So it's been like a month since we recorded, huh? Isn't that shocking? That is shocking. Do you think we're going to have to slowly build back up our listener base from zero to ten? We've got to earn the trust. Earn the trust. Earn the trust. Well, you know, traditionally, August is a month of doing nothing, which was 16 days ago. So, you know, I'm always late for everything. So I think I'm on track here, so to speak. But that said, so I was at DevOps Days Auckland or New Zealand or something a couple weeks ago now, which is great. You know, it's always fun to go out and see the people. But I noticed in New Zealand. I don't know if you noticed that you were recently in Europe, too. I don't know if you're like me. Whenever you travel pretty much anywhere, you like to go tour the grocery stores. I'm guessing you don't like that, but you should start going to grocery stores. It's fun to see what they're doing there. And it is a, what with the thrilling per diem we have, you can get yourself quite a meal if you shop at the grocery store. Or maybe if the CFO is listening, you can be very economical in feeding yourself on the road. I don't want to be like a 80s Walmart advocate of sharing rooms with people because I feel like we're adults and not in summer camp. But anyways, go to the grocery store. So I go, you know, go to the grocery store and they have like five or six different types of bacon in the grocery stores there. And they don't even call it, I mean, they don't even call it bacon. It's specifically like, maybe they do, but it's like the type of thing like center cut bacon. And of course, being New Zealand, Australian people, they put a Y at the end of the edge of everything. So they call it streaky bacon. That's the kind of bacon we have. Nice. But it is, I mean, it is, it is a, it's a different perspective on the world when you've got like five or six different types of bacon. And I have to admit, you know, when you go to an American grocery store, you basically select bacon on two vectors. Well, three vectors, let's say brand and price. And if you want extra honey and pepper on it, that's pretty much your selection.
2:02I thought you were going to go with thickness. All right. I guess there is thickness. Maybe, maybe there's four. I think, I think thickness might be a variation on price. I would argue cheap bacon is thin, expensive bacon is thick. And if, if you're, I guess maybe for all of our West Coast listeners out there, if you're all into like, you know, avoiding nitrates, that's another vector. I don't know what a nitrate is, but it sounds delicious. It does. Well, I would really like to see the coffee table book of Coté visits international grocery stores. Like that would be just, I'd plop down a pre-order right there. Well, you know, you know, my overarching retirement goal, the only way I work at places is so I can slowly amass knowledge and the capital or reputation to borrow capital to fund my ultimate project called How to Eat, which I envision as a multi-channel media empire, kind of like Oprah Winfrey. Where basically I, or maybe more like Anthony Bourdain. Is that how you say his word? Where, where basically I go to different cultures and, and I describe, you know, if the food is good or bad. I don't care about that. But the mechanisms of eating it. Have I ever told you about this? No, I feel like you need to say more. Well, let me, let me say a little more. Cause I was talking to someone, uh, uh, about this over there in New Zealand and, uh, and throughout the day he looked over at me. He was like, you should really do that project except in that delightful accent, but maybe I can get some support here. I'll start a Kickstarter. Is that still a thing or a pay trio? Yeah. Yeah. I'll start a Da Vinci or something. I guess it wouldn't be Da Vinci. I'll start a Medici. Anyways. Uh, the idea is every single culture, they have a slightly different way of eating various dishes. Like I'll give you a couple of examples. One, if, if you had never encountered a taco, right. And I put a taco in front of you, you would probably be like, what do I do with this? Right? Like, do I pick it up or not? Now to complicate it even more, if you go to some of the more, um, old school places here in, in Austin and the rest of Texas, they will bring you a taco.
4:02If you buy a con, get a combo plate, that would be a whole other thing. What's a combo plate when I, if I get something called a federale, what am I in for? But they'll bring you as a first coat, first, uh, first course, a crispy taco. And sometimes the taco is pre-deconstructed. So you'll get like the shell and the meat and the lettuce and the tomato and the cheese, like all on the side. And you can put it all together on your own. Right? So then you got to explain that. Now, what if you have a salsa bar? What are you gonna do with the salsa bar? Now, have you ever been to a salsa bar where they've got those little pickled carrots? I forget what those are called. What do you do with those? Anyways. So then another one, like, I don't know, you've probably been to Korean barbecue places. Sure. So you order your Korean food and then they bring you out like, uh, I don't know, 10 or 20 of these little bowls with things in them. I have no idea what you're supposed to do with that. Like, is it like a chutney in India that I think is a compliment is a condiment or is it a thing on its own? Right? Like how do, how do, what's the, what's the, how do you operate all this stuff? And I've asked our Korean colleagues who I also saw one of them in New Zealand, like what you do with all those bowls and very indeterminate answer. So we, we got to get to the bottom of this. So all I'm saying is after the, as part of the, how to eat project, I think it makes sense. You'll go to the grocery stores sort of like, you know, when you're trying to fill your, uh, 20 to 25 minutes of minutes of content on a basic cable TV show and you need to have kind of like the sidebar discussions. Like while we were here, we also went to the grocery store, which is technically not part of the theme of this show, but we ran out of time. So we had to put some stuff in here. So we'll, we'll do that. That's a show. Yeah. Maybe I should do this in, uh, in, uh, what's it called? We chat. What do the kids do? Snapchat. Yeah. Yeah. Well, I'll, I'll round out my trip report. We also took my niece, my 16 year old niece and speaking, easing in to the exciting world of digital computers. She had a funny quote. We were asking her, uh, if she uses Facebook and you know, she's 16, she said, no, I've never really used Facebook. I feel like if I was in Facebook, I would just see pictures of my mom.
