1 00:00:00,000 --> 00:00:02,820 Well, this is not a tremendous amount of news this week. 2 00:00:03,160 --> 00:00:08,820 Thankfully, we have a guest who I think can fill the void of interesting factoids. 3 00:00:09,200 --> 00:00:10,800 Do you want to introduce yourself, guest? 4 00:00:12,100 --> 00:00:12,700 Sure. 5 00:00:13,520 --> 00:00:20,920 So my name is Mark Fisher, and I am the technical lead for Project Riff at Pivotal. 6 00:00:21,920 --> 00:00:28,680 And prior to that, I've spent the last decade, more than a decade now, on the spring team. 7 00:00:29,860 --> 00:00:51,840 And started with the spring team in 2006, and then worked on a number of projects there, including spring integration, and then spring cloud stream, spring cloud data flow, and spring cloud function as well, before shifting over to Riff and serverless and functions as a service and all those fun things about a year ago. 8 00:00:52,020 --> 00:00:53,860 And so a decade, what's that like? 9 00:00:55,200 --> 00:00:56,320 It's all been a blur. 10 00:00:56,960 --> 00:00:59,120 Yeah, I guess that happens. 11 00:00:59,120 --> 00:01:07,760 I mean, that's, let's see, if you get a laptop for three years, that means you've gotten three different machines from work? 12 00:01:08,080 --> 00:01:08,560 Is that right? 13 00:01:08,720 --> 00:01:09,160 Or is it different? 14 00:01:09,240 --> 00:01:10,580 That sounds about right, yes. 15 00:01:10,800 --> 00:01:11,200 Yeah? 16 00:01:11,320 --> 00:01:12,140 Maybe on its fourth. 17 00:01:12,520 --> 00:01:12,760 Yeah. 18 00:01:12,920 --> 00:01:13,080 Yeah. 19 00:01:13,280 --> 00:01:14,920 You got the one with the touchpad on it now? 20 00:01:15,320 --> 00:01:17,500 I do, and that I find annoying. 21 00:01:17,500 --> 00:01:23,960 And I'm just waiting for my chance to get the 32 gig of RAM on the 15-inch MacBook. 22 00:01:23,960 --> 00:01:24,160 Okay. 23 00:01:24,540 --> 00:01:26,300 Yeah, that would be crazy. 24 00:01:26,300 --> 00:01:26,540 Yeah. 25 00:01:26,700 --> 00:01:32,260 My wife has an older laptop, and I'd forgotten how nice those old buttons are. 26 00:01:32,460 --> 00:01:34,680 I guess you would call them keys in the modern era. 27 00:01:35,280 --> 00:01:37,120 But the keys are nice. 28 00:01:37,400 --> 00:01:38,920 They're not all like, I don't know. 29 00:01:38,980 --> 00:01:40,040 I mean, this keyboard's fine. 30 00:01:40,180 --> 00:01:40,920 It's very functional. 31 00:01:41,200 --> 00:01:50,720 But I feel like it's sort of like, it's like, I don't know about your grandparents, but you would go into their house, and they would have this formal living room that no one, it was, you know, I always thought it was funny. 32 00:01:50,780 --> 00:01:53,640 It's a living room that no one lives in, and you couldn't touch anything. 33 00:01:53,880 --> 00:01:56,800 And these keyboards feel kind of like you're not supposed to touch them. 34 00:01:57,120 --> 00:01:57,980 They're just... 35 00:01:57,980 --> 00:02:01,160 Do you, Coté, do you have one of the newer MacBooks? 36 00:02:01,160 --> 00:02:03,760 Because I have an older one, so I still have like HDMI and USB. 37 00:02:03,900 --> 00:02:05,740 I don't have to carry around the 50 dongles. 38 00:02:06,060 --> 00:02:07,720 Do you have to carry around tons of things? 39 00:02:08,020 --> 00:02:09,580 I'm on dongle style here. 40 00:02:09,680 --> 00:02:10,560 Full dongle mode. 41 00:02:10,560 --> 00:02:10,880 Okay. 42 00:02:11,120 --> 00:02:12,140 I got the four out. 43 00:02:12,140 --> 00:02:12,580 Yeah, makes sense. 44 00:02:13,580 --> 00:02:13,980 Oof. 45 00:02:14,340 --> 00:02:16,040 What a problem. 46 00:02:16,560 --> 00:02:17,200 What a world. 47 00:02:18,060 --> 00:02:18,560 Yeah, yeah. 48 00:02:18,560 --> 00:02:20,560 The dongles now have their own firmware, so. 49 00:02:21,740 --> 00:02:23,080 Yeah, your dongles have dongles. 50 00:02:23,120 --> 00:02:24,100 There's too much going on here. 51 00:02:24,220 --> 00:02:24,660 I can't get it. 52 00:02:25,080 --> 00:02:28,320 Yeah, you know, when Pivotal sent me this, they sent me one of those. 53 00:02:28,440 --> 00:02:34,540 It's a very popular dongle for the USB-C that's got like three USB hubs and an HDMI internet port. 54 00:02:35,280 --> 00:02:37,780 And it also has a micro SD card. 55 00:02:37,940 --> 00:02:39,780 And it actually is... 56 00:02:39,780 --> 00:02:42,300 I enjoy the functionality of it, right? 57 00:02:42,360 --> 00:02:48,560 Because you can put in like a little USB, SD card and, you know, have storage on there. 58 00:02:48,560 --> 00:02:50,080 And it's fun to have the internet. 59 00:02:50,080 --> 00:02:52,200 But the problem is you get the 15 inch one. 60 00:02:52,420 --> 00:02:55,560 And I think the power pass through doesn't work, right? 61 00:02:55,980 --> 00:03:00,960 So, because that would be awesome if I actually just had one thing to plug in. 62 00:03:01,580 --> 00:03:02,020 Sure. 63 00:03:02,300 --> 00:03:03,020 But I don't. 64 00:03:03,440 --> 00:03:07,180 So, you know, it doesn't really pay off there. 65 00:03:07,940 --> 00:03:08,420 Anyhow. 66 00:03:08,840 --> 00:03:11,120 Well, there is no weather this week. 67 00:03:11,400 --> 00:03:15,820 But I did want to do one small update from Amsterdam. 68 00:03:16,280 --> 00:03:18,380 I finally ate some herring. 69 00:03:18,920 --> 00:03:21,320 Traditional Amsterdam food. 70 00:03:21,320 --> 00:03:24,500 Now, I didn't realize that it was raw herring, which is fine. 71 00:03:24,760 --> 00:03:25,680 It's no big deal. 72 00:03:26,060 --> 00:03:29,920 But, you know, you get this herring and they've deheaded or beheaded. 73 00:03:30,080 --> 00:03:34,980 I don't know what the proper fish and gutted it and taken off the skin. 74 00:03:35,100 --> 00:03:39,300 And then, but they leave the tail on and you just kind of like drop it in your mouth and eat it. 75 00:03:39,360 --> 00:03:41,840 You don't actually just like put it, you know, you chew it. 76 00:03:41,840 --> 00:03:47,260 And you kind of like put it in a bed of onions to have it attached there and eat it. 77 00:03:47,420 --> 00:03:48,480 And it was good. 78 00:03:48,760 --> 00:03:49,500 It's delicious. 79 00:03:49,660 --> 00:03:50,980 It's basically like sushi. 80 00:03:51,320 --> 00:03:54,880 Like it's not really that much of a different thing. 81 00:03:55,240 --> 00:03:57,600 So I think, and then it's quite affordable too. 82 00:03:57,720 --> 00:04:00,940 You can make a little lunch out of it without vegetables. 83 00:04:01,120 --> 00:04:02,680 I guess, does onion count as a vegetable? 84 00:04:02,860 --> 00:04:04,120 That's like a root, isn't it? 85 00:04:04,120 --> 00:04:04,900 No, that seems fair. 86 00:04:05,320 --> 00:04:06,860 Yeah, that'll get your vegetable. 87 00:04:07,480 --> 00:04:10,040 That picture of you that you're including here is probably, 88 00:04:10,040 --> 00:04:14,240 we have to turn that into a Slack emoji or my new profile picture or something. 89 00:04:14,560 --> 00:04:15,260 But it's good. 90 00:04:15,360 --> 00:04:15,580 It is. 91 00:04:15,660 --> 00:04:18,240 I do have a small video, so maybe I can load Giphy up. 92 00:04:18,980 --> 00:04:21,220 Get ready. 93 00:04:21,320 --> 00:04:24,220 Yeah, I like it. 94 00:04:24,280 --> 00:04:24,560 It's good. 95 00:04:24,640 --> 00:04:26,800 And that was from the local grocery store. 96 00:04:27,000 --> 00:04:28,420 Imagine if you got it fresh. 97 00:04:29,020 --> 00:04:32,600 And, you know, another just before we get started with serious stuff that we get paid for. 98 00:04:32,980 --> 00:04:39,060 Another fun factoid about herring is that was, I think, the first way back, you know, way back when, 99 00:04:39,320 --> 00:04:45,660 way, way back when, that Amsterdam got wealthy is they invented a new technology, some new kind of boat, 100 00:04:45,660 --> 00:04:48,560 so they could go out into the North Sea for like three weeks. 101 00:04:49,040 --> 00:04:57,380 And then they also figured out that if you behead and gut and prepare the herring and salt it and stick it in barrels at sea, 102 00:04:57,500 --> 00:04:58,980 it would last a long time. 103 00:04:58,980 --> 00:05:04,560 So they could make a lot of, they could bring a lot of herring in and that made them lots of money. 104 00:05:04,560 --> 00:05:11,760 And what I like about that little factoid is it makes you realize that like even back in like the 1400s, 105 00:05:11,760 --> 00:05:15,660 there was just trade of barrels of herring just being sent around. 106 00:05:16,020 --> 00:05:21,000 People would, I mean, can you imagine eating some raw fish out of a barrel that was four weeks old? 107 00:05:22,440 --> 00:05:25,460 Man, I can't even come up with some stupid quip about that. 108 00:05:25,600 --> 00:05:26,360 I don't know if I could. 109 00:05:26,920 --> 00:05:27,100 Yeah. 110 00:05:27,600 --> 00:05:28,840 So times have improved. 111 00:05:29,380 --> 00:05:30,640 This is the golden age. 112 00:05:31,300 --> 00:05:32,040 That's right. 