Predictions past and predictions future, 2018/2019 edition

Episode 105 · 45:14 · SRT · archive.org · all Tanzu Talk

All right. Well, we're coming up on the end of the year. I mean, I'll reveal the obvious secret. We're recording this a little bit ahead of the end of the year or whatever. That's an odd phrase, but I thought we should have a, we don't have any news this episode, but we should have a little predictions thing. Whenever I was an analyst, I was low doing predictions, but whatever, you can figure out a way to do them in a fun way. So I thought, you know, you had the good suggestion that we should look back at previous predictions and kind of rate them. And I think, I think what you've done is more or less picked a bunch of other people's predictions. So that's good. I think maybe there's one prediction you made, none that I made that we can review. So that's, that's good. And we'll kind of look back on those. And, and I think, I think mostly what that exercise is fun for is to sort of get a sense of the, the aggregate of what people were thinking a year ago. And we can kind of talk about that. And then, and then we'll wrap up with some predictions of our own and I'll make some so that, uh, you know, assuming we're here next year, which hopefully we will be, uh, you know, we can comment on those as well. What's your, let me, let me ask you just one very small, uh, system of working question. Now you, you've been around, uh, for a while. What's your, what's your model when you're doing predictions? Like how do you, what do you do in terms of how do I come up with them? Yeah. Yeah. Like, like, so for example, uh, in some more formalized places, if you have a group, uh, you'll have a spreadsheet and people will write a bunch of one-liner, uh, predictions. And then you might allocate each person in this group gets 10 points and they can assign however many points they want. So you wait it and you sort it. And then you're like, these are, this is our belief in the predictions or something. That's an overly complicated way. I think usually what you do is you sit in a room over some like, you know, sandwiches or some catering and you just sort of like, as a group, figure out the ones you want or don't want before you get bored doing the task. And then a junior person has to write it up, but I

don't know what, uh, what do you do? I don't think I'm creative enough to come out with the completely left field predictions. I think what I typically end up doing is seeing where do I see like a, an early indicator, right? Is there some sink signal in the place that says like, wow, this, this kind of thing, is this a canary in a coal mine? Let me use a lot of metaphors. Uh, and so, you know, that's what, that's what I ended up doing is I look at that and then try to think like, how realistic is this? And for purposes of like our podcast, it's, it's probably better to be less realistic. Cause that's more fun and hot take like, uh, but yeah, no, I don't have a fancy flow charts or sticky notes or scoring systems or, or, you know, internet polls to help me, help me pull this down. How about you? Do you do some, uh, consult the magic eight ball or? Oh, I think, I think probably similar to you. I just pay attention to what's going on week to week and just kind of think about the, uh, the trajectory of it. I mean, I think maybe the only, the only thing different, at least in what you said is, uh, I usually go back and like, look at stuff I've bookmarked or talked about just to remind myself of things that had happened. And, uh, and then it's also like, uh, when you're, when you're all fancy pants, like, uh, like we are someone, someone pays for you to have access to analyst sites. So it's fun to go look and see what their, their predictions are present and future. Yeah. I think that's the weird thing that probably both you and I end up doing is we, for better for us, we end up absorbing a lot cause we just keep in touch with what's going on. So I just think our brains kind of grind through on this stuff and eventually certain things kind of just squirt out as like, Oh, this is probably what's happening. Cause you've been chewing on it for a year. And so it just seems like that's probably what happens is these are just insights that come from go going through gobs of information every day. Yeah. And it's also a bit of a, uh, I don't know what, what the word is like a spoiled advantage of, of like, uh, you know, it's kind of like, well, my prediction is it's more of this. Like when you, when you have too much like ongoing commentary about what's happening, it's, uh, it's hard, it's hard to like have not checked in for a year and then therefore see how much is advanced. It looks

like a, uh, a slow encroachment. Uh, I don't know the closer you look at it anyways. So what, uh, what have you gathered together of predictions for this past year? What were people saying a year ago that would happen? Yeah. So we'll share, uh, the list of ones I'm looking at. I picked up some more of high profile sort of forums, not, you know, weird guy just coming up with predictions, which is what I would write, but what are, what did we find in some larger publications? So you know, one of the first ones I'm looking at, uh, is, Hey, 2018 is going to be the year of the voice interface. Like, Hey, Alexa, and you've got, you know, Google home and you have, we all love IVR systems. So maybe that was also supposed to be more of it, but that was the first one. Like did voice really take off this year? I don't know. What do you think? Do you think that voice, I mean, you know, I've Google homes in my house, you know, obviously those sort of devices are big. I think I saw there's 10,000 people at Amazon who work on Alexa, which is just unbelievable. They must, they must be actually typing answers to the questions. I think that's what it is, right? Yeah. What should we have for dinner? And there's like three people quick swarming around a cubicle going, I don't know, maybe some stew. Uh, that might be it. I like, I like that the first thing he thought of was stew. Uh, well, first of all, I think, I think I had never connected those two things, but I think the idea of like, uh, you know, Alexa and Google home and all these things is basically like, uh, a self-service IVR system is awesome. It's just, it's just like a, uh, a decision tree of things to do. But yeah, that, that was, that was one of the things I was looking at when I was going, looking through past things is, uh, I, uh, I had forgotten how big of a deal that was not big of a deal as in the phenomena, but like, that's all anyone talked about for a while was like all those voice things and what's going to happen with it. And I don't know, I haven't gotten to go see, you know, the IDC supply chain reports or whatever, but, uh, I don't know. I think, I think people kind of like think it's interesting