6:06So there you go. There's a lot of moms on there. It has crossed 2 billion users. A lot of moms. Yeah. Yeah. I don't know, but it was interesting that, uh, and we kept, we kept giving her, uh, the business about like, so are you going to take another picture that disappears in 30 seconds? So that's fun for us. Well, I think over the past month, there's been a, uh, bevy, which is a popular word in my young days, a bevy of news. And we of course won't summarize all of it here at all. I mean, you had, uh, you had a Oracle open world. There were some other conferences going on, all sorts of stuff, uh, that, that resulted in stuff, but we've selected a few things here. So first of all, setting some context, I saw that, um, Gartner came out with their, uh, is it application infrastructure middleware? I forget what aim stands for, but, uh, they came out with some market sizing for that. I have to admit, I haven't read it in detail. Like I usually do, but I think, I think sizing the market that pivotal and people who listen to this too, and people that do application development, uh, that we operate in historically is not difficult, but it's difficult to know what it is projected out to the future where you're like multi-cloud and cloud native. And you're figuring out if you want your, uh, JEE contain your JEE app platforms versus whatever other crazy stuff. So it's a little crazy now, but I think it's good to track that number to get a sense of the footprint. And their, their estimate, we'll, we'll just go with the 20 when 2016 was 25.5 billion. And in 2017, it increases up to 27.3 billion. I love the decimal places like, you know, this amount of precision. It's very exact. Yeah. Although I was thinking I should round this up and then I thought that would be equally worse. So what are you going to do? Uh, all the way to 2020, it's about, it goes up a 10 or so billion. Well, not a little less than that to 33.6 billion, which, you know, that's encouraging. It's good to know that the, uh, the way that I pay my rent is increasing in market size instead of, uh, instead of shrinking. Yeah. I need to read that as well to see what that all encompasses.
8:06Cause is that, is that good? Do we expect it to even be faster? Is that all encompassing? I mean, I don't think you, you or I, or I don't go by a week without some salesperson asking me for market sizing information. Everyone loves those sorts of numbers, I guess, to help customers even justify some of their purchases and things like that. Well, it is a, it has a few things that like, if I was doing some, um, strategy deck, I would subtract out. Like it has, uh, I think it might even have some business processing stuff and, and maybe some, some old school rad stuff, which anyways, knowing what to do with those would be weird, but yes, you would want to make a Mecco chart of the, uh, wouldn't a Mecco chart work? Whatever. You would want to make a chart breaking out the, uh, well, a multi-year Mecco chart. Those are, I don't know if you've ever seen those, those are great breaking out the different categories over the span of, of years covered. And that would give you, that would give you, that would be the best chart. Cause it would give you a good sense of the growth, the, the sub-segment growth and, uh, shrinkage as it were. Gotcha. I need to spend more time on the Gartner site than I already do. Thanks a lot. Well, it, I did. I don't know about you, but I got an email that at the end of October, now it's down to 10, but I have 10 more PDFs I need to download. Likewise. Maybe your tokens are expiring. No, no. Yeah. Mine are too. So I'm grabbing random things. I may not read just to feel like I'm using it all. Well, you've just described my entire approach to getting, so, so also, uh, more closer to home, but narrowing down from there, the, uh, a new version of the spring framework came out that of course is the, uh, the longstanding, uh, you know, I mean, I don't know why I'm explaining it to this, this audience base, but you know, it's, it's a, a layering on top of, uh, and it's own, it's own things as well, uh, Java and adding out all sorts of middleware and, um, as the name would imply frameworks, uh, that many companies use to do their, the enterprise app dev. Now I'll admit that just like with the Gartner stuff, I haven't dug deep into it, but I think, I think, you know, a little bit more, but it's got a whole bunch of reactive stuff, which as I'm fond of saying, I had Josh long explained to me once, and there's something about backflows and something or another, it sounded wonderful and, and like good stuff.
10:07Uh, but, but there's a whole lot of stuff, uh, in there, but what, what's, uh, what's your quick highlights of what's in the spring framework five? Yeah, lots of good stuff in there at five. It was, as you mentioned, it's the reactive programming stuff, giving you an alternate programming model to just a traditional blocking request response sort of thing, having this more of asynchronous callback model. It's meant to be more scalable, can be more complicated, which is why you look to frameworks to simplify things like thread management and things that I don't want to deal with as a developer because it's hard and I can screw things up. So spring takes care of a lot of that. What's cool is that a lot of that flows down throughout the platform, down to the data tiers, down to other components. So it's not just like front end stuff all the way down to data queries. So spring data also got its first big update in like eight years at the same time. So that was cool not to be sort of don't bundle together things like spring boot. Spring boot two is coming out here in the coming months, which builds on spring framework five spring framework five really are foundational bits. And then things like boot and spring cloud and other things all extend beyond that. So the base is now updated. Everybody's now using that to package up all the new latest and greatest stuff on top and all that. You know, the other things that you mentioned kind of this month of September, we had PCF 112, which was a big update. We had, I just did a blog post last week on all the things pivotal shipped in September. It was pretty wild stuff. So yeah, September was an awesome month. Spring framework five, probably the highlight. Same with green plum five, which was a huge update for that project as well. Well, we can do a little behind the curtain for our listeners. And I was just talking with our buddy, Jared, to do a little mini podcast on the new version of PCF since he wrote a blog post. So we'll have that out there sometime soon. I'll see if I can get the pivotal blog desk people to publish two of these in a week. I'll see what happens. It'll be exciting. You know, also, I was thinking if, uh, when and if I ever start some little VC fund to, uh, do middleware, I'm going to call it foundational bits and that'll, that'll be good. We'll just invest in. It's kind of what heavy bit does. If you've ever looked at their portfolio, it's basically a developer tools that they have.
