2018 analysts predictions round-up, plus, our own, with Rita Manachi (Ep. 89)

Episode 74 · 52:53 · SRT · archive.org · all Tanzu Talk

Well, we have a guest on. This is probably the last podcast we'll record this year. I actually might have one spring-loaded to post some other week. We'll see what the editorial people around here want to do, but I guess it'll be in your feed because I control the RSS. But anyhow, this is the last live-to-tape recording we'll be doing, and appropriately, we should do predictions. Everyone needs to get ready for 2018. January's coming up. You're going to come back to work. On the 4th or the 5th, I don't know, whatever you like. And you've got to start cracking. You've got to start using up that budget you just finalized and planned out. So maybe this is a little late then, if you have to plan things. But never mind that. Never mind that. So why don't you introduce yourself, guest? Hi, I'm Rita Minacci, and I head up Analyst Relations at Pivotal. And Richard had the good idea that we should have you on because, as you are an analyst relations person, you know what all the analysts are predicting, right? At least that are relevant to our interests. And you can help us understand and run through what analysts are doing prediction-wise. And then, you know, we'll give some of our own predictions at the end here, if we fancy that. Because we are a regular, I don't know, that guy in a box at the toy store. Zoltar, is that his name? You're making some wishes. That's good. Yeah. Have you noticed that? Do you all have this over there on the West Coast? Do you have Zoltars everywhere? We have those, like, where, as they say, lousy with them down here in Austin. They're everywhere. No, I don't see Zoltars around here. Yeah, I don't either. Really, I would think in San Francisco, they'd just be on every corner with, you know, get a free IPA with your Zoltar. But I guess not. Well, so I thought, you know, we gathered together a bunch of them. And I think for the point of what we care about here, we've clustered them into three buckets.

We've got development, as in software development, infrastructure, and then everyone's favorite category, miscellaneous. Because not everything fits into there. Potpourri. Why don't you give us an overview, first of all, Rita? What's the deal with, like, analyst prediction PDFs? Like, you know, you've consumed this stuff a lot. I've written some. I'm sure Richard has written some prediction stuff. But, like, when you're looking at, like, a, not to be like my other podcast, where we exegesize this thing too much. But, like, when you're consuming an analyst prediction PDF, like, what's going through your mind? Like, how do you think about it and consider it? Where does this fit into the grand scheme of figuring out to do with one's work life? There's so many things. You know, I think it depends on where you're coming from. If you're in the marketing side, I think you think about how you will be messaging some of the developments we have going on next year. I mean, when I read the Forrester one on cloud computing, I was like, check, check. Like, we are addressing this, addressing this, addressing this, right? So I think about how these predictions will impact our business and our product strategy and our marketing strategy. So that's kind of how I think about it when I read these. And, of course, there's the, like, I call BS on this one. This is good. Oh, my God. You know, there's the other voice, the snarky voice that is judging these predictions. But from a business perspective, it's really about how it will shape our strategy for next year. You know, both product, or actually not both, but product, marketing, business. And do you think that these are like, I mean, like, how seriously should these things be taken? Because a lot of them are like, you know, this is going to happen by 2020. And I was even, I forget where it was. It was in the big Gartner's curated selection of all their favorite predictions. And it was, basically, there was a line of something like, we feel confident that we can predict the nature of work in 2025. And I was thinking, like, whoa, that's a bold statement to go on.

Yes, I agree. So, yes, the Gartner ones tend to look further out. So, I think they do tend to be a little more bold. And I think they're a little more dramatic. And the Forrester ones that I've read seem to be more based on what they're hearing today. And really looking at the next year. Like, it's really for 2018. You know? Whereas I think Gartner looks at, as you said, three years out, five years out. Like, a lot of the reports that I've read from Gartner are, in 2020, for example, you know, 40% of new applications will use, you know, artificial intelligence for co-development. You know? Do you think that this impacts their research agenda, Rita, for a given year? Or does the research agenda impact their predictions? Like, what do you see the relationship there? I'm assuming if a Gartner analyst says this is going to happen, they're probably going to be writing about it a lot over the next year, right? Correct. Correct. I think about that as well. How does it impact their editorial perspective? And what they're going to be looking for, right? Like, they're going to be looking for stories and customer examples that support this. And I think, I mean, that's separate from what they get in inquiry as well, right? And let's just face it, research drives inquiry. So, they may be, it's kind of a circle, right? You know, let's make these, and this is, there's no judgment on this, but let's make these predictions. And now everybody's going to be curious, so they're going to start asking these questions. Right. Yeah, sometimes you're actually predicting that you're actually causing the future to some extent. Like, if you just say X will happen, and that's all you keep talking about, at some point, your customers, if you're one of these analyst firm customers, you think you're going to start going that way. Like, it's an interesting way that you can actually make that prediction come true yourself. I like that. I feel like every day, all of us are causing the future. That's a positive outlook for the end of the year. It's how I wake up. We are the butterfly, right?

