WEBVTT

1
00:00:01.080 --> 00:00:04.799
<v Speaker 1>How'd you like to listen to dot NetRocks with no ads? Easy?

2
00:00:05.320 --> 00:00:08.519
<v Speaker 1>Become a patron for just five dollars a month you

3
00:00:08.599 --> 00:00:11.320
<v Speaker 1>get access to a private RSS feed where all the

4
00:00:11.359 --> 00:00:14.599
<v Speaker 1>shows have no ads. Twenty dollars a month will get

5
00:00:14.599 --> 00:00:17.679
<v Speaker 1>you that and a special dot net Rocks patron mug.

6
00:00:18.160 --> 00:00:22.879
<v Speaker 1>Sign up now at Patreon dot dot NetRocks dot com. Hi,

7
00:00:23.000 --> 00:00:24.600
<v Speaker 1>this is Carl Franklin.

8
00:00:24.239 --> 00:00:25.480
<v Speaker 2>And this is Richard Campbell.

9
00:00:25.800 --> 00:00:30.000
<v Speaker 1>We've got two special shows coming up soon, episode nineteen

10
00:00:30.120 --> 00:00:32.079
<v Speaker 1>ninety nine and two thousand.

11
00:00:32.359 --> 00:00:35.039
<v Speaker 2>For episode nineteen ninety nine, we're collecting people's y two

12
00:00:35.159 --> 00:00:37.520
<v Speaker 2>k stories what did you do to help the Y

13
00:00:37.600 --> 00:00:39.719
<v Speaker 2>two k event not actually happen?

14
00:00:40.200 --> 00:00:42.960
<v Speaker 1>And for episode two thousand, we're going to be sharing

15
00:00:43.000 --> 00:00:45.479
<v Speaker 1>stories about how dot net shaped your career.

16
00:00:46.000 --> 00:00:48.679
<v Speaker 2>We have a special page at dot netroocks dot com

17
00:00:48.719 --> 00:00:52.159
<v Speaker 2>slash voxpop where you can record messages for us that

18
00:00:52.200 --> 00:00:55.000
<v Speaker 2>we can play on these special episodes. So tell us

19
00:00:55.039 --> 00:00:57.479
<v Speaker 2>what you did for Y two k and what dot

20
00:00:57.479 --> 00:00:59.560
<v Speaker 2>net means to you, and of course how long you've

21
00:00:59.560 --> 00:01:02.320
<v Speaker 2>been listening to dot net rocks. So go do dot

22
00:01:02.399 --> 00:01:05.280
<v Speaker 2>NetRocks dot com slash vox pop now and leave us

23
00:01:05.319 --> 00:01:07.799
<v Speaker 2>a message before the thought of operates like whiskey left

24
00:01:07.799 --> 00:01:09.640
<v Speaker 2>in a glass overnight. Do it.

25
00:01:22.560 --> 00:01:25.719
<v Speaker 1>Hey, it's dot net Rocks. I'm Carl Franklin, I'm Richard Campbell,

26
00:01:25.760 --> 00:01:27.879
<v Speaker 1>and Ed Sharbern knows here. We're gonna be talking to

27
00:01:27.959 --> 00:01:30.799
<v Speaker 1>him in a little bit. But first we got our

28
00:01:31.000 --> 00:01:33.519
<v Speaker 1>opening business to do. Ed, feel free to jump in

29
00:01:33.519 --> 00:01:37.319
<v Speaker 1>if you need to. We'll do all right. This is

30
00:01:37.799 --> 00:01:41.040
<v Speaker 1>being episode nineteen ninety seven. Let's start with what happened

31
00:01:41.040 --> 00:01:44.640
<v Speaker 1>that year, so world events and cultural things, the handover

32
00:01:44.920 --> 00:01:48.799
<v Speaker 1>of Hong Kong to China. In July, the UK officially

33
00:01:48.840 --> 00:01:51.480
<v Speaker 1>transferred sovereignty of Hong Kong to China one hundred and.

34
00:01:51.439 --> 00:01:52.159
<v Speaker 2>Fifty six years.

35
00:01:52.200 --> 00:01:54.000
<v Speaker 1>One hundred and fifty six years.

36
00:01:54.359 --> 00:01:58.120
<v Speaker 2>Yeah, and you remember China promised one country, two systems

37
00:01:58.120 --> 00:01:58.799
<v Speaker 2>for fifty years.

38
00:01:58.840 --> 00:02:01.280
<v Speaker 1>You know, promises don't mean anything on the world stage.

39
00:02:01.359 --> 00:02:06.159
<v Speaker 2>Apparently that event had huge impact in Vancouver because in

40
00:02:06.159 --> 00:02:08.840
<v Speaker 2>the couple of years leading up to it, a huge

41
00:02:09.000 --> 00:02:12.639
<v Speaker 2>number because they were actually British subjects as well of

42
00:02:13.159 --> 00:02:16.199
<v Speaker 2>well to do Hong Kong fo bought homes in around

43
00:02:16.199 --> 00:02:16.919
<v Speaker 2>the Vancouver area.

44
00:02:17.039 --> 00:02:18.360
<v Speaker 1>Wow, huge numbers. Wow.

45
00:02:18.520 --> 00:02:21.360
<v Speaker 2>So drove up our housing prices. And then when the

46
00:02:21.400 --> 00:02:25.080
<v Speaker 2>handover was kind of a non event back then. They

47
00:02:25.159 --> 00:02:27.560
<v Speaker 2>largely didn't come until a few years later. When their

48
00:02:27.599 --> 00:02:29.439
<v Speaker 2>kids get old enough, they're like, we want to send

49
00:02:29.479 --> 00:02:31.319
<v Speaker 2>a broad to school. Hey, we own a house in Vancouver,

50
00:02:31.360 --> 00:02:33.879
<v Speaker 2>so they just send their kids, you know, and a Lamborghini,

51
00:02:34.080 --> 00:02:35.759
<v Speaker 2>you know, Creosa.

52
00:02:35.360 --> 00:02:38.360
<v Speaker 1>Sers and I bet the Chinese restaurants got better too.

53
00:02:38.520 --> 00:02:40.759
<v Speaker 2>Chinese restaurants are always good here. They call it Honcouver

54
00:02:40.800 --> 00:02:41.360
<v Speaker 2>for a reason.

55
00:02:43.319 --> 00:02:45.599
<v Speaker 3>My memory of the politics at the time is very,

56
00:02:45.879 --> 00:02:49.280
<v Speaker 3>very rusty because I was a scrappy eighteen years old

57
00:02:49.319 --> 00:02:51.759
<v Speaker 3>with my daughter being born that year.

58
00:02:52.319 --> 00:02:55.680
<v Speaker 2>Yeah, you were busy. I was busy, Yeah, no kidding.

59
00:02:56.159 --> 00:03:00.000
<v Speaker 1>Princess Diana died in August thirty first. I remember. Everybody

60
00:03:00.120 --> 00:03:02.080
<v Speaker 1>remembers where they were when they heard that, But I

61
00:03:02.120 --> 00:03:05.319
<v Speaker 1>was actually playing a gig in downtown Mystic, Connecticut.

62
00:03:05.400 --> 00:03:06.159
<v Speaker 2>Did you call it out?

63
00:03:06.400 --> 00:03:09.639
<v Speaker 1>Yeah? Yeah, yeah, you have to. You have to. Plus

64
00:03:09.680 --> 00:03:13.759
<v Speaker 1>it was on the TV, like everybody was stunned. The

65
00:03:13.840 --> 00:03:18.159
<v Speaker 1>Kyoto Protocol was adopted in December. That's an international agreement

66
00:03:18.199 --> 00:03:20.159
<v Speaker 1>to reduce greenhouse gas emissions.

67
00:03:20.240 --> 00:03:22.800
<v Speaker 2>Sorry, adopted? How optimistic of you?

68
00:03:23.000 --> 00:03:26.599
<v Speaker 1>Yeah, well it was adopted by Kyoto, wasn't it.

69
00:03:27.280 --> 00:03:29.280
<v Speaker 2>No, that was the location with it. It was supposed to

70
00:03:29.280 --> 00:03:31.319
<v Speaker 2>be an international treaty that everyone.

71
00:03:31.080 --> 00:03:33.120
<v Speaker 1>Yeah, well it was adopted at first and then it

72
00:03:33.159 --> 00:03:34.080
<v Speaker 1>got unadopted.

73
00:03:34.120 --> 00:03:35.800
<v Speaker 2>Well, everybody says they're going to do things, is this

74
00:03:35.919 --> 00:03:37.479
<v Speaker 2>question of whether they do things.

75
00:03:37.560 --> 00:03:39.840
<v Speaker 1>Yeah, Like I said, promises don't really mean anything on

76
00:03:39.879 --> 00:03:40.879
<v Speaker 1>the world stage, do they?

77
00:03:40.960 --> 00:03:44.719
<v Speaker 2>I would argue, And when I talk about climate change

78
00:03:45.120 --> 00:03:48.400
<v Speaker 2>that the projection is that in nineteen ninety seven at

79
00:03:48.439 --> 00:03:52.520
<v Speaker 2>the Kyoto curR we're last meeting, like four degrees temperature

80
00:03:52.520 --> 00:03:57.039
<v Speaker 2>increased by twenty fifty and that's actually off the table.

81
00:03:57.159 --> 00:04:00.680
<v Speaker 2>So we did some stuff nowhere near enough. Currently the

82
00:04:00.680 --> 00:04:03.639
<v Speaker 2>projection is somewhere between two and a half and three,

83
00:04:03.680 --> 00:04:06.400
<v Speaker 2>which is, by the way, still way too much. Yeah,

84
00:04:06.560 --> 00:04:08.599
<v Speaker 2>but like we have me progress.

85
00:04:08.719 --> 00:04:08.960
<v Speaker 1>Yeah.

86
00:04:09.039 --> 00:04:11.680
<v Speaker 3>I was hearing something I think it might have been

87
00:04:11.719 --> 00:04:14.439
<v Speaker 3>Neil de Grasse Tyson talking about it, and he said

88
00:04:14.560 --> 00:04:18.480
<v Speaker 3>there was some success out of that, that it greatly

89
00:04:18.519 --> 00:04:22.959
<v Speaker 3>reduced like the CFCs that were being put in the Ozone.

90
00:04:23.040 --> 00:04:26.000
<v Speaker 2>Well, CFCs were largely eliminated by the Montreal Protocol in

91
00:04:26.079 --> 00:04:31.000
<v Speaker 2>nineteen seventy seven. Although you know, the reason the Montreal

92
00:04:31.079 --> 00:04:33.040
<v Speaker 2>Protocol went so well is because everybody could make money

93
00:04:33.040 --> 00:04:35.680
<v Speaker 2>off of it. Every refrigerator needs to be replaced. Every

94
00:04:36.079 --> 00:04:38.079
<v Speaker 2>you know, air condition needs to replace. Like that was

95
00:04:38.199 --> 00:04:40.079
<v Speaker 2>very exciting for certain parts of the mark.

96
00:04:40.519 --> 00:04:42.920
<v Speaker 3>I might I might have swapped the two in my mind.

97
00:04:43.240 --> 00:04:46.040
<v Speaker 2>Yeah, I do. When I talk, you know, I'm talking

98
00:04:46.079 --> 00:04:49.839
<v Speaker 2>to students about, you know, about climate change. I mentioned

99
00:04:49.879 --> 00:04:52.040
<v Speaker 2>the Montreal Protocols obviously before they were born. But it's

100
00:04:52.040 --> 00:04:55.160
<v Speaker 2>also like we can as a civilization get together and

101
00:04:55.160 --> 00:04:57.519
<v Speaker 2>agree to do something and actually do it.

102
00:04:57.639 --> 00:04:58.720
<v Speaker 3>I was born a year later.

103
00:04:59.160 --> 00:04:59.600
<v Speaker 2>There you go.

104
00:05:00.079 --> 00:05:05.360
<v Speaker 1>The Asian financial crisis began in nineteen ninety seven. The

105
00:05:05.399 --> 00:05:12.720
<v Speaker 1>collapse of Thailand's currency spread all across Asia. Recessions, unrest, Yeah,

106
00:05:13.319 --> 00:05:19.720
<v Speaker 1>not good. The Oklahoma City bombing trial verdict happened. That

107
00:05:19.959 --> 00:05:24.120
<v Speaker 1>was two years prior, nineteen ninety five, but the verdict

108
00:05:24.160 --> 00:05:28.759
<v Speaker 1>was handed over in ninety seven. There's this guy named

109
00:05:28.839 --> 00:05:33.199
<v Speaker 1>Andrew Cunanan who went on a killing spree and murdering

110
00:05:33.319 --> 00:05:38.120
<v Speaker 1>several people, including Gianni Versachi died by suicide after a

111
00:05:38.199 --> 00:05:43.720
<v Speaker 1>nationwide man hut. Happy Happy Happy Joy Joy Joy. Mother

112
00:05:43.839 --> 00:05:44.560
<v Speaker 1>Teresa died.

113
00:05:45.160 --> 00:05:48.680
<v Speaker 3>Yeah, I did I catch the most somber like look

114
00:05:48.759 --> 00:05:49.600
<v Speaker 3>back day or.

115
00:05:51.399 --> 00:05:54.800
<v Speaker 1>Yeah, it'll look up looking up I do the bad news.

116
00:05:54.920 --> 00:05:58.279
<v Speaker 1>Richard does the good news. Oh okay, the Heaven's Gate

117
00:05:58.399 --> 00:06:00.240
<v Speaker 1>mass suicide. Remember the Nikes.

118
00:06:01.000 --> 00:06:05.279
<v Speaker 2>Yeah yeah, they were web developers too, they were in JavaScript. Yeah,

119
00:06:05.360 --> 00:06:06.879
<v Speaker 2>I'm sorry I say that. A lot of that's normal.

120
00:06:06.959 --> 00:06:13.160
<v Speaker 1>Well, anyway, it was a you know, a cult, and

121
00:06:13.439 --> 00:06:17.279
<v Speaker 1>thirty nine members died by drinking poison or something and

122
00:06:17.319 --> 00:06:20.439
<v Speaker 1>going to sleep with their Nikes on. Yeah, Tiger Woods

123
00:06:20.439 --> 00:06:24.519
<v Speaker 1>won the Masters. That's good news, the youngest Masters champion.

124
00:06:24.879 --> 00:06:32.439
<v Speaker 1>So Mike Tyson, Vander Holyfield's ear off, jeez, Harry Potter

125
00:06:32.480 --> 00:06:36.160
<v Speaker 1>and the Philosopher's Stone came out, and yeah, what do

126
00:06:36.199 --> 00:06:39.560
<v Speaker 1>you got for technology and science?

127
00:06:39.800 --> 00:06:46.000
<v Speaker 2>Let's start in space. There are how many eight Shuttle

128
00:06:46.079 --> 00:06:50.000
<v Speaker 2>launches in ninety ninety seven, which is an unusually large amount.

129
00:06:50.079 --> 00:06:52.120
<v Speaker 2>Three of them are shuttled mirror missions. It just happens

130
00:06:52.120 --> 00:06:53.759
<v Speaker 2>to line up that way. You know, they're roughly every

131
00:06:53.800 --> 00:06:57.399
<v Speaker 2>six months. So there's Atlantis in January, again in May,

132
00:06:57.959 --> 00:07:01.000
<v Speaker 2>and then again in September. Flew all of the Shuttle

133
00:07:01.040 --> 00:07:03.199
<v Speaker 2>MIRR missions. And that's taking crew up, taking crew down,

134
00:07:03.240 --> 00:07:04.439
<v Speaker 2>and experiments and things like that.

135
00:07:04.879 --> 00:07:07.160
<v Speaker 3>I was living in Florida at the time. I caught

136
00:07:07.199 --> 00:07:10.800
<v Speaker 3>a couple of those. Yeah, from the parking lot at work,

137
00:07:10.920 --> 00:07:13.040
<v Speaker 3>you could you could see a little flicker in the

138
00:07:13.079 --> 00:07:14.639
<v Speaker 3>sky from where we lived, and.

