30 Years of Tech Transitions

February 13, 2018 Pivotal Software

Watch Pivotal CEO Rob Mee and Michael Dell, CEO of Dell Technologies talk to Michal Lev-Ram, Senior Writer at Fortune, about the last thirty years of tech and what’s in store for the next thirty.

“I’ve actually come to believe that most people don’t take on nearly the risks that they should, and are generally way too cautious, and so I think there’s a lot of unrealized human potential because people are scared and afraid to take risks.”
–Michael Dell

Transcript

Michal Lev-Ram: I wanted to start out with just getting a little bit of your respective origin stories, and I think a lot of you in the room are probably familiar with how both companies got started but it’d be great to hear your first-person takes. Take us back to the glorious 80’s when you’re the hard-ware guy, and you were really on the software development side.

What did you guys see? What sparks and little glimpses did you have that gave you the confidence to know that you were onto something big? Who wants to start?

Rob Mee: I think way back in the 80’s when I started developing software, it was an interesting time in that I would do any kind of programming that anyone would let me do when I started, and then was developing software non-stop. Day and night, just for fun or for profit, or for anything. Occasionally people would pay me for it and that was good. I started doing some interesting things and got some partners involved, and grew Pivotal a little bit. But I always felt the status quo of software development felt a little bit unsatisfying.

It felt like things were not really as good as they could be in a number of ways, and that dissatisfaction grew until sort of in the late 90’s, I think there were a number of like minded people who were really feeling like this isn’t the way that software ought to be developed. With these really long planning cycles, and attempts to limit any kind of change due to the fact that it might introduce something dangerous, or something. Some kind of bug, or something like that. The really great thing is that in the onset of the late 90’s there was this sudden explosion of new thinking about how to do software development. Really none of the individual pieces of it were entirely new, but there were lots of really great techniques and if you put them together and kind of super-charged them and took them to the limit, you could actually get to a point where you’re attempting to change software continuously. Which is the polar opposite of where it had been, which is lock down change and don’t do anything else. You know, 18 month development cycles, and so on.

That was really the excitement. That, hey, this is actually the way that we can solve all these problems, and that feeling that, while software development was one of the most exciting and funnest things I’d ever discovered in my life, there was something desperately wrong with the way it was done. I think if you take that forward it’s really interesting because the early adopters of this kind of thing were the internet companies, Google and others like them, and the venture backed startups and all those folks in Silicon Valley. But this movement actually started in larger enterprises. They weren’t really ready to adopt that back then, but now it’s coming full circle.

Large enterprises around the world are rediscovering software development in a way, and adopting a lot of these techniques very, very rapidly. The transformation is happening at such a pace that having been in it since that origination it’s super exciting again. Right now to see the pace of change, and the rate of adoption of all of these things that we’ve been working on for quite a while.

Michal Lev-Ram: Michael?

Michael Dell: I started in my dorm room, and the initial observation and insight for the company was that I was sort of there at the beginning of the microprocessor age and going through junior high school and high school reading about microprocessors and floppy disk drives, then hard disk drives. To me it was all very exciting that an individual could have a computer. All right? Today that sounds pretty ridiculous. I tell my kids I started when dinosaurs roamed the Earth. They believe me.

I like to take things apart because you can’t know how they work unless you take them apart. I get all these computers and take them apart. IBM was, in the late 70’s and early 80’s, dominant in field of technology unlike any other firm has ever been dominant in any field, at any time, in any history, right? So when IBM decided to make the IBM personal computer that struck me as incredibly important. I got one of these things and took it apart, and what I found was that every part could be identified. There were no, sort of, black box parts, and, to the best I could tell, it was about five hundred dollars worth of parts that they were selling for three thousand dollars, which to my 18 year old mind seemed a bit like a criminal enterprise.

Then the other thing was, you go to the computer stores that were popping up all over the landscape, and they had been selling shoes or car stereos the week before, and they heard computers were going to be big, so they set up a store. They didn’t know anything about it, and that seemed to be kind of an additional crime. The idea was if you could make one of these things and sell it for seven hundred dollars you’d still have a very nice profit, and you wouldn’t need all the dealers or anything. That’s how the company started.