and, uh, it's a really good speaker. And, um, but I don't, I, I haven't really heard any like great follow on stories of like how, how, how those voice things have like changed people's lives or businesses in, um, I mean, you see some, you see some spot stories of things happening at certain companies who voice enabled certain stuff, but I don't know. A lot of times it's still faster to do things. However, I did it before than talking to a machine. Now they've gotten much better. I don't know if you ever had a car that had those first generation sort of voice activation features and you had to say it the exact certain way you had to have the, you know, it sounded like a robot when you talked to it. So I, I, you stopped using it after the first day. So I mean, these are better now, which is great. My, my four year old can scream at the Google home and it somehow knows what song to play. Yeah. That's amazing. I think maybe that is the most popular use in our house, uh, which, which for me, which is, it makes me wish it didn't exist because like last night I listened to the same, what are these kids listening today type song like 20 times. And, uh, and then even better, like my son would, my son and daughter would be trying to ask for it again and it would take them three or four times to do it. But yeah. So, so the, the potential of stuff is there. And I guess, I mean, one last thing for me, at least on that topic is, uh, you know, now that I live in a, uh, um, what would you call it? A big city where you walk around a lot in Amsterdam, right? Versus being in your car, being isolated, you're kind of like frequently out in the environment. Uh, like I, there's this idea to use another one of Siri, where if you've got the full Apple kit going on, like your earbuds and your phone and you're connected to your internet, you know, there's also 5g. I always wonder what happened to that. I've been hearing about that forever. I think 5g and BDI are in the same bucket of, of, uh, the future anyways. Yeah. And, and in theory, like you, you hook up your Siri things so that at any point I could say like, Siri, when does the next train leave from Amsterdam central to go to Muta port? And it would like, tell me something and you know, it would solve my

problem. But I just like, like, even if that did work, I just can't imagine myself walking around in a throng of people. And all of a sudden I'm like Siri and like talking to the phone. It seems so weird, but you know, I mean that idea of, and also I could never get that thing to work where I just say Siri, but I I'm still, despite like being kind of a voice downer or whatever, like I'm still attracted to this idea that I could just sort of like say that all of a sudden, no matter where I was and something would happen. Uh, that sounds cool, but I'm willing to wait the five to 30 years for it to actually work. Yeah. Especially now that we're all in open floor plans where it would just be constantly going off all the time. So I couldn't use that at the pivotal office. That's right. All right. So that was the first one. Uh, another one in this particular set of ones was talking about retailers kind of closing stores to focus on e-commerce stuff and try to fight back on Amazon. I mean, I think we saw at least the first part where a lot of retailers were closing their doors. I don't know if we saw them then kind of come back in force to actually be better online presence as some of them are, you know, one of our retail customers had just blew out black Friday, like three times what they ever did before. Of course, everything worked perfectly because it was on PCF, but it was like these online presences are, have absolutely kicked up for some, but did you see, or did you notice anything around here, around retailers kind of up in their game digitally while they closed down brick and mortar? No, I mean, not this year, at least like I remember in writing up, working on finishing out my, uh, my little book, which I forget if I told you this, we came up with the clever title monolithic transformation. We'll see how, maybe it's just monolith, but that's a good, I think that'll work. That's better than what is, what is digital transformation. Anyways, like, you know, in reading that, like I've looked at a few things, a few examples and retailers have sort of like already done as much e-commerce stuff as kind of possible. And I mean, just, I don't study retail very well, but so I guess that's a way of saying like this prediction is correct, but old.

