S 1 Ep 3 - Grant Rochelle: HP had CAD. Who Knew?

We trace a living history of CAD through Grant Rochelle’s 30 year career: HP’s software origins, CoCreate’s direct modeling, PTC’s Creo synthesis, Autodesk’s PLM tensions, and Siemens’ push into subscriptions and cloud. The throughline is simple: data, process, and change control decide winners.

• HP’s early CAD stack and life inside a hardware-first culture
• Direct modeling’s value for late-stage change on plastic parts
• PTC’s integration strategy and Creo’s blend of direct and parametric
• NX as a model for low-friction product consolidation
• Why PLM “stickiness” locks in processes and data
• Autodesk’s volume DNA versus enterprise PLM needs
• Siemens’ subscription shift and careful customer on-ramp
• The enduring role of release control, configurations, and traceability

Transcript

Welcome And Dedication To Weisberg

Roopinder 0:11

v Hello and welcome to the History of CAD podcast. My name is Roopinder Tara and I will be your host. On this podcast, we will explore the untold stories behind the tools that shape modern engineering and design. Each episode features conversations with the pioneers of computer-aided design, the innovators who built the software, and those who worked alongside them or observed or reviewed their work. We dedicate this podcast to the late David Weisberg, who wrote a 650-page treatise, The Engineering Design Revolution, which we will refer to often.

Meet Grant Rochelle

Roopinder 0:47

My pleasure to introduce Grant Rochelle. I've known Grant for many years, primarily from his days at Autodesk. Grant invited me to ride the MS-150, 150-mile-long charity ride for muscular dystrophy through the Wlliamette Valley in Oregon. Ride 100 miles with someone is to make a lifelong friend, I say.

Stumbling Into Computer‑Aided Engineering

Roopinder 1:08

Grant earned a degree in computer-aided engineering in 1991 from Staffordshire University in the UK. He worked for Autodesk the longest, but cut his teeth at HP and after Autodesk worked at Siemens. So he has been a witness and participant in the history of CAD.

Roopinder 1:24

I always associated you with Autodesk because that's how I knew you. But your history goes way further than you went, you were with HP. Can you tell me about your start with CAD? And that was with HP, was it called EME10?

Grant 1:35

Yeah, back in the late 80s, early 90s. I got connected with Hewlett-Packard through in German they call it a practicum. So it's basically an internship. And I guess really for some people, the interesting thing. Back in the 80s, so I actually ended up on a class at a course at college called Computer Aided Engineering, which I didn't even know existed. I showed up for mechanical engineering and on the first day sat down next to another young guy and we got to talking. And I asked him, Are you here to do mechanical engineering? And he said, Oh no, I'm doing computer aid engineering. And I looked at him and went, What's that? And he said, It's like what you're doing. We don't have to do thermodynamics and all that other boring stuff. He said, and there's a bunch of computer stuff in it. That sounded a bit more interesting. So I walked up to the front of the class or the registration room and said, Can I change? And they're like, Oh yeah, it's not too late. And that is literally how my path started.

Roopinder 2:25

Wow. So had you now, what was your experience with computers before? Did you have a home computer or something? What made you interested?

Grant 2:30

I saw a Sinclair 81 in my cousin's house. That was my experience with computers.

Roopinder 2:36

Ah, okay. Yeah.

Grant 2:37

And that was it. So I'd never actually seen one, not for real.

Roopinder 2:40

And now, did you say you had already graduated or you were in a company?

First Labs: VAX, Apollo, And Wireframes

Grant 2:43

No, no, I started at [...] Polytechnic in 1987. And so when I started, we had access to a VAX mainframe for doing machine programming. And there was one class elsewhere on the college which we never used, but it had very early Macs in it for doing documentation. And I think there was one PC in one of the engineering labs that ran a very early version of AutoCAD. So that was what it was like to begin with. And then sometime during that four years, they put a new computer lab in which was equipped with HP Apollo workstations. My first exposure to Unix. It was running some kind of primitive CAD program using primitives from a company, or I think the company was called Pafec, P-A-F-E-C. And had really strange names for the products. There was a 2D product and a 3D product. I think the 3D one was called Boxer or something like that, mainly because it could just make boxes from what I remember. We were learning the basics of computer-based drawing and modeling. A lot of that was primitive-based Boolean additions and subtractions. There was nothing really sophisticated at that point. But it was 3D or whatever. Yeah, and it was all wireframe. It was just like insane trying to figure out what was going on the screen. There was no rendering, nothing like that. And then for my first internship, I went to work for a