The flap of the wings that caused that. No, that, that, yeah, I think, and I think a lot of the Gartner research, particularly from the IT leader side, is way more forward looking. Um, so, because I think if you, you know, read the research and then ask them about what kinds of inquiries they're getting from end users, they're often very, they're often very different. The research tends to be more forward looking, inquiries are what's happening now. I mean, that's not to say that there isn't value in their, in their research, but, um, it's all related. Yeah, and, and, and, uh, I mean, just to wrap up the, the meta analysis, I think, uh, one, I, oddly enough, uh, I hadn't thought of the idea of using, uh, analyst predictions to sort of like get inside their head about what they think. And, uh, they're, uh, they're, they're, they're ideas about things. So that's, that's fun. And then two, uh, um, you know, in, in, in reading through these and thinking about it historically, uh, this is, you know, uh, my pet peeve about life. Like I, I noticed kind of like two broad approaches. One of them was the, um, the, uh, here's things that will happen. Let's call it the qualitative predictions. Uh, and, and, and, uh, those are fun, but I think in particular, like the Forrester ones pull in a lot more, like, survey data and things like that to sort of, uh, sort of point, point their direction to things. Now, of course, uh, I am no stranger to the fact that there's a lot of data out there and you can just selectively pick the ones that match what you want, but it's at least nice to have some, uh, percentage, uh, symbols involved in, uh, you know, I don't know, predictions that people make. So you can sort of choose your, your approach to things if you want it based on, uh, spreadsheets people have, or, or just, uh, stuff off as they used to say their dome piece. But all of them are equally fun. So, so to that end, you alluded, we'll go into the development, Rita, but you brought up, uh, you brought up, I think the first one that I read, uh, was, was the one that had a lot of predictions. It was basically, if I remember, it was like the role of AI and software development.

So what, what, and, and, and I have to admit, maybe, maybe I'm just behind the times or like the, the, uh, the pony express wagon hasn't gotten here to Austin yet. But I had never really thought about AI and software development that, that, uh, that's, that seems, that's some good stuff right there. So, so what are, what are the analysts talking about there? Actually, both Forrester and Gartner believe that app dev teams are going to be using tools that have AI to help do things like, uh, just kind of, um, repetitive tasks, QA, things like that. Which makes sense. I, I just don't think we're there yet. Um, frankly, maybe because I just value application developers and engineers too much. And, uh, I feel like their work is so critical that, you know, replacing them with a boss, this is unheard of. Now, I can, I'm, I just contradicted myself because one of the comments that Gartner says, and even Forrester actually, is that it's going to be used for the low level things, which actually feeds into our story, which is let developers focus on the. The, the, um, value, uh, driving, uh, activities. So there could be some truth to that one. I don't know how far along the app dev tools and platforms market is and incorporating AI and ML and things like that into the tools to help with that. But, you know, I, the more I think about it, the more I feel like it could be legit. But when I first read it, I rolled my eyes because I feel like AI is coming, you know, it's everywhere. It's coming everywhere. Everyone's so excited about it. And I just, I'm a little dubious about, uh, how, uh, how much it's going to be incorporated into every, uh, all over the stack. Right. Right. At least two years from now. I mean, that's, what's interesting is I agree that, I mean, there can be some uses. There was one piece of this prediction that had mentioned that AI would be used to do some interesting things like actually be like a scrum master, like a helper, like a voice driven sort of AI to actually act as sort of a coach. Like a virtual AI coach.

Like, I, I just don't see having a chat bot that, you know, bot that asks, Hey, is this a good diagram or not? Like I, I mean, maybe. But, but at the same time, you know, they had some other ones that look like iPads. Like integration can use AI to make integration easier and try to map like, Hey, I think this is supposed to connect to this. Like that makes total sense to me. Right. I can, I can see using data sets to help me make smarter decisions on some of those tasks, like connecting to data sources, maybe even doing some code test coverage or, you know, some things to help make my application development better. But I don't see it replacing architects or business analysts or things like that, at least for the foreseeable future. Yeah, that's, that's some good framing for all this stuff, which is, uh, uh, oftentimes things get used in a more pragmatic unintentional or unintentional way. And so, I mean, just to make something up, like I was, uh, I was, I was, uh, reading up on the concepts of tech debt recently, cause someone else asked me to write an interviewer thing on that topic. And, uh, you know, it reminded me of all the like code scanning and, and sort of like, would you call that static analysis of, uh, of code that you can do that gives you insights. And, you know, maybe in like 1948, that would be considered AI or something, but like nowadays it's just like code scanning. And so it does seem like you could use some sort of machine learning thing to find like good and bad patterns in code, especially execution wise. And I don't know, would you call that AI beats me, right? I mean, it's obviously it's machine learning since I just said that, and I am the source of truth on these things. Uh, but, but it, it, it is plausible that people who do analysis would start to, uh, go back to all those, all those little funny static analysis things they do. And, and then, you know, maybe even in real time, start to, uh, as, as, as you reminded us, Rita, start to be a clippy of, of just like, you know, are you sure you want to do that? Why would you implement string tokenization?