139
00:07:14.639 --> 00:07:16.920
<v Speaker 2>Then it would zip away and be gone. Right. It's

140
00:07:16.959 --> 00:07:19.879
<v Speaker 2>like you wait a long time for about ninety seconds

141
00:07:19.920 --> 00:07:22.720
<v Speaker 2>of wow, look at that, and then it's over. The

142
00:07:22.759 --> 00:07:25.519
<v Speaker 2>most interesting thing on the shuttle side in ninety seven

143
00:07:25.759 --> 00:07:28.639
<v Speaker 2>is Columbia. So Columbia and April does a mission STS

144
00:07:28.720 --> 00:07:31.560
<v Speaker 2>eighty three, which is Space Lab mission, so they're just

145
00:07:31.600 --> 00:07:33.879
<v Speaker 2>going up doing experiments and so forth. Columbia didn't have

146
00:07:34.040 --> 00:07:38.279
<v Speaker 2>the power because it was so overweight to actually get

147
00:07:38.279 --> 00:07:41.000
<v Speaker 2>to mirror, so it didn't never had that option. But

148
00:07:41.079 --> 00:07:43.720
<v Speaker 2>there's a problem with the fuel cells on board Columbia,

149
00:07:43.720 --> 00:07:45.959
<v Speaker 2>and they cut the mission short to only three days

150
00:07:45.959 --> 00:07:48.319
<v Speaker 2>to turn around and come back, yeah, and then decide

151
00:07:48.319 --> 00:07:51.920
<v Speaker 2>to refly it. So they fixed the problem, refuel and

152
00:07:52.079 --> 00:07:53.920
<v Speaker 2>only a few months later in July fly it as

153
00:07:54.040 --> 00:07:58.199
<v Speaker 2>STS ninety four, same crew, So three months turnaround fairly unusual,

154
00:07:58.399 --> 00:08:02.879
<v Speaker 2>wow for shuttles. They usually take longer than that. On

155
00:08:02.959 --> 00:08:05.519
<v Speaker 2>the Mere Space Station, they was a very im in

156
00:08:05.560 --> 00:08:07.160
<v Speaker 2>a full year. Not a good one either, you tell

157
00:08:07.480 --> 00:08:11.360
<v Speaker 2>me all good news. So in February a fire in space.

158
00:08:11.879 --> 00:08:14.759
<v Speaker 2>So one of the ways that Mirror maintained its oxygen

159
00:08:14.800 --> 00:08:18.199
<v Speaker 2>levels is they actually burned lithium per chlorate. This is

160
00:08:18.199 --> 00:08:22.160
<v Speaker 2>something they do in submarines as well, because the emissions

161
00:08:22.199 --> 00:08:26.720
<v Speaker 2>of burning that is straight oxygen. Wow. Well, something went

162
00:08:26.839 --> 00:08:29.920
<v Speaker 2>wrong and it caught fire. Yeah, that's kind of kind

163
00:08:29.959 --> 00:08:32.120
<v Speaker 2>of flammable. Yeah, it's dangerous, but they did it all

164
00:08:32.159 --> 00:08:34.639
<v Speaker 2>the time. But something went wrong and it took They

165
00:08:34.720 --> 00:08:36.639
<v Speaker 2>got the fire out in ninety seconds, because fire in

166
00:08:36.679 --> 00:08:39.799
<v Speaker 2>the space station is very, very serious. Yeah, but it

167
00:08:39.879 --> 00:08:42.519
<v Speaker 2>took minutes to get the smoke cleared out, so it's

168
00:08:42.519 --> 00:08:44.240
<v Speaker 2>a question of whether it was breathing or not. And

169
00:08:44.360 --> 00:08:47.120
<v Speaker 2>the reality is if they evacuated, probably never never be

170
00:08:47.159 --> 00:08:49.039
<v Speaker 2>able to get back into it again. So that was

171
00:08:49.159 --> 00:08:53.600
<v Speaker 2>very scary. Quickly overwhelmed in June when M threety four,

172
00:08:53.639 --> 00:08:58.039
<v Speaker 2>which was a progressed cargo supply emission, did an experimental

173
00:08:58.240 --> 00:09:04.519
<v Speaker 2>approach for docking, lost control and collided with the spector module,

174
00:09:05.120 --> 00:09:08.480
<v Speaker 2>damaging the solar panels and breaching the hull of the

175
00:09:08.559 --> 00:09:12.639
<v Speaker 2>spector module. The crew. It was a slow leak, so

176
00:09:12.679 --> 00:09:15.559
<v Speaker 2>the crew had minutes to do something about this, not seconds,

177
00:09:15.600 --> 00:09:17.720
<v Speaker 2>which was good because they needed to close off the

178
00:09:17.720 --> 00:09:20.320
<v Speaker 2>hatch on the Spector module. The Spector module having four

179
00:09:20.440 --> 00:09:23.480
<v Speaker 2>very large solar panels. The power cables that put that

180
00:09:23.519 --> 00:09:25.759
<v Speaker 2>power into the rest of the station actually ran through

181
00:09:25.799 --> 00:09:29.919
<v Speaker 2>the hatch, so they had to disconnect them without electrotying themselves,

182
00:09:30.320 --> 00:09:33.039
<v Speaker 2>push them into the Spector module, and close up the

183
00:09:33.120 --> 00:09:36.080
<v Speaker 2>hatch to maintain air pressure, which they succeeded in doing.

184
00:09:36.519 --> 00:09:38.440
<v Speaker 2>A couple of months later in August, they will have

185
00:09:38.480 --> 00:09:41.519
<v Speaker 2>an internal spacewalk, so they'll put on the space suits

186
00:09:42.080 --> 00:09:45.240
<v Speaker 2>and reopened that hatch, find that the computers and things

187
00:09:45.279 --> 00:09:48.840
<v Speaker 2>are all running fine in vacuum, and replace the hatch

188
00:09:48.879 --> 00:09:50.879
<v Speaker 2>with one that had the power line passed through so

189
00:09:50.919 --> 00:09:53.360
<v Speaker 2>they can hook the power back up again, and then

190
00:09:53.399 --> 00:09:55.840
<v Speaker 2>eventually in October to another one of those internal spacewatch

191
00:09:55.879 --> 00:09:58.559
<v Speaker 2>to add in remote control on some things and so forth.

192
00:09:58.600 --> 00:10:01.120
<v Speaker 2>Spector will never be repressure. It used to be the

193
00:10:01.159 --> 00:10:04.720
<v Speaker 2>place that the Americans slept in for the rest of Mirror,

194
00:10:04.759 --> 00:10:07.399
<v Speaker 2>which of course is already two thousand and one Yanks.

195
00:10:07.879 --> 00:10:10.799
<v Speaker 2>One more bad news on the road on the rocket side,

196
00:10:10.840 --> 00:10:13.559
<v Speaker 2>and that is a Delta two failure. This is only

197
00:10:13.840 --> 00:10:17.759
<v Speaker 2>I only mention this because Delta two had literally hundreds

198
00:10:17.799 --> 00:10:20.480
<v Speaker 2>of successful launches. It was a non event for a

199
00:10:20.519 --> 00:10:23.600
<v Speaker 2>Delta tow to fly like Falcon is today. This was

200
00:10:23.639 --> 00:10:28.039
<v Speaker 2>a GPS satellite and twelve seconds into the launch, one

201
00:10:28.039 --> 00:10:32.600
<v Speaker 2>of the solid rocket boosters exploded, blew through the tank

202
00:10:32.679 --> 00:10:36.039
<v Speaker 2>and destroyed the entire vehicle. But it was almost entirely fueled,

203
00:10:36.080 --> 00:10:38.559
<v Speaker 2>so you can find a video clip of this. It

204
00:10:38.679 --> 00:10:42.960
<v Speaker 2>is the most spectacular failure you can imagine. Wow, And

205
00:10:43.000 --> 00:10:44.919
<v Speaker 2>it was a reminder of why they have what they

206
00:10:44.960 --> 00:10:49.960
<v Speaker 2>call blockhouses, which are these reinforced concrete boxes that the

207
00:10:50.120 --> 00:10:53.080
<v Speaker 2>folks actually operating the mission work from on the pad

208
00:10:53.480 --> 00:10:57.960
<v Speaker 2>because debris rained down everywhere it was only twelve seconds

209
00:10:57.960 --> 00:11:01.080
<v Speaker 2>into its flight, including setting fire to cars in the

210
00:11:01.120 --> 00:11:07.360
<v Speaker 2>parking lot. Like geez. It was quite an unusual event. Okay,

211
00:11:07.399 --> 00:11:11.679
<v Speaker 2>some cooler stuff. The Mars Pathfinder mission launched in ninety six.

212
00:11:11.840 --> 00:11:14.799
<v Speaker 2>It lands in ninety seven. This is the Pathfire lander

213
00:11:14.840 --> 00:11:17.799
<v Speaker 2>which used the air bag landing systems. When they pressurize,

214
00:11:17.799 --> 00:11:20.039
<v Speaker 2>the airbag bounced to a stop and then it unfolds

215
00:11:20.080 --> 00:11:23.879
<v Speaker 2>itself and inside the Pathfinder lander. This so jurner rover

216
00:11:24.639 --> 00:11:27.039
<v Speaker 2>about the size of a shoe box, not very big,

217
00:11:27.279 --> 00:11:29.200
<v Speaker 2>but it was the first rover to ever move around

218
00:11:29.240 --> 00:11:31.480
<v Speaker 2>another planet. Now to be clear, not the first rover.

219
00:11:31.600 --> 00:11:36.279
<v Speaker 2>The Soviets put rovers on the Moon, you know planet.

220
00:11:36.279 --> 00:11:39.480
<v Speaker 2>Thing supposed to operate for thirty days, lasted eighty three days.

221
00:11:39.600 --> 00:11:43.159
<v Speaker 2>Very successful mission and started. They called the Pathfinder reason

222
00:11:43.200 --> 00:11:47.159
<v Speaker 2>it was a path to lower cost missions to Mars.

223
00:11:48.279 --> 00:11:52.679
<v Speaker 2>Cassini launches. This is the huge bus size spacecraft going

224
00:11:52.759 --> 00:11:56.120
<v Speaker 2>to Saturn. Launches on a Titan four B. It won't

225
00:11:56.120 --> 00:11:57.919
<v Speaker 2>get to Saturn until two thousand and four. Will be

226
00:11:57.919 --> 00:12:00.399
<v Speaker 2>the first, i think, the only thing that ever orbit

227
00:12:00.440 --> 00:12:03.320
<v Speaker 2>around Saturn and also will end the Huygens prob on Titan.

228
00:12:03.919 --> 00:12:08.320
<v Speaker 2>And finally the comet hail Bob so Alan Hale and

229
00:12:08.360 --> 00:12:13.879
<v Speaker 2>Thomas Bop both amateurs, find this astro this commet in

230
00:12:13.960 --> 00:12:17.759
<v Speaker 2>nineteen ninety five and do some calculations and realize it, Well,

231
00:12:18.000 --> 00:12:20.759
<v Speaker 2>it's a huge, absolute massive comet. Good thing he didn't

232
00:12:20.799 --> 00:12:23.200
<v Speaker 2>hit the Earth because it got quite close. It'll be

233
00:12:23.320 --> 00:12:26.519
<v Speaker 2>visible to the naked eye for eighteen months. He reached

234
00:12:26.559 --> 00:12:29.919
<v Speaker 2>his peak brightness in nineteen ninety in April ninety seven. Yeah,

235
00:12:29.960 --> 00:12:32.480
<v Speaker 2>and they calculated its orbital period at twenty five one

236
00:12:32.600 --> 00:12:34.759
<v Speaker 2>hundred years. So hang on to your hats. It'll be

237
00:12:34.799 --> 00:12:35.960
<v Speaker 2>back in forty three eighty five.

238
00:12:36.919 --> 00:12:40.559
<v Speaker 1>I remember that very clearly. In fact, I was speaking

239
00:12:40.720 --> 00:12:44.840
<v Speaker 1>at a Microsoft event in Paris and we were at

240
00:12:44.840 --> 00:12:47.919
<v Speaker 1>euro Disney and they shut down the whole park and

241
00:12:48.000 --> 00:12:51.000
<v Speaker 1>it was just hanging there over the roller coaster.

242
00:12:50.840 --> 00:12:54.480
<v Speaker 2>Just to watch this amazing thing, right, it was so cool. Yeah.

243
00:12:54.559 --> 00:12:56.679
<v Speaker 1>Yeah, And my daughter who was too got to see

244
00:12:56.720 --> 00:12:58.039
<v Speaker 1>it at too at that point.

245
00:12:58.159 --> 00:13:00.840
<v Speaker 2>So on the tech side of things, the web is

246
00:13:00.879 --> 00:13:05.000
<v Speaker 2>moving along nicely. The HTPP one point one specification is released.

247
00:13:06.519 --> 00:13:10.240
<v Speaker 2>Two guys Sergei Brandon Larry Page, who have a search

248
00:13:10.279 --> 00:13:13.879
<v Speaker 2>engine called BackRub, register a new domain Google.

249
00:13:14.159 --> 00:13:14.879
<v Speaker 1>Yeah, I heard of it.

250
00:13:15.080 --> 00:13:18.399
<v Speaker 2>The Wall Street Journal introduces the very first paywall onto

251
00:13:18.399 --> 00:13:23.600
<v Speaker 2>the Internet, and the Macromedia ships dream Weaver.

252
00:13:24.320 --> 00:13:25.679
<v Speaker 1>Yeah, fun memory.

253
00:13:25.480 --> 00:13:27.799
<v Speaker 2>Aquired by Adobe in two thousand and five. Intell raises

254
00:13:27.799 --> 00:13:30.759
<v Speaker 2>the Pentium two seventy seven point five million transistors depending

255
00:13:30.799 --> 00:13:34.960
<v Speaker 2>pro was only five point five so obviously progress. Rico

256
00:13:35.120 --> 00:13:38.720
<v Speaker 2>releases the CDRW the rewriteable CD.

257
00:13:39.080 --> 00:13:40.360
<v Speaker 1>Yeah, they didn't work too well.

258
00:13:40.799 --> 00:13:43.480
<v Speaker 2>There are problems with original CD readers because the reflectivity

259
00:13:43.559 --> 00:13:45.759
<v Speaker 2>roll this is relatively low, and eventually newer models have

260
00:13:45.799 --> 00:13:48.919
<v Speaker 2>a thing called MultiRead so they can use CDRWS properly

261
00:13:49.000 --> 00:13:53.440
<v Speaker 2>rewriteable one hundred thousand dollars. The I Tripoli introduces the

262
00:13:53.440 --> 00:13:56.559
<v Speaker 2>specification eight O two dot one you know it as

263
00:13:56.720 --> 00:14:01.159
<v Speaker 2>Wi Fi. Nobody cares until couple of years later when

264
00:14:01.279 --> 00:14:04.519
<v Speaker 2>the Modifkai did the specification called eight two dot one

265
00:14:04.519 --> 00:14:07.279
<v Speaker 2>one B. Yeah, it said a lot less expensive implement.

266
00:14:07.360 --> 00:14:10.679
<v Speaker 2>That's in ninety nine really, so of win amp. It

267
00:14:10.759 --> 00:14:14.879
<v Speaker 2>whips the lamos ass. That's Justin Frankel and Dmitri Boldiev

268
00:14:14.960 --> 00:14:18.200
<v Speaker 2>from Nelsoft make an MP three player for your PC.

269
00:14:18.600 --> 00:14:20.360
<v Speaker 1>The plugins were the thing I remember.

270
00:14:20.440 --> 00:14:22.600
<v Speaker 2>Yeah, they built a real ecosystem around it was.

271
00:14:22.679 --> 00:14:26.879
<v Speaker 1>I remember Jeff macy Lex showed me milk Drop, which

272
00:14:26.960 --> 00:14:30.559
<v Speaker 1>was a graphical, amazing graphic plug in for win Amp.