Michal Lev-Ram: So you dropped out of school?

Michael Dell: Dropped out of school. Started with a thousand dollars in my dorm room, and here we are today. It’s worked out pretty well so far.

Michal Lev-Ram: Hasn’t been too bad. So, as you guys look back on your first few years, I mean it’s interesting you said that the pieces were there, but people weren’t really putting them together for this new way of developing software. Do you think that you—and obviously you had signs, you just mentioned some of them—that this was potentially going to be something big? Were those signs enough for you guys to just kind of throw it all in and do it, or do you think that there’s some entrepreneurial dumbness that goes along with it also? Where you’re blind to what the actual risks are and you just go.

Michael Dell: I think it’s a combination of one, naïveté.

Michal Lev-Ram: That’s another word for it.

Michael Dell: If you’re naive you don’t actually know that there are risks. Okay, so that’s a good thing. And, of course, if you have nothing, you have nothing to lose, right? I’ve actually come to believe that most people don’t take on nearly the risks that they should, and are generally way too cautious, and so I think there’s a lot of unrealized human potential because people are scared and afraid to take risks.

Michal Lev-Ram: Rob?

Rob Mee: Yeah, I think it was kind a boiling a frog situation. I didn’t know I was in hot water, so I thought it was a nice bath. I was definitely naïve, and learned a lot. It just kept racking up and eventually it got bigger, we needed bigger offices, and the bank said, “Hey, that’s a pretty big office. What do you have for collateral?” And I’m like, “Well, what do you mean? We make money.”

They’re like, “No, no, no. How about your house?” I thought okay, sure, and so I put my house down as the collateral, and you end up taking risks like that as an entrepreneur. You kind of don’t think about it because it’s just a little bit more every time. But looking back I don’t think I would have done anything differently. I think they were good risks to take, and I think we were backing it up with a business that really made a lot of sense to me. Then I think by working with customers and clients, it makes sense to them. Right?

I think that’s one of the things about our business that’s interesting for us, and very rewarding, and it’s still true, which is that we work really closely with our customers to make them understand our business the way that we understand it and why it’s important for their business. And I think that’s highly rewarding to have that kind of high touch experience with customers still.

Photo by Wesley Verhoeve.
“It’s kind of a secret to our success that our customers constantly make us better by working with us, and we’re incredibly fortunate to be in that position.”
—Rob Mee

Michal Lev-Ram: You both talked about risk, and you probably have relatively high tolerances for risk. When you look at the people in this room, and the companies that they represent, some of the largest companies in the world, there’s a lot at stake here. What suggestions do you have about how they should approach risk when it comes to technology? Do you think that there is a healthy level of risk associated with what their mandates are?

Michael Dell: What I’m seeing inside business today is speed is become a major topic, and it is so because the slowest time of change relative to the future is right now. It’s only going to go faster, right? And it’s much easier to create new companies, and so all these companies are working to say, “Hey, how can we take market facing requirements that we’re learning about from our customers and our users, and turn them into competitive advantage and software faster?” What happens if you don’t do it?

Well, probably not something good. And so I think the speed element is driving a lot of what our customers are doing. What I’d also tell you is I think we, as companies, all learn a tremendous amount from our customers. I mean, you could have this idea that there’s some mad genius, you know, sitting off in a corner somewhere figuring everything out. I don’t think that’s the way it works. I think we listen, and we have tons of data, and all of our companies are able to take that information and, in some ways, intuit the answer based on the behaviors that our customers exhibit, some of it explicit, some of it implicit, and take all that and we create a better product, and a better service, and a better outcome based on the collective intelligence of everything that’s happening. I think that’s a far better model than trying to sit in a cave and come up with the answer, you know?