That already exists. I mean, I think the, the key would probably be getting rid of all these hard coded rules, right? I mean, that's what in 1946 or 1997, it was based on like, Hey, here's some quick rules to tell you your code looks good or not. And they're fixed, right? You're not really changing it versus something that's learning and saying, Hey, we've, we've discovered these things over time. And like, that does get super interesting. I just, you know, Rita, you, you touched on this. I just don't see this making its way into many of these tools by 2020. I mean, it's two years from now. Is that going to be in my visual studio? Is that going to be in my, my dev suite, my CICD tool? I mean, maybe, but that seems like that's a massive leap in a short time period. And one thing I did think about too, is it, is this sort of the evolution of automation as well? Maybe that's more relevant on the infrastructure side, but using AI to automate more, again, low level things, you know, uh, it's something that I thought that popped into my head when I was reading these. Yeah. Yeah. Did anyone cover that idea of like use it, whatever it's AI or ML or whatever it, the automation layer. I don't remember reading that either. I don't think I saw that. I don't think so either. Oh my God, there's my prediction. There you go. Walk it in. Yeah. No, you know, it's a part of the, it would have come up in the IT operations predicts piece. Um, but they really led with the skills gap, but I'm getting ahead of myself because we're going to talk about infrastructure later. Jump to the next one on serverless. Yeah. And then, and then, so also in the, uh, I forget which it was the, the compute one that Gartner did. They, uh, let's see, I'll, I'll, I'll, it's basically says by 2020. So, so as, as Richard keeps reminding us in just two years. Uh, 90% of serverless development will occur outside. Basically will be what we used to call shadow IT. It'll, it'll, it'll be outside of the, uh, they call it I and O organization. I forget what that stands for, but the IT organization. Uh, and, and so I guess, I guess like it won't be managed by them or something. I should remember the details of what I was reading here, but, but I mean that for any new technology and you see this over and over again, it does seem like it, uh, takes it several years to be standardized.

And thus in the purview of, of the, uh, the people who, who run something and standardize it. Right. And then remember, I mean, to this point, there is no option really for serverless on premises. So if I am using public cloud, I am either swiping the credit card or using my account. So it's not shocking that a lot of that work would happen outside of the ticketing based provisioning system that happens on on-prem systems. Cause that's not a choice. Now I get pivotal function service, things like that will make that more interesting, but I get this one. This one actually doesn't surprise me too much. And it seems to be lumped with serverless microservices and event driven. Uh, those, those three things are mentioned often in the same, you know, if not same sentence, same paragraph. And to Richard's point, I think we're in a good place. Yeah. Now building on this, I think we put this in the infrastructure one. Maybe it's related to this though, but this idea of function paths and things like that, this assumption that serverless in general is going to double triple again, that doesn't seem shocking given it's starting off just now. So it should be doubling and tripling every couple of years in terms of usage, in terms of popularity. I think that, that also doesn't feel too shocking to me. That's good point. That's a good point. If it's from, yeah, if it, you know, going from a small number of doubling is not that big. So yeah, I think it was Gartner that says that by 2020, the number of applications and serverless function paths will more than double from 2017. Right. If you're saying IaaS is going to double over the next two years, that would be interesting to me. Like that would seem like crazy given it's a bigger number at this point. Totally. And I guess, I guess to transition into the infrastructure there, there's a, as part of this, there's a, I think it's even from the same prediction thing. I just pulled that out. Cause you know, serverless must be a developer concern. Uh, but there's also a, uh, commentary about like, uh, how much, or a prediction about how much, uh, uh, what do they call it? Mission critical containerized cloud native applications. They still use a hyphen in cloud native. So we should send them a memo, Rita, that the hyphen doesn't exist anymore. But, uh, anyways, some people still do.

I know. I know. I, I remember a couple of years, I remember a couple of years ago, spending a lot of mental brain points on, on a hyphen in there or not. And so now I feel like I can make fun of it anyways, uh, there, there's kind of a side note. So their prediction is that in two years by 2020, more than 50% of enterprises will run mission critical containerized cloud native applications and production. Now, as I'm always fond of reminding people like penetration is not the same as market share. So what that means at the minimum is that there will be at least one application that an enterprise is running. Not necessarily that all of them, but whatever, it's still good momentum. And I think, I think that to me, like the other thing that's interesting about that is I left off the, uh, there's a comma and then it says up from less than 5% today, which to read into that means that five, five, five, less than 5% of enterprises are running mission critical containerized cloud native applications in production, which, you know, it's probably no footnote that defines the size of an enterprise, right? Like IDC used to define an enterprise as more than 999 people, uh, whereas Gartner might be more than 10,000 or whatever. So, so who knows? Uh, but, um, that is, it kind of matches up with, with, uh, the serverless thing. See, uh, going back to the thread there is that they probably don't tax a mind taxonomy and eyes at this way, but I'm pretty sure like all serverless thing runs in a container, right? So to some extent, like, uh, 5%, you've already got less than 5% of things running in a container. And then if serverless is the type of container thing, there's probably even less, it's sort of like the shrinking egg thing, but, uh, that is some crazy growth. I can't do that kind of math because as we discussed, I have a philosophy degree, but that seems like, uh, that's a big CAGR as it were, uh, over the next two years. It is. My question would be of that 50%, hopefully, I mean, in my ideal world, 80% don't realize they're doing it right.