113 00:05:33,480 --> 00:05:39,020 Well, speaking of the golden age, it's been a quiet week for infrastructure software news. 114 00:05:39,120 --> 00:05:43,340 I think there was some relevant to our guest here. 115 00:05:43,700 --> 00:05:50,300 There was some coverage of like native or native or whatever, just some kind of follow-up and analysis, 116 00:05:50,300 --> 00:05:53,220 which, which also brought in some, some serverless things. 117 00:05:53,220 --> 00:05:57,620 And I don't know, I'll put, there's some links in there for people who in the show notes, 118 00:05:57,780 --> 00:06:01,120 if you want to look at it, but they're just, we've kind of talked about those things, 119 00:06:01,120 --> 00:06:07,160 but have you, have you caught anything else there and your, your nets and your Dutch cog ships? 120 00:06:08,420 --> 00:06:11,180 I don't know how I do that kind of lead in. 121 00:06:11,560 --> 00:06:16,460 So a few minutes before we recorded, we actually announced that steel toe 2.1 ship. 122 00:06:16,460 --> 00:06:20,700 So if you're a .NET developer and you're using the latest .NET framework, .NET core, 123 00:06:20,780 --> 00:06:24,420 and trying to build these more observable systems, so you're doing distributed tracing, 124 00:06:24,640 --> 00:06:28,860 or you're trying to have better actuator support for .NET devs, that's great. 125 00:06:29,000 --> 00:06:30,100 That's fully available now. 126 00:06:30,160 --> 00:06:30,960 So that was nice news. 127 00:06:31,480 --> 00:06:38,420 And then I included one in here about that new-ish Intel chip foreshadow, the name of the flaw. 128 00:06:38,760 --> 00:06:40,900 That was another one of these things that had to quickly get patched. 129 00:06:41,380 --> 00:06:45,320 And another one of these that you can either run around with your, you know, kind of head cut off, 130 00:06:45,380 --> 00:06:51,020 trying to figure out how to patch systems and contact your OS vendor and roll out updates and figure out what breaks. 131 00:06:51,380 --> 00:06:54,300 We didn't really make a big deal of it because Ubuntu patched it. 132 00:06:54,300 --> 00:06:58,420 We rolled it out to customers and many of them are already updated because of Bosch and Comfort. 133 00:06:58,760 --> 00:07:04,680 So again, we like to talk about the operating system doesn't necessarily matter a lot to the end person 134 00:07:04,680 --> 00:07:05,880 who just wants to run software. 135 00:07:05,980 --> 00:07:08,860 It matters a ton as a baseline for what you're running. 136 00:07:08,980 --> 00:07:12,840 But when it's embedded in platforms, this stuff becomes boring even when they're critical updates. 137 00:07:13,080 --> 00:07:17,480 So it was kind of neat to see that get rolled out again and keeping you safe if you're running Cloud Foundry. 138 00:07:18,360 --> 00:07:21,380 I wonder if there's a committee where they name viruses. 139 00:07:21,600 --> 00:07:22,660 I'd like to be on that committee. 140 00:07:23,020 --> 00:07:26,080 Maybe they're less ominous, just like, you know, happy time. 141 00:07:26,480 --> 00:07:27,200 It's the happy time. 142 00:07:27,200 --> 00:07:31,320 Yeah, I mean, I'm not super worried about the Intel daisy problem or something like that. 143 00:07:31,380 --> 00:07:34,420 It really just makes me a buttercup, something that kind of makes me smile. 144 00:07:34,700 --> 00:07:36,120 The sunflower exploit. 145 00:07:36,560 --> 00:07:38,540 Yeah, that can't be too bad. 146 00:07:39,780 --> 00:07:40,980 That would be nice. 147 00:07:41,080 --> 00:07:46,520 Now, the other question, I don't want to ask this every week, but maybe every now and then. 148 00:07:46,560 --> 00:07:47,380 How's the book going? 149 00:07:47,660 --> 00:07:48,260 Almost done? 150 00:07:49,580 --> 00:07:49,940 Oh, me? 151 00:07:50,020 --> 00:07:51,080 Yeah, I'm more than halfway done. 152 00:07:51,160 --> 00:07:52,700 I'm feeling pretty good about that. 153 00:07:52,800 --> 00:07:57,440 I have had the obligatory one contact with the editor to tell them it will not be on the original schedule. 154 00:07:57,840 --> 00:07:59,660 I think it'll only be one time. 155 00:08:00,000 --> 00:08:02,020 I think I'll finish it when I'm supposed to now. 156 00:08:02,220 --> 00:08:03,500 So it's going pretty well. 157 00:08:03,880 --> 00:08:06,040 Well, I figure book deadlines. 158 00:08:06,180 --> 00:08:09,960 Any kind of deadline is kind of like, you know, a serving suggestion. 159 00:08:10,360 --> 00:08:12,080 It's sort of, you know, add salt to taste. 160 00:08:12,360 --> 00:08:14,800 That's kind of what the deadline is, as I recall. 161 00:08:15,580 --> 00:08:17,160 Yeah, plus or minus three months. 162 00:08:17,660 --> 00:08:18,300 Three day time. 163 00:08:18,820 --> 00:08:19,260 Exactly. 164 00:08:19,940 --> 00:08:22,980 You might want a pinch of salt or a tablespoon. 165 00:08:23,300 --> 00:08:23,880 It depends. 166 00:08:24,900 --> 00:08:25,040 Yeah. 167 00:08:26,800 --> 00:08:27,240 Excellent. 168 00:08:27,660 --> 00:08:31,540 Well, so I'm always trying to finish up a little booklet that I have. 169 00:08:31,540 --> 00:08:33,620 And I published an excerpt. 170 00:08:35,000 --> 00:08:44,120 I've gotten a lot of, I wouldn't call it faint praise, but it's sort of, some people have told me that I have too many links. 171 00:08:44,380 --> 00:08:46,280 And so it takes a long time to read things. 172 00:08:46,580 --> 00:08:51,940 So I tried to post a smaller excerpt that maybe still has a lot of links. 173 00:08:52,040 --> 00:08:59,460 But anyways, it's basically this part on like, if you're doing your, we should work on phrases for this. 174 00:08:59,520 --> 00:09:05,460 If you're digital transformation, something, you know, and you're trying to be the sort of leaders of it. 175 00:09:05,580 --> 00:09:09,940 It's probably a good idea not to go it alone because that doesn't really work out very well. 176 00:09:09,940 --> 00:09:18,540 So you got to take a page from some classic change management or just really corporate life and build up some alliances and friendships and things like that. 177 00:09:18,600 --> 00:09:26,220 So it's very brief, but it's just a little, little write up of doing that and how some other people in the, I don't know, the pivotal world have done that. 178 00:09:26,400 --> 00:09:33,420 And also it gave me the chance over in medium, they have these things, you know, everyone is, has like snap clone things where they make these story cards. 179 00:09:33,660 --> 00:09:36,620 And wow, they're, they're doing something over there. 180 00:09:37,300 --> 00:09:38,020 Love it. 181 00:09:38,900 --> 00:09:39,780 I'm using those. 182 00:09:39,840 --> 00:09:45,960 It's unfortunate because it's, it's done for a mobile form factor, which renders perfectly in your browser on the desktop. 183 00:09:45,960 --> 00:09:51,480 But if you open it on the actual mobile device, uh, you have to install the medium app. 184 00:09:51,820 --> 00:09:56,840 So I don't know if that really works out well, but it doesn't know, no, I don't like it. 185 00:09:56,920 --> 00:09:59,260 Well, no, you're, you're, uh, your progress has been good though. 186 00:09:59,280 --> 00:10:00,860 I've liked following the last couple you've done. 187 00:10:00,900 --> 00:10:04,640 There's, I think you did one recently around even some measurement of things, right? 188 00:10:04,660 --> 00:10:06,620 Of kind of progress and things like that. 189 00:10:06,620 --> 00:10:07,560 So I really liked that one. 190 00:10:07,580 --> 00:10:10,020 I'll probably use some of that for an upcoming talk. 191 00:10:10,180 --> 00:10:10,620 So a nice one. 192 00:10:10,920 --> 00:10:11,240 Measurements. 193 00:10:11,380 --> 00:10:12,280 Well, that's good to hear. 194 00:10:12,640 --> 00:10:14,580 I, I, I love a little measurements. 195 00:10:14,740 --> 00:10:16,380 Measurements are, are mysterious. 196 00:10:16,840 --> 00:10:20,420 I, as, as I've joked about, I think I might even mention this here. 197 00:10:20,420 --> 00:10:23,940 Like whenever you're a, I don't know if you've had this, if you two have had this experience 198 00:10:23,940 --> 00:10:27,800 when you write something, but measurements or metrics are one of those things where you 199 00:10:27,800 --> 00:10:31,220 ask people, uh, and you know that they know the answer and you're like, well, tell me what 200 00:10:31,220 --> 00:10:32,260 metrics that they use. 201 00:10:32,520 --> 00:10:36,380 And then they like go on poetically about like, well, what is a pudding? 202 00:10:36,600 --> 00:10:38,920 You've got a Yorkshire pudding and like a pudding cup. 203 00:10:38,980 --> 00:10:39,620 What is a pudding? 204 00:10:39,860 --> 00:10:41,220 And they get all philosophic. 205 00:10:41,220 --> 00:10:44,940 And then you're like, okay, I gotta, I got, you know, I got stuff to do my life. 206 00:10:45,560 --> 00:10:48,580 And then, and then you write something and you put it in front of them. 207 00:10:48,580 --> 00:10:51,620 And they're like, let me write you 50 pages on exactly what metrics are. 208 00:10:51,700 --> 00:10:55,500 And you're like, come on, how come you didn't just tell me about this in the beginning instead 209 00:10:55,500 --> 00:10:56,240 of putting talk. 210 00:10:56,640 --> 00:10:59,620 So hopefully that will lick some of that there. 