And I mean, I guess, I guess what I have seen kind of anecdotally is, uh, well, one on the negative side, and this doesn't really have anything wholly to do with this prediction on its own, but some of the, as they say, forces behind it, like, I think Sears is shutting down like completely, right? Which is, that has a lot of its own problems besides computers that were driving that from the lengthy fun, fun stories I've read. But then also like Toys R Us, I think went out of business this year, as I recall. But, uh, yeah, I mean, I think, I think with some exceptions, I think most retailers who are going to be online are sort of like, given their late start, usually are like at par with like selling stuff online. And I think, I think the one, the one thing as far as like replacing to the point of the prediction I've seen is, uh, like at least at grocery stores, um, you know, down in Texas, there, there's only one grocery store that matters unless you're, uh, you're one of the other grocery stores that is interesting and interested in being a pivotal customer. Then there's many grocery stores that matter. But the, the biggest grocery store down there is HEB and over the past year they've been adding, uh, or at least I've noticed them adding omni-channel stuff, right? Where they'll deliver your groceries. You can pick up your groceries. And my sense is that like, that's additional business that they're adding, right? It's not like replacing in-store stuff. It's just like another way to get your groceries. So, I don't know. I mean, I mean, I feel like I, I would change it slightly, I guess, to be like, I think, I think online things are augmenting and growing retailers' business and not sort of like sucking away from the, uh, in-store experience, if you will. Yeah. No, it's interesting. I mean, the, I, the amusement isn't lost on me that, you know, around my office is an Amazon Go store. Like now they've opened, they've actually opened up physical stores and my kids get a catalog in the mail from Amazon. Like what the, like you're literally doing the Sears game plan. I know that, that was, that was one of my, uh, my main winsomey kind of thoughts when Sears was going out of business is

I read somewhere that they stopped doing the big catalog some years ago. And, and I was thinking like that used to be one of my favorite things as a kid was like, you get the Sears catalog and you go circle all the things you want for, uh, for Christmas. And man, even now I'm, I'm taking like a few extra seconds to just enjoy that memory. That was always a fun time. So that'd be great if Amazon did that. Just print that up and go buy things. Yeah. Destroy all the physical companies and then they'll send me catalogs. I think it's a plan to success. Uh, the other one in this list, I'll wrap this one up with, uh, besides the CMO, it says we'll be more important after the, uh, the most important C level, like after the CEO. And now that I've recently come out as a marketer, uh, I like that one. That's, uh, that's good. I think marketing is more important than ever. So yeah, that sounds great to me. How about you? Yeah. Way to suck up to your boss. Be like the, the, the head of marketing. Very important. Yeah. Yeah. This game is played. Yeah. It was so, so wait, what is the prediction again? The CMO? So the prediction was that the, yeah, it'd be the most important C level after the CEO. And I think, you know, joking aside, I think some of that is about like this era of actual getting messaging figured out. It's very noisy in this marketplace, understanding how you actually connect people that actually is really meaningful nowadays. Yeah. Yeah. So I think, I think, I think, uh, to, to be, to be Weasley about it, when I think of this prediction, it's kind of like more of a, uh, you know, my favorite metaphor, not one, one of my favorite metaphors is like something's a Rorschach test, right? And this is like a Rorschach test of like your perspective on things. And I think in some enterprises, some organizations, marketing is the primary value generator, right? Like, um, you know, I use a lot of shaving cream and the difference between any giving shaving cream is not much, right? Like it might be a different color. Maybe I put it on with a brush or, you know,

whatever. Like it's really, it's a commodity. And so in that business, like marketing is incredibly important, right? Like how you market something is a huge part of what drives your margin. And so if you're in that kind of business and you're doing digital transformation or trying to learn how to computer better than sure, your CMO is probably going to drive a lot of those projects, determine the budget for it. And more importantly, they're the ones who are going to get the value from it. And so it's kind of like their deal, if you will, uh, versus like, I think other businesses, like if you were, um, I don't know, uh, manufacturing tractors or cars, then like marketing is also important, but it's not as important as, as like, well, our tractors are 20% more awesome because of this thing. Right. Right. Or, or, you know, like the classic thing in agro tractors and agro agricultural stuff is like, we have satellite technology and IOT stuff that makes like programming your tractor and gives you, you know, crop yields of like five to 10% more, which means you don't go out of business as a farmer or whatever. I don't know that industry. So yeah, I mean, I think, I think more of, of, I don't know. I mean, I think it's more indicative of like, you have to figure out how to apply like this function of being able to customize the software you have and whoever the ultimate wielder of that is what becomes important after the CEO. And maybe there's one exception like, and well, I don't know, like when, in talking with retail people, I've noticed that there still is a distinct like IT department and, and the, the managers of like, you know, the store experience versus the rental tool thing to use a hardware thing. Like they're still kind of independent and, uh, they get serviced or helped by IT. So maybe that's the distinction is like, who is IT's primary customer that is not IT in the business. And, um, I don't know. I think that's more of what's interesting is