HP Internships And Early HP CAD

Grant 4:00

regular engineering company called Faggle in the Basque country in northern Spain. So I worked in a drawing office, which cemented my view that I didn't want to be an engineer because I was bored. This company made big heavy rolling machines for sheet steel and aluminum to make washing machines and things like that. I love engineering. I just realized pretty quickly, I think I've got what it takes and it doesn't interest me. But what we were doing at college really interested me because we were doing robotics and automation and the CAD stuff and some very basic CAE. So we were super fortunate. It was a small polytechnic, but they had associations with BMW, Hewlett-Packard, and a few other companies. I met with a representative from HP. I got picked for my internship and spent six months working in the learning products group doing documentation work for the mechanical design division. Nobody had any idea HP wrote software. And so back then, yeah, they had a drafting package called the 3D solids modeling package called ME30. They had data management called HP DMS, document management software or something like that. That was pretty much it. But they developed a sheet metal modeling module for the 3D solids modeling.

From ME30 To Solid Designer

Grant 5:11

Once I went back to work full-time, which was after I graduated in 91, they'd been busy developing a next generation 3D solids modeling tool called Solid Designer and a next generation data management product called Work Manager. The early 90s were beginning to heat up a little bit as far as the market for software.

Roopinder 5:29

While you were working on these early programs, I think of the graphics make me think of an Etchosketch, but primitive. Did you have to change your skills instead of using a slide rule? You were using coding, probably. You had to learn programming.

Grant 5:42

There was a little bit of that for me. If I go back to the college experience, there was some of that. Actually, so with ME30, it really was set up a bit like AutoCAD. The user could do almost anything through the user interface with Pix. We were using a tablet and stylus.

Roopinder 5:56

The menu was laid out on a tablet, which is how AutoCAD used to work as well.

Grant 6:00

The first time I used AutoCAD, it was already like a mouse and keyboard.

Roopinder 6:04

Before that, everyone had tablets and menus, elaborate menus, big tablets, lots of cascading menus, but you could drive it through a menu.

Grant 6:12

That was the graphical user interface on the screen. You just couldn't pick from that. We didn't have mouse.

Roopinder 6:17

HP had that interface. Did it eventually develop into it?

Grant 6:20

More modern UI? Yeah. I was reading up about it when I

Motif, Early GUIs, And The First Web

Grant 6:23

posted that thing you saw about this old stuff I wrote. During that intervening period between my internship and when I went back, they developed motif for the Unix workstation, which was early Windows for Unix. Once things progressed to this next gen product, Solid Designer, it had a proper graphical user interface. It was Windows-based. Bearing in mind it was still running on a Unix workstation and a much easier user experience. But they both offered, just like with AutoCAD, so ME30, the first generation product and Solid Designer, a way to program. You could customize the you could make it do clever things that weren't accessible. You could automate a lot of tasks.

Roopinder 7:00

So that comeback. Motif is now an architectural program. It's started by one of the ex-CEO of Autodesk, Amar Hanspal. His group has left to create a program called Motif. So interesting how history repeats itself.

Grant 7:13

I still have these very fond memories, Roopinder. It's only going to mean things to all guys and gals like us who've been around this long. It makes me smile. My daughter's 21. She's never known anything but technology the way it is today. I still remember sitting with a guy who was the systems administrator for the mechanical design division, who was this lovely American guy called Gary Stowe. Bless him. And he showed me he was like the guru. Anything that was going on, he knew about it. If there'd been a dark web back then, he would have been in it, finding things out all the time. Gary showed me the very first Motif on his workstation. No one else where I worked in Berlingen Germany had seen it. And Gary showed me, he said, This is incredible, dude. He goes, Look, no more. Let's literally write Unix commands at a prompt to make it do things. And then he said in this really

Life Inside HP: Software In A Hardware Giant

Grant 8:00

cool thing called the World Wide Web. I'm like, what? And then he started showing me he was putting URLs in manually and taking me to pages of writing back then. It's just insane.