211 00:11:01,020 --> 00:11:05,560 So, uh, I was gonna, I was gonna say, speaking of putting talk. 212 00:11:05,820 --> 00:11:06,020 Yeah. 213 00:11:08,700 --> 00:11:13,960 So, so why not, you know, first of all, I was, uh, I was, I was emceeing, uh, which happens 214 00:11:13,960 --> 00:11:15,580 to me, uh, some of my initials. 215 00:11:15,580 --> 00:11:22,680 Uh, I think the spring, uh, the spring tour, uh, let's see Dallas some time ago and spring 216 00:11:22,680 --> 00:11:24,100 platform tour, I think we call it. 217 00:11:24,440 --> 00:11:26,800 And the number one, the, the, the, thank you. 218 00:11:26,860 --> 00:11:31,320 The number one thing that people were asking, wanted to ask about that we didn't get to discuss 219 00:11:31,320 --> 00:11:33,440 with them was, uh, was project riff. 220 00:11:33,560 --> 00:11:34,920 They were all like, what is riff? 221 00:11:35,180 --> 00:11:40,180 And so, I mean, I mean, I'm putting the clock a little bit back, but back then, let's say 222 00:11:40,180 --> 00:11:45,140 this was in the spring, what, what would, how would you have answered them and told them 223 00:11:45,140 --> 00:11:45,820 what that was? 224 00:11:46,180 --> 00:11:46,560 Sure. 225 00:11:46,820 --> 00:11:54,960 Um, well, riff is a, uh, it's basically a function as a service platform that can run 226 00:11:54,960 --> 00:11:57,520 on any Kubernetes installation. 227 00:11:57,520 --> 00:11:59,840 So it can run on prem. 228 00:11:59,920 --> 00:12:01,560 It can run in, in a public cloud. 229 00:12:01,980 --> 00:12:13,920 And, um, we, I think that one of the key differentiators for riff is that we have focused on, um, an event 230 00:12:13,920 --> 00:12:15,420 driven model at the core. 231 00:12:15,520 --> 00:12:21,860 So every function is invoked based on, um, an event. 232 00:12:21,860 --> 00:12:26,620 And back in the spring, that would have meant an event, um, being sent over Kafka. 233 00:12:27,100 --> 00:12:33,840 We were basically coupled to Kafka at the time, but, um, building on Kubernetes and, and running 234 00:12:33,840 --> 00:12:42,040 that Kafka code in a sidecar so that what the developer needs to provide is just a function 235 00:12:42,040 --> 00:12:47,100 that basically acts as a callback and the sidecar then, um, invokes that. 236 00:12:47,540 --> 00:12:55,320 So it was kind of the ideas behind some of the spring projects that we had built in the 237 00:12:55,320 --> 00:13:00,660 years leading up to that, um, such as, especially like spring cloud stream, where you would have 238 00:13:00,660 --> 00:13:05,240 a binder to Kafka or rabbit or some other event driven messaging middleware. 239 00:13:05,240 --> 00:13:10,880 And then the user just writes a function, just writes a callback to be invoked. 240 00:13:11,480 --> 00:13:17,500 Um, but we took that model into, um, Kubernetes and therefore made it polyglot. 241 00:13:17,540 --> 00:13:24,720 With, um, an extensible model for the actual invokers that, that handle that inversion of control. 242 00:13:25,820 --> 00:13:30,720 So, so I, I have a, I have a slightly, uh, off topic question cause you reminded me of something 243 00:13:30,720 --> 00:13:34,820 I'm always, I'm always wondering, and especially since you, you, you know, uh, been, been working 244 00:13:34,820 --> 00:13:39,940 on this for like 10 years or in this area when, when, when you're, when you're working on, 245 00:13:40,060 --> 00:13:44,400 I mean, let's, let's, let's say spring is, well, I guess it's called the spring framework. 246 00:13:44,400 --> 00:13:48,300 Spring is a framework that, that you use to build other things. 247 00:13:48,300 --> 00:13:52,720 Uh, it's sort of like, I, I guess you could just run spring on its own. 248 00:13:52,720 --> 00:13:53,960 That would be sort of fun. 249 00:13:54,260 --> 00:13:57,020 Be sort of like running an air conditioner with no one in the room. 250 00:13:57,740 --> 00:13:59,400 It'd be an interesting experiment. 251 00:13:59,400 --> 00:14:05,200 But, uh, so when you're building that and you decide to have like, um, a dependency, I 252 00:14:05,200 --> 00:14:08,780 guess, on something like Kafka, like, how do you think through that architecturally? 253 00:14:09,000 --> 00:14:13,580 Like, cause you can't, you can't build everything in it. 254 00:14:13,580 --> 00:14:18,060 Like you, you have to use some other things and like, what, I don't know, like, how do you 255 00:14:18,060 --> 00:14:20,000 think through to do that or not do it? 256 00:14:20,000 --> 00:14:24,100 And, uh, the sort of compromises you made or make or the benefits, like what's the, uh, 257 00:14:24,100 --> 00:14:25,740 the decision tree for that kind of thing? 258 00:14:27,900 --> 00:14:34,640 Well, it's, it's tough because I think usually it involves defining an abstraction and getting 259 00:14:34,640 --> 00:14:39,860 that abstraction to be the right level where you're not creating the lowest common denominator, 260 00:14:39,860 --> 00:14:47,180 but you're actually, um, allowing people to choose what they plug in behind that abstraction 261 00:14:47,180 --> 00:14:54,040 for certain reasons and be able to expose the value of the reason they chose that thing, but 262 00:14:54,040 --> 00:14:57,100 in a way that is still, uh, pluggable. 263 00:14:57,100 --> 00:15:02,800 And so that usually in a, in a spring based application for something like the binders and 264 00:15:02,800 --> 00:15:06,660 spring cloud stream, that means that you can plug in Kafka, you can plug in rabbit, you 265 00:15:06,660 --> 00:15:12,820 can plug in, um, say Redis streams or, or, uh, if you're in a public cloud, you can plug 266 00:15:12,820 --> 00:15:15,640 in Kinesis, uh, on Amazon, for example. 267 00:15:16,320 --> 00:15:21,980 Um, and they're similar enough that an abstraction can define the common model there. 268 00:15:22,080 --> 00:15:28,380 And so that's what makes it pluggable, but then whichever one you choose, um, there may 269 00:15:28,380 --> 00:15:32,320 be certain parameters that you could specify in the declarative configuration that allow 270 00:15:32,320 --> 00:15:34,780 you to take advantage of some of its core features. 271 00:15:35,280 --> 00:15:41,620 So, so essentially like the, what you want to go for is, uh, you can swap out as, as they 272 00:15:41,620 --> 00:15:46,260 say the, the implementation underneath there, or, or you want to get as close to that as 273 00:15:46,260 --> 00:15:50,900 possible so that you're not sort of stuck with just the one thing that you, you would 274 00:15:50,900 --> 00:15:51,320 need to run. 275 00:15:51,500 --> 00:15:51,640 Right. 276 00:15:52,460 --> 00:15:52,820 Yeah. 277 00:15:52,880 --> 00:15:53,100 Yeah. 278 00:15:53,100 --> 00:15:58,240 I asked that cause I think the, uh, the first time I encountered this was way, way back when, 279 00:15:58,240 --> 00:16:03,180 when I was working, uh, on a systems management piece of software and, uh, you know, there's, 280 00:16:03,240 --> 00:16:06,920 there's, there's, uh, there's, there's, uh, there's, there's a joke about like, you know, 281 00:16:06,920 --> 00:16:11,000 the, the first thing you need to do is invent the world is sort of like how you set up 282 00:16:11,000 --> 00:16:11,540 any system. 283 00:16:12,020 --> 00:16:15,420 Uh, but, but the, the first thing you'd have to do for this system is install. 284 00:16:15,720 --> 00:16:18,760 I forget which, which, uh, which TIBCO thing it was. 285 00:16:18,820 --> 00:16:21,340 I don't think they even owned it back then, but it was just like this. 286 00:16:21,340 --> 00:16:23,340 I, I think it was a Corbra thing. 287 00:16:23,540 --> 00:16:24,880 You had to set up this orb. 288 00:16:25,080 --> 00:16:29,180 And I remember when we went to the training, about two hours was spent setting up the orb 289 00:16:29,180 --> 00:16:31,000 before you even installed the software. 290 00:16:31,540 --> 00:16:33,160 So it's sort of stuck in my mind. 291 00:16:35,560 --> 00:16:41,100 So let me ask you, Mark, what you quick mentioned there kind of in passing was the event driven 292 00:16:41,100 --> 00:16:46,060 sort of foundation of riff and why you felt like that was something that was important. 293 00:16:46,140 --> 00:16:51,180 Do you see it as different than how kind of traditional FAS has been up till this point? 294 00:16:51,260 --> 00:16:52,820 So what is that, you know, what is that difference? 295 00:16:52,980 --> 00:16:54,600 Why do you, why do you call that out uniquely? 296 00:16:56,360 --> 00:17:06,040 Well, I, I do think, um, there's a difference in the, uh, extent to which it can support messaging 297 00:17:06,040 --> 00:17:07,040 patterns, right? 298 00:17:07,080 --> 00:17:14,440 So like in riff, um, we, by, by having the sidecar communicate with Kafka, one of the 299 00:17:14,440 --> 00:17:19,080 things that we could do between the sidecar and the container that hosted the developer's 300 00:17:19,080 --> 00:17:29,900 function is to manage that incoming, um, stream of events as a, an actual stream, um, through 301 00:17:29,900 --> 00:17:34,880 a gRPC, um, service binding, right? 