I mean, every PCF customer today is running containerized production workloads in a cloud native platform. You might not know because who cares, right? I mean, that's the point of some of this is that, you know, when I'm using things like Kubernetes and we're going to talk about that in a moment, but you know, that's not the end game. The end game is the things that are making it easier to ship applications to those sorts of platforms. And those are good platforms for platforms. So hopefully, sure. A lot of these apps are containerized and hopefully nobody cares because that's just the way you, you deliver the software. And that's an implementation detail. And this, this reminds me or doesn't remind me, but evokes something, uh, be interesting to hear your take on what the analysts think about this, Rita is, is in the, uh, I don't know what our cutesy name for this anymore is. But in, in this sort of like Twitter world of like infrastructure-y cloud native thought lords and ladies, there's this ongoing trend now of like, hey, you should pay attention, uh, to the, uh, the stuff above Kubernetes and the past layer and the actual applications that you have. And like, uh, uh, I think in the past, let's say 60 days or so, like that's been something that, uh, everyone comes up with a 280 character or less kind of quiff about and posts. And then it gets retweeted and it's good, good thought, thought, uh, magistrating. Uh, but are you, are you seeing analysts make that point or are they, uh, are they still obsessed with, uh, just container orchestration and things like that? Uh, I, it's, it's, it's, it's a case by case basis. Uh, they, I haven't had anybody say it that bluntly, but I know that, uh, for example, I think that's actually probably more the IO foot, like the, I feel like Forrester has been more aggressive in, in, in that than, uh, others. Uh, and then actually something that, uh, Red Monk wrote recently kind of, kind of, kind of to me echoed the whole value of line, um, piece. I think it was Fenton Ryan's article, uh, most recent article, which I can't pull up right now, but, uh, but that kind of, that, that addresses that.

Um, I want to go back to this whole, like serverless microservices event driven thing. And it's something that Richard says, and I haven't seen anybody put this out in writing and, uh, I haven't gotten like strong agreement verbally from the analysts, but that it's not a zero sum game, right? It's not that, you know, containers eat paths and, uh, serverless eats containers or whatever. Um, you know, Gartner is kind of rotating towards something called the hybrid application platform strategy. So like different, different platforms for different levels of application. So we'll be interested to see how they evolve that thinking and that research. But I think that might be going in that direction. Yep. Great point. I mean, hopefully this is all additive stuff and that's, that's, what's interesting, right? Is that we're, we're trying to build more things in the right places and we have more choices than ever, which is awesome. Yeah. And, you know, actually, back to your point about this, this, this, this kind of reads, it feeds into the infrastructure and operate in IT operations, um, predictions. And it's what, how IT and operations folks need to, which is by the way, IO is infrastructure operations, uh, how they need to up themselves, right. For this, for these new ways of, of doing things. But again, we'll address that later. Yeah, that was, that was a extensive thing about, uh, training that, that, that, that did come up a lot that, you know, there, there was also like skipping around a little bit. There was, there was a, as always a good emphasis on like changing the way you think about IT and strategy. In fact, I think, I think I was, I was kind of subtly jabbing at the idea of, uh, needing to go over that earlier this morning in that, like, it's, you know, you always have to reinvent how you use IT. Just, just, just as a reminder to people out there, you got to evolve, evolve the way that your, your tools pan out. And speaking of, of reinventing. So there was, uh, in one of the, the Forrester pieces, I think, I think, I think like most people, but in a very, well, concise, definitive way, they were basically like in 2018 Kubernetes wins. So you don't have to worry about those other things.

And, and, and I think, I think they drew out, I'll try to summarize it like four reasons why, uh, why this is the case. One, because it's got 15 years of product, of, you know, experience, so to speak, since it was based on Google things. And it's got all the necessary features today. They say number two, and, uh, has the largest and fastest growing community. And then, uh, and then also that it's a part of, uh, lots of pass offerings, including, uh, cloud foundry. So it's sort of, uh, overall it, uh, I think, I think what they're saying is that, uh, they believe it works and, uh, it's cemented into existing things. And, uh, so there you go. It's available everywhere. Yeah, sure. Great. What's next? I think that's the thing, right? That's cool. Like that, that's great that, you know, that there's a lot of people contributing to something that will power a lot of things. I hope it to some extent, again, for the end user and again, pivotal container service. I'm excited about that's going to be great for folks who want to use Kubernetes, but in a lot of cases, that's still a good fit for certain types of workloads. And other people will still keep using serverless and they'll keep using IaaS and they'll use bare metal and they'll use cloud foundry, you know, pivotal application service. And that's all great stuff. So that's awesome that we have a kind of common scheduler that people have rallied around. Now let's make sure that it's great to use all this tech in a secure fashion. I can deploy continuously and people like doing IT again. Yeah, see, that's, that's, that's a good summary of the aforementioned shift in the thought, the thought lording wins right there. And actually the, the Finch and Ryan piece I was talking about, it's called commodity container. It was the commodity container story that post feeds into what we're talking about here. And then also about this whole value line, right? Like where it becomes like, who cares about what's happening underneath? Who cares about the container stuff and container orchestration stuff? It's the value of it. So he actually, I think Artic articulates that very well in that post. I'll put a link there, but yeah, yeah, it is, it is a fun post overview. So then, so then back to the infrastructure stuff, what, what, you know, private clouds, always a perennial thing to figure out where people stand on and what's going on.