273
00:14:31.159 --> 00:14:33.200
<v Speaker 1>I used to It's just the coolest thing I ever saw.

274
00:14:33.639 --> 00:14:39.559
<v Speaker 3>Used to let the visualizers run on the monitor. Super psychedelic.

275
00:14:40.039 --> 00:14:40.279
<v Speaker 1>Yep.

276
00:14:41.279 --> 00:14:44.200
<v Speaker 2>Just give it a look and let's see what else.

277
00:14:44.240 --> 00:14:45.960
<v Speaker 2>A couple a couple of other good ones. Oh yeah,

278
00:14:45.960 --> 00:14:47.720
<v Speaker 2>they'll sell the AOL for in nineteen nine nine for

279
00:14:47.720 --> 00:14:49.559
<v Speaker 2>eighty mil which is good play because it wouldn't be

280
00:14:49.600 --> 00:14:53.360
<v Speaker 2>worth that much later. So world done. This is the

281
00:14:53.440 --> 00:14:57.000
<v Speaker 2>year that IBM's Deep Blue after two years of trying

282
00:14:57.080 --> 00:15:00.480
<v Speaker 2>beats Casper off. Ye, it's an RS six thousands computer,

283
00:15:01.080 --> 00:15:04.519
<v Speaker 2>twelve gigafaughts of processing power, which is exactly the same

284
00:15:04.519 --> 00:15:07.279
<v Speaker 2>as an iPhone seven. A few years later, and on

285
00:15:07.320 --> 00:15:11.000
<v Speaker 2>the Microsoft side, Visual Studio the very first version, which

286
00:15:11.000 --> 00:15:14.399
<v Speaker 2>of course included VB five, they announced Windows ninety eight

287
00:15:15.240 --> 00:15:17.960
<v Speaker 2>and Bill Gates becomes the world's richest man for the

288
00:15:18.000 --> 00:15:21.840
<v Speaker 2>first time, and in probably a PR bit of a

289
00:15:21.879 --> 00:15:26.080
<v Speaker 2>pr blender shows up on a giant screen at Macworld

290
00:15:26.159 --> 00:15:28.519
<v Speaker 2>to announce that Microsoft will invest one hundred and fifty

291
00:15:28.559 --> 00:15:31.159
<v Speaker 2>million dollars to Apple, and people freak out.

292
00:15:31.279 --> 00:15:31.919
<v Speaker 1>Yeah, it was.

293
00:15:31.919 --> 00:15:35.440
<v Speaker 3>About the same time I started building my first Windows

294
00:15:35.480 --> 00:15:40.559
<v Speaker 3>desktop application. There you go in VB and wind forms, and.

295
00:15:40.559 --> 00:15:43.240
<v Speaker 2>I'll mention two games, the first version of Grand Theft Auto,

296
00:15:43.240 --> 00:15:45.240
<v Speaker 2>which is sort of an overhead perspective game which is

297
00:15:45.279 --> 00:15:49.639
<v Speaker 2>still killed people and stole cars. And Ultima Online, one

298
00:15:49.679 --> 00:15:53.000
<v Speaker 2>of the first graphical, massive multiplayer online games.

299
00:15:53.519 --> 00:15:56.320
<v Speaker 1>All Right, I'm just gonna we're running out of time

300
00:15:56.360 --> 00:15:58.279
<v Speaker 1>for the beginning here, so I'm just going to read

301
00:15:58.360 --> 00:16:03.320
<v Speaker 1>off the top ten albums of nineteen ninety seven. Radiohead, Ok, Computer,

302
00:16:04.600 --> 00:16:09.879
<v Speaker 1>The Verve, Urban Hymns, The Prodigy, The Fat of the Land, Puff,

303
00:16:09.960 --> 00:16:14.840
<v Speaker 1>Daddy in the Family, No Way Out, Garth Brooks sevens,

304
00:16:15.240 --> 00:16:19.519
<v Speaker 1>Spice Girls, Spice and that was continuing the massive global

305
00:16:19.559 --> 00:16:25.399
<v Speaker 1>sales because it was there last year too. Andrea bos

306
00:16:25.480 --> 00:16:32.639
<v Speaker 1>Shelley Ramanza, the Titanic music from the motion picture Celine Dion,

307
00:16:32.759 --> 00:16:37.799
<v Speaker 1>Let's talk about love and Shania Twain come on over top.

308
00:16:37.840 --> 00:16:42.399
<v Speaker 1>Ten movies of nineteen ninety seven, Hercules, The Fifth Element,

309
00:16:43.399 --> 00:16:47.600
<v Speaker 1>My Best Friend's Wedding Liar, Liar, Oh my god, that

310
00:16:47.679 --> 00:16:51.559
<v Speaker 1>was hilarious. Great movie, as good as it gets. Air

311
00:16:51.600 --> 00:16:57.360
<v Speaker 1>Force One, Tomorrow Never Dies, Men in Black, The Lost World,

312
00:16:57.480 --> 00:17:03.320
<v Speaker 1>Jurassic Park, and titan As. If you didn't know, Yeah,

313
00:17:04.079 --> 00:17:06.160
<v Speaker 1>all right with that, let's roll the music for a

314
00:17:06.160 --> 00:17:07.000
<v Speaker 1>better no framework.

315
00:17:07.119 --> 00:17:16.599
<v Speaker 2>Awesome, man, what do you got? It's funny you were

316
00:17:16.599 --> 00:17:17.519
<v Speaker 2>talking about.

317
00:17:17.279 --> 00:17:21.519
<v Speaker 1>Games because I went looking in the you know, the

318
00:17:21.960 --> 00:17:25.240
<v Speaker 1>trending repos on GitHub, and one of them is for

319
00:17:25.279 --> 00:17:30.839
<v Speaker 1>a thing called robust Toolbox, which is a homegrown gaming

320
00:17:31.000 --> 00:17:35.039
<v Speaker 1>engine written in C sharp, and that led me to

321
00:17:35.599 --> 00:17:38.720
<v Speaker 1>the game that they wrote it for space Station fourteen,

322
00:17:39.880 --> 00:17:44.160
<v Speaker 1>which is a remake of space Station thirteen, but it

323
00:17:44.279 --> 00:17:46.440
<v Speaker 1>runs in C sharp on Robust Toolbox. So then I

324
00:17:46.480 --> 00:17:48.839
<v Speaker 1>went looking for Space Station thirteen to see what that

325
00:17:48.920 --> 00:17:52.400
<v Speaker 1>was all about. And it's an overhead view kind of

326
00:17:53.720 --> 00:17:57.039
<v Speaker 1>you know, survive and thrive in space. You know, all

327
00:17:57.039 --> 00:18:01.000
<v Speaker 1>sorts of problems happen. I don't know exactly what, but

328
00:18:01.079 --> 00:18:02.960
<v Speaker 1>in the little demo that they were doing, there was

329
00:18:03.240 --> 00:18:07.359
<v Speaker 1>digital blood on the floor, so something nasty must have happened.

330
00:18:08.880 --> 00:18:11.680
<v Speaker 1>It looks fun. But the cool thing is that this

331
00:18:12.279 --> 00:18:16.920
<v Speaker 1>Robust Toolbox, the homegrown engine written in C sharp that

332
00:18:17.079 --> 00:18:21.680
<v Speaker 1>was primarily for Space Station fourteen, is released as its

333
00:18:21.720 --> 00:18:28.839
<v Speaker 1>own thing. So spetistration. Yeah, cool. So that's what I got.

334
00:18:28.880 --> 00:18:30.039
<v Speaker 1>Who's talking to us today?

335
00:18:30.079 --> 00:18:32.799
<v Speaker 2>Richard grabbed a commental for show nineteen ninety one, The

336
00:18:32.799 --> 00:18:35.359
<v Speaker 2>One We Do Andrew Murphy talking about leading teams in

337
00:18:35.359 --> 00:18:37.000
<v Speaker 2>the time of AI. I got a lot of reaction

338
00:18:37.119 --> 00:18:38.960
<v Speaker 2>on this show, and I'm going to read one of

339
00:18:38.960 --> 00:18:40.599
<v Speaker 2>the comments. Here's from Adam right. He says, I'm in

340
00:18:40.599 --> 00:18:42.000
<v Speaker 2>the middle of listening to this episode and I feel

341
00:18:42.039 --> 00:18:44.920
<v Speaker 2>a bit conflicted. It's possible on going through the grief process.

342
00:18:44.960 --> 00:18:47.119
<v Speaker 2>As Andrew mentioned, there are quite a few predictions about

343
00:18:47.119 --> 00:18:50.079
<v Speaker 2>the future based on this arguably short time frame in

344
00:18:50.079 --> 00:18:52.799
<v Speaker 2>which these AI tools have been somewhat useful. Richard asked

345
00:18:52.839 --> 00:18:54.480
<v Speaker 2>at one point, who are the people that aren't adopting

346
00:18:54.480 --> 00:18:56.279
<v Speaker 2>this technology yet? And I would suggest that it's the

347
00:18:56.359 --> 00:18:58.920
<v Speaker 2>late adopters. There are sets of developers and companies that

348
00:18:58.960 --> 00:19:00.960
<v Speaker 2>don't hop on every band in the first few years

349
00:19:00.960 --> 00:19:02.559
<v Speaker 2>of its use. It takes time for new tech to

350
00:19:02.599 --> 00:19:06.000
<v Speaker 2>be overwhelmingly adopted, absolutely true. Aside from that grip, there's

351
00:19:06.079 --> 00:19:07.960
<v Speaker 2>likely another grip that cares about the potential and actual

352
00:19:07.960 --> 00:19:11.799
<v Speaker 2>sociological impacts of technology in the real world. You've seen

353
00:19:11.839 --> 00:19:13.880
<v Speaker 2>the price of computer components lately. One of my favorite

354
00:19:13.880 --> 00:19:16.160
<v Speaker 2>YouTube channels, Gamers Next is freakily calls out the RAM

355
00:19:16.240 --> 00:19:18.839
<v Speaker 2>shortage as being due to the reservation of silicon waivers

356
00:19:18.960 --> 00:19:21.039
<v Speaker 2>don't exist to be used in GPUs that also don't

357
00:19:21.039 --> 00:19:25.039
<v Speaker 2>exist for AI tosay centers that don't exist. I think

358
00:19:25.079 --> 00:19:28.160
<v Speaker 2>it's undeniable that developers using this technology are literally training

359
00:19:28.200 --> 00:19:31.599
<v Speaker 2>their own replacement. Maybe in some sense it's more like

360
00:19:31.640 --> 00:19:33.759
<v Speaker 2>sharpening an axe, but I think more of us should

361
00:19:33.759 --> 00:19:35.599
<v Speaker 2>be at least concerned about where the acts will land.

362
00:19:36.680 --> 00:19:38.440
<v Speaker 2>The best hope for the technologies. Rigidal lid out in

363
00:19:38.440 --> 00:19:40.079
<v Speaker 2>one of his talks. Is the models get smaller and

364
00:19:40.079 --> 00:19:41.960
<v Speaker 2>more specialized and work well enough that we put more

365
00:19:41.960 --> 00:19:43.759
<v Speaker 2>trust in them. In my opinion, that would be the

366
00:19:43.799 --> 00:19:46.400
<v Speaker 2>time when mass adoption would be warranted. Perhaps it will

367
00:19:46.400 --> 00:19:48.400
<v Speaker 2>take the current enthusiasts working out the flaws. Do you

368
00:19:48.440 --> 00:19:51.839
<v Speaker 2>already get there? Though? Thanks for the thought promoting episode.

369
00:19:52.160 --> 00:19:54.400
<v Speaker 2>I mean, it's a whole conversation coming around now about

370
00:19:54.440 --> 00:19:57.240
<v Speaker 2>this idea that the job wasn't to write the code anyway,

371
00:19:57.440 --> 00:20:01.880
<v Speaker 2>it was to provide solutions to customers. And you know,

372
00:20:01.920 --> 00:20:04.559
<v Speaker 2>the solution is much about data collection, is it is

373
00:20:04.599 --> 00:20:07.799
<v Speaker 2>about implementation, so that we have tools that will do

374
00:20:07.839 --> 00:20:10.400
<v Speaker 2>implementation is an important part of the equation. It just

375
00:20:10.519 --> 00:20:11.880
<v Speaker 2>makes us all the faster.

376
00:20:11.839 --> 00:20:14.359
<v Speaker 1>And also guiding of the AI. You know that to

377
00:20:15.119 --> 00:20:18.319
<v Speaker 1>achieve the goal that the customer wants, that's not an

378
00:20:18.400 --> 00:20:19.000
<v Speaker 1>easy task.

379
00:20:19.079 --> 00:20:22.039
<v Speaker 3>I think we're also seeing it drive demand. So when

380
00:20:22.319 --> 00:20:24.960
<v Speaker 3>demand increases, how much software hasn't been built because the

381
00:20:24.960 --> 00:20:28.960
<v Speaker 3>barrier was too high. Yeah, So now we're seeing you know,

382
00:20:29.160 --> 00:20:33.440
<v Speaker 3>more work because there's more demand. So that's keeping things

383
00:20:33.480 --> 00:20:35.440
<v Speaker 3>a little bit more stable than people predicted.

384
00:20:35.480 --> 00:20:38.960
<v Speaker 2>I mean, certainly from the machine learning perspective, this was

385
00:20:39.000 --> 00:20:42.160
<v Speaker 2>true for radiologists you know when the first image models

386
00:20:42.200 --> 00:20:44.880
<v Speaker 2>really got popular, and there's like seven hundred proved by

387
00:20:44.880 --> 00:20:49.119
<v Speaker 2>the FDA now for doing recognition. I think even hinted

388
00:20:49.200 --> 00:20:51.400
<v Speaker 2>in self count said you're a fool if you're a radiologist.

389
00:20:51.440 --> 00:20:53.559
<v Speaker 2>That job is over. And in reality, of course, a

390
00:20:53.599 --> 00:20:55.319
<v Speaker 2>demand for medical energy is gone through the roof, and

391
00:20:55.319 --> 00:20:57.519
<v Speaker 2>the demand for rideologists has gone through the roof, right

392
00:20:57.559 --> 00:21:00.039
<v Speaker 2>because they can move faster with these new tools and

393
00:21:00.079 --> 00:21:02.400
<v Speaker 2>there's far more people to test. So I think that

394
00:21:02.519 --> 00:21:04.160
<v Speaker 2>cycle is going to go on for quite some time.

395
00:21:04.200 --> 00:21:06.799
<v Speaker 2>And it just seems like, you know, history doesn't necessary

396
00:21:06.799 --> 00:21:08.559
<v Speaker 2>repeat itself, and I'm talking about the history of like

397
00:21:08.680 --> 00:21:12.000
<v Speaker 2>five years ago, but it does rhyme, and I think,

398
00:21:12.039 --> 00:21:15.160
<v Speaker 2>you know, you're onto something there to ed with. There

399
00:21:15.400 --> 00:21:18.400
<v Speaker 2>is increasing demand. There's more kinds of software that need

400
00:21:18.440 --> 00:21:20.200
<v Speaker 2>to be built. I mean, get into this sort of

401
00:21:20.240 --> 00:21:22.279
<v Speaker 2>sense of are we all going to surround ourselves with

402
00:21:22.319 --> 00:21:24.240
<v Speaker 2>our own custom software? Because I know a lot of

403
00:21:24.240 --> 00:21:26.240
<v Speaker 2>developers are going to get with these tools, that's what

404
00:21:26.279 --> 00:21:29.160
<v Speaker 2>they're doing. I already do that, and where the developers go,

405
00:21:29.359 --> 00:21:29.920
<v Speaker 2>others go.

406
00:21:30.119 --> 00:21:30.359
<v Speaker 1>Yeah.