Roopinder 8:10

So that was how you first learned of the WWW. Amazing. So what happened to all of HP's products? They became what? They HP sold that part of the business, correct?

Grant 8:21

Yeah, I went through an interesting evolution. This CEO, Tilman Schad, and the German management. Yeah, we were on a course. It was very difficult to be a software business inside of a gigantic hardware company. A very tangible way that materialized for me was after three years working in Germany, I took a new job. I stayed with the mechanical design division, but I moved up to the UK with my girlfriend at the time, with Nancy. And we were based inside a HP sales office in Bracknell in the UK. And we had just ported a solid designer, work manager to silicon graphics because it was important, if you were a software vendor, to be on multiple platforms or at least two. And so I remember me and another colleague showing up one day in the HP sales office, which most of the people working there were selling workstations and other hardware with some SGI boxes. There was a lot of arguing going on in the office about what was seriously, not with me and my colleague, like with our boss and with some of these sales managers, about what the hell was going on and why were we showing competitive hardware inside of HP sales office.

Roopinder 9:24

You should have been showing HP software on HP workstations. That was the company line.

Grant 9:29

Instead, from the perspective of HP salespeople who were bringing

CoCreate Spinout And Dot‑Com Era

Grant 9:33

clients into our offices, bear in mind these were not on show. This was just us working at our desks, but it rankled. Clearly, this was not okay. Tillman and the management team, I'm not saying it's because of that, but they recognized that they were going to need more independence from the mothership. And so in 2000, they initiated a management buyout. So we became a wholly owned subsidiary. HP still had an interest in it. I don't know how that was structured. But around about 2000, we became a wholly owned subsidiary with a new name, CoCreate Software.

Roopinder 10:04

Co-Create. Based in Sindelfingen?

Grant 10:06

Yeah, Sindelfingen, Germany. So we yeah, we moved out of the HP buildings, at least in in Europe and in Germany, and started on this path, which, bearing in mind, again, if you think about the time in 2000, this was the sort of the beginning, or was sometimes in that dot-com bubble. Everyone was talking about was we're doing a management buyout, and then we're gonna we're gonna be more successful, and then we're gonna IPO. So that was the plan. And we were all given founder shares in the company and things like this.

Roopinder 10:34

This was during so this was part of the dot-com craze. Everybody wanted to be a known as an internet company and cash out, right?

Grant 10:40

That's right, from a timing point of view. I joined in 91. Between 91 and 94, I was in Germany, 94 to 98, I was in England. In 98, my wife and I moved to Colorado. Not long after we got here, we became a wholly owned subsidiary. Ultimately, I think it might have been two years after I left CoCreate

PTC Acquisition And Direct Modeling’s Value

Grant 11:00

and joined Autodesk. So maybe around 2007-ish, 2008, CoCreate got acquired by PTC. And that was not a shock to anybody because for several years before that, and before I left, we got a new CEO called Bill Gascoigne, who I think was ex-SDRC, I don't remember exactly his background. He'd made it very clear his mission was to get the company sold because we were never going to IPO. I don't know if I'm talking out of turn. It's a long time ago, so I think it's fine. But it was very much out in the open inside the company that at some point we were going to get acquired. For me personally, I didn't want to sit around and wait for that to happen because I thought there could be some negative consequences for some of us. And I was in marketing at that point, which is quite often the first group in line for reductions because you've got duplication of roles and things like that.

Roopinder 11:46

So normally what happens when somebody gets a company ready for sale, they try to make it look as lean as possible, they try to make it as profitable as possible, R&D kind of just goes away. You're just putting out in the market.

Grant 11:58

And there was definitely a focus on the profitability part and not a lot going on in terms of development. And we were struggling because SolidWorks was out, Inventor was out.

Roopinder 12:07

Competition was really heating up. Oh, yeah. CAD market.

Grant 12:10

Yeah, in the mid-to-those early to mid-2000s. So yeah, there was a lot going on. And really, our business was tied to pretty large customers. We did have some resellers around the world, and our biggest customer was Hewlett Packard, not surprisingly. They developed the software, or we developed the software, and it was in use for designing all of the printers, all of the workstations, like any kind of hardware device at that time was being designed with CoCreate software.