So how, how would you summarize what, what the analysts are saying, predicting about private cloud, Rita? Well, this is hilarious because they're so, these are very different. You know, Forrester is quite bullish. They think there's going to be kind of a renaissance for private cloud and, and hybrid cloud as they, as they, as they call it. And they actually credit Azure Stack for being one of the drivers of that, of the, particularly the hybrid cloud. So they think that it's, it's, it's, it's going to be, I think the actual prediction is private cloud gets new life as an app creation and modernization platform. And so one of the points they make is that it's a great place for developers to kind of get their chops, right? To learn how to move more quickly and to do things, do things like perhaps, perhaps serverless and things like that. And then enterprises can kind of move into the cloud on their, at their own pace, which also echoes something that James Governor said, which is, you know, premature optimization, right? So when they're ready, they can move to the public cloud. So I'm getting private cloud is a good place for people to practice rapidly, rapidly building new applications and modernize implementations and processes and handle the scale of new application development. Yeah, well, as you say, so fascinating, given there's such a difference between even some of the Gardner and Forrester opinion. And I think it's, I don't know, as we look at the landscape, Coté, you talk to plenty of shops who deal with this stuff too. I mean, for me, it's private cloud historically has been awful. And that's because it's mainly been a ticketing system on top of virtual machine provisioning. Like, that's it. That's what a private cloud has been. And if I can actually deliver application services, data services, you know, I can get that agility that we all loved about public cloud in the first place. I mean, I'm going on 10 years of being a public cloud crazy fan. But if private cloud is actually interesting now, and I can actually get the agility I want, it's on demand and date managed services, the only true benefit of public cloud is going in locations I can't get into myself.

You know, I want a server in Germany in four minutes. That's awesome. Or I'm doing crazy scale. You know, I need 10,000 containers in the next 10 minutes. Great. Trying to use some of these really interesting novel services like around AI and machine learning. But a lot of the other agility stuff is now available on-prem in some of these private solutions. So if I have sovereignty concerns or other just workloads I don't want to move to public, I don't feel this sort of complete fear of missing out anymore if I do private right. So I kind of agree with Forrester, which I would have never said a year or two ago. But I get that kind of renewed interest on actually making cloud just a destination versus, like, I have to move to public or else I'm a complete, you know, dinosaur and I'll go out of business. You know, and I think there's a nuance, right? And Gartner's beef with private cloud is that they think that it doesn't offer a true cloud experience. And again, to your point, that may have been true a couple of years ago, but it's come a long way. And you can get a more cloud experience than before. And I think also Gartner points out that people are really using private cloud quotes, if you will, or cloud quotes for DevOps and things like that, for those kinds of experiences as opposed to a true cloud experience. Right. What do you think, Coach, are you a private cloud renewed interest guy or you can't wait for it to die a fiery death? Well, I think the way you phrased it is the, I mean, intellectually, I've always been hoping for a fiery death. But, you know, maybe it's more of a phoenix. And it's just that, as you said, like most private cloud things have been, to rephrase the better way you put it, they're just not private cloud, right? They're just like regular virtualized clusters. And I think that's always been the intellectual problem with private cloud is that the point of cloud is that it's not private, right? Like, I mean, there was a big debate many years ago about, like, why do we even say public or private? Like, this is crazy.

And I think all the OpenStack people and also VMware's involvement in people, you know, like they and Microsoft, like they sort of defined what these things would be. And then, but then, you know, just, I mean, two more things. Like, one, like, it is notable that most of the reasons for doing private cloud are rarely about the efficiency and productivity of public cloud. They're like something that public cloud doesn't do or doesn't have as far as regulations and whatever that you need to do for private cloud or because you have all this stuff or whatever. And so it's very rare that you come across a private cloud pitch that's like, because it's awesome, right? It's true. Because it, like, has all this functionality that's fantastic and things like that. And I think what, I don't know, does Gartner still use confidence things? I haven't seen that in a long time, but I don't know what my zero point whatever number of confidence would be. But I think, I think there is an interesting observation to make that have, has, has the overall vendor world figured out private cloud that is awesome and, you know, hopefully is, is similarly priced. Like, I'm not sure it'll ever be like the same price, but like, does something like an Azure stack actually get you all of the sort of public cloud experience? Or is it just another case of like, sort of a fallback, if you will? And I think, I mean, that's all sort of like, to your point, like many of the people that I talk with when I'm traipsing about, like, they all want the awesomeness of public cloud, but they have very valid reasons for not doing it. And, and, and then many of them, like, you know, Home Depot is a notable public example. They, they really do strategically want to move to public cloud as much as possible. They're just, you know, I don't know, building their passageway to that. I got lost there. But yeah, so, so I think, I think it's, it's not quite the craziness of, of even a couple of years ago to think that like private cloud might be something it's, and it's, it's to go on a little bit.