12:10Got it. I don't know what a heavy bit is. Maybe that's some sort of nuclear thing. No, it's your bad name. There you go. But, you know, I spoke at their place once and, uh, they gave me a t-shirt and that's probably the t-shirt that I wear the most. One of those, uh, tri-blend things. Cause their logo is not too, um, nerd identifiable, very small. So I, you know, feel better. So speaking of nerd identifiable, uh, like there's a, there's a new book I started reading all about the nerd space, so to speak, um, called, called the four. And I don't know if you've ever come across this, uh, Scott Galloway. I don't know if it's Galloway or Galloway guy, but, um, I don't know. He's, he's one of those people who like makes a lot of inappropriate analogies in NBA speak. So he can be very entertaining or off-putting. Like, I mean, we'll get to talking about it, but, uh, in one of my recent columns, uh, someone had just thought after reading my stuff, they would advise the register that they should start hiring adults to write for them, which. I saw that. I like, I like that comment that made me laugh. So, so I think, I think, I think our buddy Scott here is clearly an adult, uh, but, uh, anyways, so he's a little bombastic, which can take a style to get over. And I haven't finished reading the book, but I think, I think what's why I bring it up here is so the four, I mean, whether you want to people come up with names for this, like, you know, Fang or whatever, uh, he doesn't cover Netflix. He alludes to them, but this one is basically Amazon, Facebook, Google, and, uh, Apple. And he sort of makes the case that these, obviously the tech's important and these, these four companies are basically coming to get everyone's lunch and or redefining what the competitive landscape looks like for business. Like I said, I haven't finished it yet, but it's, it's a nicer approach to that than the usual kind of throwaway business book that I've liked so far. And I think if, uh, if I can finish reading it, uh, by the end of this week, I've got another podcast I do where we, uh, spend way too, way too much time, excuse me, talking about business books and maybe we'll review this one, but, uh, I'll, I'll, I'll mention it again when we, uh, when I finish it up, it's, it's definitely worth looking into if you're interested
14:12in why you would care about technology and kind of in, as long as you've read the halo effect and you know what a halo effect is and how to filter, uh, analysis through that. It's interesting, uh, to learn from, uh, various assertions in there. Awesome. So then also just to, uh, this is more of a, um, what would you call it? Momentum indicator. Also there was, there was a puppet and chef to two of the, uh, venerable and only the best meanings of that term, uh, configuration management and devopsy and cloud companies. They both came out with some releases that I would grossly oversimplify as, uh, the build pipeline is really important. And so, so puppet, uh, acquired a company sometime back that's got a name that's funny enough that I don't remember it. Um, and, uh, essentially it adds a lot of, of functionality. They had an announcement that they were bringing to, to, they were GA-ing, if you will, a lot of, uh, pipeline management and packaging and all sorts of things like that. Right. And, and then chef has a similar one with habitat or product they've had for a while, which has cloud foundry support into, in it. And I think what's interesting about both of those, which is not to say that, that this way of doing configuring didn't do this, but it's, it's, you're starting to see the shift where because of containers, I guess, instead of going out to configure the server, you're now configuring the image or the container and then handing it off to another platform to install and do things for you, which again, is not like an absolute statement or whatever, but that's kind of in the configuration management space. It's I think an interesting technological trend to watch is when, when do your configuration management things stop going to the server or whatever and actually messing around with it and instead creating like whatever it is, an image or a container that then gets installed on it, which again, I mean, people have done that with VM images and stuff like that for a while, but it's a, it's still a somewhat of a shift from how things have been done.