like IT providing someone, finding that role and, and having them be the customer and who that, what that letter after C is before O is just, it just depends on what the moneymaker in the business is. There you go. That was my dissembling. If I'm using that word correctly. Good dissembling. Uh, I'll move up to the, uh, the new stack next. Actually it was written by Craig McLuckie, who I guess is about to be one of our colleagues when, uh, VMware's intent to acquire Heptio is completed. But so he wrote up some stuff, uh, for the new stack last year, looking at kind of cloud. And, you know, one of his first points was, Hey, multi-cloud is going to be imperative. Open source is the foundation for it. That seemed like a safe bet. I think we saw a lot of people talking multi-cloud this year, realizing multi-cloud. It seemed like software vendors recognize that their future probably lied, you know, lay in being a multi-cloud provider versus saying, we're just going to be, Hey, we're just going to couple to a single cloud and hope they don't eat us. Instead. It seemed to be more of like, Hey, let's make sure our software runs in multiple places on prem and off. So that seems like a pretty safe prediction, but I file accounts that has been true this year. What do you think? Yeah. Yeah. It reminds me of another, I think maybe this was in one of the Gartner or Forrester things, maybe Forrester, but it was essentially, yeah, it was Forrester and it was, Oh, I forget. It's what's his name. It's the guy who is Twitter picture is black and white and he's got a good smile and a little, and a beard. What's last name starts with a B anyways. What, one of, one of his predictions was, uh, that private cloud was going to, you know, in, in, uh, 2019 be more fleshed out in a bigger deal. And he had a, he had a pretty, um, coherent writeup of it. And so, yeah, I mean, I, I often, uh, I don't know. I have a lot of like philosophic angst about like multi-cloud and lock-in and all these kinds of things. But I think, I think, I think what, when I hear multi-cloud, I think, I think that people are interested in, in two things. Well, one, first of all, I think that prediction is accurate over the past year.

Uh, yeah. And, and I think, I think what people are talking about, let's take us vendors off the table. Cause we're like clever and conniving when we use words, we're just trying to help people. But, uh, anyways, I think what people are talking about is like, I don't want to be restricted to functionality of some cloud I choose. So I want to be able to have, I want optionality that in the future, like prices don't go up and I, I don't regret, uh, choosing this cloud because now I'm locked into this pricing and even, and equally worse, I now regret, regret choosing this cloud because there's, I have like feature envy of some other cloud that's out there. And then I always, I think there's another case of like, I just, you know, literally have to be able to run on multiple clouds for various reasons. Uh, and yeah, I mean, I think, I think that is a very real, like architectural concern people have. Like it's hard to, it's always been hard when you're making it decisions to choose like one platform from top to bottom. Like you always want some variety in there and the optionality to try different things. So I think, uh, yeah, I mean, multi-cloud makes a tremendous amount of sense if only because you want to, uh, take advantage of as many different types of functionality as possible. And then also, uh, man, there's a lot of choice out there at the moment. And yet, and yet I feel like you have to make a choice. So when there's, when there's a, um, what would you call it? When it, when it's a buyer's market, like you want to buy as much optionality as possible, right? Like you don't want to restrict yourself and, uh, you don't want to lock into things. So, and lock in in the sense of like, I've committed to this decision and therefore I can't decide to do other things later. Uh, right. Like, yeah, I mean, that definitely seems a multi-cloud. That's a thing. I think also like to add some more dissembling is actually to conceal your, your true intentions with a word wall, which is actually not what I was doing. I think I was

more babbling. Uh, but to, to add some more babbling, I think, I think, and this is a little like, I think one thing that will be interesting to see over the next year is how much, uh, like Amazon takes, takes the place that like a, uh, early two thousands Microsoft had of this like dominant, like big Island of its own versus all the other players and how that kind of conceptually fits in. And I, and I think that kind of situation would drive a tremendous amount of multi-cloud think as we saw with open source and Linux versus windows, uh, in the early two thousands. Yeah. Good. Uh, well, the last one I'll pull from this one, uh, Craig had mentioned, Hey, this will be the year of reliability engineering, you know, is Google's SRE sort of stuff picked up. And, you know, I think that there's a lot that we saw probably in pivotal around that is more people wanted to treat their platform as a product that offering from us remains extremely popular. as people realize that anything they're running, just having, you know, a shotgun approach of upgrades and Hey, when do we evolve this thing? How do we run it? And a bunch of admins just support 50 different products, but instead like treat our core platforms as actual products with life cycles and do reliability engineering and make sure that we're thinking about SLOs and who's using this thing. And how do we think of error budgets and everything won't be a hundred percent online and how do we build reliability in all those sorts of concepts seem to come to the forefront. There were definitely some high profile outages this year. Like there will always be, but I don't know. Did you see an interesting uptick in people thinking about reliability, especially as you know, you just brought up again, a lot of multi-cloud, we're building more complicated systems. I mean, the monolith for all of its warts and maybe downsides also was probably more predictable. Like if it did crash, you kind of knew how to take it back online. Now we're building systems that are almost emergent. Like you don't know how this thing will behave. So have you seen an uptick in that or is just, just par for the course? Like, yes, make your systems reliable. Yeah. Yeah. I think that, I think the way you explained it, I was thinking of it one way when