407
00:21:30.440 --> 00:21:31.880
<v Speaker 2>So Adam, thank you so much for your comment. In

408
00:21:31.880 --> 00:21:33.319
<v Speaker 2>to copy of music coba Is on its way to you,

409
00:21:33.359 --> 00:21:34.680
<v Speaker 2>and if you'd like a copy of music Cobe, I

410
00:21:34.680 --> 00:21:36.559
<v Speaker 2>read a comment on the website at dot net rockst'll

411
00:21:36.559 --> 00:21:38.160
<v Speaker 2>comment on the facebooks. We publish every show there, and

412
00:21:38.160 --> 00:21:39.480
<v Speaker 2>if you comment there and ever read it on the show,

413
00:21:39.559 --> 00:21:40.759
<v Speaker 2>we'll send you copy of music Cobo.

414
00:21:40.839 --> 00:21:42.640
<v Speaker 1>And if you can't wait for that, go to music

415
00:21:42.680 --> 00:21:44.759
<v Speaker 1>tocode by dot net and you can get the whole

416
00:21:44.759 --> 00:21:48.359
<v Speaker 1>collection in MP three wave and flat formats. Okay, we've

417
00:21:48.400 --> 00:21:50.440
<v Speaker 1>wasted enough time, and it's not a waste of time,

418
00:21:50.480 --> 00:21:54.640
<v Speaker 1>of course, let's bring on Ed. I'll officially introduce him.

419
00:21:54.759 --> 00:21:58.960
<v Speaker 1>Ed Scharboneau is a principle developer advocate for Progress Software,

420
00:21:59.359 --> 00:22:03.279
<v Speaker 1>a ten times Microsoft MVP, and video author on dome Train.

421
00:22:04.039 --> 00:22:06.640
<v Speaker 1>Ed works at the intersection of modern dot net developer

422
00:22:06.680 --> 00:22:11.400
<v Speaker 1>experience and artificial intelligence at Progress I'm Sorry, Richard Progress

423
00:22:11.559 --> 00:22:15.680
<v Speaker 1>Nice Ed helped Ed help drive early work around Blazer

424
00:22:16.000 --> 00:22:20.480
<v Speaker 1>and the creation of telleric for TELLERICUI for Blazer, helping

425
00:22:20.599 --> 00:22:24.519
<v Speaker 1>bring modern dot net web development to a broader developer audience.

426
00:22:25.079 --> 00:22:28.039
<v Speaker 1>More recently, his work has focused on AI and agentic

427
00:22:28.119 --> 00:22:33.559
<v Speaker 1>RAG systems, where he helps guide product strategy, evaluate emerging technologies,

428
00:22:33.920 --> 00:22:38.119
<v Speaker 1>and leads development of developer tools that bring advanced AI

429
00:22:38.319 --> 00:22:41.839
<v Speaker 1>capabilities into the dot net ecosystem. I can't wait to

430
00:22:41.880 --> 00:22:42.880
<v Speaker 1>hear what you got to say.

431
00:22:43.039 --> 00:22:43.119
<v Speaker 3>Ed.

432
00:22:43.559 --> 00:22:44.839
<v Speaker 1>Welcome, Thanks for having me.

433
00:22:45.039 --> 00:22:47.680
<v Speaker 3>Appreciate you guys having me on. I've been a longtime

434
00:22:47.720 --> 00:22:51.720
<v Speaker 3>listener and we've all bumped into each other a few

435
00:22:51.720 --> 00:22:53.000
<v Speaker 3>times and it's been a pleasure.

436
00:22:53.200 --> 00:22:54.279
<v Speaker 1>Sure, yeah, sure, you.

437
00:22:54.279 --> 00:22:56.400
<v Speaker 2>Brought me a lovely bottle of whiskey. I think it

438
00:22:56.400 --> 00:23:00.319
<v Speaker 2>was last year actually from the was it red ry?

439
00:23:00.880 --> 00:23:01.000
<v Speaker 1>Well?

440
00:23:01.079 --> 00:23:06.039
<v Speaker 3>That was a JEP decreed red bloody, bloody butcher corn,

441
00:23:07.759 --> 00:23:11.480
<v Speaker 3>red corn. Yeah, it was a weeded bourbon. But it's

442
00:23:11.519 --> 00:23:14.920
<v Speaker 3>absolutely one of my favorites. And it's gotten. It's got

443
00:23:14.960 --> 00:23:17.559
<v Speaker 3>an odd nose on it. It's it smells a little

444
00:23:18.279 --> 00:23:19.039
<v Speaker 3>cologne like.

445
00:23:19.279 --> 00:23:21.640
<v Speaker 2>It's threatening on the nose. But boy, was good in

446
00:23:21.680 --> 00:23:25.119
<v Speaker 2>the mouth. And I don't have it anymore. I drank it.

447
00:23:25.599 --> 00:23:27.960
<v Speaker 1>Okay, that's the best testimony right there.

448
00:23:28.079 --> 00:23:31.480
<v Speaker 2>Yeah, all right, what have you been working on? My friend?

449
00:23:31.519 --> 00:23:33.319
<v Speaker 2>You've been up to stuff. I get that.

450
00:23:33.920 --> 00:23:40.359
<v Speaker 3>I've been been a busy, busy person, so I I

451
00:23:40.480 --> 00:23:42.960
<v Speaker 3>love to be on the bleeding edge of of all

452
00:23:43.119 --> 00:23:47.200
<v Speaker 3>tech all the time. It's, you know, part of being

453
00:23:47.240 --> 00:23:49.599
<v Speaker 3>a developer advocate I think is keeping your ear to

454
00:23:49.640 --> 00:23:54.279
<v Speaker 3>the ground. But over the years, I've done a lot

455
00:23:54.319 --> 00:23:57.599
<v Speaker 3>of a lot of technical sessions, and I love this

456
00:23:57.720 --> 00:24:01.759
<v Speaker 3>quote by H. G from War of the Worlds that

457
00:24:01.359 --> 00:24:05.359
<v Speaker 3>I that I always use that kind of describes what

458
00:24:05.599 --> 00:24:08.920
<v Speaker 3>drives me to keep up with all of these technologies.

459
00:24:08.920 --> 00:24:11.000
<v Speaker 3>And I don't know if you're familiar with the beginning

460
00:24:11.039 --> 00:24:15.039
<v Speaker 3>of War of the World's, but it's with infinite complacency.

461
00:24:15.160 --> 00:24:18.519
<v Speaker 3>Men went to and fro over this globe about their

462
00:24:18.519 --> 00:24:22.920
<v Speaker 3>little affairs, affairs, serene in their assurance of their empire

463
00:24:23.000 --> 00:24:26.920
<v Speaker 3>over matter, and the bit about complacency always got me.

464
00:24:27.720 --> 00:24:31.880
<v Speaker 3>It's like, even when HD Wells was writing about this,

465
00:24:32.599 --> 00:24:35.799
<v Speaker 3>you kind of realize that people go about their day

466
00:24:36.079 --> 00:24:38.119
<v Speaker 3>and put their head in the sand, and we get

467
00:24:38.119 --> 00:24:41.319
<v Speaker 3>in these patterns of doing the same thing day in

468
00:24:41.359 --> 00:24:44.599
<v Speaker 3>and day out, and we might lose sight of things

469
00:24:44.599 --> 00:24:46.680
<v Speaker 3>that are happening around us that we should be watching

470
00:24:46.680 --> 00:24:50.839
<v Speaker 3>out for. And with my career, it's always been what's

471
00:24:50.920 --> 00:24:54.720
<v Speaker 3>the technology that's going to come take my food away?

472
00:24:54.960 --> 00:24:55.960
<v Speaker 1>Right? Yeah, what's the.

473
00:24:55.920 --> 00:24:58.279
<v Speaker 3>Next big thing that's coming. I need to pay attention to.

474
00:24:58.440 --> 00:24:59.920
<v Speaker 2>Who's going to move your chee?

475
00:25:00.160 --> 00:25:00.559
<v Speaker 1>That's it.

476
00:25:01.799 --> 00:25:03.680
<v Speaker 3>That's kind of what's driven me over the year. So

477
00:25:03.759 --> 00:25:08.599
<v Speaker 3>I think that quote I started using when responsive Web

478
00:25:08.680 --> 00:25:12.279
<v Speaker 3>came out and I started talking about responsive web and

479
00:25:12.319 --> 00:25:15.400
<v Speaker 3>how people needed to watch out for the mobile factor

480
00:25:15.599 --> 00:25:20.640
<v Speaker 3>of doing web development and everything's going to need to adapt.

481
00:25:20.839 --> 00:25:22.720
<v Speaker 2>Yeah, the mobile funder, it wasn't just a mobile finder,

482
00:25:23.160 --> 00:25:25.759
<v Speaker 2>it was a tablets. We were happily doing our m

483
00:25:25.799 --> 00:25:27.880
<v Speaker 2>dots back in the day, and then the flip and

484
00:25:27.960 --> 00:25:30.799
<v Speaker 2>iPad comes out. You're like, what a third factor?

485
00:25:31.720 --> 00:25:36.160
<v Speaker 3>Oh no, now we can't predict the screen size. And yeah,

486
00:25:36.279 --> 00:25:38.039
<v Speaker 3>I tried to get ahead of that trend at the

487
00:25:38.039 --> 00:25:41.079
<v Speaker 3>company I was working at at the time, and we

488
00:25:41.079 --> 00:25:44.519
<v Speaker 3>were like one of the first industrial manufacturers that had

489
00:25:44.559 --> 00:25:48.279
<v Speaker 3>a web presence that worked on any screen size. Nice,

490
00:25:48.480 --> 00:25:52.599
<v Speaker 3>and it caught some attention because of that fact. There

491
00:25:52.599 --> 00:25:55.160
<v Speaker 3>were plenty of you know, marketing sites that were starting

492
00:25:55.240 --> 00:25:58.240
<v Speaker 3>to do this type of thing, but nobody in the

493
00:25:58.279 --> 00:26:00.799
<v Speaker 3>sector I was in was paying attention to it.

494
00:26:00.880 --> 00:26:04.039
<v Speaker 2>Yeah. So the oldest media query setups I ever saw

495
00:26:04.279 --> 00:26:07.440
<v Speaker 2>was Audi. Audi was on it really early on. It

496
00:26:07.519 --> 00:26:09.279
<v Speaker 2>used to be my went I talking points. It's like

497
00:26:09.359 --> 00:26:10.759
<v Speaker 2>look car company.

498
00:26:11.680 --> 00:26:14.599
<v Speaker 3>So I've always always kind of been a full stack dev.

499
00:26:14.720 --> 00:26:18.720
<v Speaker 3>But a lot of the stuff that I've done publicly

500
00:26:19.480 --> 00:26:23.000
<v Speaker 3>as a developer advocate has always been UI facing. So

501
00:26:23.079 --> 00:26:25.440
<v Speaker 3>I've done a lot of a lot of web development,

502
00:26:25.519 --> 00:26:29.119
<v Speaker 3>a lot of UX talks, and a lot of UI stuff.

503
00:26:30.039 --> 00:26:33.400
<v Speaker 3>But you know, I've always been a full stack dot

504
00:26:33.480 --> 00:26:36.920
<v Speaker 3>net dev prior, so you know I would own the

505
00:26:36.920 --> 00:26:40.640
<v Speaker 3>whole thing. So from time to time you'll see me

506
00:26:40.680 --> 00:26:44.240
<v Speaker 3>talking about you know, odd odd back end things like

507
00:26:44.799 --> 00:26:49.359
<v Speaker 3>the internals of Enity framework and that type of thing.

508
00:26:49.720 --> 00:26:51.759
<v Speaker 1>Hey, why don't we take our break now and then

509
00:26:51.799 --> 00:26:55.319
<v Speaker 1>get into retrieval augmented generation. Is that a good idea?

510
00:26:55.880 --> 00:26:57.759
<v Speaker 1>All right, sure, we'll do that. We'll be right back

511
00:26:57.759 --> 00:27:03.839
<v Speaker 1>after these very important messages, and we're back. It's dot

512
00:27:03.920 --> 00:27:06.960
<v Speaker 1>net ROX. I'm Carl Franklin, That's Richard Campbell, and that's

513
00:27:07.079 --> 00:27:10.920
<v Speaker 1>Ed Shaberneau. We're about to dive into this topic here

514
00:27:11.240 --> 00:27:19.000
<v Speaker 1>of RAG in agentic RAG. So let's just define it.

515
00:27:19.039 --> 00:27:21.759
<v Speaker 1>I mean, I'll take a stab at it. I haven't

516
00:27:21.839 --> 00:27:25.759
<v Speaker 1>used it yet, but from what I understand, it's taking

517
00:27:25.960 --> 00:27:31.160
<v Speaker 1>an existing LLM and adding data to it and coding

518
00:27:31.200 --> 00:27:34.599
<v Speaker 1>it in a database so that it can be retrieved.

519
00:27:34.640 --> 00:27:37.079
<v Speaker 1>Data can be retrieved from it and you can think

520
00:27:37.079 --> 00:27:41.039
<v Speaker 1>of it like taking you know, user manuals, help files,

521
00:27:42.039 --> 00:27:45.880
<v Speaker 1>documents about your company, anything that you would want to

522
00:27:45.920 --> 00:27:48.480
<v Speaker 1>be able to look up later in index. Is that

523
00:27:48.720 --> 00:27:49.559
<v Speaker 1>a good It's.

524
00:27:49.400 --> 00:27:53.200
<v Speaker 3>A pretty good primer. So when we talk about retrieval

525
00:27:53.240 --> 00:27:59.119
<v Speaker 3>augmented generation, we're talking about augmenting the context of the

526
00:27:59.160 --> 00:28:02.960
<v Speaker 3>model that you're working with and injecting some information and

527
00:28:03.000 --> 00:28:07.359
<v Speaker 3>then reshaping the prompt based on that information. So there's

528
00:28:07.400 --> 00:28:11.240
<v Speaker 3>there's actually a couple of different processes that are used

529
00:28:11.240 --> 00:28:15.839
<v Speaker 3>within a RAG system. And it's kind of interesting because

530
00:28:16.039 --> 00:28:18.799
<v Speaker 3>the whole RAG part, the thing that gives it its title,

531
00:28:19.119 --> 00:28:22.200
<v Speaker 3>is only just a small piece of it. So, uh,

532
00:28:22.759 --> 00:28:26.440
<v Speaker 3>the RAG concept on its surface is you know, injecting

533
00:28:26.519 --> 00:28:31.519
<v Speaker 3>that that context for you. But the retrieval part has

534
00:28:31.559 --> 00:28:33.200
<v Speaker 3>a lot of steps to it, and in order to

535
00:28:33.240 --> 00:28:37.079
<v Speaker 3>get retrieval, you have to have something to retrieve, So

536
00:28:37.119 --> 00:28:40.119
<v Speaker 3>there's a lot of a lot of different subprocesses in here.

537
00:28:41.599 --> 00:28:46.880
<v Speaker 3>So on the retrieve side of it, you have a database,

538
00:28:47.279 --> 00:28:52.319
<v Speaker 3>and there there's quite a few different vector databases out

539
00:28:52.359 --> 00:28:55.960
<v Speaker 3>there that that can be used for this. But with

540
00:28:56.079 --> 00:28:59.240
<v Speaker 3>the product that I work on, we use the Nuclear dB,

541
00:28:59.480 --> 00:29:03.759
<v Speaker 3>which is a database that we own it's an open

542
00:29:03.759 --> 00:29:09.119
<v Speaker 3>source database, but it's our vector database, and you want

543
00:29:09.160 --> 00:29:14.720
<v Speaker 3>to store your documents in that database and they get vectorized.