16:16Yeah, no, it's, it's a good take. I mean, I think I saw those same things with Habitat and the things Puppet are doing. And others is just doing configuration management. There's more you can impact if you're these companies. So yeah, the build pipeline, actually dealing with app constructs and not just infrastructure, also things you're seeing them talk about more now is actually getting the apps to these environments, not just configuring the environments. I mean, we, we talk about a lot with Bosch, you know, managing things at scale and there's just bigger problems we're starting to solve now. So it's cool to see these, you know, venerable DevOps companies take a, take a swing at it. Well, so also in news, but we can focus a little bit more on it since it's closer in our world. The last week was the Cloud Foundry Summit EU, you know, about the entire ecosystem of, of, of Cloud Foundry, the open source thing and everything. So I think, I mean, there, there were many things, there's an overall press release and there's a few other ancillary announcements around it, but I don't know for, for, for my, for my money, not that I'm really spending any, cause I'm trying to fund that how to eat project, but sure, of course. And now, and now you've got me worried about budgeting for going to a grocery store. Maybe, maybe I can get the grocery stores to pay me to go there. Uh, as long as I talk with like a regional assistant manager or something, but anyhow. Uh, so there was basically, um, there's a lot of, um, uh, coalescing around basically taking Kubo, which is the way of, of running Kubernetes on top of, uh, not on top of with Cloud Foundry and, uh, making there, there's a, there's a fancy longer name. I, I think, I think either you give something a funny name nowadays, or it has to have four words in it. So that's exciting, but, uh, you know, it's, it's, what, what is it? The, the Cloud Foundry container runtime, which I think fits with the general naming scheme of things. So, well, they, they did it with the renaming as well of the Elastic runtime for Cloud Foundry application runtime, which really, I think finally helps people, I think, visualize how these things work together. That's right. That's right. And I've started to use this one slide at DevOps days that, uh, that, that we had, or,
18:19or was to be had by whoever the, the, the owner was, uh, back at VMworld that shows how, um, the, uh, the, the, the app part runs alongside, uh, the content, the container for that, excuse me, the PKS part, which I think, I think is a useful way of illustrating it. And, uh, there's another good, uh, excellent illustration of this by, by our friend, Ian Andrews, uh, Bart and George did a film with them. And, uh, I think Ian did his best James water impersonation, perhaps the best I've ever seen where he got, he got a, he got a sheet of paper and he drew out, he whiteboarded, uh, how this all fits together. And it is, um, Oh, I have a link to it here in the show notes, but it's a good general overview of all of cloud foundry. Um, and especially it's pretty detailed. Like it goes down to Bosch and talks about it, but it also talks about how, uh, not only the cloud foundry container runtime, but the pivotal container service, you know, of course fits into it and everything else. So it gives you a full sense of essentially with, with these two parts and then the, uh, the Bosch managed, um, I don't know, infrastructure use below, you really have like pretty much all the, the cloud native option, the modern options you would want for running, running applications, whether it's sort of like a super cloud native 12 factory stuff that, that most all of our customers have been using, uh, thus forth, or you want to go down the, uh, get, get your hands more dirty by running it in Kubernetes type approach that you have. Yeah, no good summary. I think if it's, I mean, Jules from IBM had a good post today talking about as we think of the application runtime versus the container runtime and talking about, like, it's not even just about who builds the container when you pick those two things. It's also his point was cloud foundry is optimized for stateless apps and you use some of these other technologies for more stateful apps. It's in that, you know, cloud foundry CF push is still one of the most powerful constructs available in technology today when it comes to application development that does a lot more than just build a container. It obviously handles some of the security components and hooks up monitoring and logging all these fun things. So it's neat when you have these sort of different runtimes all managed by the same infrastructure.
20:22That's a really powerful thing versus all these bespoke systems that you're stuck kind of stitching together with different security models, different patching cycles, different op skills. You know, this gets pretty interesting when you have the application runtime, container runtime, sitting here with that same foundation. Yeah, yeah. It's, I mean, it starts to be as, as we've called it a platform and I'll, I'll have to go find a link to this, but I was looking over some, uh, some of our road showed slides and, um, I think, uh, there's a couple of people who give it. Um, but there's one presentation that we have, uh, that, uh, that goes over like all these different things you were saying and how they kind of reduced down to a platform. And, and I think, uh, I think everything's kind of coalescing around that. You get the whole stack of stuff, the platform that you need to run everything. The, the other two things is there was, uh, there were several new members to the, the cloud foundry foundation. I think, I think it's always notable when they, uh, I mean, I love a tech vendor, but like whatever, you know, tech vendors will do tech vendor stuff. So it's great when they join and, uh, it's always enjoyable, but even more valuable is when people like home Depot, uh, actual enterprises join because they have, uh, such, such a, I don't know, such, such a usage and commitment therefore to the cloud foundry ecosystem that they want to go through the trouble of ponying and, you know, one goes through the trouble of ponying it up extra money. But then also as, uh, as, as hopefully most of them do getting involved in having a closer relationship with the, uh, the software that they use. It's almost like we're getting to that idea way back when from the, uh, rainbows and sandals air where people, uh, people really care about their source code and they want to contribute to it. Not, not just as certain people have suggested to the federal government recently that they should just pay someone to do that for them. And not worry about it. Right. No, I love it. You brought that up. That was the point I wanted to make was, you know, that was the thing I know governor talked about, James governor of red bunk talked about as well with, yeah. I mean, home Depot joining, you look at, what's it? 40% of the CFF foundation members are quote unquote enterprises that aren't software companies. They're not coming up with distributions. They're not building add ons, which is also all awesome, right? You love a foundation that also has investors who are contributing code and making sure you