you, when you read it, but no, the way you explained it. Yeah, I agree completely. That like, I think, I think there is, uh, there's so much discussion of like site reliability engineering and, uh, and that kind of thing that there's, uh, there's even like delightful or annoying, depending on your perspective, debates about like, is SRE real? And how is it different than DevOps? And like you and I have talked about this a little bit from time to time. And, uh, I, I, I, I must've been late at night when I should have been going to sleep instead of, you know, being on the internet. I was, I remember I was Twittering something that like, uh, you know, you've reached success as a thought technology when people are like debating what it is, right? Like debating what SRE is and if it's real and what's not. And so, yeah, I think, I think that idea as, as we would say it as thinking of a, uh, a, uh, a platform as a product, which is kind of the SRE approach. Yeah. I think that definitely like came to a early fruition in this past year and we'll see even more of it. And just one more data point coincidentally on that is, so I've been trying to finish watching all of, uh, the Rabobank presentations that are out there, which is to say all the presentations that this guy Vincent at Rabobank has done. Maybe there's other ones I should look up, but it's interesting to look at the timeline of, so they have a platform that they put in place maybe a year ago. I mean, pivotal cloud foundry, obviously, but they have put in place this idea of a platform as a product that they use to run their online bank. And also that they use as their microservices platform for the entire bank. And what's interesting to think about their timeline is it took them, I think, three years to get to that point. Like it took them two years. Once they decided that they needed to decide on something, so to speak, it took them two years to stand something up. So I would have to imagine a year before that they were like, we should decide to decide something. And, and so, you know, it took them two or three years to get to the point where they

stood that up, I think, earlier this year, let's say at the beginning of this year. And so if you think about it, like most all other large organizations like them would be on that same timeline or a similar timeline. And so they would have had to start about two or three years ago to in 2018, have it be the year they have like SRE and platform as a product. And so I think, I think most people, if you think back two or three years ago, they would have started thinking about this stuff. And if they're as sort of lucky and with it as Robbo is, they, they would have been able to deploy it this year. So yeah, I think that lines up. And it's definitely what's in the discussion. Everyone talks about what's DevOps versus SRE versus there's no other words, just those two. That's it. All right, good. I got four more for you. So the one I'm looking at CIO online, CIO.com. One of theirs was, Hey, Google gets real. The prediction was Google becomes the number two public cloud in 2018, outpacing Amazon and Microsoft to customer growth, beating Azure in revenue. That has not happened. And obviously Google had some, some turnover at the top with Thomas Kurian coming in over Diane Greene's tenure and being in, I don't know, I'm personally still bullish on what's happening at Google cloud. It's still great tech. They're expanding at a great pace. You know, it's a talented team there. I think they'll continue to try to figure out how do they get the enterprise traction when they don't have that initial foothold that Microsoft has by default by being already a trusted provider there. And Amazon just has its own ubiquity that they get to sell. And Google's still seen, I think, as a, as a consumer sort of company. So that one didn't land, but you know, what did you think of this year in Google? Well, I think, I think, uh, you know, I'm always sore over Google reader, but that's fine. That's a whole other part of the company. Me too. Yeah. We're not going to talk about that right now. But, uh, yeah, I mean, I think, I think, I think what, what we in our infrastructure, enterprise software, whatever it is we do, let's say cloud neck of the woods, what we learn all the

time is like, uh, Google can be, can seem like they're kind of like aloof, but really it's a bad idea to think that they're not going to figure it out. Right. Like, and, and, and as part of that, um, like I think something that I I've heard over the past year to the point of this prediction is pretty much every developer type person I've talked about is like Google cloud is awesome. Like they, they, I've heard it characterized over and over again that Google is like a cloud built for developers. Uh, and that would make sense. So yeah, I mean, I think, I think Google is constantly well positioned to do very well. And I mean, they're, they're top three, right? Like I, I remember when Thomas Kurian was going in, there was something like they had, I don't know, eight, 10% or something of the public cloud market, which seems small, but I guess it would, they were the third tranche of, of, of, uh, market share ownership. Maybe it's bigger than that. Yeah. I probably got that all wrong, but, uh, but yeah, I mean, they have a tremendous amount of potential and I think, I think, uh, yeah, I mean, I mean, they, they just, I don't know. They, they, uh, I think they'll do well. Yeah. I think that's the safe path. And, and I think, you know, as an example, right? Like, like, uh, maybe two or three years ago, like they were talking about Kubernetes and no one even really understood what that was. And like the leading thought leadership on Kubernetes was literally called Kubernetes the hard way, which was sort of like, this is a very difficult thing, uh, to do. And yet that's sort of like, uh, I don't know, a big deal in that space nowadays. So yeah, they're, they're pretty good at that. And then, and then as a, uh, as a sidecar, like no distant pun intended, uh, for that, like, I think, I don't know. I mean, I think hiring the, the, you know, the, the guy Thomas from, uh, from Oracle, it's sort of like all we've been saying,