544
00:29:15.039 --> 00:29:18.119
<v Speaker 3>So what that means is when we feed the documents

545
00:29:18.160 --> 00:29:23.000
<v Speaker 3>into the system, those documents are going to be broken

546
00:29:23.039 --> 00:29:26.759
<v Speaker 3>down into bite sized pieces so we can manage the

547
00:29:26.799 --> 00:29:30.400
<v Speaker 3>size of those documents better. So they get chunked, and

548
00:29:30.440 --> 00:29:34.400
<v Speaker 3>those chunks get broken down and then sent through an

549
00:29:34.440 --> 00:29:37.880
<v Speaker 3>embedding model. So the embedding model is going to take

550
00:29:37.960 --> 00:29:43.480
<v Speaker 3>those chunks of text and is going to look for

551
00:29:43.519 --> 00:29:47.079
<v Speaker 3>the semantic meaning in that chunk of text and store

552
00:29:47.119 --> 00:29:51.440
<v Speaker 3>that semantic meaning as a number, So that's your vector

553
00:29:51.839 --> 00:29:56.880
<v Speaker 3>that you can then search on. These vector searches are

554
00:29:57.000 --> 00:29:59.960
<v Speaker 3>very powerful because they search on meaning. It's not like

555
00:30:00.079 --> 00:30:04.519
<v Speaker 3>a keyword search. You can find things that are relative

556
00:30:04.839 --> 00:30:08.799
<v Speaker 3>to the thing that you're searching for. One of one

557
00:30:08.839 --> 00:30:13.599
<v Speaker 3>of the easy ways to describe the semantic searches. If

558
00:30:13.599 --> 00:30:17.160
<v Speaker 3>you are storing things like milk and cheese in your

559
00:30:17.240 --> 00:30:19.680
<v Speaker 3>database and you search for dairy, you should be able

560
00:30:19.720 --> 00:30:21.880
<v Speaker 3>to pull back those items that are milk and cheese

561
00:30:22.000 --> 00:30:27.160
<v Speaker 3>because they are semantically similar. But what's really interesting about

562
00:30:27.279 --> 00:30:31.400
<v Speaker 3>the algorithm for that is the cosine similarity. You can

563
00:30:31.880 --> 00:30:35.599
<v Speaker 3>invert that as well. So if you say, what you know,

564
00:30:35.640 --> 00:30:38.759
<v Speaker 3>I want the opposite of dairy, you might get something

565
00:30:38.799 --> 00:30:41.720
<v Speaker 3>like capsaicin. Right, So you have hot peppers and things

566
00:30:41.759 --> 00:30:43.880
<v Speaker 3>like that. So if you think about it in a

567
00:30:44.000 --> 00:30:48.720
<v Speaker 3>larger query, you might say, what, you know, what can

568
00:30:48.759 --> 00:30:52.480
<v Speaker 3>I use to put out this fire in my mouth

569
00:30:52.559 --> 00:30:55.880
<v Speaker 3>after I've eaten hot peppers? And the answer might be

570
00:30:56.000 --> 00:31:00.440
<v Speaker 3>to drink milk because that is the opposite of the

571
00:31:00.480 --> 00:31:03.279
<v Speaker 3>pepper and it will put out the fire. So it's

572
00:31:03.319 --> 00:31:06.079
<v Speaker 3>a really powerful search tool. And when we run that

573
00:31:06.160 --> 00:31:09.440
<v Speaker 3>search and the vector database comes back with an answer,

574
00:31:09.599 --> 00:31:13.880
<v Speaker 3>it retrieves that chunk of text and it augments the

575
00:31:13.920 --> 00:31:17.920
<v Speaker 3>original question and then feeds that to a large language

576
00:31:17.920 --> 00:31:23.519
<v Speaker 3>model to generate a new answer. So you're taking the

577
00:31:23.519 --> 00:31:26.279
<v Speaker 3>context and you're expanding it with this knowledge that you've

578
00:31:26.359 --> 00:31:29.319
<v Speaker 3>just retrieved from your vector database. And that's where that

579
00:31:29.440 --> 00:31:30.960
<v Speaker 3>term rag comes from.

580
00:31:31.359 --> 00:31:33.880
<v Speaker 2>So what vector databases are we talking about here? As

581
00:31:33.920 --> 00:31:37.480
<v Speaker 2>a particular product like ionomatically think something like elastic, But.

582
00:31:38.759 --> 00:31:41.039
<v Speaker 1>I was saying that they have their own so you guys.

583
00:31:41.119 --> 00:31:44.559
<v Speaker 3>Yeah, we have one called the Nuclear database, right, And

584
00:31:44.640 --> 00:31:49.119
<v Speaker 3>the reason it's called nuclear dB is because the company

585
00:31:49.200 --> 00:31:53.759
<v Speaker 3>that created it was Nuclear, and Progress acquired that company

586
00:31:53.839 --> 00:31:58.759
<v Speaker 3>last year and we named the product Progress agentic RAG.

587
00:32:00.200 --> 00:32:03.920
<v Speaker 3>So you will see the term Nuclei used within the

588
00:32:03.960 --> 00:32:06.920
<v Speaker 3>APIs and things like that in the database itself because

589
00:32:06.960 --> 00:32:11.400
<v Speaker 3>it originally came from the company name. But yeah, we

590
00:32:11.480 --> 00:32:14.079
<v Speaker 3>use the nuclei a dB for our database.

591
00:32:14.960 --> 00:32:17.480
<v Speaker 1>Okay, So is there any limit to the size of

592
00:32:18.319 --> 00:32:20.599
<v Speaker 1>data or documents that can be stored or is it

593
00:32:20.680 --> 00:32:22.119
<v Speaker 1>just limited by disk space?

594
00:32:22.440 --> 00:32:27.200
<v Speaker 3>So this is one of those situations where our product

595
00:32:27.279 --> 00:32:31.000
<v Speaker 3>actually has has an advantage. So we actually have had

596
00:32:31.039 --> 00:32:35.119
<v Speaker 3>customers with millions of documents in their database. So it's

597
00:32:35.160 --> 00:32:38.559
<v Speaker 3>a RAG as a service. So you're talking about something

598
00:32:38.599 --> 00:32:42.839
<v Speaker 3>that you're deployed to in the cloud, so you can

599
00:32:42.920 --> 00:32:45.680
<v Speaker 3>do a hybrid mode on prem things like that with

600
00:32:45.759 --> 00:32:48.079
<v Speaker 3>it as well, but generally people use it in the cloud,

601
00:32:48.680 --> 00:32:52.200
<v Speaker 3>so you're really just limited by the cloud itself, which

602
00:32:52.640 --> 00:32:55.960
<v Speaker 3>you know, we can just keep adding storage too, so

603
00:32:56.200 --> 00:33:01.160
<v Speaker 3>the storage can get quite large. There are other ways

604
00:33:01.200 --> 00:33:06.119
<v Speaker 3>to do RAG. There's like open AI for example. You

605
00:33:06.160 --> 00:33:09.680
<v Speaker 3>can do some file activities with and I think their

606
00:33:09.720 --> 00:33:12.920
<v Speaker 3>limit is like ten k of files, so ten thousand

607
00:33:12.920 --> 00:33:18.200
<v Speaker 3>files and then it's maxed out with our solution like

608
00:33:18.200 --> 00:33:21.759
<v Speaker 3>I said, you can. You can just keep scaling. And

609
00:33:22.519 --> 00:33:26.559
<v Speaker 3>what's interesting about the architecture that we're using is it

610
00:33:26.599 --> 00:33:33.440
<v Speaker 3>takes a lot of the complexities out of RAG. Jeff

611
00:33:33.440 --> 00:33:36.920
<v Speaker 3>Fritz and I were having a conversation on one of

612
00:33:36.920 --> 00:33:39.880
<v Speaker 3>his shows the other day and I was telling him,

613
00:33:39.920 --> 00:33:43.039
<v Speaker 3>you know, the best way that I like to explain

614
00:33:43.079 --> 00:33:46.160
<v Speaker 3>this is, have you ever have you ever rolled your

615
00:33:46.160 --> 00:33:47.400
<v Speaker 3>own off system?

616
00:33:47.799 --> 00:33:51.400
<v Speaker 2>Only? Yeah, I'll do that again.

617
00:33:51.640 --> 00:33:56.160
<v Speaker 3>And it was a mistake, all right. There are plenty

618
00:33:56.160 --> 00:33:58.559
<v Speaker 3>of tools out there that will let you build RAG

619
00:33:58.599 --> 00:34:03.039
<v Speaker 3>systems from. There's a lot of moving pieces. You've got

620
00:34:03.240 --> 00:34:06.799
<v Speaker 3>large language models, embedding models that you have to choose from.

621
00:34:07.039 --> 00:34:10.000
<v Speaker 3>You have to choose a database, you have to have

622
00:34:10.079 --> 00:34:14.039
<v Speaker 3>a system to ingest data, and you have to have

623
00:34:14.079 --> 00:34:17.800
<v Speaker 3>a storage place to store that vector data. And you

624
00:34:17.840 --> 00:34:21.360
<v Speaker 3>can go off the shelf and get solutions from every

625
00:34:21.840 --> 00:34:24.239
<v Speaker 3>big tech vendor out there for each of those pieces

626
00:34:24.280 --> 00:34:27.480
<v Speaker 3>and glue them together and make your own solution. But

627
00:34:28.000 --> 00:34:31.440
<v Speaker 3>there's a lot of complexities to each one of those pieces,

628
00:34:32.079 --> 00:34:34.599
<v Speaker 3>and unless you have a data scientist or two on

629
00:34:34.639 --> 00:34:37.639
<v Speaker 3>your team, it's probably just the better idea to go

630
00:34:37.760 --> 00:34:42.400
<v Speaker 3>with something that's already been prescribed. So that's something that

631
00:34:42.440 --> 00:34:46.079
<v Speaker 3>we do well with this product. It's a RAG as

632
00:34:46.119 --> 00:34:49.840
<v Speaker 3>a service platform with a lot of different ways to

633
00:34:50.000 --> 00:34:56.400
<v Speaker 3>ingest data. So you can ingest PDFs, word documents, text data,

634
00:34:56.639 --> 00:34:59.559
<v Speaker 3>all sorts. But you can also do things like MP

635
00:34:59.599 --> 00:35:03.400
<v Speaker 3>four five files, m P three's. So we have providers

636
00:35:03.400 --> 00:35:05.480
<v Speaker 3>for all sorts of data types that's going to take

637
00:35:05.519 --> 00:35:09.119
<v Speaker 3>that data into the system and vectorize it so you

638
00:35:09.199 --> 00:35:14.400
<v Speaker 3>can perform not just search, but uh, you know, intelligence

639
00:35:14.599 --> 00:35:15.320
<v Speaker 3>from that data.

640
00:35:15.320 --> 00:35:18.239
<v Speaker 1>All right, So you said before that most people are

641
00:35:18.239 --> 00:35:19.960
<v Speaker 1>going to use it in the cloud. But you know,

642
00:35:20.039 --> 00:35:22.880
<v Speaker 1>one of the one of the great things about using

643
00:35:22.960 --> 00:35:28.760
<v Speaker 1>RAG systems is they that you can vectorize your your

644
00:35:28.840 --> 00:35:32.079
<v Speaker 1>sensitive documents. You know that should be kept on premises.

645
00:35:32.639 --> 00:35:37.679
<v Speaker 1>And so would a great big o Lama server be possible,

646
00:35:38.000 --> 00:35:39.840
<v Speaker 1>be possible to use with this product.

647
00:35:39.960 --> 00:35:42.199
<v Speaker 3>Yeah, you could do hybrid, So you can you can

648
00:35:42.199 --> 00:35:44.320
<v Speaker 3>do hybrid, or you can do it fully on prem

649
00:35:44.360 --> 00:35:45.079
<v Speaker 3>if you wanted to.

650
00:35:45.159 --> 00:35:46.639
<v Speaker 1>Now what does hybrid look like?

651
00:35:46.840 --> 00:35:50.199
<v Speaker 3>So you could hybrid, you would you would take your

652
00:35:50.280 --> 00:35:53.880
<v Speaker 3>data and store it locally, okay, so you could host

653
00:35:53.960 --> 00:35:56.599
<v Speaker 3>just the database portion of it and then you could

654
00:35:56.679 --> 00:36:00.719
<v Speaker 3>run the models in the cloud. Okay, so you have

655
00:36:00.800 --> 00:36:04.880
<v Speaker 3>your choice of models as well. So that's another aspect

656
00:36:05.039 --> 00:36:09.280
<v Speaker 3>of this being an entire solution that's end to end.

657
00:36:09.679 --> 00:36:11.320
<v Speaker 1>Is uh.

658
00:36:11.360 --> 00:36:13.960
<v Speaker 3>There there's actually a back office for it that you

659
00:36:14.000 --> 00:36:18.800
<v Speaker 3>can log into and then you can control various aspects

660
00:36:19.000 --> 00:36:22.199
<v Speaker 3>of the product. So I can go into my back

661
00:36:22.239 --> 00:36:25.559
<v Speaker 3>office and I can choose all the models, you know,

662
00:36:25.639 --> 00:36:28.800
<v Speaker 3>from all the models that are out there, from Azure

663
00:36:29.000 --> 00:36:34.239
<v Speaker 3>to AWS and Gemini and all of those things. And

664
00:36:34.280 --> 00:36:38.079
<v Speaker 3>I can provide custom endpoints and keys if I wanted

665
00:36:38.119 --> 00:36:41.800
<v Speaker 3>something that's completely customed and hosted by my you know,

666
00:36:41.960 --> 00:36:42.800
<v Speaker 3>my own servers.

667
00:36:43.400 --> 00:36:45.320
<v Speaker 1>So yeah, that was that was going to say you

668
00:36:45.400 --> 00:36:49.000
<v Speaker 1>could just run it all locally if you wanted to,

669
00:36:49.320 --> 00:36:51.079
<v Speaker 1>with something like a Lama, right.

670
00:36:51.000 --> 00:36:54.519
<v Speaker 3>Yeah, So it's it's plug and play with with whatever

671
00:36:54.559 --> 00:36:55.760
<v Speaker 3>you would like to bring to it.

672
00:36:56.239 --> 00:36:56.519
<v Speaker 1>Cool.

673
00:36:57.000 --> 00:37:00.960
<v Speaker 3>So it is extension extensible and modular, and like I said,

674
00:37:01.000 --> 00:37:04.320
<v Speaker 3>you could you can fully take it offline or you

675
00:37:04.360 --> 00:37:06.719
<v Speaker 3>can host the pieces that you want in the cloud.

676
00:37:07.760 --> 00:37:10.280
<v Speaker 3>So it does have a hybrid aspect to it as well.

677
00:37:10.400 --> 00:37:13.320
<v Speaker 1>And this is a free tooled is it all encapsulated

678
00:37:13.320 --> 00:37:15.440
<v Speaker 1>in nuclear dB or what is it?

679
00:37:15.440 --> 00:37:18.000
<v Speaker 3>It is a commercial product, so it does have a

680
00:37:18.039 --> 00:37:21.519
<v Speaker 3>license involved with it. It is open source.

681
00:37:21.599 --> 00:37:25.079
<v Speaker 1>But so RAG is a service. Is the commercial product? Yeah,

682
00:37:25.079 --> 00:37:28.199
<v Speaker 1>it's called nuclear dB. Is the devector database part of it? Correct?

683
00:37:28.639 --> 00:37:32.239
<v Speaker 3>Yeah, The product itself is called Progress agentic RAG. The

684
00:37:32.320 --> 00:37:36.800
<v Speaker 3>database portion is nuclear dB, and the RAG is a

685
00:37:36.840 --> 00:37:42.039
<v Speaker 3>service is is a commercial product. There there are open

686
00:37:42.079 --> 00:37:45.639
<v Speaker 3>source all of it is open source though it is.

687
00:37:47.719 --> 00:37:51.400
<v Speaker 3>All of the code is up on GitHub and various.

688
00:37:51.079 --> 00:37:54.039
<v Speaker 2>Repos But that doesn't mean, you know, it's not trivial

689
00:37:54.079 --> 00:37:56.519
<v Speaker 2>to operate this stuff. Is the reason it's a service.

690
00:37:56.239 --> 00:38:00.320
<v Speaker 3>No, Right, So you know, just just the data gestion

691
00:38:00.440 --> 00:38:02.760
<v Speaker 3>alone is a big portion of it. Like I said,

692
00:38:02.760 --> 00:38:05.840
<v Speaker 3>it takes in all of these different sources of material.

