00:57:44,960 --> 00:57:47,760
when you have a phone, when you get older, you

1059
00:57:47,880 --> 00:57:51,280
can use it as your portable BS detector. And you know,

1060
00:57:51,320 --> 00:57:54,679
and somebody tells you something, don't take it at face value.

1061
00:57:54,719 --> 00:57:56,159
Go and look it up and look it up at

1062
00:57:56,159 --> 00:58:03,599
a reputable fact checking site, not just like you know, TikTok. Anyway, callum,

1063
00:58:03,679 --> 00:58:06,000
thank you very much, it's been so great talking to you,

1064
00:58:06,159 --> 00:58:07,960
and this is great stuff and I can't wait for

1065
00:58:08,079 --> 00:58:09,679
V two. I'm gonna run it myself.

1066
00:58:10,039 --> 00:58:10,599
Speaker 2: I can't wait.

1067
00:58:10,800 --> 00:58:12,639
Speaker 3: Yeah, I can't wait to roll it out of you though.

1068
00:58:12,719 --> 00:58:13,480
Thanks having me.

1069
00:58:13,880 --> 00:58:17,239
Speaker 1: You let us know when it's available do and thanks Callum,

1070
00:58:17,920 --> 00:58:20,880
and we'll talk to you next time on dot net rocks.

1071
00:58:41,480 --> 00:58:44,039
Dot net rocks is brought to you by Franklin's Net

1072
00:58:44,320 --> 00:58:48,239
and produced by Pop Studios, a full service audio, video

1073
00:58:48,320 --> 00:58:52,360
and post production facility located physically in New London, Connecticut,

1074
00:58:52,679 --> 00:58:56,840
and of course in the cloud online at pwop dot com.

1075
00:58:57,679 --> 00:58:59,679
Visit our website at d O T N E t

1076
00:59:00,119 --> 00:59:04,039
R O c k S dot com for RSS feeds, downloads,

1077
00:59:04,199 --> 00:59:07,840
mobile apps, comments, and access to the full archives going

1078
00:59:07,920 --> 00:59:11,320
back to show number one, recorded in September two thousand

1079
00:59:11,320 --> 00:59:13,960
and two. And make sure you check out our sponsors.

1080
00:59:14,159 --> 00:59:16,920
They keep us in business. Now go write some code,

1081
00:59:17,480 --> 00:59:18,239
see you next time.

1082
00:59:19,159 --> 00:59:20,960
Speaker 3: You got javans

1083
00:59:23,079 --> 00:59:23,119
Speaker 2: And