Roopinder 12:35

It reminds me of

Creo’s Blend Of Direct And Parametric

Roopinder 12:36

what Siemens actually does. Siemens software is used quite a bit, if not exclusively, by all of its divisions for its production and manufacture of products. They used to eat their own dog food, as people say. So that was happening at HP. Now, didn't it eventually become acquired by PTC?

Grant 12:52

So it was acquired by PTC. My perspective from the outside, looking at what was happening over the next several years and still being in touch with some former colleagues, they did a pretty good job of figuring out what they were going to do with these multiple CAD offerings. So that this was what was different about Solid Designer was it was not a parametric modeler. It was a freeform modeler, which was why the printer division at HP valued the way it allowed their engineers to operate. In that business, at least at that time, they were making last minute changes all the way up to the point where they were releasing to production. They had a lot of injection moldy plastic parts. The feedback loop between engineering and factoring when they were prototyping, they were having to make lots of little changes, which in a parametric model are really hard to do when the design is 80-90% locked in.

Roopinder 13:40

Oh yeah, you'll break it.

Grant 13:41

Correct. Especially for plastic parts with all the reinforcing ribs and features and all the tapers required to pop it out of a mold. The designers at HP loved Solid Designer for its ability to do those kinds of things.

Roopinder 13:52

What you call freeform modeling. They came to be known

Lessons From UG, SDRC, And NX

Roopinder 13:56

as direct modeling.

Grant 13:57

Or dynamic modeling. So you could pretty much do what you want when you wanted. But then they also fought with a little bit because that those are just the shells on these products. Everything that was inside them was just like any other product. There was a chassis, a frame, there were PCBs, there's a lot of mechanical parts where actually parametrics can be really helpful, right? Because there's not a lot that changes and you make thousands of these things. So I think the evolution of Creo was PTC figuring out what were the best aspects of this ability to do what you wanted when you wanted, and bring them together with the ability to regiment how you design things in a parametric form.

Roopinder 14:34

I have to interject a little bit here because I remember that time, and we as press were introduced to Creo. And I pressed PTC, they didn't take this very well. I said, You had Co-Create and you have Pro Engineer. Why don't you make a product called Procreate? And they were like, oh stupid journalists. We can't do that. We're a respectable company. Fast forward to now, there's actually a product called Procreate. And nobody thinks much of it. It's used not for CAD, of course. I thought, oh man, I thought of that. This episode of The History of CAD is sponsored by Shaper3D, which could quite possibly be the future of CAD. I first saw Shaper 3D being demonstrated in San Francisco in 2018. There was István Csanády on his first trip to the US from his native Hungary. His startup Budapest was going to show a group full of CAD insiders

Stickiness, PLM, And Switching Costs

Roopinder 15:24

how to do CAD. But where was his computer? He stood in front of us with an iPad in one hand, an Apple Pencil in the other. He proceeded to create a solid model. And where I stood, I could see no UI. He had no keyboard, no last. It was based on Parasolid. What Shaper3D finally going to give engineers a natural way to sketch? Finally able to capture ideas as easily as drawing on a notepad. This was missing from every CAD program. Shaper has grown since those early years. The easiest to use MCAD program ever. Check it out. S-H-A-P-R 3D.

Roopinder 16:08

See you're ahead of your time. Did that replace Pro Engineer for a PTC.

Grant 16:13

I believe so, yeah. At some point, but I think there was a long on-ramp for Creo. We had these conversations inside of Autodesk and also inside of Siemens as well. If you're going to do an acquisition of a competing technology, first of all, you've got to figure out is it the right thing to do? What is the net gain versus the net subtract of doing these things? You're also capturing an installed base of customers that are using that product that you're buying. Now, from the outside in and not knowing all the details and really how it went, I think the company that engineered a way to do that in the least offensive or least intrusive way, in a way that really embraced the customers and brought them over with them was when when UG acquired SDRC.

Roopinder 16:52

I-DEAS software.

Grant 16:54

Correct. So two products at the end of many years, what you ended up with was NX.

Roopinder 16:59

Yes.

Grant 17:00

I don't know how they got there. I've no clue what they did to get there with the products, but they ended up meeting somewhere.