Rob Mee: I think to that point, and then not to mention risk in a minute, but something that we’ve always done with co-development with our clients is really pairing up with them in person and building software. And usually the notion of that is “hey, people are engaging with Pivotal to learn this more modern way of software development from us by pairing up with us,” but the reality is it’s a bi-directional relationship. And so, every time we engage with a new client they’re actually sharpening us, and teaching us through that bi-directional relationship. So we keep getting better by doing that. It’s one of the great driving secrets, it’s not an intentional secret, but it’s kind of a secret to our success that our customers constantly make us better by working with us, and we’re incredibly fortunate to be in that position.

I think the issue of risk is really interesting, because what a lot of people are looking at like, “Should I take a risk to do this? To do something in a new way?” And it feels risky to change the way you develop, or to change the way you do platforms, or to change the way you do security, especially since the new ways of doing these things, the new technologies we’re using, are generally optimized for continuous change.

They’re saying, “Okay, well we have to go faster, and we have to do continuous deployment.” That sounds scary, right? Because shouldn’t we be checking it and triple checking it for a few months, and then put it into production? What do you mean continuous deployment, right? That seems very risky, but it’s actually the reverse, which is not doing it, is where the risk lies. Working through all of the learning and potentially some of the pain and some of the breakage to get to the point where you can do things continuously gets you to a much safer place in terms of building software that is resilient, and available, and secure because you’re testing everything incrementally as you go. These thousands of small changes are really adding on top of each other incrementally and building along on the safety that you’ve been building. Instead of waiting and putting something huge into a production scenario where you’re not sure if it satisfies the users or it really works very well, and you’re actually saying, “La, la, la, just please work.” That old style of development.

It takes a sort of a leap of faith to go through and get to the point where you can do continuous development security, I think. That sort of with technology and would process how you get to this continuous deployment kind of a model, which is actually safer. Approaches to security are going through a similar evolution where traditionally IT groups who said, “Don’t change anything. Lock it all down.” Right? “This server’s been running for four years continuously, with no down time. How about that? We haven’t changed a thing on it.”

I think these days, if you think about a server that’s been running steadily without ever having gone down for four years, you wonder what’s kind of been incubating in it, and you’re looking for Sigourney Weaver. Something’s gonna pop out that you really don’t like. The way that you really approach that now is you say, “Let’s take that virtual machine, and kill it all the time, and then restart it at a known good state and have the system rebalance all the applications and no one knows that it happened.” But what you’ve done is you’ve wiped it clean, so you can’t get any infection that lives there. That’s just one example of how the approach to security is evolving toward a continuous change in order to foil threats and have a continuously clean environment.

I think a lot of this, you know the metaphor or the trend to go toward continuous change feels risky, but it’s the reverse. Not doing it is the risky part.

Photo by Wesley Verhoeve.
“What a lot of the large organizations are lacking is the substrate in terms of platform, process, culture that allows them to connect all of their different applications, and all of their capabilities in a consistent way with APIs, with processes that allow them to do continuous evolution in a culture that embraces that change.”
—Rob Mee

Michal Lev-Ram: Okay, I have more questions for you guys, but the first person to ask a question from the audience gets a free Dell. No—I’m kidding. Or a year of storage virtualization software. What would it be today?

Michael Dell: All of the above.

Michal Lev-Ram: Yes. Seriously, any questions? Yes.

Audience #1: We started the conversation off looking backward, but if you could look into a crystal ball thirty years out what would you see for you and for the company?

Rob Mee: Should I start with that?

Michael Dell: Yeah, go ahead.

Michal Lev-Ram: How many years out did you say?

Rob Mee: Thirty.

Michal Lev-Ram: I was going to ask them five to ten, but let’s go with thirty.

Rob Mee: No, no one ever asks for thirty, that’s why this is such an interesting question. It’s always five.

Michael Dell: Yeah, I’m looking forward to hearing the answer.

Rob Mee: Because it’s really been about thirty years since I first started meddling around with software, and the funny thing is the more things change the more they stay the same. I’m going to think I’m going to give a bit of a counter intuitive response, but you’ve seen 2001 A Space Odyssey, right? And you’ve got the HAL 9000, which by the way in the story was supposedly activated in 1992—so that’s 1992 technology. Well, in 1968-69 how did they get to what should HAL be? Well, talk to the folks at MIT and all the AI researchers and said, “Well, what are computers going to be like at this time?” Okay, well here’s what we project they’re going to be like.