I mean, I remember, I remember an epic meeting back when I was at Dell that I had with a room full of Forrester people at their, their fantastic new offices with guitars in the lobby. And I got a, live at Altmont, Allman Brothers CD is a parting gift, which was cute and nice. I don't think I've ever opened that, but I should probably listen to it. But, you know, I remember there was a very feisty debate amongst the analysts about if private cloud was BS or not, and that everything was going to go to public cloud. And this must have been in 2011. And so it is like, it is, it is interesting to watch over time what sentiment is amongst analysts and others about, about that. And it seems like people are kind of open to private cloud. Like there's not really a big debate about it anymore, which doesn't mean private cloud's valid, but it means that it's not that controversial of a thing. It seems a lot more real than back in those, those theoretic days of getting an Allman Brothers CD. There's my epic commentary on private cloud. Just mic drop moment. That's right. You know, actually, it's funny you say that because I need to get a new little arm that holds up my microphone because it's broken and it does like slowly lower. So as I talk more and more, my, my mic literally is dropping. That's why every now and then you can hear me pulling it back up. But so, so let's see. Well, we've, we've alluded to it many times, but in all of this, right? Like whether it's the actually getting a real private cloud or figuring out serverless, like there's this ongoing emphasis in all of these pieces about you need to fix your skills and the way you're thinking about things and organizing. And give us, give us, give us an overview of, of what the DevOps kids would call sort of like the culture commentary that's going on in all of these, Rita. I think it's, it's, it's still, it's still an issue, which surprises me.

But that's because, you know, of where I work, I think. But it's still an issue and it still comes up. And, you know, the whole thing is like the technology may be there, but if people don't use the tools, they have to change the way they use the tools. And also, and in the case of operations folks, they have to evolve themselves and evolve their roles to be more versatile and in some cases be more like developers. Forrester calls out this concept of immersive training, I think is what they called it or something like that. You know, so pivotal labs, an IBM garage kind of style things as taking off in 2018. So we like that one a lot, but that was one they called out specifically in the top 10 predictions for 2018. And then the IT operations 2018 predicts from Gartner, again, talks about a skills gap for the I&O folks. Having more versatile, which to me echoes some things that Forrester has been writing earlier in the year, which is that, again, operations folks need to evolve themselves, perhaps even be more like developers. So it's still an issue. And I think both both firms understand that and see that. Well, I'm glad that's what immersive is, because I was thinking it was going to some sort of augmented reality thing, which is good to hear they didn't go down that path. Well, I lost my thought. Oh, yeah. Even I was going to say, we even get, you know, when we talk to the analysts, they tell us they get they get a lot of questions from user clients, like, how do we deal with this? So it's still a thing. It's going to be a thing this year. Yeah, that is I mean, I'm I'm I'm a little bit like you in the sense of it is an obvious problem that we've always had in the IT world. And so so so it's maybe maybe it's maybe it's an instructive like thought hole to go down to be like, why? Why is it the case that this is always a problem? Like to go on another little jag, I finally finished reading this book called Programmed Inequality about the the the the way the way that that gender was used.

Kind of destructively in the history of British the British IT world, specifically in the government, how the continual sort of like, I don't know, not thinking as of women as full people caused all sorts of problems. And a lot of it was basically like, as we always have in the IT world, like staffing is hard, like it's hard to find people. And so it is it is a little weird how we haven't solved that problem yet. Maybe maybe it's the typical like human thing of like, you know, if if if you give me a 25 cents today, I'll happily pay it back on Tuesday. Like we're just happy to build up debt and not really address the problems we have so that we can pay for the hamburgers on our own. And I can't remember which one, which which report was that they talked about looking to address the skills gap is looking for people with diverse backgrounds and diverse experiences. So and I think it's something we also do and say at Pivotal, you know, you know, especially in the case of app dev or engineering, like, can they demonstrate critical thinking? You know, they may not have a degree in computer science or software engineering or whatever. But if they can demonstrate critical thinking, you can teach them some of these skills. Yeah. Oh, hey, we have philosophy majors and poli sci majors on this podcast. So why not? That's right. So that's right. So and then and then I think I think finally in that area. Well, sorry, you're going to go on. What I was going to say, maybe you're maybe you're going to say this, but one of my favorites from the Forrester one is the integrated security and the cloud forms, which they bring up. And that's that's the only place that I saw that. And what do they mean by that? I didn't read that one very closely or perhaps at all. Let me go here. I mean, I mean, cloud security becomes integrated with an integral to cloud platforms. So rather than bolting on security features, have it built into the platform.

Right. So which I think, you know, making the right making the right thing easy sort of sentiment. And, you know, and they go into more details of like zero trust and all that stuff like that. That's kind of over my head a little bit. But I like that. I like that. I like where they're going with it in that it has to be part of the program or part of the platform as opposed to something you bolt on because that hasn't worked. Yeah, you know, when I when I was kind of glancing at that, it was reminding me of of another like idle thought I've been messing chewing on recently. And like I think back at Springwind Platform, I was in our CIO summit track for a little while. And, you know, Justin Smith, our security person, came in and was just giving his talk. And I've been involved in a couple of conversations here and there about like security and patching. And it's sort of what you're alluding to is it's it's one of these problems, to put it in my snarky way of like, well, the way you fix this problem is to stop doing dumb stuff. Right. Like there's there's no like there's not a magic DVD I can ship you that you're going to install and it's going to fix all your security things like you need to when you're doing your your sort of new things or you're refactoring to use that term sort of precisely your old things. You should probably make them secure. And here here's here's a bunch of stuff to do it. Right. And, you know, if integrated security is is a good way of putting it, then that's fantastic. But it seems like everyone everyone's always looking for like a quick fix with security. But it's just like that doesn't really happen very well. They just need to introduce healthier processes that everyone follows so that it's it's, you know, just a regular old feature of it, which I think I think maybe even more. More so than the the the drive to do digital transformation. The fact that like CEOs get fired over security probably motivates a lot of people to think about it more.