302 00:17:34,880 --> 00:17:39,560 So then that means that the developer's code can actually handle streaming input. 303 00:17:39,780 --> 00:17:43,880 If you're using Java, you could, you could depend on a reactor flux. 304 00:17:43,880 --> 00:17:46,740 And if you're using JavaScript, you can depend on a node stream. 305 00:17:47,240 --> 00:17:52,640 Um, so there are some use cases that where that's valuable because then the, the, the 306 00:17:52,640 --> 00:17:58,500 function developer can, can perform windowing operations and map reduce type operations on 307 00:17:58,500 --> 00:18:04,240 the stream of data, as opposed to just single input, single output, which we also support, 308 00:18:04,240 --> 00:18:10,100 but that's, that's almost like the degenerate case, even if it's, even if it's the, uh, 80% 309 00:18:10,100 --> 00:18:10,580 case. 310 00:18:11,020 --> 00:18:15,960 So being able to support both of those, it's, it's very hard to build a model that takes the 311 00:18:15,960 --> 00:18:16,740 other direction. 312 00:18:16,740 --> 00:18:21,600 I think where it assumes from day one, everything is just a single HTTP request. 313 00:18:21,900 --> 00:18:24,800 And then somehow we're going to add streaming on top of those. 314 00:18:25,020 --> 00:18:25,380 Right. 315 00:18:27,060 --> 00:18:30,720 I mean, that architecture, which is, as you say, it's kind of been, I don't know if to 316 00:18:30,720 --> 00:18:32,260 me, it's been, that's the dominant one, right? 317 00:18:32,300 --> 00:18:38,620 I mean, functions get spun up via mostly HTTP requests and there's not much state sharing 318 00:18:38,620 --> 00:18:40,620 or any sort of messaging paradigm there. 319 00:18:40,700 --> 00:18:45,140 So if you were doing this in a typical FAS today, would you be using an event stream processor 320 00:18:45,140 --> 00:18:50,100 and then bolting functions on so that you're still passing the data through something that 321 00:18:50,100 --> 00:18:51,000 has some of those primitives? 322 00:18:53,000 --> 00:18:53,360 Yeah. 323 00:18:53,420 --> 00:18:54,740 Well, I think, I mean, 324 00:18:56,060 --> 00:18:58,960 I'm thinking, for example, I'd have to use Kinesis with Lambda, right? 325 00:18:58,980 --> 00:18:59,800 Or something like that. 326 00:18:59,800 --> 00:19:01,680 Is that how I would stitch things together today? 327 00:19:02,840 --> 00:19:09,460 Well, okay, so that's an interesting one because in the Lambda example, if you choose to use 328 00:19:09,460 --> 00:19:15,700 Kinesis as your trigger, then you really end up with a different input type as opposed 329 00:19:15,700 --> 00:19:22,260 to if you're connecting your function directly to the API gateway to handle HTTP requests. 330 00:19:22,260 --> 00:19:32,460 And so that's the kind of consistent model that we're aiming for to where you can take a function 331 00:19:32,460 --> 00:19:39,020 and either write it for single input, single output, or streaming in, streaming out, and 332 00:19:39,020 --> 00:19:48,260 be able to plug that in to any, you know, independently of the upstream source of events, right? 333 00:19:48,260 --> 00:19:48,620 Right. 334 00:19:49,080 --> 00:19:49,840 That makes sense. 335 00:19:50,200 --> 00:19:53,260 So tell me a little bit about how we got to Knative. 336 00:19:53,700 --> 00:20:00,120 So, you know, if we were, we launched Riff at SpringOne Platform next year, kind of took 337 00:20:00,120 --> 00:20:02,160 the wraps off of you, did a great keynote presentation. 338 00:20:02,360 --> 00:20:04,160 You and Dave Sire did a fun breakout. 339 00:20:04,280 --> 00:20:06,560 People can go back and watch, which is well worth their time. 340 00:20:07,140 --> 00:20:09,980 But, you know, as we were going through the spring and all of a sudden, you know, the 341 00:20:09,980 --> 00:20:13,980 team wasn't doing as much Riff stuff and this other work was emerging, what was kind 342 00:20:13,980 --> 00:20:16,980 of the, you know, I guess we don't have to get the whole backstory, but more 343 00:20:16,980 --> 00:20:18,100 importantly, why? 344 00:20:18,220 --> 00:20:19,700 Like, why did these efforts merge? 345 00:20:19,800 --> 00:20:23,120 What are you excited about with Knative that might have not happened if Riff had stayed 346 00:20:23,120 --> 00:20:24,040 on its current path? 347 00:20:25,300 --> 00:20:25,520 Right. 348 00:20:25,680 --> 00:20:33,060 So I think a big part of it is that, you know, at Pivotal, we talk a lot about the value 349 00:20:33,060 --> 00:20:33,400 line. 350 00:20:33,660 --> 00:20:36,880 And that's an important concept to our customers. 351 00:20:38,080 --> 00:20:41,940 And it's just as important when we're building things internally. 352 00:20:41,940 --> 00:20:46,680 So anytime that we discover that there's another effort where we could be part of a larger 353 00:20:46,680 --> 00:20:53,220 community that's building the foundation of something, then it's worth considering that 354 00:20:53,220 --> 00:20:56,660 as opposed to building it completely on our own. 355 00:20:56,660 --> 00:21:01,600 And so when we first heard about Knative, it seemed like there was a significant amount 356 00:21:01,600 --> 00:21:07,520 of overlap in the sense that they were adding custom resource definitions to create higher 357 00:21:07,520 --> 00:21:14,760 levels of abstraction on top of Kubernetes in order to support serverless workloads and functions. 358 00:21:16,180 --> 00:21:23,000 Now, to the extent that we looked at what, I mean, obviously, we were doing a lot of that 359 00:21:23,000 --> 00:21:24,180 in Riff ourselves. 360 00:21:24,560 --> 00:21:32,120 And it sort of triggered us to evaluate what extent of what we were doing in Riff was more 361 00:21:32,120 --> 00:21:40,500 generic and what was more specifically in our differentiation, like the event-driven paradigm. 362 00:21:40,780 --> 00:21:50,260 So a lot of the lower-level concerns are things that, like being able to scale automatically, scaling 363 00:21:50,260 --> 00:21:57,840 back to zero, providing metrics for the functions that are running, providing the observability 364 00:21:57,840 --> 00:21:58,620 of those functions. 365 00:21:58,940 --> 00:22:02,340 There are a lot of common concerns there. 366 00:22:02,600 --> 00:22:06,040 And so it made sense for us to at least consider that option. 367 00:22:06,760 --> 00:22:12,960 And then as we started collaborating with the team at the time, it was Google and then others 368 00:22:12,960 --> 00:22:14,720 also started to come on board. 369 00:22:16,320 --> 00:22:19,940 And we shared some of our ideas around eventing. 370 00:22:20,580 --> 00:22:27,620 And our goal was basically to see if we could get enough of what we needed to support our 371 00:22:27,620 --> 00:22:33,900 model into Knative so that we could still build on top of it and then take advantage of what 372 00:22:33,900 --> 00:22:38,120 we hope becomes a larger ecosystem for the common foundation. 373 00:22:40,120 --> 00:22:49,480 So can you, like, throw a good illustrative or illustrative example of an application you 374 00:22:49,480 --> 00:22:51,760 might write in this style? 375 00:22:53,360 --> 00:22:54,000 Sure. 376 00:22:55,680 --> 00:23:09,800 I think, so with Knative, there's actually the idea that it's a unified model for scaling. 377 00:23:10,120 --> 00:23:12,280 Regardless of the type of workload. 378 00:23:12,480 --> 00:23:20,800 So they want to be able to support anything on the spectrum from applications to functions 379 00:23:20,800 --> 00:23:29,920 in a way that it can be auto-scaled and can be, the lifecycle can be managed from the developer's 380 00:23:29,920 --> 00:23:30,920 source code. 381 00:23:31,260 --> 00:23:37,080 So if you opt into having the build managed by the cluster, then that step is also in common. 382 00:23:38,180 --> 00:23:44,260 And, but for us on the Rift team, we were very focused on function-based runtime, the actual 383 00:23:44,260 --> 00:23:46,360 function-based use cases, right? 384 00:23:46,420 --> 00:23:57,400 So things like, I mean, there's the classic examples of something that runs relatively infrequently, 385 00:23:57,400 --> 00:24:05,600 so you don't want to have it consuming resources continuously, like the, when an image is uploaded 386 00:24:05,600 --> 00:24:09,560 to a directory, then you create thumbnails for it, or a video is uploaded and you create 387 00:24:09,560 --> 00:24:14,320 different formats for that video using a converter. 388 00:24:15,000 --> 00:24:22,740 But then there's also, in our world, many more kind of enterprise use cases where we see that 389 00:24:22,740 --> 00:24:28,400 developers are breaking down these monolithic web applications into microservices and some 390 00:24:28,400 --> 00:24:35,540 of the most fine-grained pieces that fall out from that end up being good examples for running 391 00:24:35,540 --> 00:24:36,080 as functions. 392 00:24:36,220 --> 00:24:44,300 Maybe they're called infrequently or they're things that, that can take advantage of streams 393 00:24:44,300 --> 00:24:47,040 of events with the model that I was describing earlier. 