Roopinder 17:06

They

Joining Autodesk: Vault And ProductStream

Roopinder 17:07

met without a lot of friction. There was not a lot of outcry. Ideas, people.

Grant 17:11

And I would imagine not a lot of leakage, which is a huge risk when you do these kinds of things.

Roopinder 17:16

By leakage, do you mean that people would migrate to something else?

Grant 17:19

Yeah, if you disenfranchise a significant portion of the install base of this company that you're acquiring, whatever projections you were putting in front of the people who were signing off on it were just going to go straight down the toilet. So I think, again, as an outsider at that time, it looked to me like they did a really good job. And for all the talk inside of competitors, when things like that were happening, I was in some other company and everyone would be like, Oh, we've got a great opportunity to go and capture some of these customers because they're going to be really upset. We never did. You might get a few. We would go in and agitate and try and inject fear and uncertainty and doubt into the mind of the customer. But it's part of their process, they're just not going to give it up.

Roopinder 17:57

Exactly. Yeah. What's a good word for that? People are upset in their ways. You heard that term go prying hard.

Grant 18:02

Yeah, but there's this term that we use on the vendor side of things, which is stickiness. Data management, PLM, is the ultimate sticky tool. It's very hard once you've committed. CAD, less, but still, you've either hired people who are trained in that tool or they've cross-trained and now they're using that tool. Your processes, everything that you know about how efficient you are at that point in time is based around the usage of that tool. Everything's revolving around it. And this is, I think this is one of the great ironies is I think still how underappreciated the role of these tool sets is in the big picture, even inside of these customers themselves. The focus gets put on a lot of other things like ERP and CRM, and those are all important too. But what goes in engineering and manufacturing as it relates to these tools and processes gets about this much mind share compared

Autodesk’s PLM Tensions And Volume DNA

Grant 18:50

to all these other things.

Roopinder 18:51

Yeah, absolutely. I think that's where all the action is. That's where certainly our interest is, and I think that's the part of the product. All the other things are just boring. PRP, my eyes just glaze over. So you went to, is that the time you said you went to Autodesk? Now then you got involved with uh inventor mechanical desktop was dead by then, correct?

Grant 20:27

Yeah, MDT was pretty much done. I joined in 2005. I was primarily hired because of my background with data management, coming from HP's Co-Create. At that time, Autodesk had launched Vault for work in progress management of CAD files, and we're on the point of within a few months of releasing ProductStream, which was this add-on which would allow you to do release management and workflows. They wanted more expertise in-house from people that had worked in that area. I joined someone I used to work with at HP, Jim White, a lifelong friend since the late 80s. He's here in the Portland area, and we still work together on and off now.

Roopinder 21:04

So who was heading the group? That was Buzz Cross.

Grant 21:06

It was Jim, actually. So Jim was heading up the product management team. Uh, and really it consisted of Jim and then Brian Roepke, who had co-founded True E Vault, which Autodesk had acquired and then commercialized it into Vault, and then basically included it with all of the CAD products, so at no cost. So it was in Inventor, AutoCAD, AutoCAD Mechanical, AutoCAD, Electrical, all of those kinds of things. And then there was me. It was very small. That's how we got started. And I was unaware at the time of the history of Autodesk that this was maybe their fourth attempt at making a go of it with data management.

Roopinder 21:38

From the media side, it always looked like Autodesk was trying to put down PDM, even though it was having some PDM. I think Carl Bass, even at one point, said the only people that have a PLM problems are PLM companies.

Why Enterprises Need Real PLM

Roopinder 21:54

By the way, it seemed to us like he was not giving PLM its proper due. It's that Autodesk wanted to be known just as a CAD company. How did that make your group feel? Things like that.

Grant 22:05

Yeah, there was a lot of tension internally between these competing viewpoints. On the one hand, from some parts of leadership, there was recognition that if you wanted to be successful in the market, you were gonna have to figure out a strategy for what to do with these thousands of CAD files that you were producing. Because without some form of data management, you were creating chaos and inefficiencies with duplicate parts and drawings and random part numbers, and all of these processes were being managed in spreadsheets.

Roopinder 22:34

You knew from your previous experience with large enterprises that this is how the big boys did it. The big boys had PLM. PLM was almost like more central to them than DAD was. Like this is what the rug that tied the room together. This was this was PLM.