Oh, boy. Yeah. And, you know, I think so. So basically what what Forrester is predicting is that more past platform providers and IaaS are going to offer integrated and stream streamlined alternatives to, you know, point solutions for security. And then they go on to recommend that their clients ask vendors, you know, how they're going to guarantee that there's their security solution will work across IaaS, PaaS and SaaS platforms. So to me, that's, again, a good one for us, for Pivotal, given our our security strategy and our secure by default. Awesome. Right. Mm hmm. I mean, I don't think we're saying I don't think we're saying don't worry about it as long as you have us. Right. And Richard, you can maybe type up on this one, but it's a nice kind of, you know, base foundation for some security things. Yeah, I just I just I think that base is getting more and more important because it's not like people don't know best practices. I think I saw Fintan actually do a survey on Twitter. Ryan from from Redmonk yesterday on kind of how honestly how important is security in your enterprise? And it was still like less than half. We're saying, yep, it's an actual top priority. So you can know all the best practices in the world. It's just not happening. So I either need to have platforms to do it for me or assume I'm just going to be insecure because I just don't see a third avenue at the moment because we people know what to do. They're just not doing it. So, you know, at that point, then I need systems to be smarter and protect me for myself. So how about we there's there's a lot more. Well, maybe like four or five. That's that's a lot compared to zero. But there's there's there's there's a lot more that we could go over here that I'll will put references to in the show notes. And also, I wasn't always good at ferreting out the URLs, but most of these are good. I mean, one one thing on the metal layer about prediction stuff is they are also a great marketing thing for analysts.

So they tend to be more public about them. And like there's a lot of blog post summaries of these if you search around for them. But so we'll put some summaries there. But Rita, pick out one more for us to go over before we get to attempting to put ourselves on our own petards and try to make predictions on our own. All right. Actually, we didn't get to this one. And I had it written down because it's something I'd heard about for a while. It's that SaaS providers are going to extend like are going to include PaaS capabilities in their offerings or something along those lines. It's the whole, you know, PaaS everywhere kind of thing. Oh, yeah. I remember that one. Yeah. And I think both firms mentioned that. Right. So anyway, I think that's an interesting one. Yeah. We've seen that Workday Salesforce. I mean, there's another number of these companies that try to offer a platform to run their stuff next to it that's more custom. So I can imagine that train continuing. Yeah. No, go ahead. As I said, the other one is, and it's related, is the whole low and no code platform. Again, both firms are bullish on those platforms. Yeah. Just give a brief summary of the whole. I haven't looked into this, but what is the high code, low code, citizen developer thing going on in the analyst world? Mixed terms, right? Like, so I think there's the high productivity platforms, which are slash low code. So OutSystems, Mendix, Heroku, that kind of thing that kind of, you don't have to be a developer to use and build little applications, quick applications. There is one former Gartner analyst who used to call them crappy little apps, craps. That sounds horrible. I know that sounds super judgy and horrible. I don't get in trouble for that one. But because I think people are writing more business critical applications using these platforms, not to devalue our partners there. And then the high control would be something like a cloud foundry.

Like a pivotal application service, right? Where, you know, where it's like actually requires you to write code and all that stuff. Am I wrong to think that this concept of a citizen developer has reemerged? I remember reading that from Gartner like three or four years ago. And I could have sworn I've seen it like many times. I guess I could go do Google Trends thing. But like, have you seen that phrase used a lot recently? No, I haven't. It used to be used way, way more a few years ago. I totally am with you on that. Like, it was everywhere. And I was like, oh, no. But they're not, I don't think they're using it. I don't think either of them is using it as much. And I'm wondering if that has a question. I think it comes up once in a while. It'll, you know, it'll kind of slip out. But they're not using it as much as a thing. Yeah. It sounds like maybe in, how do you say his name? In a Robert Heinlein universe. That's what you call programmers. Citizen developer. If your code compiles for six months in the build pipeline, you could become a citizen. And I think the point where the high productivity versus high control is like, you know, again, it's also not a zero-sum game, right? I think both firms believe that you have to have both. You know, and if you buy in, I don't know, Forrester definitely does that. You've got to be a software. So, you know, software, you're going to be a software. Every business is a software business, which means it's going to be a lot of software that needs to be written. And we just don't have the people for it, right? So that's why, again, low-code platforms will be more important because they kind of can help you scale a little bit, you know, the people part. So, yeah, it's not a zero-sum game. So let's close out with our own predictions. Since I'm the host, I will buy myself time by going last. So how about we'll start with you, Rita? Now, you already came up with one. You can just reprise that and detail it.