and I said this on a software defined talk podcast a while ago, and I think reprised it here a little bit is like, everyone's been saying for the past two or three years that Google should have more enterprise oriented sales. So they hired an enterprise oriented person. And right. I mean, we'll see what happens, right? I mean, there's always cultural problems and things like that, but, uh, I don't know. I think it'll probably turn out well. There you go. Good. Uh, the last one in the CIO one was actually tongue in cheek that block blockchain will literally transform everything in the cloud. And just saying, cause I have to do something about blockchain, but you know, non-joking aside, do you feel like blockchain changed the world in, uh, in 2018, you know, Amazon at reinvent announced their own ledger database and a blockchain sort of service, but it seems like obviously parts of that, and then you get into sort of Bitcoin and some other aspects of that universe, which kind of crashed a little bit this year, but like the blockchain world, did you see applicability in actual enterprise places of people trying to use that sort of model successfully? Nope. I didn't see a ton. So I don't, I don't, I don't think there was really, uh, there's always lots of potential in the, the only, I shouldn't say the only, the most interesting stories I was here is like supply chain management. Um, so yeah, I mean, it seems interesting for that, but it seems way overblown. Um, and, and, and I, and I think, I mean, not to, well, not to, let me keep babbling on some more here. Like, I think the problem is, and this is from a very non-intelligent stance is like all blockchain does is say you had two parties that agreed to something and it's verified by a third party. And I'm sure it's a lot more mysterious than that, but it's sort of like, well, if two parties can agree on a third party to agree on something, do we need a bunch of fancy math? Like what, like you still need this third party. And I think that's kind of what AWS announced is like, we will be your third party. So there you go. Right. But if you can't like recruit that third party, then it doesn't

matter. I don't know. I mean, I'm sure there's more nuance to it, but it's sort of like, it seems like a lot of over-engineering to trying to avoid picking someone as your third party verifier. Like it's really weird. So, uh, I don't know. I mean, I mean, as far as like verifying that stuff happened in your supply chain and a few things here and there, it seems interesting, but like, I don't really see how it's going to disrupt the global banking system. Yeah. I mean, we'll see. I'm not like I'm down on it. It just seems like it's so overhyped, but I guess we'll keep an eye on that. 2018 or 2019. Yeah. And then we can talk about our own stuff was, uh, it might got to hit machine learning AI stuff. So I'm looking at the pivotal software trends and one was like, Hey, you're going to see, you know, even more applicability of things around machine learning and AI. You're going to see this come into more ways and more places in enterprise, not just against startups doing the chat bots or whatever. So are you seeing a IML pick up the pace? I think I've definitely seen more people interested, maybe doing still some initial stuff versus transforming their business based on machine learning or doing certain things. But it seems like it's becoming a bit more mainstream. I don't know if we're going to have the robots actually running companies in two years. Like that. What do you think? You ever see that twilight zone episode where like a mainframe ends up eventually firing the CEO, I guess, I guess that's an early version of Elon Musk paperclip problem, right? Where, uh, you tell an AI to make a paperclip and it ends up eradicating the human race. Cause that's the most efficient way to make a paperclip or something. That makes a lot of sense to be a downer. Yeah. I mean, I was going to mention, I'm watching Bojack Horseman and a robot just took over our company, but that's very different. Oh, nice. I haven't watched that in a long time. Uh, yeah. I mean, I think, I think maybe this is one of those where like, uh, we're like, it's not that it's not happening. It's that there was no significant movement in 2018 versus 2017, if that

makes sense. So yeah, I mean, I agree with what you said. I see a lot of talk and conversation about potential, but I haven't read, I haven't read enough stories of actual usage in the enterprise to make me think I've read enough stories of actual usage in the enterprise. Right. Besides the, like, like, Hey, we used the image recognition, like package services from the cloud providers. Yeah. Yeah. Yeah. Yeah. That's cool. But yeah. Exactly. Like we talked about, uh, Amazon part finder a while back, right. Where it's just like, and I played around with this again today where you can just like point your phone, uh, you know, using the Amazon app at like some product and it'll try to say, Oh, this is what it is. So you can buy more of it. And, and it worked, I tried it on two things or I tried it three times and it worked two times on a, uh, a plug adapter and a pin. And then I tried it a third time on the pin and it got confused, which is fine. Right. Like two out of three is pretty good. Uh, but, um, yeah, I mean, I think the most interesting thing I always hear is like, so if we wanted to do insurance claims, this stuff could help, which sounds awesome and pragmatic, but I haven't heard any more beyond that. Right. Yeah. Another thing I'm bullish on, but how do, how do companies take what we're actually encouraging them to do, which is distributing more of their data. So you don't have these massive data lakes and all these single massive databases instead. Hey, have a database per microservice. And then by the way, tomorrow glean the information from all of them and make quality predictions. So, I mean, it's interesting. Is the architecture actually going to make some of this harder? Maybe not, but I don't know if we've, uh, we've actually landed on something that's repeatable yet. So I guess we'll see. Yeah. Yeah. Yeah. Yeah. I mean, and, and I forget this must be, and I think I mentioned this a while back there was, there was this great article that was basically like, if you're doing digital transformation and you're not changing your business around, then you're wasting your time. Right. Like if there's no change to what you're doing, then it doesn't matter. And for example, like, I think there's a lot of practical things I think that machine learning could apply to. Like, uh, I stay at