22:26can plug, make things pluggable. That's all super exciting. But when you also have stakeholders, when you do have these enterprises who say, look, I'm betting my business on this tech and I want it to be successful. I think that's pretty exciting. And even back to, to something you and I talked about before we started to hear with even SpringOne, you know, we've, we've shared more of the agenda coming up with that. And that is just full of, again, more name brand enterprises than I've ever seen speak at a, speak at a tech conference, not a 10, because there's always been enterprise folks who attended conferences, but more of these name brand, like, oh my gosh, my parents know who these are companies who are speaking about this sort of deep investment in this technology that's changing how they operate. I think that's exciting stuff that we didn't see five or 10 years ago. All these tech conferences were full of vendors, sometimes talking to vendors. And now you've got enterprises saying like, look, we're, we're kind of software companies in a sense, and we're betting on some of these technologies at scale. Let's talk about it. I, that's, that's super great. And seeing that at CF summit was good too. Yeah. I just, to do a little ad for the SpringOne platform. I, uh, we were, uh, I think, I think we're supposed to help finalize some track thing, putting together all the, uh, the track stuff for conferences is always delightful. And I'm glad I basically don't have to do much there, but so, so we'll see how it evolves, but it is, I, you know, I, I'm, I'm one of the track chairs for the, um, the, uh, I don't know, DevOps and build pipeline and, uh, and miscellaneous track. But I know for sure that in that track, we, we spend a lot of time sourcing, uh, what you might call customer cases, people actually talking about how, how they've done this stuff there. And looking across the other ones, just as you were saying, like, I think, you know, I think the emphasis of the conference is very genuinely like, here's all the technical stuff for sure. We got plenty of that, but then also let's actually bring some actual users, uh, who can talk about their experience with it and share tips on putting it in place, which you don't always see in a lot of conferences. Most, I think most conferences like this, they, they skew more towards the technology part as they probably should. But, uh, I, I think everything in the cloud native area area is early enough that it's,
24:26it's important to start talking about, uh, all, all the people actually using it, which, which will be nice. And, and I think you can still use this. If you want to get $200 off registration, you can use the code S1P200 underscore. If you remember that key, uh, Coté, C O T E. I don't think anyone signed up for that, but I like this fantasy land where, uh, where I give out a code and people would actually get $200 off. Anyways, it's the beginning of December. Yeah, keep flogging it. It's it's no, it's, it's sincerely, I mean, as someone who attends a lot of conferences, it's the best technical program I've seen at a conference in years looking at the whole thing. So whether you're a spring developer or not, it's awesome stuff. And you're saying, again, you're seeing this at the CF summit as well, where it's not just cloud founder, it's also distributed systems concepts. It's Kubernetes, it's other things. So I think we're in this fun space right now where we're not just single, single focused on like one type of thing, but we're thinking about how do companies improve business performance through software? That's, I don't know, that's a lot of fun for me. And I think for a lot of our listeners, absolutely. All right. Well, uh, so then the last thing there, before we get to, uh, a little, a little overview of stuff we've been working on recently, um, speaking of self-promotion there, there's a, when we get to that part, there's also a new survey. The, uh, the cloud foundry foundations have been putting out a lot of surveys recently. And this one is a, uh, a user survey, which by its title and reading through demographics, I presume means people who use cloud foundry, which, which sounds good. And so I wrote, I wrote a quick summary on my own blog that I'll link to about my, uh, my highlights for it. But, you know, essentially the, I think, I think there's maybe, um, I would say three main takeaways. One of them is, and I don't know the exact methodology that they did this with, but it's fine. I'm sure is you can, there, there's a, some good before and after analysis of, uh, of release speed, right? Your ability to release software frequently. And as always like nearly releasing software, being able to release software frequently is not that big deal, right? Like you might be just changing the background color of a header in your 500 page application. But if you're using that as a way to, uh, one, keep things secure because you can release
26:30patches and, and, and fix things up, but to continually experiment and try out new ways that your software can accomplish whatever goals it is, then I think that is pretty indicative of a, uh, a helpful and healthy attribute that you have. So you can see in the, there's, there's kind of a hard to read chart there, but it gets summarized well about how, how people have, when they start using cloud foundry, they're able to speed that release cycle up, which, which is nice. And then I think, I think the second and third thing is there's, there's some analysis of surveying and therefore analysis of basically, uh, people do use multi-cloud things. So they, they run their cloud foundry instance on more than one type of infrastructure, which I think is, is interesting to see. Yeah. Or, and I guess intuitively it's kind of obvious, but it's always good to have validation. And then also as part of that is, um, there's some surveying about what kind of container orchestration frameworks people use, which, which is, uh, which is interesting as well, just from a informational point of view. And then finally, this is kind of small, but like there's some mention of, uh, the number of applications that a few different, uh, customers use. And so you can get a sense of the maturity of, of users and enterprises using a cloud foundry here and there. Like for example, the notes that Comcast has a 1500 developers using it and Home Depot has more than 2,500 developers. And then there's, um, there's some good, uh, survey results in pie charts about, uh, I think they break it out into maybe like 10 to 20 and then 50 and a hundred and then 500 plus, and then a thousand plus, uh, apps that people run in it. So it's a good, it's the title. I would have titled it something more like the state of the, uh, the cloud foundry user base. That would, that would be my first suggestion that would be worked on quite a bit, but, uh, to that. And I think it's a good, it's a good snapshot of what's going on with cloud foundry usage out there, uh, which. Yeah, that's a good summary. I thought, yeah, like you said, the multi-cloud was interesting.