And obviously Siri is not HAL, right? Now, we’re 25 to 30 years after HAL was supposedly first brought online. So, I think people get really, really optimistic about what’s going to happen over that time frame. It doesn’t happen in the way that people think. I think there are threads and paths that go off, and software technology advances really fast in ways that people don’t expect, but in all the ways they’re predicting it’s not nearly as fast as you would think.

So, for example, thirty years ago, you know I was programming in languages that people still program in now. People still program in assembly. People still program in C. All these things. These things haven’t really changed all that much in the way that people do things, and I have heard from way back in the early 90’s, and I still hear today, “In five years, no one’s going to program. It’s going to be blah.” Right? “It’s going to be completely different.” So I guess what I’m going to say is that in thirty years, I think there’s going to be a bunch of stuff that you still recognize, and maybe that’s the surprising prediction. I think there’s still going to be a whole bunch of stuff that you still recognize because I go back thirty years, and I’m like, “Yup. Most of it’s the same.” Then there’s some really interesting new stuff.

Answering without answering.

Michal Lev-Ram: So, it’s pretty much going to be like today.

Michael Dell: Yeah, I think well one is; Computers are machines. The human brain is an organism. Those are different things. So, maybe, you know, 15, 20 years from now we’ll have computers that have more power than 10 billion brains in terms of computational power, but it’s still not an organism. It does different things. What we’ll be able to do with that? I don’t really know.

What’s always interesting about technology are the combinatorial inventions and the permutations where people say, “Okay, now you have this, and this, and this. You put this together, now you can do this other thing.” Those are really more business model inventions on top of the technology, but certainly the pace of improvement in all the underlying technologies is not going to slow down at all. It’s only going to accelerate, and that’s super exciting, super interesting. I think will generally yield tremendously positive outcomes for humans.

There’s all sorts of scary scenarios and science fiction movies, and that sort of thing, but if you look at the actual results they’ve been way, way more positive for humans than anyone would have predicted. I think that’ll continue.

Michal Lev-Ram: As you guys look at what’s going on now with younger companies, smaller companies in Silicon Valley, and Austin, and other tech hubs, what do you find exciting and potentially game changing out there? It can be an enterprise, consumer, whatever areas you see that are exciting to you. Or is there nothing?

Rob Mee: No. I’m generally a little bit of an AI skeptic, I think, compared to a lot of folks in my field.

Michal Lev-Ram: Like Michael?

Rob Mee: Yeah. Maybe.

Michal Lev-Ram: Well, we can get to that.

Rob Mee: I’m not sure. I’m not sure where we fall on the various… but it doesn’t mean that I don’t think a lot of the AI technology isn’t extraordinarily useful. I do, I just don’t necessarily think that they’re suddenly going to rise up against us, or eclipse humans, or have volition, or what have you. I think they’re very useful. But I think one of the things that’s going on right now— AI, for a lot of large organizations, is used in fairly narrow applications and people can apply it to a particular type of problem. What a lot of the large organizations are lacking is the substrate in terms of platform, process, culture that allows them to connect all of their different applications, and all of their capabilities in a consistent way with APIs, with processes that allow them to do continuous evolution in a culture that embraces that change.

When you’ve got all of that there, now you start weaving in the really interesting technologies onto that, and you can start taking advantage of all of these things that express their state and their intention to the artificial intelligence technology, and you can take what insights you glean from that technology and you can weave it back into what you’re developing on a continuous basis. That gets really exciting. Then you can apply these technologies across a large organization in ways that you would never think because right now most of those applications of it are very silent, very isolated.

Michael Dell: I think the artificial narrow intelligence is much easier, and more promising, in the near term. It’s much harder to do artificial general intelligence. I still see all these new businesses being created as people combine things together. That, to me, is super interesting, and, of course, what’s happening is IT, as some of our customers have told us, has gone from a chore to core. You can’t actually do anything without IT, you can’t design a product, you can’t have a customer relationship, buy, sell, whatever, big company, small company.