like Marriott and Starwood hotels and about half of the time I have to email to get my invoice sent to me for whatever reason. And like, that would be a great machine learning thing to be like, Hey, this one person emails us half the time to get their invoice. So maybe we should just make sure we can send them an invoice. Right. I mean, there's kind of like that proactive stuff that would seem to be really simple to do, but I don't know. So maybe hopefully next year we'll sort this out and I won't have to wait to file my expenses. Awesome. So let's take a couple of minutes then and, uh, have our own hot take predictions. All right. Well, I'll, I'll, uh, I'll start with one to give you time to read your notes or think, I think, I think, uh, building on the AI and ML machine learning thing here, here's my bold prediction. I don't know if this is hope or prediction, but I think, I think, I think it pans out using my complicated model for doing predictions. I think, I think people will stop talking about artificial intelligence and they'll just talk about machine learning and they'll be like, Oh, AI is some Arthur C. Clark stuff. And that's not what we're talking about. We're just talking about machine learning. And, and I think even the economist has kind of front run, like, I think they'll have some cover story of like some machine that's like sitting at a desk learning. And, uh, and then that'll be it. We'll call it machine learning forever. No one will say AI AI again. Hmm. Yeah. All right. That's a hot take. Uh, good. So what do I have for you? Uh, actually I think that, and maybe you'll like this one. I think architects are going to come back in vogue. Oh, nice. Cause, cause we're going to look at arguably more modern software development. You cannot, I mean, in your monolithic book, I can't just heads down on a stack anymore. I really do have to understand how to glue things together, stitch things together, understand big picture. And more importantly, like, I don't know if it's ever been easier to make absolutely catastrophic decisions in tech just because we're so empowered on stuff. And so, you know, that choice of multi-cloud might be great for this project, but I've just added a gob of technical debt potentially for the next person.

So architects being able to come over the top and make informed decisions. So not sort of just these helicopter architects who just kind of, you know, fly around and dispel wisdom, but people who actually know these systems, it seems like that's going to be increasingly important as I'm maybe doing less raw coding. Cause I shouldn't be writing my own identity system anymore. And instead I'm gluing together a lot of managed services in the cloud. I need architecture. That's not just a dev skill. So I actually think that's going to be a hotter job next year. That's good. That's like to use another old sci-fi reference of Frank Herbert. One will have the, uh, the sleeper will awaken. No, I think that makes total sense. And, and to plug it again, right? Like if you, if you watch some of Vincent from Bravo banks talk, he's talking exactly about enterprise architecture and thinking through that. So I think, uh, I think that'll come to fruition. I like that one. So here, here's another one kind of, uh, something I complained about. I think, I think there'll be a lot of complaining about the online world. And in the U S I've already heard lots of people talking about, like, we should have some sort of data protection laws and things, but I think absolutely nothing will happen. And the government won't even talk about it except in a crazy way. Cause just like there will be no U S GDPR stuff. Cause really like, I don't think we care and there's other things going on. So, uh, I think we'll, we'll just have the, the extra thing to click on free existence in the U S that we've enjoyed so well, so much, and that Europe no longer has. That's basically how it is in Europe. Now, if you want to look at anything on the internet, you're required to click a different button than you used to. We've added a new click and sometimes it's even more confusing. There's like a radio button you have to select things on. Uh, but I don't think we'll have that next year. Got it. All right. So I guess my last one would be, I think you look at the absolute mounds of cash that a lot of companies have on their balance sheets right now. I think you're going to see even more M and a next year and potentially the big stuff, maybe not just a little tuck ins of like, Hey, here's a cool company that did, you know, a series a funding

round. Let's pick them up. But you might see more of this IBM red hat sort of world of, you know, some of these mega cloud vendors who are just printing money right now saying, Hey, how do we tuck in more software companies that might still be multi-billion dollar acquisitions, but we have the cash and let's keep consolidating the market. So I don't know. I just think M and a is going to be even bigger next year as software companies have more esoteric questions of like, can we beat the service providers? What's this open source versus cloud provider debate going to result in? Like, can these companies actually not be a cloud provider and still make significant money on open source? I think there's going to be a lot of good questions we're going to be asking and then probably some M and a as a result. Yeah, no, that's good. I mean, you know, I am no financial prognosticator, but, but modeling it out in my head, you're right. There's a lot of cash around, right? Like we're in this weird, in the tech world, this weird situation where like, like, for example, Microsoft, every, every one seems extremely successful and there's limitless potential and lots of money around, but people are like all weirded out in the general business world. I don't know. It's always a strange situation. And, and then also, uh, to be all macro person, like, uh, I think next year will be the beginning of the next U S presidential campaign cycle, right? Like, cause it'll be, and so like, that'll be crazy. And, and then, so, you know, in such a situation and it's sometimes what happens is you've got a great asset, right? Some tech thing, and you have the ability to take advantage of that and not in a bad way. And so there's a, uh, there's a flight to like, um, I don't know, the castle to consolidation and doing things to like, hold on to something, build the value and then do something with it. So sure. That would be interesting. That's a further tech industry consolidation. That, that makes sense. I like that. Yeah. Should be a fun year either way. So then the last one, and then, uh, and then that'll be it. We'll see how these pan out next year.