28:30You still have 40 something percent on, on vSphere and really with another 22 on open stacks. You're still talking almost two thirds is sitting on premises or, but then you also see, you know, cause the numbers don't add up, uh, plenty of people still using public cloud with it, which is great, but I liked the point you made. Cause I also jumped out at me that you had what 28% of organizations with 50 or more developers using this. And to your pie chart number, it was, I think it was a third of these accounts are using more than 50 apps in the platform. So to me, it means like, this is not just, we installed this departmentally. We're now on some vendors logo slide because we signed a contract and pushed it to dev. Like, no, no, this is like, this is running a big part of our business and we have now dozens or hundreds of people using it. To me, that's when this tech starts to hit critical mass. It's not just kicking the tires. This is probably running at least one core business app at a lot of these companies, which is cool. Yeah. I was reading a summary of a summary of, uh, you know, as you noted before, every, uh, every Monday we get these notes about what's going on in the field and to kind of unhelpfully anonymize it, there was one anecdote I was hearing where, um, on a weekly basis, there was one of, one of our customers using pivotal cloud foundry that was updating one of their, uh, their, their core applications, I don't know, six or seven times a week. And no one, no one noticed in a good way, which is to say no one noticed that things might be broken or down or whatever. They were able to upgrade, uh, without impact, which, which that's, that's always heartening to hear. We'll get all these, these effects of, uh, you know, the, the crazy.com people in the real world. Well, you joked about it. Like, Hey, sometimes it's just changing a color scheme, but I think we've all worked in places where even doing that required like a quarterly release. Yeah. Yeah. Right. Just everything was so bundled up. It was a pain in the neck. If there was a bug or a feature problems, like, well, we'll fix it next six months instead of like, yeah, I fixed it now. By the time we finished talking, it's in production. Like that does change the game for a lot of these companies. Yeah. And, and so you, you wrote a post recently velocity as a service. I haven't ever heard the Vaz thing. I think we need to follow up with the, uh, with the Gartner Paz people to see if that's
30:30a new category. Don't, don't, don't do that. Don't make it in there. But, uh, so what, what were you, it was, it was five ways to ship like a software company. So what are these, uh, what are the tips you were throwing out there? Yeah. Cause I didn't want to make it just nothing but a humble brag post about all the software we shipped because my, my, my really conclusion point was, I think it was more like a helpful brag post. Yeah. Maybe that's out. I mean, my concluding point was, look, every one of the enterprises I think that uses PCF or at least most of them have more developers than Pivotal does, which if you make some sort of extrapolations, you could say every one of our customers should out ship us, right? I mean, we're much smaller than our, our customers are. And so for many of these enterprises, they have enough people. They should be shipping software constantly. While we shipped absolutely tons of stuff in September, I looked at kind of how have I noticed that we've gotten better since I joined here a year and a half ago. And, you know, it was more and more about making the pipeline. It's kind of step one, which figuring out where to automate. You know, I've spent more time studying into value stream mapping and studying into lean and over these last few years. And again, theory of constraints, if I'm, if I'm optimizing the wrong places, I'm wasting my time. I need to optimize bottlenecks. I need to elevate constraints. And so looking at where you automate, but then incrementally doing it, it can't be all or nothing because that'll take you forever. So step one was legitimately investing in that. I think for Pivotal, that's, that's done a huge difference where we have our master pipeline and we have these components that fundamentally build PCF on every cloud all the time. That that's helped us a lot. And then the second one was improving some of the cross team visibility, right? You walk around our Pivotal offices and there's just concourse pipelines on every monitor up on the, up on TVs. And it's look, teams can see how everyone's build is doing. We get these giant sets of drumbeat emails in every Friday afternoon. So we see what the highlights are. I, this morning jumped into the product backlog for a number of teams to see what the next major release had. So this sort of hyper-transparency sometimes makes people uncomfortable when it's about fiefdoms and local optimizations of teams.
32:32But when you take it as, are we making our customers' lives better? Then you kind of improve your transparency and want people to see what you're working on. The third step was upgrading your tech. And I think your, your, uh, register posts touched on this sometimes a little bit as well, which is sometimes you can almost over-rotate on culture and focus so much on meatware, as you would say that you, you leave your kind of sometimes the tech stuff wanting. And if you aren't actually operating like a software shop with your tech, sometimes those cultural things are wasted. And so you've got to actually upgrade your tech stack. Right. Uh, number four, I looked at the self-service stuff. I mean, again, I've watched everything isn't automated within Pivotal, but I've, I was on a, an internal page last week where it was, Hey, if you want to join the master pipeline, here's this whole set of steps and here's your prereqs before you contact us. And here's what you can expect. And this is how your team can now build its product on this pipeline. And it wasn't completely self-service, but it was also a lot, very much more empowering than let me hunt down a person who can tell me some tribal knowledge of how to do something. So documentation, APIs, all these kinds of things that get individual teams away from blocking you. And then the last thing I pointed out was the release train. Again, when I joined here, I don't know, a year and a half ago, it was, you know, four months for a release to maybe seven months and three months. It was just a bit choppy. And the team said, look, we're going to ship quarterly and what's ready is going to ship. And if something's not ready, it ships the next time. And you have this sort of release train, I get on the train, you missed it. Fine. You'll go on the next one. We do not hold up releases. And we've all, again, probably listeners here, you, yourself, me, have been on these sort of death march projects that keep missing their date because, you know what, you didn't finish that last feature or this testing of this whole thing isn't done. And so this idea of just saying it's about getting things out the door so customers can use it, give you feedback and improve it, it's a discipline. But I think for us, that's changed a lot. A spring team does that. Rabbit team does that. PCF team does that. All of this sort of model of just keep shipping, keep learning. And, you know, I think that's helped a lot. So those five things, I try not to be too contrived.