693
00:38:05.920 --> 00:38:08.199
<v Speaker 3>If you were to try to roll something like this yourself,

694
00:38:09.239 --> 00:38:11.559
<v Speaker 3>you know, this is something that demo's great, you can

695
00:38:11.599 --> 00:38:13.880
<v Speaker 3>build over a weekend type of a thing, but then

696
00:38:13.920 --> 00:38:16.760
<v Speaker 3>when you try to scale it, you know, as soon

697
00:38:16.800 --> 00:38:20.400
<v Speaker 3>as you know, somebody from upper management's like, all right,

698
00:38:20.440 --> 00:38:23.599
<v Speaker 3>now we want to ingest some MP three's and then

699
00:38:23.639 --> 00:38:25.840
<v Speaker 3>this video would be really nice, and then you're you're

700
00:38:25.840 --> 00:38:29.480
<v Speaker 3>off trying to figure out solutions to ingest that chunk

701
00:38:29.519 --> 00:38:33.039
<v Speaker 3>it up and get into your vector database. Right. Another

702
00:38:33.079 --> 00:38:37.360
<v Speaker 3>thing that this does is you can scrape web content

703
00:38:37.519 --> 00:38:42.119
<v Speaker 3>very easily. I can put in a site map and

704
00:38:42.320 --> 00:38:44.320
<v Speaker 3>it will take in that entire site map, And I

705
00:38:44.320 --> 00:38:50.360
<v Speaker 3>can also specify selectors like CSS selectors and XML selectors

706
00:38:51.119 --> 00:38:55.119
<v Speaker 3>so that it only ingests the pertinent information. Then you're

707
00:38:55.159 --> 00:38:57.920
<v Speaker 3>not re ingesting like a bunch of menus and footers

708
00:38:57.960 --> 00:39:00.000
<v Speaker 3>and all that stuff to kind of cloud up your data.

709
00:39:00.880 --> 00:39:04.360
<v Speaker 3>So there's a lot of infrastructure built on just the

710
00:39:04.519 --> 00:39:09.159
<v Speaker 3>ingestion process alone. And then also you probably noticed it

711
00:39:09.239 --> 00:39:12.119
<v Speaker 3>has the word agentic in it. It's not just a

712
00:39:12.159 --> 00:39:16.199
<v Speaker 3>regular RAG system. There are agents that are working alongside it.

713
00:39:17.679 --> 00:39:21.079
<v Speaker 3>One of those types of agents is on the ingestion side,

714
00:39:21.119 --> 00:39:23.480
<v Speaker 3>So as you ingest things, you can assign agents to

715
00:39:23.559 --> 00:39:27.280
<v Speaker 3>do tasks with that data. So these are kind of

716
00:39:27.320 --> 00:39:32.159
<v Speaker 3>like task runners that might summarize or categorize or tag

717
00:39:33.360 --> 00:39:36.719
<v Speaker 3>and do various activities on the data as it's being ingested,

718
00:39:36.920 --> 00:39:41.440
<v Speaker 3>and stores that metadata in the database as well. So

719
00:39:42.039 --> 00:39:44.920
<v Speaker 3>one of the interesting things that we have a demo

720
00:39:45.039 --> 00:39:52.039
<v Speaker 3>of is we ingested a bunch of financial documentation from

721
00:39:52.559 --> 00:39:55.000
<v Speaker 3>you know, the big Fortune five hundred companies. So we

722
00:39:55.119 --> 00:39:58.679
<v Speaker 3>got their annual reports in their m PDF format, pulled

723
00:39:58.679 --> 00:40:01.679
<v Speaker 3>them in ingested those and then there's an agent that

724
00:40:01.800 --> 00:40:05.639
<v Speaker 3>runs on that PDF, and it asks the large language

725
00:40:05.639 --> 00:40:09.280
<v Speaker 3>model to inspect the PDF for data that would go

726
00:40:09.280 --> 00:40:14.840
<v Speaker 3>good in charts, and it extracts from that unstructured data

727
00:40:16.159 --> 00:40:20.440
<v Speaker 3>into a format that is easily easily chartable in bar charts.

728
00:40:21.480 --> 00:40:24.440
<v Speaker 3>And then when you retrieve that information out, you can

729
00:40:24.480 --> 00:40:28.519
<v Speaker 3>just ask it for structured data and you don't have

730
00:40:28.559 --> 00:40:30.639
<v Speaker 3>to do any kind of special plumbing or anything. You

731
00:40:30.719 --> 00:40:33.039
<v Speaker 3>just give the large language model the shape that you

732
00:40:33.079 --> 00:40:36.719
<v Speaker 3>want out, and it recognizes that it has this jacent

733
00:40:36.840 --> 00:40:40.400
<v Speaker 3>data that's chartable in its database, and it pulls that

734
00:40:40.440 --> 00:40:42.320
<v Speaker 3>out and puts it in your structured data for you.

735
00:40:43.199 --> 00:40:45.559
<v Speaker 3>So now you've got these charts that are displaying on

736
00:40:45.679 --> 00:40:48.719
<v Speaker 3>screen that were generated by an agent just by ingesting

737
00:40:48.760 --> 00:40:53.320
<v Speaker 3>a PDF document that had no structured financial data in it.

738
00:40:53.599 --> 00:40:57.280
<v Speaker 3>It was just data that was pulled out. And what's

739
00:40:57.400 --> 00:40:59.760
<v Speaker 3>nice about doing it at ingestion time is you don't

740
00:40:59.760 --> 00:41:02.519
<v Speaker 3>have to keep repeating that over and over again. You

741
00:41:02.519 --> 00:41:04.559
<v Speaker 3>know it's stored in the database now, so you don't

742
00:41:04.559 --> 00:41:09.000
<v Speaker 3>have to run that generation every single time somebody retrieves

743
00:41:09.199 --> 00:41:10.039
<v Speaker 3>charting information.

744
00:41:10.320 --> 00:41:12.000
<v Speaker 1>What are some of the things that can go wrong

745
00:41:12.039 --> 00:41:13.119
<v Speaker 1>with RAG systems?

746
00:41:13.800 --> 00:41:19.199
<v Speaker 3>So the search portion of it is difficult. The search.

747
00:41:20.239 --> 00:41:22.159
<v Speaker 3>When we talk about vector search, I gave you the

748
00:41:22.280 --> 00:41:25.719
<v Speaker 3>very simple example of you know the dairy and whatnot,

749
00:41:25.800 --> 00:41:32.519
<v Speaker 3>but would you search in vector database? The vector search

750
00:41:32.719 --> 00:41:36.719
<v Speaker 3>is very much a semantic search. So when you phrase

751
00:41:36.760 --> 00:41:40.679
<v Speaker 3>your question, if that question isn't phrased that the same

752
00:41:40.719 --> 00:41:45.679
<v Speaker 3>way the data is written and ingested, the semantic meaning

753
00:41:45.800 --> 00:41:48.800
<v Speaker 3>might be different, but it still might contain the answer

754
00:41:48.880 --> 00:41:53.119
<v Speaker 3>that you want. So if you try to roll your

755
00:41:53.119 --> 00:41:56.639
<v Speaker 3>own you're just using an off the shelf semantic search

756
00:41:56.760 --> 00:42:01.119
<v Speaker 3>and nothing else, you might not get really great search results.

757
00:42:02.199 --> 00:42:04.159
<v Speaker 3>You might might let a lot of answers slip through

758
00:42:04.199 --> 00:42:04.679
<v Speaker 3>the cracks.

759
00:42:04.960 --> 00:42:06.800
<v Speaker 2>Interesting because you know, one of the things we talked

760
00:42:06.800 --> 00:42:08.800
<v Speaker 2>about here was that a whole sort of milk and

761
00:42:08.920 --> 00:42:11.840
<v Speaker 2>cheese versus dairy thing like this is supposed to be

762
00:42:11.880 --> 00:42:13.519
<v Speaker 2>the strength it is.

763
00:42:14.320 --> 00:42:17.760
<v Speaker 3>But depending on how you phrase that question. You know,

764
00:42:17.880 --> 00:42:21.760
<v Speaker 3>I phrased it as what might help, you know, put

765
00:42:21.800 --> 00:42:25.000
<v Speaker 3>out the flames of a hot meal or something like that.

766
00:42:26.400 --> 00:42:30.800
<v Speaker 3>For a rudimentary vector search, that probably wouldn't return the

767
00:42:30.840 --> 00:42:35.639
<v Speaker 3>results we want For a system like this, we we

768
00:42:35.679 --> 00:42:38.199
<v Speaker 3>wanted to have that in depth knowledge to be able

769
00:42:38.239 --> 00:42:42.239
<v Speaker 3>to pull those results back. So the search that we

770
00:42:42.280 --> 00:42:46.880
<v Speaker 3>implement isn't just a standard vector search. It's actually a

771
00:42:47.000 --> 00:42:51.800
<v Speaker 3>multi part search. So we use a keyword search on

772
00:42:51.920 --> 00:42:55.480
<v Speaker 3>top of the vector search, and then we also implement

773
00:42:55.559 --> 00:43:02.880
<v Speaker 3>something called named entity record nition, and this gives us

774
00:43:03.119 --> 00:43:09.679
<v Speaker 3>a knowledge graph. So knowledge graph is really important in

775
00:43:09.719 --> 00:43:15.239
<v Speaker 3>a system like this because it associates other concepts that

776
00:43:15.559 --> 00:43:18.119
<v Speaker 3>a large language model might not be able to infer

777
00:43:18.199 --> 00:43:23.079
<v Speaker 3>on its own. So for example, it might have named

778
00:43:23.239 --> 00:43:28.760
<v Speaker 3>entities in it, like a company, a person. You could

779
00:43:28.760 --> 00:43:32.280
<v Speaker 3>think of it that way as other nouns that it

780
00:43:32.440 --> 00:43:35.440
<v Speaker 3>is associating with the data that's there. So you might

781
00:43:35.440 --> 00:43:38.800
<v Speaker 3>have like a company sells a product, and you might

782
00:43:38.840 --> 00:43:40.920
<v Speaker 3>have the company name and the product name in there.

783
00:43:41.800 --> 00:43:46.400
<v Speaker 3>And it's kind of similar in a way to relationsships

784
00:43:46.440 --> 00:43:50.360
<v Speaker 3>in a relational database, but instead of tying things together

785
00:43:50.639 --> 00:43:53.639
<v Speaker 3>by a key, you're tying them together by a concept.

786
00:43:54.119 --> 00:43:57.719
<v Speaker 3>So you might have a concept of the company progress

787
00:43:58.760 --> 00:44:03.920
<v Speaker 3>and it sells the product TELLERQUI for Blaser, and the

788
00:44:04.000 --> 00:44:08.159
<v Speaker 3>thing that connects those two dots is the selling portion

789
00:44:08.320 --> 00:44:12.599
<v Speaker 3>of it. So this company sells X. And then you

790
00:44:12.679 --> 00:44:15.079
<v Speaker 3>may have a query later that you're trying to find

791
00:44:15.079 --> 00:44:21.559
<v Speaker 3>out what companies have products that compete with this product,

792
00:44:22.000 --> 00:44:24.800
<v Speaker 3>and since it has that knowledge graph built in, it

793
00:44:24.840 --> 00:44:29.119
<v Speaker 3>can reverse that concept and figure out what other companies

794
00:44:29.280 --> 00:44:34.400
<v Speaker 3>it has in its database that have similar products, where

795
00:44:34.599 --> 00:44:38.000
<v Speaker 3>a basic semantic search probably wouldn't pick those things up.

796
00:44:38.800 --> 00:44:42.360
<v Speaker 3>So this helps with things like people in places, locations,

797
00:44:42.880 --> 00:44:45.480
<v Speaker 3>all of that type of thing. All those types of

798
00:44:45.599 --> 00:44:48.679
<v Speaker 3>things can get connected through the knowledge graph, and that

799
00:44:48.920 --> 00:44:52.400
<v Speaker 3>really increases the accuracy of the search, so you're going

800
00:44:52.440 --> 00:44:53.840
<v Speaker 3>to get a lot better results.

801
00:44:54.079 --> 00:44:56.920
<v Speaker 2>Do you get into situations where stuff coming out of

802
00:44:56.960 --> 00:44:59.960
<v Speaker 2>the vector base is conflicting with the large language model,

803
00:45:00.119 --> 00:45:06.000
<v Speaker 2>like does it become a competition for accuracy or proper responses?

804
00:45:06.280 --> 00:45:12.119
<v Speaker 3>So you have full control over the large language model

805
00:45:12.199 --> 00:45:16.480
<v Speaker 3>and the prompting, but by default the prompts don't allow

806
00:45:18.079 --> 00:45:22.320
<v Speaker 3>the training data to conflict with the data that's being

807
00:45:22.400 --> 00:45:25.360
<v Speaker 3>pulled from the vector database. So there's some guardrails that

808
00:45:25.400 --> 00:45:28.280
<v Speaker 3>are already in place to help prevent this sort of thing.

809
00:45:29.440 --> 00:45:32.239
<v Speaker 3>If you go creating custom prompts, you might be able

810
00:45:32.280 --> 00:45:38.079
<v Speaker 3>to trip that up a little bit, but we yeah,

811
00:45:38.159 --> 00:45:41.719
<v Speaker 3>so we do have some things in place already for

812
00:45:41.760 --> 00:45:45.880
<v Speaker 3>you to try to to eliminate that. So generally if

813
00:45:45.920 --> 00:45:50.079
<v Speaker 3>you ask a question to progress agentic RAG and it

814
00:45:50.119 --> 00:45:54.679
<v Speaker 3>doesn't have that data in its storage anywhere. So it

815
00:45:54.719 --> 00:45:56.840
<v Speaker 3>goes into the vector database that can't find anything. It

816
00:45:56.880 --> 00:46:00.039
<v Speaker 3>may have plenty of stuff in training about it, but

817
00:46:00.119 --> 00:46:02.719
<v Speaker 3>it doesn't have documents to support it. And that's something

818
00:46:02.760 --> 00:46:04.039
<v Speaker 3>we probably should touch on to.

819
00:46:04.760 --> 00:46:07.880
<v Speaker 1>Another reason to use a hybrid model. Keep your data.

820
00:46:08.599 --> 00:46:11.880
<v Speaker 3>Yeah, it'll come back and it'll say, you know, I

821
00:46:11.880 --> 00:46:14.800
<v Speaker 3>don't have a context to provide an answer for that,

822
00:46:15.320 --> 00:46:19.360
<v Speaker 3>rather than pulling hallucinating something, right, And hallucination is a

823
00:46:19.360 --> 00:46:25.000
<v Speaker 3>big part of this as well, right, So with so

824
00:46:25.119 --> 00:46:27.719
<v Speaker 3>that's something to be argued about large language models. Actually,

825
00:46:28.320 --> 00:46:33.599
<v Speaker 3>I always say that hallucinations. One person's hallucinations is another

826
00:46:33.639 --> 00:46:37.320
<v Speaker 3>person's creativity. So if you want.

827
00:46:37.239 --> 00:46:40.559
<v Speaker 2>Good line, I don't know that it's true, but excellent line.

828
00:46:41.119 --> 00:46:43.880
<v Speaker 3>If you want to tell us creative story, you know

829
00:46:43.920 --> 00:46:47.440
<v Speaker 3>you might might want to, you know, create a fictional story.

830
00:46:47.519 --> 00:46:50.360
<v Speaker 3>With a large language model, you want to hallucinate as

831
00:46:50.400 --> 00:46:52.519
<v Speaker 3>much as possible. You don't want it to ground itself

832
00:46:52.519 --> 00:46:57.079
<v Speaker 3>in reality. But if you are asking it facts about

833
00:46:57.760 --> 00:47:01.280
<v Speaker 3>data that you've ingested, you don't want it to do that.