394 00:24:48,660 --> 00:24:53,900 Yeah, it seems like, I mean, one of the, one of the examples, one sees a lot is like a, or 395 00:24:53,900 --> 00:24:57,680 I, I guess I should say, I, I see a lot as like a reservation system. 396 00:24:57,820 --> 00:25:06,000 And, uh, I don't know, even, even at least, uh, I assume from a, uh, a consumer, a person 397 00:25:06,000 --> 00:25:10,500 point of view, I don't know if there's someone who reserves things not on this model, but like 398 00:25:10,500 --> 00:25:15,080 even, even when I like go to, to kayak or whatever, and just search for flights, like that seems 399 00:25:15,080 --> 00:25:19,980 like a sort of one-time operation that might work well, like something just scurries off 400 00:25:19,980 --> 00:25:21,600 and looks for things and then returns it. 401 00:25:21,680 --> 00:25:26,200 And then, you know, crawls off to die somewhere until it's brought back to life again. 402 00:25:26,660 --> 00:25:33,460 But, uh, yeah, I, I mean, I mean all of those, those, uh, that, that kind of, you know, it's 403 00:25:33,460 --> 00:25:38,700 interesting, like, like you mentioned, uh, the idea of, as we said, whether it's like, you 404 00:25:38,700 --> 00:25:44,040 know, from, from zero to in or whatever, or scaling up or being bursty, like that seems to 405 00:25:44,040 --> 00:25:50,500 be like a primary characteristic of interest in, in doing serverless and K native stuff. 406 00:25:50,540 --> 00:25:53,540 I mean, is that, am I, am I, am I overemphasizing that? 407 00:25:54,520 --> 00:25:56,360 No, I think that's definitely true. 408 00:25:56,360 --> 00:26:01,320 I mean, on one end of the spectrum, you're, you're scaling to zero for, for applications 409 00:26:01,320 --> 00:26:07,300 where you don't, or at least specific functions within a larger application that, that don't 410 00:26:07,300 --> 00:26:09,680 need to be continuously serving traffic. 411 00:26:09,940 --> 00:26:12,820 Uh, some of those might be, there's always a trade-off. 412 00:26:12,900 --> 00:26:18,340 It might be that the latency requirements are not, um, very important, right? 413 00:26:18,360 --> 00:26:22,120 So it's like something that happens offline in reaction to something else, like updating 414 00:26:22,120 --> 00:26:27,280 your, your, uh, flight miles or something after a flight, like it doesn't need to happen 415 00:26:27,280 --> 00:26:28,040 instantaneously. 416 00:26:28,460 --> 00:26:31,840 Um, other examples would be things that are very bursty. 417 00:26:31,840 --> 00:26:38,620 And so then you're, you're kind of on the, um, the scaling to inside, um, and that might 418 00:26:38,620 --> 00:26:44,980 be because there's irregular traffic or there might be something like, um, a sale for an 419 00:26:44,980 --> 00:26:45,740 online retailer. 420 00:26:45,740 --> 00:26:51,800 And so obviously they have, um, very inconsistent traffic and times when they need to scale much 421 00:26:51,800 --> 00:26:53,980 higher than the, than the steady state. 422 00:26:54,480 --> 00:26:55,000 Yeah. 423 00:26:55,180 --> 00:27:01,600 I think that more and more we see that, that even more traditional enterprise use cases start 424 00:27:01,600 --> 00:27:08,580 to fall into these categories because operations are running 24 seven and, um, things are, are 425 00:27:08,580 --> 00:27:10,900 moving online that didn't used to be. 426 00:27:10,900 --> 00:27:15,780 And you've got more and more fine grained ways of interacting with customers and chat bots 427 00:27:15,780 --> 00:27:22,300 and like there's more and more entry points into the system where, um, having, being able 428 00:27:22,300 --> 00:27:26,240 to break things up into more fine grained components is, is useful. 429 00:27:26,880 --> 00:27:27,400 Yeah. 430 00:27:27,460 --> 00:27:33,220 I mean, that definitely seems the, uh, it's, it's sort of like the issue, the issue, the, 431 00:27:33,220 --> 00:27:37,880 the fun mind problem I've always had around like, like, uh, serverless and functional programming 432 00:27:37,880 --> 00:27:42,840 stuff is to think about like, what are some types of applications that would be a poor fit 433 00:27:42,840 --> 00:27:43,300 for that? 434 00:27:43,380 --> 00:27:48,380 And, and as you're kind of saying that the reason why it's kind of a weird or a delightfully 435 00:27:48,380 --> 00:27:54,160 endless question is, well, if you can break things down into components, then almost anything 436 00:27:54,160 --> 00:27:58,720 can, I don't know, maybe not like video game playing or something like maybe, maybe if there's 437 00:27:58,720 --> 00:28:03,080 things that are like extremely real time and high latency, like you just have an ongoing 438 00:28:03,080 --> 00:28:07,860 process or whatever, but most of life isn't like this, this phone call. 439 00:28:07,880 --> 00:28:14,800 I don't know if this would work out serverless wise, but it's, it seems like a huge amount 440 00:28:14,800 --> 00:28:18,800 of things you could break down into a series of components that are linked together. 441 00:28:19,420 --> 00:28:19,820 Right. 442 00:28:20,340 --> 00:28:24,220 If you're starting from scratch though, I think what's always interesting there is if you were 443 00:28:24,220 --> 00:28:25,620 building some of these systems from zero. 444 00:28:25,840 --> 00:28:26,000 Yeah. 445 00:28:26,120 --> 00:28:31,460 But I don't think Mark, have you seen much that migrates just natively to a function platform? 446 00:28:31,460 --> 00:28:34,460 It seems like that's often at least a refactor or rewrite. 447 00:28:34,460 --> 00:28:39,320 So, you know, sometimes I know, Coté, you're interested in also some of the legacy app story 448 00:28:39,320 --> 00:28:39,580 here. 449 00:28:39,840 --> 00:28:44,900 I guess there's that decision of, should you refactor an app into a series of interconnected 450 00:28:44,900 --> 00:28:45,880 functions? 451 00:28:46,060 --> 00:28:51,540 Or is this something you see for new types of use cases or kind of what's your take on serverless 452 00:28:51,540 --> 00:28:55,200 as a whole and how this is supposed to fit into somebody who probably already has a lot 453 00:28:55,200 --> 00:28:56,240 of investments in software? 454 00:28:58,100 --> 00:28:58,560 Yeah. 455 00:28:58,580 --> 00:29:08,000 I mean, I think that's, it's definitely more of a, it's, it's more appealing as a runtime 456 00:29:08,000 --> 00:29:11,420 model for a greenfield development. 457 00:29:11,420 --> 00:29:17,960 I think that on the just like pragmatically looking at the way people break down a monolith 458 00:29:17,960 --> 00:29:24,480 is often following patterns like the, the strangling the monolith pattern. 459 00:29:24,480 --> 00:29:24,800 Right. 460 00:29:25,260 --> 00:29:28,780 And usually there's a point where you reach diminishing returns. 461 00:29:28,780 --> 00:29:33,780 So you're not going to literally break that monolith down into a thousand functions and then, you 462 00:29:33,780 --> 00:29:34,380 know, you're done. 463 00:29:34,440 --> 00:29:40,860 It's a, it's often just some of the, the pieces get chipped off as, as functions where 464 00:29:40,860 --> 00:29:41,620 it's most obvious. 465 00:29:41,620 --> 00:29:47,760 And at some point you realize that the rest you can leave as it is, but the, the, some of 466 00:29:47,760 --> 00:29:53,420 the drivers for that are not only latency requirements or things that run offline or things that are 467 00:29:53,420 --> 00:30:00,680 called in frequently, but also once you do have these very fine grained kind of single responsibility 468 00:30:00,680 --> 00:30:05,880 endpoints that can be exposed independently as functions, then you know, that the scaling 469 00:30:05,880 --> 00:30:10,760 characteristics can be determined based on the traffic to that thing, because it's only 470 00:30:10,760 --> 00:30:15,760 called to do that one thing, as opposed to some giant web app that is called for a thousand 471 00:30:15,760 --> 00:30:16,520 different reasons. 472 00:30:16,620 --> 00:30:21,620 And it's not easy to, to determine, um, how to scale that. 473 00:30:21,620 --> 00:30:26,460 So once you get to that really fine grained function, then things like auto scaling become, 474 00:30:26,460 --> 00:30:35,160 uh, possible where with large web apps with larger surface areas, it's, um, it's not always 475 00:30:35,160 --> 00:30:39,880 a, something you can actually drive to reality. 476 00:30:39,880 --> 00:30:41,640 Auto scaling, right. 477 00:30:42,620 --> 00:30:42,940 Yeah. 478 00:30:42,940 --> 00:30:48,400 I mean, do you, uh, from, from talking to you and, and even your answers here, it sounds 479 00:30:48,400 --> 00:30:52,840 like you think that you can be quote unquote serverless anywhere, right? 480 00:30:52,840 --> 00:30:56,220 I mean, even the K native, I guess, lends itself to saying it's not just a public cloud 481 00:30:56,220 --> 00:30:57,100 only thing. 482 00:30:57,100 --> 00:31:00,780 On one hand, I see plenty of debate that, Hey, serverless is really something that can only 483 00:31:00,780 --> 00:31:02,340 happen in public cloud. 