The rule of what we all do collectively, together, has gotten way, way more important, and IT has sort of broken out of IT. Right? Then when you think about this whole topic of digital, it becomes core to how an organization thinks about its strategy, how it evolves, how it serves its customers, and so it’s center stage in the evolution of strategic thinking of a business. That’s a big deal, whether you’re a big giant company, or a small company, and every thing that has the potential for some super big change also can be over-hyped, and there will be mistakes along the way.

Photo by Wesley Verhoeve.
“When you think about this whole topic of digital, it becomes core to how an organization thinks about its strategy, how it evolves, how it serves its customers, and so it’s center stage in the evolution of strategic thinking of a business.”
—Michael Dell

Michal Lev-Ram: Other questions?

Audience #2: Could you please elaborate a bit about ecosystems, because I think that corporations tend to employ people in order to ensure that they have the entire training completely in their hands. I think now pure software and pure products are focusing instead on marketplaces and combining solutions—will this continue in your eyes?

Rob Mee: Yeah, I think so very much. In being a platform company, essentially, we’re absolutely reliant on having a vibrant ecosystem, and I think discussing recently as a company that’s building up to be a platform company if you take us back four and a half years ago, it’s a chicken and an egg situation. You really need an ecosystem and a lot of partners, and systems integrators, and ISV’s to come in and make your ecosystem rich, but until you have momentum it’s hard for them to come in and commit to that. We really, I think, spent a lot of effort working for the first few years and eventually gave live birth to a chicken, and now the eggs are appearing. It’s a bit of a strained metaphor, but …

Michal Lev-Ram: That’s beautiful.

Rob Mee: But now the eggs are all hatching, and …

Michal Lev-Ram: Are you just going to keep going with that?

Rob Mee: Yeah, I’m not going to let it go. But I think we’ve recognized the importance of that ecosystem for us as a platform company, and I think that in other open source communities—and then we see that the really interesting thing is we’re very open source based ourselves. You see these communities start to merge, and have convergent evolution, and swap bits of DNA, and it gets really interesting. So, I think in an open sourced world the ecosystem just becomes doubly important.

Michael Dell: Back to the speed, and it’s easier to connect things together so you can do it in a community ecosystem fashion. So, yes.

Michal Lev-Ram: Other questions?

Audience #3: So how long do you think that it takes for a larger company to go and move from where they are into this modern world that we’re talking about? Because it’s a massive challenge for a big company to move from that spot to this spot.

Rob Mee: Yeah, it is. Well we’ve seen some companies that are doing it. Instead of moving from zero to literally fifty percent of their development done an entirely new way with the new technologies, really, essentially, a transformed organizational structure. So, a lot of the roles that they previously had are gone or have morphed into new roles in an organization, which actually leads you to the how of that question. Sure, it takes several years at least to get to a say fifty percent level, what we’ve said, but in order to do that you might need to create a new organization. Not try to simply modify the old organization, or try to infect it with a virus and hope that it spreads, but you actually have to create a new organization and start pulling things in both from the outside and from the old organization.

Building that up until that’s the new normal, and that’s one of the ways that you can actually achieve that transformation of all of the mechanisms of software development, and the way that IT is structured, and the way that the product groups are put together that have adopted all of these new things; agile and lean and DevOps, and design thinking, and all of these things put together in an organization with new roles.

“If you have a big company, there are a lot of people running around that tell you, you can’t do stuff.”
—Michael Dell

Michal Lev-Ram: Have you guys seen some… could you share some best practices out of what you’ve seen in companies that have made that transition relatively fast? Especially as it pertains to the cultural change and organizational changes, like you mentioned, side of things. Anything that you see that seems to work universally?

Rob Mee: I’m not sure if I could claim universality, but we’ve definitely seen a lot of really good patterns. In addition to potentially creating an organization and building that up, I think there’s also—really from the start—adopting a beginners’ mindset. Instead of saying, “Okay, I’m going to let everything go, and work with other people that have done this before, and then based on that start propagating that.” Often times when we work with people we say, “Hey, work with us at first to do things in a new way, and learn how to use this new technology platform and so forth.” And then we step back and go out of the picture, and hopefully that group within the company actually acts as the change agent now.