834
00:47:01.320 --> 00:47:05.639
<v Speaker 3>You want it as grounded as possible to hallucinates least

835
00:47:06.079 --> 00:47:10.800
<v Speaker 3>as it can. So with progress agentic RAG in general

836
00:47:10.880 --> 00:47:16.960
<v Speaker 3>RAG systems, you want citations, so that is something that's provided.

837
00:47:18.440 --> 00:47:21.800
<v Speaker 3>So when you do a vector search, it will pull

838
00:47:21.800 --> 00:47:24.199
<v Speaker 3>a bunch of information up and it may not even

839
00:47:24.360 --> 00:47:27.400
<v Speaker 3>use that information in the final answer. So what you

840
00:47:27.480 --> 00:47:30.960
<v Speaker 3>end up with is a list of resources. These resources

841
00:47:31.039 --> 00:47:34.280
<v Speaker 3>are things that hit the search but might not have

842
00:47:34.320 --> 00:47:37.239
<v Speaker 3>been relevant enough to include in the answer. And then

843
00:47:37.519 --> 00:47:41.320
<v Speaker 3>on top of resources, you'll have citations, and citations are

844
00:47:41.360 --> 00:47:45.199
<v Speaker 3>the resources that did get used in the answer. And

845
00:47:45.239 --> 00:47:50.920
<v Speaker 3>then you will also get back information about what string

846
00:47:50.920 --> 00:47:55.000
<v Speaker 3>of text was used in that answer, so it'll give you,

847
00:47:55.679 --> 00:48:00.159
<v Speaker 3>for example, it pulled the answer from this pdf, and

848
00:48:00.159 --> 00:48:03.519
<v Speaker 3>it it's coming from this paragraph in this pdf, so

849
00:48:03.559 --> 00:48:06.800
<v Speaker 3>you can really trace back, you know, the resource.

850
00:48:06.840 --> 00:48:08.800
<v Speaker 2>And that's I think what people want is like when

851
00:48:08.800 --> 00:48:11.519
<v Speaker 2>you when the offer admits it a definitive statement of

852
00:48:11.559 --> 00:48:14.679
<v Speaker 2>some kind, it's like, here's where that was said. I'm

853
00:48:14.679 --> 00:48:17.920
<v Speaker 2>not fabricating this. That's powerful stuff.

854
00:48:18.639 --> 00:48:23.920
<v Speaker 3>And all of the quality of metrics are driven by

855
00:48:24.280 --> 00:48:27.840
<v Speaker 3>the agentic framework that's inside of the system as well.

856
00:48:27.920 --> 00:48:31.280
<v Speaker 3>So I mentioned the agents are there and multiple capacities.

857
00:48:31.400 --> 00:48:34.400
<v Speaker 3>One of them is for something that we call remy,

858
00:48:34.519 --> 00:48:40.880
<v Speaker 3>which is the system that continuously monitors the information and

859
00:48:40.960 --> 00:48:44.480
<v Speaker 3>the queries that are being ran in the system, and

860
00:48:44.719 --> 00:48:50.400
<v Speaker 3>these are weighing in the groundedness the accuracy of the

861
00:48:50.440 --> 00:48:55.400
<v Speaker 3>information there, and it's it's displayed to you on a

862
00:48:55.480 --> 00:49:01.079
<v Speaker 3>dashboard with a graph that shows where the those UH

863
00:49:01.280 --> 00:49:03.000
<v Speaker 3>scores are being hit.

864
00:49:03.119 --> 00:49:05.079
<v Speaker 2>Now, this seems like the thing you'd have to build

865
00:49:05.119 --> 00:49:07.880
<v Speaker 2>yourself if you're doing right from scratch. Absolutely.

866
00:49:08.159 --> 00:49:13.280
<v Speaker 3>Yeah, if you want to validate how your system is working,

867
00:49:13.320 --> 00:49:14.840
<v Speaker 3>you're gonna have to come up with a bunch of

868
00:49:14.920 --> 00:49:18.840
<v Speaker 3>evaluators and you're going to have to have those evaluators

869
00:49:18.880 --> 00:49:22.079
<v Speaker 3>continuously running against your data as you ingest new data

870
00:49:22.159 --> 00:49:25.440
<v Speaker 3>and as users are hitting queries on it. Yeah, so

871
00:49:25.480 --> 00:49:27.320
<v Speaker 3>that's that's part of the system.

872
00:49:27.559 --> 00:49:29.559
<v Speaker 2>I see a lot of value in that, just like

873
00:49:29.679 --> 00:49:32.719
<v Speaker 2>there are patterns to how you do those evaluations.

874
00:49:32.800 --> 00:49:34.360
<v Speaker 3>So you've got being able to.

875
00:49:34.280 --> 00:49:36.159
<v Speaker 2>Get them ready to go for you.

876
00:49:37.519 --> 00:49:41.679
<v Speaker 3>Difference, You've got context relevance, which is how relevant your

877
00:49:41.679 --> 00:49:45.840
<v Speaker 3>information was based on the query. You've got the answer relevance,

878
00:49:46.039 --> 00:49:49.000
<v Speaker 3>and it's how how close was the AI generating an

879
00:49:49.000 --> 00:49:52.880
<v Speaker 3>answer that's quality, and then groundedness is all the context off.

880
00:49:54.039 --> 00:49:56.840
<v Speaker 2>As soon as you're ingesting a bunch of data from

881
00:49:56.840 --> 00:50:00.559
<v Speaker 2>a bunch of places, like, you're going to get conflicting results,

882
00:50:01.760 --> 00:50:04.679
<v Speaker 2>and so it's like, what does the tool do when

883
00:50:05.079 --> 00:50:08.440
<v Speaker 2>present when it retrieves two facts that are not the

884
00:50:08.480 --> 00:50:13.039
<v Speaker 2>same about the same query, right, So whatever tooling I

885
00:50:13.079 --> 00:50:15.920
<v Speaker 2>have to help surface that problem like that, that's what's

886
00:50:15.960 --> 00:50:17.159
<v Speaker 2>going to get you to travel here.

887
00:50:17.639 --> 00:50:20.119
<v Speaker 3>It's it's part of a hybrid search to try to

888
00:50:20.119 --> 00:50:21.199
<v Speaker 3>shake those things out.

889
00:50:22.239 --> 00:50:23.679
<v Speaker 1>So what's the pricing?

890
00:50:23.800 --> 00:50:28.440
<v Speaker 3>Like the pricing I'm never the sales guy in this situation,

891
00:50:28.639 --> 00:50:30.719
<v Speaker 3>so I don't remember the pricing off that's top of

892
00:50:30.800 --> 00:50:33.719
<v Speaker 3>my head. I think it's I think there's a first

893
00:50:33.719 --> 00:50:36.760
<v Speaker 3>of all, there's a fourteen day free trial, so you

894
00:50:36.800 --> 00:50:38.559
<v Speaker 3>can kick the tires on it, which doesn't.

895
00:50:38.360 --> 00:50:40.679
<v Speaker 2>Seem very long when you're putting together rag data like

896
00:50:40.719 --> 00:50:42.159
<v Speaker 2>that's not a trivial.

897
00:50:41.719 --> 00:50:43.920
<v Speaker 3>Effort, So it isn't it is? That we'll talk We

898
00:50:43.920 --> 00:50:46.039
<v Speaker 3>could talk about that in a minute. I'll go over

899
00:50:46.119 --> 00:50:48.760
<v Speaker 3>some of the pricing tiers that I can kind of remember.

900
00:50:48.800 --> 00:50:51.159
<v Speaker 3>I think it's I'm looking it free for fourteen days,

901
00:50:51.239 --> 00:50:53.119
<v Speaker 3>six hundred or something like that.

902
00:50:53.800 --> 00:50:57.920
<v Speaker 1>Seven hundred dollars is the introductory price per month, right,

903
00:50:58.239 --> 00:51:01.000
<v Speaker 1>and then there's like an almost teeth a month for

904
00:51:01.119 --> 00:51:01.880
<v Speaker 1>another level.

905
00:51:02.079 --> 00:51:06.280
<v Speaker 3>Yeah, so you would think fourteen days isn't a long time,

906
00:51:08.079 --> 00:51:10.199
<v Speaker 3>and I would like to see this be longer myself,

907
00:51:10.239 --> 00:51:13.679
<v Speaker 3>as a developer advocate, I'm always airing on the side

908
00:51:13.800 --> 00:51:18.000
<v Speaker 3>of the consumer that's using it, So I do agree

909
00:51:18.039 --> 00:51:20.280
<v Speaker 3>with you there. But one of the things that supports

910
00:51:20.320 --> 00:51:23.360
<v Speaker 3>the fourteen days is how quickly you can actually turn

911
00:51:23.400 --> 00:51:26.920
<v Speaker 3>this around. It's actually really surprising, and I'm not just

912
00:51:26.960 --> 00:51:30.679
<v Speaker 3>saying that because I work for the company that has it.

913
00:51:31.280 --> 00:51:34.199
<v Speaker 3>I'm actually a big fan of the product. Like it.

914
00:51:34.199 --> 00:51:37.519
<v Speaker 3>It's very quick to get your data in and as

915
00:51:37.519 --> 00:51:40.760
<v Speaker 3>soon as that data is ingested, there is that back

916
00:51:40.800 --> 00:51:44.559
<v Speaker 3>office that I told you about that has a search

917
00:51:44.639 --> 00:51:47.840
<v Speaker 3>function built in that you can just start keying in

918
00:51:47.880 --> 00:51:50.679
<v Speaker 3>searches to see what type of results you're going to

919
00:51:50.719 --> 00:51:53.440
<v Speaker 3>get back, and it's got the citations and everything right

920
00:51:53.440 --> 00:51:55.719
<v Speaker 3>there where you can see it. And there's even a

921
00:51:55.840 --> 00:51:58.840
<v Speaker 3>rag lab inside where you can test different prompts against

922
00:51:58.880 --> 00:52:01.800
<v Speaker 3>different models, sorts of stuff. So the turnaround time is

923
00:52:01.960 --> 00:52:04.880
<v Speaker 3>very very quick on this, and you can even deploy

924
00:52:05.960 --> 00:52:08.639
<v Speaker 3>like a JavaScript widget on your page to get started

925
00:52:09.400 --> 00:52:14.800
<v Speaker 3>and have the search functionality raised like immediately on on

926
00:52:14.840 --> 00:52:18.639
<v Speaker 3>your web platform. But when you want to want to

927
00:52:18.840 --> 00:52:24.639
<v Speaker 3>want to customize this, there's also SDKs in every flavor,

928
00:52:25.599 --> 00:52:28.000
<v Speaker 3>and we are on dot Net rocks, so you probably

929
00:52:28.039 --> 00:52:30.079
<v Speaker 3>want to hear about the dot Net SDK a little bit.

930
00:52:31.239 --> 00:52:33.840
<v Speaker 3>But once you ingest your data, you can just hop

931
00:52:33.880 --> 00:52:37.239
<v Speaker 3>on one of the SDKs. For the dot Net SDK,

932
00:52:38.599 --> 00:52:44.679
<v Speaker 3>it's I wrote it. I wrote it to be very

933
00:52:44.719 --> 00:52:46.599
<v Speaker 3>dot Net friendly. Have been a dot net developer for

934
00:52:46.639 --> 00:52:51.199
<v Speaker 3>twenty years, so I'm using all of the patterns that

935
00:52:51.199 --> 00:52:55.199
<v Speaker 3>dot net developers are used to. The platform itself is

936
00:52:55.320 --> 00:52:59.400
<v Speaker 3>written in Python, but since it's a software as a service,

937
00:52:59.639 --> 00:53:01.639
<v Speaker 3>you don't need to concern yourself with that. As a

938
00:53:01.679 --> 00:53:05.400
<v Speaker 3>dot Net developer, You're you're mainly working with the SDK,

939
00:53:05.719 --> 00:53:10.199
<v Speaker 3>which is talking to the rest APIs and on the

940
00:53:10.280 --> 00:53:12.519
<v Speaker 3>dot neet side, you've got all of the types that

941
00:53:12.599 --> 00:53:16.719
<v Speaker 3>come back from the rest APIs and nice C sharp

942
00:53:16.840 --> 00:53:23.239
<v Speaker 3>strong types, and the APIs follow similar methodologies to things

943
00:53:23.280 --> 00:53:29.519
<v Speaker 3>like Microsoft Extensions AI. You have dependency injection, so you

944
00:53:29.559 --> 00:53:33.280
<v Speaker 3>go into your RAG system, you ingest your data, you

945
00:53:33.280 --> 00:53:35.199
<v Speaker 3>go to your SDK, you pull in the new get

946
00:53:35.239 --> 00:53:41.480
<v Speaker 3>package and you do dependency injection. You say, use nuclear dB,

947
00:53:41.960 --> 00:53:45.719
<v Speaker 3>here's my keys, here's my endpoint, and then you have

948
00:53:45.760 --> 00:53:49.400
<v Speaker 3>a service that you can call search on and that's it.

949
00:53:49.480 --> 00:53:51.320
<v Speaker 3>Just like three or four lines of code and you're

950
00:53:51.880 --> 00:53:55.360
<v Speaker 3>talking to this system that's giving you intelligent answers.

951
00:53:55.960 --> 00:53:58.880
<v Speaker 1>So this sounds really cool. A couple more things about

952
00:53:58.880 --> 00:54:03.039
<v Speaker 1>the pricing lane layers. The starter layer that's seven hundred

953
00:54:03.079 --> 00:54:06.880
<v Speaker 1>monthly is text files only max seven hundred and fifty

954
00:54:06.920 --> 00:54:13.119
<v Speaker 1>megs per file, five gig index data or fifteen thousand resources.

955
00:54:14.639 --> 00:54:18.880
<v Speaker 1>The pro and these are cloud offerings. The pros nineteen

956
00:54:18.960 --> 00:54:23.039
<v Speaker 1>twenty five monthly and that has all file types, twenty

957
00:54:23.079 --> 00:54:28.079
<v Speaker 1>five gig in next data or eighty thousand resources. But

958
00:54:28.199 --> 00:54:33.039
<v Speaker 1>then a customized quote or enterprise you got to contact

959
00:54:33.679 --> 00:54:37.519
<v Speaker 1>Progress and that's the only one where you can have

960
00:54:37.719 --> 00:54:40.239
<v Speaker 1>on prem options. So if you're going to do it

961
00:54:40.280 --> 00:54:42.320
<v Speaker 1>on prem, you got to call yeah.

962
00:54:42.480 --> 00:54:42.719
<v Speaker 2>Yeah.

963
00:54:42.920 --> 00:54:46.559
<v Speaker 1>The other two are in the cloud only, so there you.

964
00:54:46.840 --> 00:54:48.719
<v Speaker 3>I think on the enterprise side, you're looking at companies

965
00:54:48.719 --> 00:54:50.519
<v Speaker 3>that are doing like I said, we have some that

966
00:54:50.559 --> 00:54:54.880
<v Speaker 3>are doing millions of documents, right, and the pricing model

967
00:54:54.880 --> 00:54:57.920
<v Speaker 3>probably shifts for somebody that's scaling it that large of

968
00:54:57.920 --> 00:55:02.800
<v Speaker 3>a scale it is, yeah, but the time to market

969
00:55:03.119 --> 00:55:06.239
<v Speaker 3>on using something like this is what really shocked me.

970
00:55:06.360 --> 00:55:10.880
<v Speaker 3>When the folks at Progress were like, you know, we're

971
00:55:10.880 --> 00:55:13.039
<v Speaker 3>making this acquisition, we want you to take a look

972
00:55:13.079 --> 00:55:18.880
<v Speaker 3>at it from a developer perspective, especially given the front

973
00:55:18.960 --> 00:55:23.519
<v Speaker 3>end capabilities that you know about, and see what you

974
00:55:23.840 --> 00:55:25.559
<v Speaker 3>think of this product. And I was like, wow, this

975
00:55:25.679 --> 00:55:29.320
<v Speaker 3>is this is really amazing. The only thing it's missing

976
00:55:29.400 --> 00:55:32.480
<v Speaker 3>for me is a C sharp SDK, and I immediately

977
00:55:32.480 --> 00:55:35.360
<v Speaker 3>got to work on building a C SHARPESTDK for it,

978
00:55:35.480 --> 00:55:38.039
<v Speaker 3>not because I was tasked to, but because I saw

979
00:55:38.079 --> 00:55:38.840
<v Speaker 3>the value there.