484 00:31:02,360 --> 00:31:06,160 Cause it's the only place you can literally hide the infrastructure and just up an API 485 00:31:06,160 --> 00:31:07,860 or an endpoint versus on-prem. 486 00:31:07,940 --> 00:31:11,660 You're going to have Kubernetes pods to deal with virtual physical infrastructure. 487 00:31:11,660 --> 00:31:14,100 So stop trying to bring that paradigm on-prem. 488 00:31:14,200 --> 00:31:18,780 We should be steering everybody, the public kind of fall on that discussion of, should we 489 00:31:18,780 --> 00:31:22,840 be investing in quote unquote serverless like tech that would run in a, in your own data 490 00:31:22,840 --> 00:31:23,180 center? 491 00:31:24,360 --> 00:31:24,760 Yeah. 492 00:31:24,880 --> 00:31:29,480 So, I mean, I think obviously there's, um, probably the only thing that, that, that people 493 00:31:29,480 --> 00:31:35,620 agree about serverless is that it's a very confusing term that nobody agrees on. 494 00:31:36,160 --> 00:31:40,220 Um, and I think part of that is how literally do you take it? 495 00:31:40,220 --> 00:31:47,700 Um, and another part of that is because there, there's definitely one viewpoint that you, you 496 00:31:47,700 --> 00:31:52,520 can't talk about serverless unless it's public cloud, because otherwise you have servers and, 497 00:31:52,640 --> 00:31:57,340 um, but then you can take that to the extreme and say, well, you know, Amazon also has servers. 498 00:31:57,740 --> 00:32:03,820 Uh, eventually there are servers and, um, you know, if someone at Amazon is using Lambda, 499 00:32:04,340 --> 00:32:06,000 does that mean it's not serverless? 500 00:32:06,080 --> 00:32:06,480 I don't know. 501 00:32:06,980 --> 00:32:14,620 Um, so I think that one, one way to look at it is that if, and this is especially relevant 502 00:32:14,620 --> 00:32:20,660 for, for us, because it's our primary focal point is that there's the developer experience 503 00:32:20,660 --> 00:32:23,440 that it's somewhat separate from the operator's experience. 504 00:32:24,220 --> 00:32:30,640 And so I think if you're running serverless on a public cloud, then it ends up, um, removing 505 00:32:30,640 --> 00:32:33,880 responsibility, uh, not all of it. 506 00:32:33,960 --> 00:32:37,800 I don't think there's any such thing as no ops, but it does remove, or at least shift the 507 00:32:37,800 --> 00:32:40,760 responsibilities for operators as well as developers. 508 00:32:41,140 --> 00:32:46,000 But you can still have something that's called serverless, um, whether you like the term or 509 00:32:46,000 --> 00:32:51,640 not, that runs, uh, that provides a simplified experience for developers. 510 00:32:51,640 --> 00:32:58,320 And that's just meaning that they, something that they can focus on their code and focus 511 00:32:58,320 --> 00:33:04,440 on the smallest scope of the code that's, that's solving a business problem and be able to upload 512 00:33:04,440 --> 00:33:09,620 that in some way into the cloud without thinking about actual server processes that they need 513 00:33:09,620 --> 00:33:10,160 to manage. 514 00:33:11,620 --> 00:33:17,260 So it's very fuzzy, but, um, I think putting it between the developer and operation, uh, 515 00:33:17,580 --> 00:33:20,240 responsibilities is, is one way to help with that. 516 00:33:21,520 --> 00:33:21,920 Yeah. 517 00:33:22,140 --> 00:33:27,440 It's, it's, it's hard to nail a term like that because, uh, uh, I mean, I mean, it's because 518 00:33:27,440 --> 00:33:32,240 basically it's just like, there's less stuff developers need to worry about, but then, but then 519 00:33:32,240 --> 00:33:35,640 on the other hand, you don't want to make them all goofy and be like, you don't need to worry 520 00:33:35,640 --> 00:33:36,380 about anything. 521 00:33:37,320 --> 00:33:41,080 There's still lots of considerations that need to come into play, but I don't know. 522 00:33:41,280 --> 00:33:45,040 I think, I think it's like, like most terms is just like, it's, it's a little too late 523 00:33:45,040 --> 00:33:46,800 to come up with something, something new. 524 00:33:46,900 --> 00:33:48,900 And then, and then we would just debate that. 525 00:33:49,780 --> 00:33:50,140 Right. 526 00:33:50,240 --> 00:33:55,900 And, and I think, I mean, what's interesting about the points you make about, uh, overdoing 527 00:33:55,900 --> 00:33:59,540 it to where you don't have to worry about anything is that on the flip side, there are actually 528 00:33:59,540 --> 00:34:05,980 some things that become, uh, more of a concern or at least more of a challenge to, to worry 529 00:34:05,980 --> 00:34:10,940 about when you're running in this kind of a model, because, uh, for instance, if you're 530 00:34:10,940 --> 00:34:15,260 if your containers are going away, scaling to zero, then you need to be able to have access 531 00:34:15,260 --> 00:34:18,620 to logs for things that may be failed, but are now gone. 532 00:34:18,620 --> 00:34:22,900 Um, you need to be able to, once you've broken things up into such fine grained functions 533 00:34:22,900 --> 00:34:28,200 that all interact with each other in a number of different ways, you need to be able to see 534 00:34:28,200 --> 00:34:34,260 the trace of a, of a, uh, user transaction that spanned a dozen different functions. 535 00:34:34,260 --> 00:34:34,580 Right. 536 00:34:34,580 --> 00:34:41,320 So things like logging and all the observability and, and, um, and metrics become a bigger challenge, 537 00:34:41,320 --> 00:34:46,160 um, at least for the developers of platforms that the developers are using. 538 00:34:46,520 --> 00:34:52,380 And one of the things I noticed when, uh, I kind of got into this a little more and followed 539 00:34:52,380 --> 00:34:56,280 your work on Knative is yes, there's those things you call out, which, Hey, now you have 540 00:34:56,280 --> 00:35:00,540 to think of new things and think of instrumentation differently and monitoring and all that. 541 00:35:00,540 --> 00:35:05,780 But there's things that I liked that we added to Knative for all of us to make it also easier 542 00:35:05,780 --> 00:35:09,820 for devs to do in general, two things you, you quick touched on earlier, but could you explain 543 00:35:09,820 --> 00:35:13,740 a little more about the build system, which I think is interesting and obviously the venting 544 00:35:13,740 --> 00:35:16,580 subsystem, which has a lot of, I guess, riffs heritage in. 545 00:35:17,040 --> 00:35:21,700 So to me, those are beyond just the serving component of just run apps, scale them to zero, 546 00:35:21,920 --> 00:35:26,260 have revisions, all really good stuff, but the build and event actually pretty neat things 547 00:35:26,260 --> 00:35:30,800 that are arguably trying to make the developer experience more powerful and more interesting. 548 00:35:30,920 --> 00:35:32,380 Could you, could you explain those a bit? 549 00:35:33,020 --> 00:35:33,220 Sure. 550 00:35:33,400 --> 00:35:33,580 Yeah. 551 00:35:33,660 --> 00:35:39,800 So with the build, um, the basic idea in Knative is that you, you enter the system, 552 00:35:39,820 --> 00:35:45,580 with a, uh, component that it's a custom resource in Kubernetes called configuration. 553 00:35:46,020 --> 00:35:52,480 And that's the thing that points to the source code or a source image. 554 00:35:52,480 --> 00:35:55,200 And so that's basically, those are the two options that you have. 555 00:35:55,240 --> 00:36:00,060 You can, you could have already built an image, whether you did it manually on your laptop and 556 00:36:00,060 --> 00:36:04,560 pushed it somewhere, or as part of, uh, some CI pipeline, right? 557 00:36:04,580 --> 00:36:05,700 There's already an image. 558 00:36:05,700 --> 00:36:11,220 Then you just point to the image and the configuration, uh, that you either create or update points 559 00:36:11,220 --> 00:36:12,560 to that image. 560 00:36:12,560 --> 00:36:14,580 And that triggers a deployment. 561 00:36:15,320 --> 00:36:20,260 And every time you make a change to a configuration, whether you've changed the image tag or something, 562 00:36:20,260 --> 00:36:23,280 then that will trigger a new revision to be deployed. 563 00:36:23,500 --> 00:36:26,460 And so that part is the serving core. 564 00:36:26,900 --> 00:36:32,580 Now, what's interesting is that if you choose to, then instead of pointing directly to an image in the 565 00:36:32,580 --> 00:36:35,940 configuration, you can point to a build, uh, process. 566 00:36:35,940 --> 00:36:40,520 You basically, um, choose a build template and then provide the parameters for that template. 567 00:36:40,800 --> 00:36:45,120 And so that could be anything ranging from, um, something that we've been working on. 568 00:36:45,120 --> 00:36:50,960 That's obviously, uh, in the, in the DNA at pivotal is build pack support, right? 569 00:36:50,960 --> 00:36:56,100 So we're working now on, uh, riff building on top of K native, being able to use build packs 570 00:36:56,100 --> 00:37:02,500 to create the, to ultimately create the image from just a developer's function code and the 571 00:37:02,500 --> 00:37:07,100 parameters that point to what base that needs to be added on top of and, and what other things 572 00:37:07,100 --> 00:37:08,200 you need to configure. 