Or you can come up with another one. But what do you want to predict for next year? I'm going to predict that AI is going to take a longer time for AI to become part of the app depth tool chain than both are predicting. Okay. Lame, I know, but it's good. Good. All right. That's yours. Anything else? No, I think that's it for now. All right. I'll do two because I can. I'll say that Amazon and Google make some sort of iPads play. Gardner says they're tracking over 100 providers. I'm actually intrigued as to why neither Amazon or Google are willing to challenge Microsoft's growing dominance with their products there. So when it comes to integration paths, like integrating things, I would think Amazon, I would think that Google want to have a bigger play there. I'm surprised they've been on the sidelines there. So I could see them picking up one of these smaller, you know, tail providers to just kind of give themselves a good story because that's stickiness that makes things come to your platform. So I would expect one or both of them to buy somebody this year. And then the other one I'll call out is, again, something you may have a better sense of, but I think we're going to see. I'll make a random number, at least five enterprises sell, like traditional non-software enterprises sell products this year. We saw this with Liberty Mutual. We saw this with others who they've gotten good at software. So why shouldn't they be selling data sets and products to actually make extra money as side businesses? So I expect we'll actually see more enterprises realize they can make money off IT. It's not a cost center. That's a fun one. That's like the fundamentals of the API economy theory were always interesting, right? Which I'm rephrasing in a slightly different way what you're saying, right, is that a business has some good process or data that they could sell to other people, even competitors, and make a new product out of it. And I'm not sure. I don't know. I don't really have time to think about why that didn't pan out with such a cool name as API economy. But I can see that that could be more realistic, especially since we have some cases now.

And it is, you know, it reminds me of all these B2B firms that you encounter, you know, when you sell enterprise software, like someone like a CoreLogic, like another pivotal customer who like, that's essentially their business. So it's not technically an instance of what you're saying, but you can see how you could amass all of this information and sell it, if only for like consumer ad targeting and stuff like that, right? Like, but yeah, that would be a nice new revenue source for people is to like more systematically, instead of just doing a big data dump of something, like systematically sell access to process and data that various businesses had. Well, I'm going to steal my prediction from my other podcast over at Software Defined Talk. We had a whole episode on predictions if you want to, you know, go to that area, the explicit marked podcast. Anyhow, so I've been thinking about this a lot recently. And I think even a couple episodes ago, I talked about this with Matt Curry, if I remember the publishing order of things. And that is, I think there'll be a very thorough discussion of how DevOps is a little less than real. And there's a lot of the benefits and practices that are in there. But the idea, as the name would imply, of having these two things together, working hand in hand, metaphorically speaking, like will be sort of lessened as a reality. And people will realize that all the same stuff is going on. But really what they're talking about are what we would call an SRE, which is to say a sysadmin who takes a programming first mentality to doing things. And, of course, there will be a chorus of people who say that's what we've been arguing all along. I call those people all those commenters in the register that Richard reads for me, thankfully. But, you know, they'll sort of defend that that's the way it always has been. But really, I think there's still a broad perception that DevOps is more about having those two people together, again, because it's right there in the name.

But, you know, I very rarely encounter people who actually are doing sort of like that classic idea of a DevOps thing and more of the here's a platform ops person. And the developers are just cognizant of the fact that their software will be running on the Internet. So they do some operation things. So we'll see how that pans out. Hey, you know, I feel like you guys had someone's better prediction. I've got to make a lot of talk about cloud native data, if you will, and data DevOps, the DevOps model kind of starting to be more important in the data world as well. Right. Especially with things like serverless, event driven, microservices. It's a gnarly problem. And I think a lot of people are like, we just don't want to know. We don't want to know. We don't want to get there yet. Because they know they're going to have to deal with it eventually. And they just don't want to do it yet. And I think the analysts are going to start talking about it more, which will hopefully drive more interest and visibility on it. And hopefully we'll be able to help our clients address it. I think just as with serverless, people always joke like there are servers. And I think with stateless applications, the ongoing joke is there is state. It's just really hard. So we don't like to talk about it. So that definitely seems like something. Well, also, I'll just leave you with one, you know, since you won't have our voices to listen to for several weeks or things. While you're there enjoying your eggnog, I want to recommend a book to read for people who are interested in this stuff. It's a very old book. It's called Moral Mazes, The World of Corporate Managers. And, you know, it's a bit cynical. So you might want to have a couple drinks of your bourbon-laced eggnog before you dive into it. But if you manage to read through that book, I think it will give you a very good idea of how to actually start getting things done in your large enterprises.

Whether it involves knives or candy, you know, the stick or the carrot. Is that right? I'm just thinking of a donkey. And also, there's a good companion presentation from a great American insurance company, GIAC, over at Springwind Platform. We posted all the videos, it seems like, by the way. So I'll put a link to those. But he had a great presentation along these lines called How to Navigate the Sea of No's. He's much more positive than this book is. But those are both good sources for figuring out how to actually change culture and fix those problems that we have all the time. So thanks for being on, Rita. That was fun. Yeah, thank you for having me. And by the way, that was one of my favorite sessions at Springwind was the John Osborne piece. Yeah, it was pretty good. Well, as always, this has been Pivotal Conversations. If you're bored during the holidays and you haven't listened to every single episode, you can always go to SoundCloud.com slash Pivotal Conversations and peruse our back catalog. There's a lot in there you can just listen to. It's like a time machine through infrastructure software. News of the past two or three years. I forget how long it's been. And as always, we post the full formal show notes usually every Thursday at Pivotal.io slash podcast. And if you want to reach out to us, we're all somewhere there in Twitter or the internet or something. But you can also just email podcast at Pivotal.io, which would be great because I've been getting a lot of email about putting ads in our podcast about a bunch of inane stuff. So it would be nice to actually hear from a human or a chat bot. Either way, sounds good. We'll see everyone next time. Bye-bye.

Tanzu Talk · 🍕 Get more pizza.