980
00:55:38.880 --> 00:55:40.039
<v Speaker 1>I was like, this is cool.

981
00:55:40.079 --> 00:55:43.719
<v Speaker 3>I got to bring this to dot net devs and

982
00:55:44.559 --> 00:55:49.159
<v Speaker 3>the the uniqueness of it, I think is something that

983
00:55:50.400 --> 00:55:53.239
<v Speaker 3>we're bringing to the table for dot Net. There are

984
00:55:53.440 --> 00:55:56.119
<v Speaker 3>other you know, rag type things out there, but like

985
00:55:56.159 --> 00:55:59.800
<v Speaker 3>I talked about with the open AI file APIs, those

986
00:55:59.840 --> 00:56:03.440
<v Speaker 3>are been locked under a preview and you have to

987
00:56:03.599 --> 00:56:07.840
<v Speaker 3>kind of sign off through pragma warnings to even use it.

988
00:56:08.000 --> 00:56:10.760
<v Speaker 3>It's like, this is not supported. You're going into our

989
00:56:10.920 --> 00:56:13.719
<v Speaker 3>chart uncharted territory. Put these you know, put these flags

990
00:56:13.719 --> 00:56:16.519
<v Speaker 3>in so you're allowed to even test this. Please don't

991
00:56:16.519 --> 00:56:19.400
<v Speaker 3>deploy it to production. We went V one on the

992
00:56:19.480 --> 00:56:24.039
<v Speaker 3>sd K last week. It's been fully tested and in

993
00:56:24.400 --> 00:56:30.840
<v Speaker 3>tested and it covers the entire rest API. So you've

994
00:56:31.039 --> 00:56:34.920
<v Speaker 3>you've got this solution that you can spin up extremely quick.

995
00:56:35.679 --> 00:56:38.159
<v Speaker 3>I could get a demo this running and probably less

996
00:56:38.199 --> 00:56:43.599
<v Speaker 3>than twelve to twenty minutes, and uh, you're.

997
00:56:43.679 --> 00:56:47.719
<v Speaker 2>I pulled a list of the demos from the Telleric

998
00:56:48.079 --> 00:56:51.920
<v Speaker 2>get heberpository, one for Maui, one for Kendo, and one

999
00:56:52.000 --> 00:56:53.039
<v Speaker 2>for Blazer Nice.

1000
00:56:53.199 --> 00:56:55.920
<v Speaker 3>Yeah, yeah, that's that charting example is in there. Yeah,

1001
00:56:56.000 --> 00:56:59.480
<v Speaker 3>so that one that one mashes up the progress agentic

1002
00:56:59.559 --> 00:57:04.239
<v Speaker 3>RAG with Kendo Ui or Blazer Maui as well. I

1003
00:57:04.280 --> 00:57:08.880
<v Speaker 3>think has the chart example. And when you dig into

1004
00:57:08.920 --> 00:57:10.880
<v Speaker 3>the source code for that, you're going to be surprised

1005
00:57:10.920 --> 00:57:15.519
<v Speaker 3>at how little business logic is there. It's mostly markup.

1006
00:57:16.199 --> 00:57:17.760
<v Speaker 3>Most of us is just the presentation.

1007
00:57:19.320 --> 00:57:20.599
<v Speaker 1>So that was one.

1008
00:57:20.519 --> 00:57:25.920
<v Speaker 3>Of the nice fits here when when we did this acquisition,

1009
00:57:26.199 --> 00:57:29.239
<v Speaker 3>like it was very much a back end service there,

1010
00:57:29.480 --> 00:57:32.079
<v Speaker 3>there are, like I said, there are widgets in there

1011
00:57:32.159 --> 00:57:34.800
<v Speaker 3>that do you like, some vanilla JavaScript stuff to get

1012
00:57:35.440 --> 00:57:40.079
<v Speaker 3>you started, But when you're ready to build an application,

1013
00:57:41.400 --> 00:57:45.559
<v Speaker 3>you're really going to want to customize the UI, and

1014
00:57:45.639 --> 00:57:50.840
<v Speaker 3>we have done UI for you know, almost twenty years

1015
00:57:50.840 --> 00:57:52.920
<v Speaker 3>now at progress with the teller.

1016
00:57:54.760 --> 00:57:57.480
<v Speaker 2>It's all about how you present the information, right, Yeah.

1017
00:57:57.519 --> 00:58:00.880
<v Speaker 3>And I was like, now we've got the ability to

1018
00:58:00.960 --> 00:58:06.719
<v Speaker 3>do these really intelligent queries and results and use our

1019
00:58:06.920 --> 00:58:10.679
<v Speaker 3>UIs with it, like this is bread and butter, peanut

1020
00:58:10.760 --> 00:58:14.079
<v Speaker 3>butter and jelly. Yeah, you know, these two things go

1021
00:58:14.119 --> 00:58:14.719
<v Speaker 3>together great.

1022
00:58:14.800 --> 00:58:19.000
<v Speaker 2>It also speaks to how there's less and less interface

1023
00:58:19.039 --> 00:58:21.039
<v Speaker 2>part less buttons and knobs and switches, and more and

1024
00:58:21.039 --> 00:58:25.960
<v Speaker 2>more screen space given to visualization. It's one, you know,

1025
00:58:26.000 --> 00:58:28.679
<v Speaker 2>it's just a box to describe what you want and

1026
00:58:28.719 --> 00:58:31.360
<v Speaker 2>then the rest of the screen is dedicated to showing

1027
00:58:31.400 --> 00:58:31.960
<v Speaker 2>you what you want.

1028
00:58:32.199 --> 00:58:34.239
<v Speaker 3>I know I've said search quite a few times too,

1029
00:58:34.280 --> 00:58:37.360
<v Speaker 3>but I think search really under sells what we're doing

1030
00:58:37.360 --> 00:58:40.519
<v Speaker 3>here in the RAG space in general, not just for

1031
00:58:40.559 --> 00:58:46.320
<v Speaker 3>this particular solution, but all of it. Searches the like

1032
00:58:46.400 --> 00:58:49.360
<v Speaker 3>Hella world, right, it's it's the easy go to.

1033
00:58:49.559 --> 00:58:52.960
<v Speaker 2>Well and it aamly a company doesn't have a search problem. Yeah,

1034
00:58:53.039 --> 00:58:55.639
<v Speaker 2>it's just the normal all search stocks.

1035
00:58:55.760 --> 00:58:55.960
<v Speaker 1>Yeah.

1036
00:58:56.280 --> 00:58:58.679
<v Speaker 3>I'll give you some interesting things around search too. But

1037
00:59:01.119 --> 00:59:07.400
<v Speaker 3>one of the things that this does well is it

1038
00:59:07.440 --> 00:59:12.000
<v Speaker 3>can do like product recommendation, which is kind of search like,

1039
00:59:12.719 --> 00:59:15.159
<v Speaker 3>but you can you can come at this from a

1040
00:59:15.199 --> 00:59:18.239
<v Speaker 3>different point of view instead of asking it a direct question.

1041
00:59:18.960 --> 00:59:21.639
<v Speaker 3>You say, I'm a dot net developer, what products do

1042
00:59:21.679 --> 00:59:25.159
<v Speaker 3>you have that can help me with this problem? And

1043
00:59:25.280 --> 00:59:29.360
<v Speaker 3>do product recommendation. And you can also use this as

1044
00:59:29.400 --> 00:59:33.360
<v Speaker 3>a competitive advantage if you want to say, bring in

1045
00:59:33.440 --> 00:59:36.679
<v Speaker 3>data from a competitor and bring in your own data

1046
00:59:36.760 --> 00:59:41.360
<v Speaker 3>and then do a competitive analysis and do you can

1047
00:59:41.440 --> 00:59:46.519
<v Speaker 3>do battle cards in real time for your salespeople. You

1048
00:59:46.519 --> 00:59:49.559
<v Speaker 3>can say, all right, what you know, what product is

1049
00:59:49.639 --> 00:59:53.519
<v Speaker 3>my competitor selling and what is you know? What is

1050
00:59:53.559 --> 00:59:56.239
<v Speaker 3>their audience and how do we compete against them? And

1051
00:59:56.280 --> 00:59:59.039
<v Speaker 3>this can do that analysis for you and shape battle

1052
00:59:59.079 --> 01:00:03.840
<v Speaker 3>cards in a UI and display them nice. One of

1053
01:00:03.880 --> 01:00:09.199
<v Speaker 3>the more interesting demos that I worked on with our

1054
01:00:09.239 --> 01:00:12.920
<v Speaker 3>sales folks, and I can't name names of particular products

1055
01:00:13.000 --> 01:00:16.039
<v Speaker 3>or companies, so I'll be a little bit vague, is

1056
01:00:16.400 --> 01:00:20.880
<v Speaker 3>we did one that was a food laboratory. So we

1057
01:00:20.880 --> 01:00:25.760
<v Speaker 3>were working with somebody that did a product that had

1058
01:00:25.800 --> 01:00:31.599
<v Speaker 3>to do with food additives. And what we had come

1059
01:00:31.679 --> 01:00:34.880
<v Speaker 3>up with as a solution is, you know you're in

1060
01:00:35.320 --> 01:00:40.559
<v Speaker 3>a laboratory working on the next donut, Let's say, and

1061
01:00:40.880 --> 01:00:43.280
<v Speaker 3>you want to know what you can use as a

1062
01:00:43.320 --> 01:00:48.239
<v Speaker 3>food coloring for that donut. Why not make a mobile

1063
01:00:48.280 --> 01:00:53.519
<v Speaker 3>application that you snap a photograph of your prototype, and

1064
01:00:53.719 --> 01:01:01.239
<v Speaker 3>the AI analyzes that photograph for coloringschene, whatever other keywords

1065
01:01:01.239 --> 01:01:04.519
<v Speaker 3>you want to tag that with, and then goes off

1066
01:01:04.639 --> 01:01:08.280
<v Speaker 3>to the RAG system and searches for products that you

1067
01:01:08.480 --> 01:01:13.599
<v Speaker 3>have that can be used for that result. And then

1068
01:01:13.639 --> 01:01:15.960
<v Speaker 3>it's going to come back with not only an answer,

1069
01:01:16.000 --> 01:01:18.119
<v Speaker 3>but a bunch of data sheets that you can click

1070
01:01:18.159 --> 01:01:21.400
<v Speaker 3>through and find data on, and then you can ask

1071
01:01:21.480 --> 01:01:24.079
<v Speaker 3>follow up questions on it, things like how will this

1072
01:01:24.159 --> 01:01:28.000
<v Speaker 3>coloring affect the taste of my product? And that goes further,

1073
01:01:28.280 --> 01:01:30.840
<v Speaker 3>you know, beyond a search that's not a normal search.

1074
01:01:30.920 --> 01:01:34.519
<v Speaker 3>Now you're you're doing like a research roll and a

1075
01:01:34.599 --> 01:01:37.519
<v Speaker 3>developed you know, R and D roll inside of this tool,

1076
01:01:37.559 --> 01:01:41.639
<v Speaker 3>and it's doing it in several minutes rather than you know,

1077
01:01:41.719 --> 01:01:43.559
<v Speaker 3>something that would take hours or days.

1078
01:01:44.000 --> 01:01:46.920
<v Speaker 1>Nice, it's just good stuff ed. When were we going

1079
01:01:47.000 --> 01:01:47.719
<v Speaker 1>to see you next?

1080
01:01:49.119 --> 01:01:54.039
<v Speaker 3>I am going to be at stir Trek nice, and

1081
01:01:54.119 --> 01:01:58.159
<v Speaker 3>that is on May first, and I'll also be at

1082
01:01:58.239 --> 01:02:02.760
<v Speaker 3>code Stock. Code Stock's a good one in Tennessee. I

1083
01:02:02.840 --> 01:02:06.079
<v Speaker 3>think that one's mid April. That one went away for

1084
01:02:06.079 --> 01:02:07.880
<v Speaker 3>a little bit and came back. There was just some

1085
01:02:07.920 --> 01:02:10.079
<v Speaker 3>really good people in Tennessee. I'm happy to see that

1086
01:02:10.079 --> 01:02:13.199
<v Speaker 3>they brought that back and then also be at the

1087
01:02:13.280 --> 01:02:18.039
<v Speaker 3>MVP summit. This probably is likely to air after that though,

1088
01:02:18.599 --> 01:02:20.559
<v Speaker 3>So no, you are.

1089
01:02:20.559 --> 01:02:25.400
<v Speaker 1>Corrector yeah, yeah, April ninth, Yeah, someone in this little air. Yeah,

1090
01:02:25.480 --> 01:02:28.679
<v Speaker 1>so all right, dude, good stuff, Thanks very much for sharing.

1091
01:02:28.920 --> 01:02:31.000
<v Speaker 3>Yeah, you can you can find out more about this

1092
01:02:31.119 --> 01:02:33.320
<v Speaker 3>at progress dot com. You can actually kick the tires

1093
01:02:33.320 --> 01:02:35.119
<v Speaker 3>on it right from the main screen there there's a

1094
01:02:35.119 --> 01:02:39.719
<v Speaker 3>search box that deploys the the product behind the scenes,

1095
01:02:39.880 --> 01:02:43.159
<v Speaker 3>so you can just ask questions to about anything Progress related.

1096
01:02:43.159 --> 01:02:46.840
<v Speaker 3>And it's using that we're dog fooding at hardcore right there. Great,

1097
01:02:46.880 --> 01:02:51.239
<v Speaker 3>and then you can find it at Newgate on Newgitt

1098
01:02:51.360 --> 01:02:57.639
<v Speaker 3>repository as the Progress dot nuclear package. And yeah, we.

1099
01:02:57.679 --> 01:03:00.360
<v Speaker 1>Got a few more links on the page at dot

1100
01:03:00.400 --> 01:03:02.760
<v Speaker 1>nerocks dot com. So ed, thanks very much and we

1101
01:03:02.800 --> 01:03:05.960
<v Speaker 1>will talk to you next time on dot net rocks.

1102
01:03:06.119 --> 01:03:25.119
<v Speaker 4>Thanks for having us.

1103
01:03:18.679 --> 01:03:30.159
<v Speaker 1>Dot net Rocks is brought to you by Franklin's Net

1104
01:03:30.280 --> 01:03:34.239
<v Speaker 1>and produced by Pop Studios, a full service audio video

1105
01:03:34.280 --> 01:03:38.440
<v Speaker 1>and post production facility located physically in New London, Connecticut,

1106
01:03:38.639 --> 01:03:43.440
<v Speaker 1>and of course in the cloud online at pwop dot com.

1107
01:03:43.639 --> 01:03:45.760
<v Speaker 1>Visit our website at d O T N E t

1108
01:03:46.000 --> 01:03:50.000
<v Speaker 1>R O c k S dot com for RSS feeds, downloads,

1109
01:03:50.159 --> 01:03:53.840
<v Speaker 1>mobile apps, comments, and access to the full archives going

1110
01:03:53.880 --> 01:03:57.280
<v Speaker 1>back to show number one, recorded in September two thousand

1111
01:03:57.280 --> 01:03:59.920
<v Speaker 1>and two. And make sure you check out our sponsors.

1112
01:04:00.119 --> 01:04:02.920
<v Speaker 1>They keep us in business. Now, go write some code,

1113
01:04:03.480 --> 01:04:04.239
<v Speaker 1>see you next time.

1114
01:04:05.119 --> 01:04:07.519
<v Speaker 4>You got trade Middle vans by the

1115
01:04:07.960 --> 01:04:16.480
<v Speaker 1>Sam Is Home and my taxes in line red