573 00:37:08,360 --> 00:37:09,860 And so that drives a build pack. 574 00:37:09,860 --> 00:37:15,920 You can also have something that delegates to say the Google cloud builder or something that runs, 575 00:37:15,920 --> 00:37:23,660 um, kind of in between approach, which is, uh, something like Kaneko, which can take, um, a Docker 576 00:37:23,660 --> 00:37:27,340 file and build an image in a container and then, and then push that. 577 00:37:27,560 --> 00:37:29,220 So there are a number of different options. 578 00:37:29,220 --> 00:37:35,200 And the whole idea behind the build template is that it's, it's, uh, configurable and extensible 579 00:37:35,200 --> 00:37:40,780 and can support this whole range of, of options, but you always have the ability to just come 580 00:37:40,780 --> 00:37:44,340 with an image as well, um, on the eventing side. 581 00:37:44,340 --> 00:37:52,900 So that's more about how request traffic, uh, makes its way to a route that is sitting in 582 00:37:52,900 --> 00:37:53,960 front of those revisions. 583 00:37:54,520 --> 00:38:02,540 And, um, um, the idea there is that, um, you can plug into event sources and those could 584 00:38:02,540 --> 00:38:11,300 be things like databases or cloud services or file systems, and those will be adapted so 585 00:38:11,300 --> 00:38:18,620 that the events that come off of those end up, um, passing through the, uh, route for, for 586 00:38:18,620 --> 00:38:20,140 different revisions that they target. 587 00:38:20,540 --> 00:38:26,000 And what we, uh, were most involved in on the riff team was, was helping to define a model 588 00:38:26,000 --> 00:38:29,200 around a pub sub system, basically. 589 00:38:29,200 --> 00:38:33,300 So there's a, there's a custom resource called channel, and then there's a custom resource 590 00:38:33,300 --> 00:38:37,180 called bus, which is an implementation that can back those channels. 591 00:38:37,180 --> 00:38:42,140 And that's where we brought the Kafka implementation from riff as one example. 592 00:38:42,140 --> 00:38:46,180 But now there's also a Google cloud pub sub implementation of that. 593 00:38:46,820 --> 00:38:52,060 And, um, there will be others, uh, for that bus implementation to back a channel. 594 00:38:52,680 --> 00:38:57,780 And the other resource then is the subscription, which allows you to point to a given channel 595 00:38:57,780 --> 00:39:04,800 and then, uh, have, um, you can have a one to many relationship from a channel to various, 596 00:39:04,800 --> 00:39:06,440 uh, services. 597 00:39:06,440 --> 00:39:12,340 And those services are, are basically the end points, uh, typically for your revisions that 598 00:39:12,340 --> 00:39:16,360 were triggered by the creation of a, of a K native configuration. 599 00:39:16,720 --> 00:39:23,080 So you can create a model where you maybe are listening to say a GitHub web hook as an 600 00:39:23,080 --> 00:39:29,740 input source or a Google cloud storage bucket as an input source, or any of the number of 601 00:39:29,740 --> 00:39:35,180 adapters that we think will start to grow in a, in an ecosystem that's driven by the community. 602 00:39:35,420 --> 00:39:40,080 And then you, you point those to a channel and then downstream from that channel, you can 603 00:39:40,080 --> 00:39:41,000 add different subscribers. 604 00:39:41,000 --> 00:39:46,820 And that's the part that we needed to have at a level that we could replatform riff. 605 00:39:47,140 --> 00:39:49,660 And, um, and so we were able to do that. 606 00:39:49,660 --> 00:39:52,120 And now we're continuing to make those contributions there. 607 00:39:52,620 --> 00:39:56,780 Well, well, so, uh, you know, I think, I think as you're describing to use one of my favorite 608 00:39:56,780 --> 00:40:01,640 templates for things like, like any good technology, it can be as complicated as you don't want 609 00:40:01,640 --> 00:40:02,080 it to be. 610 00:40:02,760 --> 00:40:08,720 And, and, and to that end, like, like putting aside the legacy stuff, uh, cause I mean, you've 611 00:40:08,720 --> 00:40:13,780 kind of covered that in regards to the following questions or as much as is necessary. 612 00:40:14,080 --> 00:40:18,520 Like if, if you were sitting down to like, let's say you can pick something else, but make 613 00:40:18,520 --> 00:40:23,020 a reservation system, like what kind of planning do you start doing? 614 00:40:23,420 --> 00:40:27,500 Uh, you know, if you're going to do things in a serverless way and, and, and let's assume 615 00:40:27,500 --> 00:40:32,200 the old thing, you know, of like, you validate that this is the right technology to pick and, 616 00:40:32,200 --> 00:40:33,580 and, and all that kind of stuff. 617 00:40:33,720 --> 00:40:39,840 But like, what kind of pre planning architecture stuff do you do? 618 00:40:39,920 --> 00:40:43,720 What, what, what do you think through of, we need these components and what's the big 619 00:40:43,720 --> 00:40:46,760 wall look like that you got a bunch of boxes and lines on? 620 00:40:48,020 --> 00:40:56,900 Well, I mean, I think to a large extent, it's, it's similar to any kind of greenfield microservices 621 00:40:56,900 --> 00:41:05,080 type approach where, um, most of the ideas from, uh, domain driven design apply. 622 00:41:05,220 --> 00:41:11,900 So you think about where are your bounded contexts, where does your state for each of those need 623 00:41:11,900 --> 00:41:19,000 to be stored and which things interact with each other and what makes up the aggregates 624 00:41:19,000 --> 00:41:21,880 that are being shared, uh, between those boundaries. 625 00:41:22,440 --> 00:41:31,180 And then you map those things themselves into functions and the, what, the state that they 626 00:41:31,180 --> 00:41:36,160 need would be mapped into services that, that those functions need to connect to. 627 00:41:36,420 --> 00:41:37,520 Well, that makes sense. 628 00:41:37,840 --> 00:41:38,720 I mean, it's a good question. 629 00:41:38,940 --> 00:41:43,840 I, I, it's, it seems like the key thing is always, uh, you know, where, where's your state? 630 00:41:44,760 --> 00:41:45,240 Right. 631 00:41:45,380 --> 00:41:50,260 I shouldn't say it's not the key thing, uh, but it's a, uh, it's a, what would you say? 632 00:41:50,320 --> 00:41:51,420 A necessary thing. 633 00:41:51,600 --> 00:41:53,360 You always have to have that involved somewhere. 634 00:41:54,160 --> 00:41:54,280 Right. 635 00:41:55,280 --> 00:41:55,640 Yeah. 636 00:41:55,680 --> 00:42:00,200 So one of those technologies that you may have chosen ahead of time that Coté just mentioned, 637 00:42:00,300 --> 00:42:04,540 but one that we didn't talk about here, but I've seen kind of get mixed up as we talk about 638 00:42:04,540 --> 00:42:07,080 Knative and Riff and things like that is Spring Cloud Function. 639 00:42:07,280 --> 00:42:09,340 I know you had some involvement in that. 640 00:42:09,380 --> 00:42:10,620 So that's not the same as Riff. 641 00:42:11,040 --> 00:42:12,160 It's not the same as Knative. 642 00:42:12,320 --> 00:42:16,720 Could you kind of talk about this other piece of the pivotal serverless universe and Spring 643 00:42:16,720 --> 00:42:18,800 Cloud Function and kind of summarize what that is? 644 00:42:20,080 --> 00:42:20,560 Sure. 645 00:42:20,560 --> 00:42:28,380 Um, yeah, I mean, I think from my, from my perspective, there's, there's a, um, an evolutionary 646 00:42:28,380 --> 00:42:33,740 aspect to this as well, because I started working on Spring Cloud Function when I was first 647 00:42:33,740 --> 00:42:41,620 looking at, um, what serverless could mean for Spring developers and taking the typical Spring 648 00:42:41,620 --> 00:42:47,720 philosophy of things, which is basically all about inversion of control and letting developers 649 00:42:47,720 --> 00:42:52,280 focus on business logic and, uh, we usually refer to the Hollywood principle, right? 650 00:42:52,340 --> 00:42:53,800 Like don't call us, we'll call you. 651 00:42:54,180 --> 00:43:01,080 And so extending that framework up to the level where we'll call you with the smallest amount 652 00:43:01,080 --> 00:43:04,200 of code that you could possibly need to worry about. 653 00:43:04,200 --> 00:43:09,280 Um, so that was, um, so that was, that was the original idea behind Spring Cloud Function. 654 00:43:09,280 --> 00:43:15,440 And, and in some sense, it was just an evolution itself of Spring Cloud Stream where with Spring 655 00:43:15,440 --> 00:43:20,020 Cloud Stream, you would configure the binder through properties. 656 00:43:20,020 --> 00:43:22,100 Like I want to connect to Kafka or I want to connect to Rabbit. 657 00:43:22,400 --> 00:43:27,740 And then my code is in this one class annotated with Stream Listener. 658 00:43:27,940 --> 00:43:31,880 And so Spring Cloud Function basically just said, you don't need to annotate anything. 659 00:43:31,880 --> 00:43:33,220 You can just provide a function. 660 00:43:33,620 --> 00:43:38,640 And if you're building an app that exposes a single function, it was very easy to, to 661 00:43:38,640 --> 00:43:45,960 just provide the function and then let us provide the dependencies that wrapped that up into something 662 00:43:45,960 --> 00:43:46,440 runnable. 