34:34Those are real things we do. And if you do that, gosh, I just don't see how any of our customers shouldn't be out shipping us every month. Yeah, it's a good summary of the, you know, this is a joke, I think, from Martin Fowler about microservices, the how tall you must be to ride the ride. And, you know, I think, I think the last one in particular is like one of the more valuable yet unhelpfully mystic things of, of agile software development, which is, I'm not going to promise to meet any dates, but I'll promise I'll release like at the same time, like, what's the phrasing? I'm going to release every week, but I make no promises about what's in there, which, which is really like disconcerting. If you have this trust in a traditional waterfall approach, which is to say, here are the 50 things that we're going to commit to releasing in 12 months or something, whatever it may be. And so at first blush, this, this mystic thing I'm talking about sounds like, so you're not going to release my software, but, but if you couple it together with something we were talking about earlier is if you can release frequency, then you, you're reliably doing something. And so I think, I think the, this is yet another way to trick people into realizing that software is very hard and you can't make any estimates around it. And so instead, what you want to do is you want to just know that each week you'll be releasing something. And so you'll be, I don't know, hewing to reality instead of just kind of, uh, estimating and making up what's going to happen in 12 months. You just take things as they come. Right. That's the thing is there's no 12, 18, three year estimate that's ever right ever. So all we're spending months locked in rooms with flow charts and Gantt charts when really I would rather the team optimize for maneuverability and adaptability and saying, look, we're going to ship constantly. I have no idea what we're going to ship in six months. All I know is it's going to be something our customers care about. And if you say no to that, I think you're just stuck kind of in an older mindset where
36:34you were worried about big releases where you had to promise everything. Cause after the project shipped, everybody went away and that was the final thing you had. And that's not the way for software anymore. Things shouldn't be done. Things should constantly be iterated. So as people start to recognize that, yeah, you're not going to leave me once you ship software. I can trust the fact that if I keep seeing progress, you'll get to where I want eventually. Yeah. I mean, it's sort of like, I don't know about in your neck of the woods, but it seems like the highway projects are never done. And last I checked, we've been building roads since way before the Romans. And if we can't estimate that, like in our, in our little blip of an industry of software development, which is like what, I don't know, 50, 60 years old at the best, depending on when you start counting. Like, of course we don't know how to estimate. They can't even estimate a road, which, which is pretty fundamental thing in human engineering. Yep. But progress, visible progress makes a huge difference at most of the companies. I think you and I always talk to around here is that just gets everybody amped up when they see things constantly happening and they don't wait for the big bangs. Man. Speaking of, of, uh, bacon in New Zealand, lots of construction around Auckland. You got to plan around that. If you're going to drive your two and a half hours to go see the glow room caves or the Hobbiton, it's going to be a lot slower than the maps tell you. Cause you got a bunch of construction and I could never figure out like if people drive fast or slow there, right? Like, cause it seemed to be a big emphasis on it's only ever a hundred kilometers an hour and people didn't seem to drive fast, but then you get out into the, uh, the country or whatever they call it over there and you feel like you're the slowest person on the road. But then there was like tons of, uh, police officers over there stopping people. So I don't know. I don't know what's happening. Did you drive? Cause I drove over there on the wrong side. I was just trying not to kill myself. Oh yeah. Yeah. Both, both my wife and I drove and yeah, it was, it was weird. It took like a couple of days to, uh, to adapt, especially eventually the most difficult things in a, in ascending order to be all confusing is basically, uh, doing the left turn is weird and doing, doing, doing, doing, uh, doing the right turn is really weird because the right turn is like a left turn and that you go into the, I don't know, it's
38:37just very bizarre. And then, and then backing up what a disaster that is, right? Like it's just, my brain is just wired to back up the, uh, the, uh, the American style, not like back. It's very confusing, but anyhow, lots of construction there. Uh, so then, yeah, I, speaking of, uh, uh, self-flogging as it were. Uh, so I, I, in this time, I think I've had, uh, uh, two register pieces come out. Now, one of them was like, they have a, uh, speaking of continuous, uh, delivery, they've got a conference coming up in, I don't know, early next year. Um, their second one of this. And so, uh, my editor over there, he also does a feature editing. So he's, he's done a few pieces on, um, trying to drive traffic to there. There's actually of all people, I'm sure everyone who listens here follows tech news reporters and people and as closely as I do, but of all people, they got Trevor Potts to write an overview of microservices, which I'm looking forward to, uh, to reading. He has one of my favorite tech columns ever where he spins. And I mean, this only in the most beloved way where he spins like the first two paragraphs talking about how, uh, it's like a country Western song. He's like this week, my cat died and I had to move. And then I got all these press releases I have to deal with, but it was a good, uh, good piece of gonzo journalism. Anyways. Uh, so I, I wrote one that was, um, essentially, I don't know, my, my final writing down of how we came to this, like DevOps cloud native native world, kind of starting in the, uh, starting with extreme programming, I guess. I mean, there's obviously an obligatory mention of the mythical man month, because, uh, you know, if, if you're waiting for this article to finish, it's to ensure its quality. Uh, so you got to mention that. And then, and then just sort of the idea of how, uh, how essentially this thread from, uh, from extreme programming and a little bit of the clue train manifesto, and then the, uh, the, uh, the old yield democratization, uh, that happens in businesses of the internet, but more importantly, that era of the internet where, uh, Twitter was always going down and people had to figure out how to keep things up and running and yet release features and