I think, uh, I think, you know, I think we'll, we'll figure out finally what serverless versus functions versus whatever are. And I think with, there's plenty of definition, there's plenty of technical definitions of what it is. Like it's this thing or that thing and what it looks like, but more of what I mean, even more so than what it is, is what you would use it for and why it makes and what it makes sense. Right. So on one end of the spectrum, you've got Simon Wardley, who's like serverless is everything and it's going to change the world, which is fine. And then on the other end of the spectrum, I don't know who represents this view, but it's just like, I don't know, sometimes I want to use it. And, uh, somewhere in the middle there is like, basically, I guess what I'm saying is, do you build your entire application on top of whatever serverless is, or is it just a thing that you use in the same way you use cues? And I don't really know what the answer is, but hopefully we'll sort that out as a community. And, uh, by this time next year, we'll know exactly what it is and how to deploy it. And that'll get sorted out. I think, I think that'll probably be the main focus of development think over the next year is sorting that out. I think that's a good question, you know, and a good prediction. We, I spent time at the Gartner show recently and, you know, they've even seen an uptick in sort of queries. If some people who've kind of thrown up their hands on trying to even run their own container stuff and say, Hey, maybe I should switch to something else. And that's just an early indicator of something. And other people are obviously, you know, KubeCon, a big event there's, that's not going away anytime soon, but that question of apps and containers and functions and managed services. And I think that that's a great prediction. I think we're going to get CRISPR next year versus this all or nothing. I think we'll get a little tighter on what parts of an enterprise portfolio might fit in each one of these. And what's the transition look like from one to the other? Cause nobody's tomorrow rewriting their ERP system as a set of functions. So what does the future look like? You know, I don't know why, maybe this is my, uh, De Bono lateral thinking thing, but I feel like, I feel like there's some wisdom that we had, you know, you got two types of screwdrivers,

you got a flathead screwdriver and a Phillips screwdriver. And my question is always, why do we have a flathead screwdriver? Right? Like, I mean, have you ever been confronted with a flathead screwdriver job and be like, Oh boy, that was a good choice. Right? Like, like it's always the worst situation when you have to put a flathead screw in. It's always bad, right? It slips out. It doesn't work. And a Phillips is much better. Cause it's like sticks in there. And I feel like, I feel like that's like, why have this second choice? That's no good. You need, you need to figure out the right, the right kind of way of doing this one, putting a screw in. Maybe if you have a nail, use a hammer, but like, don't, we don't want flathead screwdrivers. Let's make sure that doesn't happen. Like just pick something and move on. No flathead screwdrivers. Wow. That is a strong note to end on. Maybe you're going to switch us to metric. I mean, I mean, am I wrong? Is there a good reason to use? Maybe there's an aesthetic reason. Maybe it might look cool. Right. But like the experience of, of screwing in or worse, screwing out, is that the right phrase? A flathead screw is terrible. It's a bad experience versus a Phillips is just like, boom, done. It works. So that's it. I think the screwdriver community is going to be up in arms about this conversation, but I'm excited about it. That's a, that's a passionate punch. That's right. Well, as always, this has been Pivotal Conversations. If you want to find this show, I don't know why you'd want to find it if you're already listening to it, but if you want to find this show and all the others, you can go to soundcloud.com slash Pivotal Conversations. And despite the best wishes of our editorial board, who would like me to tell you this at the beginning of the episode, which I always forget, uh, what's going to happen over the past two years, I have a prediction, uh, that the Pivotal Insights and Pivotal Conversation podcast will merge and become just the Pivotal podcast, which I think for you, the listeners will be great because you'll get access to, uh, both access. You'll hear both of the podcasts and on Pivotal

Insights, they do a fantastic job. They'll, I have another prediction. I bet they'll have Vincent from Bravo Bank on at some point. Uh, but they'll have, uh, they, they talk with customers about how they've been doing things and, uh, both Dormain and Jeff are great at, uh, interviewing people. And then you'll also get this podcast. So, uh, hopefully you won't have to really do very much, uh, and you'll continue to get those as well. So don't get shocked when you start to be even more impressed with the, with the Pivotal podcast. And with that, we'll see everyone next time. Bye-bye.

Tanzu Talk · 🍕 Get more pizza.