663 00:43:46,440 --> 00:43:53,000 And what that meant is then, uh, because in addition to inversion and control, we also want 664 00:43:53,000 --> 00:43:57,760 to have portability so that you can write the same code locally that you're running on some 665 00:43:57,760 --> 00:43:58,200 platform. 666 00:43:58,340 --> 00:44:04,520 And what I was noticing at the time is that even though fast platforms like Lambda are 667 00:44:04,520 --> 00:44:09,260 focused on the developer writing smaller amount of code, there was still something about the 668 00:44:09,260 --> 00:44:14,860 platform in the contract that you had to implement, especially for Java, I would say it's maybe 669 00:44:14,860 --> 00:44:21,180 not as much the case for, for JavaScript, but you would need to extend the framework classes 670 00:44:21,180 --> 00:44:22,800 or implement the frameworks interfaces. 671 00:44:23,540 --> 00:44:27,420 And so what we did with Spring Cloud Function is we provided just the thinnest possible layer 672 00:44:27,420 --> 00:44:31,020 that implemented or extended the target platform. 673 00:44:31,760 --> 00:44:33,620 And then that layer calls your code. 674 00:44:33,960 --> 00:44:38,080 And so you can swap out different dependencies for different of those adapter layers to run 675 00:44:38,080 --> 00:44:44,700 on different FaaS runtimes like Lambda and Azure and, um, and OpenWisk as well. 676 00:44:44,800 --> 00:44:46,660 So we have adapters for all of those. 677 00:44:46,840 --> 00:44:53,260 And then since Spring Cloud Function could also be used as just a Spring boot app, which means 678 00:44:53,260 --> 00:44:55,320 it can be an executable jar. 679 00:44:56,000 --> 00:44:59,540 That's where things maybe start to seem blurry because I can take a Spring Cloud Function app 680 00:44:59,540 --> 00:45:02,800 and just run it, um, on its own. 681 00:45:03,080 --> 00:45:06,180 And so that's where it feels like it's not just a framework. 682 00:45:06,180 --> 00:45:07,160 It's also a runtime. 683 00:45:07,680 --> 00:45:11,360 Um, but we, we build on top of that actually for Rift. 684 00:45:11,460 --> 00:45:16,360 If you're building a Java application in Rift and you just provide the function, like the 685 00:45:16,360 --> 00:45:19,900 literal Java util function function implementation. 686 00:45:20,900 --> 00:45:29,260 And, um, and that gets laid on top of a base image that has the Java function invoker, which is actually 687 00:45:29,260 --> 00:45:31,640 a Spring Cloud Function, um, foundation. 688 00:45:31,640 --> 00:45:37,400 And it's a boot app that then spins up in the container in Rift on Knative. 689 00:45:38,180 --> 00:45:39,640 Lots of layers there, but. 690 00:45:41,560 --> 00:45:46,760 But I mean, well, well, well, as always, if you don't want like a, uh, we were talking about 691 00:45:46,760 --> 00:45:50,900 earlier, if you don't want a tremendous amount of dependency and lock-in, then you need layers. 692 00:45:51,320 --> 00:45:53,780 Like layers are the answer to that. 693 00:45:54,660 --> 00:45:59,520 I mean, you're, I mean, Spring Cloud Function, kind of you bring the FAS versus Rift as you 694 00:45:59,520 --> 00:46:00,200 bring the function. 695 00:46:00,340 --> 00:46:02,240 Is that at least coarsely fair? 696 00:46:02,920 --> 00:46:05,340 Yeah, I think that's, that's a good way to look at it. 697 00:46:06,000 --> 00:46:06,200 Yeah. 698 00:46:06,260 --> 00:46:14,120 And I mean, even taking your, the idea of inversion of control always, uh, with Spring, um, has 699 00:46:14,120 --> 00:46:21,120 always been that you, you build a jar from the dependencies, the framework dependencies. 700 00:46:21,320 --> 00:46:23,660 And then also your code, right? 701 00:46:23,780 --> 00:46:29,260 And with the difference really with Rift is that we cross over this line where you don't 702 00:46:29,260 --> 00:46:32,300 bring the, the framework dependencies in your jar. 703 00:46:32,420 --> 00:46:34,060 You just give us the jar that's the function. 704 00:46:34,180 --> 00:46:40,000 And so instead of, it's not only don't call us, we'll call you, but it's even don't depend 705 00:46:40,000 --> 00:46:41,560 on us, we'll depend on you. 706 00:46:41,920 --> 00:46:49,040 And so you just, the, the container that spins up has the user provided code as a dependency 707 00:46:49,040 --> 00:46:56,720 that it then loads and invokes, um, and you, as a developer, you can write unit tests for 708 00:46:56,720 --> 00:47:03,040 that function, but you don't even, unless you choose to use a Spring library for something 709 00:47:03,040 --> 00:47:07,980 within the scope of your code, you don't even have the dependencies on the framework, um, in 710 00:47:07,980 --> 00:47:08,560 your artifact. 711 00:47:09,840 --> 00:47:10,320 Hmm. 712 00:47:11,320 --> 00:47:12,240 That makes sense. 713 00:47:12,540 --> 00:47:15,500 Well, uh, thanks for being on to go over this stuff. 714 00:47:15,540 --> 00:47:16,020 It was nice. 715 00:47:16,280 --> 00:47:20,200 We get to talk about fish and then, uh, you know, don't depend on us. 716 00:47:20,260 --> 00:47:21,140 We'll depend on you. 717 00:47:21,580 --> 00:47:22,840 It's, that was good. 718 00:47:23,120 --> 00:47:28,520 So, uh, I mean, I mean, finally, um, what, what do you think, like, like, what are, what 719 00:47:28,520 --> 00:47:30,640 are y'all planning on doing over the next year or so? 720 00:47:30,640 --> 00:47:33,040 Like, what's, what's the sense for the roadmap that you'll have? 721 00:47:34,780 --> 00:47:38,120 Well, I mean, there's quite a bit, it's still an early project. 722 00:47:38,440 --> 00:47:47,800 Um, and there's, uh, I, I think the Knative, uh, release right now is 0.11 and, um, same 723 00:47:47,800 --> 00:47:49,040 for project riff. 724 00:47:49,040 --> 00:47:56,120 So we, we, uh, when we sort of went into a quiet period where we were working on Knative, 725 00:47:56,220 --> 00:48:01,780 but before it was public, we stopped, um, somewhat intentionally on version 007. 726 00:48:02,500 --> 00:48:06,940 And, uh, we recently, after replatforming on Knative, released version 0.1. 727 00:48:07,240 --> 00:48:10,840 So, um, in both cases, there's a zero in the first digit. 728 00:48:10,840 --> 00:48:14,120 So that shows that we have, um, we still have some work to do. 729 00:48:14,620 --> 00:48:17,780 Um, we're also focused on the, on the pivotal side. 730 00:48:17,800 --> 00:48:23,880 On the, um, building the pivotal function service on top of riff as well. 731 00:48:24,000 --> 00:48:25,500 So that will be our commercial offering. 732 00:48:25,960 --> 00:48:29,420 And that's part of the pivotal cloud foundry umbrella. 733 00:48:30,260 --> 00:48:36,660 Um, so there's, there's quite a bit going on there as well, running riff as a replatformed 734 00:48:36,660 --> 00:48:43,980 on Knative component on top of Pivotal's container service and, um, and thinking more about what 735 00:48:43,980 --> 00:48:47,260 those commercial features are, uh, for that. 736 00:48:47,320 --> 00:48:53,020 And, um, and, and definitely all of those challenges that I mentioned earlier around 737 00:48:53,020 --> 00:48:59,220 observability and tracing and metrics and things that really turn it into a full platform. 738 00:49:00,260 --> 00:49:05,560 As you say, there's 0.3 more percent work to do or not percent 0.3 more work to do. 739 00:49:06,440 --> 00:49:08,020 So, no, no, that makes sense. 740 00:49:08,220 --> 00:49:12,020 And then, and then, uh, I mean, if people want to find more, I'm sure they can, uh, you 741 00:49:12,020 --> 00:49:15,640 know, if they wanted to get ahold of it or like read through the docs or whatever, what 742 00:49:15,640 --> 00:49:16,440 would you point them to? 743 00:49:16,940 --> 00:49:20,340 So there's project riff.io for riff itself. 744 00:49:20,500 --> 00:49:29,700 Um, there's the, uh, PFS site on pivotal.io and for Knative, there's the, uh, GitHub org, 745 00:49:29,700 --> 00:49:38,380 which is Knative, um, and there is a, um, I guess that pretty much covers, uh, the entry 746 00:49:38,380 --> 00:49:38,640 points. 747 00:49:38,720 --> 00:49:41,660 Everything else is accessible from there. 748 00:49:42,280 --> 00:49:42,960 Well, good. 749 00:49:43,040 --> 00:49:46,820 Well, thanks again for being on and, uh, for the listeners as always. 750 00:49:46,820 --> 00:49:47,720 Thanks for listening. 751 00:49:47,900 --> 00:49:53,520 This has been Pivotal Conversations and you can find, uh, this episode and all the other 752 00:49:53,520 --> 00:49:56,900 ones listed at soundcloud.com slash Pivotal Conversations. 753 00:49:57,240 --> 00:50:01,920 And every Thursday or so, uh, we post the full show notes where you can see those things that 754 00:50:01,920 --> 00:50:05,800 we skipped over and, uh, maybe I'll put a picture of me eating a herring. 755 00:50:05,800 --> 00:50:10,040 That'll be fun, uh, over on pivotal.io slash podcast. 756 00:50:10,180 --> 00:50:12,400 And with that, we'll see everyone next time. 757 00:50:12,400 --> 00:50:13,080 Bye-bye.