40:39how that, how that got sucked, so to speak, into the enterprise, uh, because public cloud and other cloud things were available. And, and now how, how we have, uh, people focusing on let's, let's release the software more frequently and, and let's improve the features and make it so that we have programmable businesses. And, uh, you know, the, the headline they gave it was a little weird because I think I was arguing that like people do matter, but I don't know. Right. I should go back and reread it a third time because to be honest, I'm never really sure what I wrote when I said it. Well, I know when I finished reading it, like that felt different than what the headline said, but it was, it was still good. Yeah. I'm pretty good at like speaking out of like, uh, the multiple dimensions of my mouth and saying the same conflicting things, but yeah, yeah, I, I think it's an all right piece. So then there was another one that I think I came out with since last we talked that was, um, I finally took the effort to write down, uh, what, what I've been finding enterprise architects do, uh, in, in a DevOps world. I mean, it's, it's, I didn't really say cloud native cause that's, uh, you know, DevOps is a little more broader for the audience there, but it, it applies the same. I mean, you know, I've since given this talk a few times now, I think, including in, uh, in Latvia and, uh, yeah, I think, I think the general, like, I guess there's probably about, about three points, right? One that, that architects still need to drive strategy. And two, there's this platform we were talking about earlier, and someone's got to be like the product manager of that and even come up with the idea to, uh, well, you know, basically be the product manager of it. And, and then third, uh, from talking with some of our customers and other people, like, um, I don't know, EA should be doing this, but I, I think, I think one of the things that they need to do more of is, is basically, uh, finding and testing out new technologies that they introduce into the, uh, the organization. And I think in some of our, not all of them, but in some of our more impressive customers, you see that the enterprise architects end up doing that a lot in addition to like governance
42:42stuff that they do. But so, so there, there's, uh, there's that piece. That's one of my regular columns. And, uh, I don't know, like I said, I, I often forget what I wrote once I send it in, but they're, they're, they're good summaries of, uh, those areas that I've been looking into recently. Yeah. No, I like that one. Again, we're doing a lot of research from when we first started talking about this at the beginning of the summer. So it was cool to see some of that come together. And as usual, the commenters calling you a, an insane hippie. That's right. Insane hippie. Is that, is that one of them? I don't know about this time. Yeah. Yeah. Yeah. I, I, I've, I've fully adopted the, uh, you know, I'm, I'm basically, uh, a cross between Martin Friedman and, uh, you know, that, that Steve Bong guy in the onion loaded up on Valium. I think, I think that's a good, that sounds relaxing. It's great. So we'll see it. We'll see if my most recent column makes it. Cause I think I characterize myself as, uh, you know, freedmanning amongst conference rooms, just like a taxi in India. I love that. Uh, so, uh, what was I going to mention? Uh, well, I've forgotten, but whatever. So, yeah, I think, I think that's, that's essentially what, oh, I know. I know. Speaking of, speaking of working on this all summer, I think not a day goes by where, where I, I, I don't remember something you said last year when I finally released my 49 page PDF was like, what was it? It was something like, yeah, you've been working on that a long time. So I'm trying to see if I can reduce my release cadence down, uh, but we'll see. I gotta, I gotta write something there that that would be, uh, or publish something that would be handy. We'll see if we can round it up to a whole number, not just 49. I'll put it in, I'll, I'll go down to the, uh, the lower right-hand side of some chart and just drag it out some more. So the page goes down. All right. Well, uh, you got anything else you want to go over before we wrap up? No, I don't think so. Again, always keep your eye out for, uh, releases coming. Coté and I will be back on a more regular schedule with some cool guests coming up here shortly. And, uh, all the March towards SpringOne.
44:44So we'll have some preview shows and things coming up to that as well. That's right. That's right. Well, next week I'll be at, uh, DevOps days, Nashville. If, if you're in that area, you stop by, say hello. That's always nice. And, uh, I think one of the only, I, there's probably another DevOps days I'm going to, but there's also, uh, the FedScoop digital summit, something that I'm going to be on the panel on in, uh, DC the week after next, uh, if you're in that area. And, uh, I'm also going to, I'm, um, I think I'm giving the enterprise architecture talk at a meetup the night before I'll have to put this all in the show notes, but, uh, if you're in that DC area, just come on by there. It's about midweek week after next and, uh, check that out. So with that, with that, this has been another Pivotal Conversations. If you want to get all the latest episodes, which hopefully will be released more than 30 days, uh, between them, but, but maybe that's a reliable release train and you should just take what you can get, but we'll pretty much put them out weekly. Uh, you can go to soundcloud.com slash Pivotal Conversations. Now, pretty much every Thursday, we put up some formal show notes over there at pivotal.io slash podcast. You can look in there for that discount code I mentioned and links to other things in that conference whose dates I should remember, but I don't, I got it written down here in the computer, so it'll be fine. And, uh, as always, it's nice if you, uh, share with people that, that you like this podcast, you know, through all the usual stuff, it's always, it's always nice to be at a conference and, uh, hear that people listen to this and more importantly, uh, other than making, uh, us feel good about ourselves. Uh, it's good. It's a good way to just kind of like start a conversation about, uh, fun computer stuff. So with that, we'll see everyone next time. Bye-bye.