They go on a self sustaining transformation, and they then have to ask the rest of the organization to suspend their disbelief and come in and try things with a beginners mind, and do that again. That’s kind of a pattern that you see over and over again is imploring people to check their ego at the door, forget everything they’ve known just for a while. Right? Try something in an entirely different new way, and once you’ve internalized it you can sort of then synthesize that with everything else you know into something bigger, and that’s your own. That’s another pattern that I think we’ve seen a lot.

Michal Lev-Ram: Any thoughts, Michael?

Michael Dell: Yeah, what I’ve seen and what we’ve done inside our company that’s worked well—very similar to what Rob is saying—is you spin up two, three, four teams as kind of separate thing off to the side. You get it going, and then you go back to the main organization, you kind of call for volunteers. “Hey, who wants to be part of this new way of doing things?” You’ll get maybe 25% or a third, we’ll say yeah, sign me up. Then you gradually start moving the whole thing that way, and that’s a pretty tried and true method to get it going.

Rob Mee: I’ll elaborate on that, because I’ve seen the way that Michael talks about it to his teams, which is really powerful. It’s a really strong executive statement of, “Hey, we’re going to do this new thing, and we’re going to do it wholeheartedly. We’re really going to try it, and you have my full backing to do it.” That is another pattern that’s really critical is that support, really strong direction from the top. If you go back to when we were working with Silicon Valley startups for years, and some of them really took off, and some of them kind of worked with us and did things in this new way and then sort of entropy set in and they started going down.

Usually what happened was deadlines would come up, and scary things would happen, and maybe the CEO would say, “I know you’re writing all these tests, you’re doing all this automation, but we really have to ship next Tuesday. Can you stop writing tests until after you get in … I swear I’ll give you more time after. Heck, I’ll give you extra time after we ship to write tests.” That kind of fear reaction, it just destroys teams faster than you can imagine. I think from a leadership perspective you gotta come in and make that really strong statement of support when you’re trying to move on to something pretty radically different.

Michal Lev-Ram: Okay. Very quick question?

Audience #4: On small teams hiring people and getting prototypes working. How many rules do you allow them to break? Do you let them to go completely off the vantage for how they hire, how they bring talent on, how they do contract negotiations, or do you draw a line somewhere?

Michael Dell: Well, we adopted the Pivotal methodology in terms of hiring and screening talent because I observed a different level of talent inside the Labs than folks that had a lot of three letter acronyms on their resume, but couldn’t actually do the work. That was pretty much a copy exact of the Pivotal model. Every company has its own unique elements. At our company if we want to get something done we tell them, just get it done, if anybody gets in your way just shoot them. We don’t actually shoot people, but we are from Texas so we use colorful analogies like that. Just make sure everybody knows we’re not actually shooting people.

Michal Lev-Ram: Figuratively.

Michael Dell: But don’t let anything stand in your way. Look, if you’re going to do something new you have to be willing to break things, and sort of make stuff happen. If you have a big company there are a lot of people running around that tell you, you can’t do stuff.

Michal Lev-Ram: I think we’re out of time. I came away with a lot of chicken and egg analogy, and thank you for that lovely last note, but thank you guys so much for your time.

This video was filmed at the Built to Adapt conference in Sausalito, California. The transcript was edited for clarity.


30 Years of Tech Transitions was originally published in Built to Adapt on Medium, where people are continuing the conversation by highlighting and responding to this story.

Previous
Getting Started with Spring Cloud Services on Pivotal Application Service
Getting Started with Spring Cloud Services on Pivotal Application Service

This post explains how to set up Spring Cloud Services and Pivotal Application Service. This combination is...

Next
Steeltoe Turns 2.0, Adds Support for ASP.NET Core 2.0, CredHub, and a SQL Server Connector
Steeltoe Turns 2.0, Adds Support for ASP.NET Core 2.0, CredHub, and a SQL Server Connector

Steeltoe is a toolkit to help .NET developers build apps for Cloud Foundry using established microservices ...