Grant 22:49

And one of the things I would say is Buzz recognized that this was a challenge that needed to be figured out. I think what I heard was Carol, before Carl, Carol Bartz, was also a proponent of doing something about this. But it was like it's got to be done in the way that Autodesk does things. This is where it got really challenging because Autodesk is a volume business. Always has been, and as far as I'm concerned, always will be. But that is the DNA of the company. We used to say it's got to be something that can be shipped in a box, and obviously that doesn't exist and hasn't existed in a long time. And someone's got to be able to do it themselves or with the help of a reseller. And data management is not simple, right?

Siemens, Teamcenter, And The Full Stack

Grant 23:28

The rewards can be really big, the payoff is really big, but it takes work up front. You've got to have clean data. You can't just dump all your rubbish in there. You've got to think through how you're going to take these manual processes you're running today and somehow instantiate them into a software product with some rigorous process so that you actually get what it was you wanted and you can see an ROI. Because you used to hear all these horror stories right about poor ROI or no ROI from PLM projects. Endless services engagements costing a gazillion dollars and things like that. So there was always this push and pull inside of Autodesk because we didn't want to be a big service business. That was not the DNA of the company. But how were we going to get this kind of technology into the hands of customers in a way that they could do a lot of things themselves or with the help of a partner? There was certainly, I would say, probably some commentary coming from some parts of company leadership that weren't super helpful.

Roopinder 24:22

So large part, Autodesk lately, more lately than ever, has been pursuing large customers and enterprises. But would you say that lack of a solid PLM solution is what's keeping them from that?

Grant 24:35

If you say they have, obviously, I'll take that on face value. I don't know. Obviously, they have. I guess first of all, what I would say with my knowledge is the biggest customers that Autodesk has are not in design and manufacturing. They have gigantic construction customers, no doubt. They certainly have some large customers because there is a major account business focus inside of the sales organization

Culture Shift: Leaving Autodesk For Siemens

Grant 24:54

at it. I think there's two challenges. There is still this void. I think they would struggle to go in and tell a holistic story about how they can help a company, let's say it's a division of GE with their end-to-end problem set. Maybe they don't want to. Different sort of tap dances we would have to do when I was there was oh, we support best in breed. So we're the best CAD tool for you, and we can do this and we can do that. But yeah, the data management stuff, if you want to use Team Center or whatever, we can work well with those tools. I don't know if that's still how they handle an objection like that, but they have no offering. I think what's been going on for the last several years is they are trying to absorb some elements of information management into their underlying platform to make it as transparent as possible to the user, like what they're actually doing. And that'll get you so far, but it's not going to get you all of the way there. Because you're still going to need to release things, you need to do configurations, you need to do where use queries. There's all sorts of things that go on in these organizations that are day-to-day activities multiple times a day sometimes. I think it's really challenging. And having come from the Siemens side, and I don't have exposure to the PTC and Dassault, there is still a lot of hay to make in accounts where they have Autodesk CAD, but they need help in a broader sense because companies like Siemens PTC or Dassault Systèmes can go in and say, we can solve that for you. And once they put the it's like wrapping a BOA constrictor around something, if you buy in and you get Team Center, the next thing is they're going to be coming in and showing you how much

Driving Subscriptions And Cloud At Siemens

Grant 26:24

better it works with NX, much better with SIM, how they can leverage the Mendix platform so that you can do all your low-code stuff and get every last little widget you've ever wanted.

Roopinder 26:33

They've got the full enterprise toolbox, that's for sure. So what I was I know Brenda left before you did, I think. She went over from Autodesk to Siemens. And I saw you at I think I saw you my one visit. So you came over there for a while. It seemed like Siemens was welcoming people with experience from Autodesk. And conversely, Autodesk was not really valuing that same experience. Seemed happy to let that happen. I think maybe even forced it to happen. This is all speculation, no way of knowing. Where Siemens valued experience, and that's why they got people like you and Brenda.

Grant 27:09

Does that fit with your I think some of your observations align with my own? The way I would describe it is in terms of how these companies were evolving, they were in different places. When I left Autodesk in 2018, the culture of Autodesk was very welcoming of brand new employees who could come in with non-industry experience, different perspectives. So they know that it's not that they so they might be coming from software, but different software. It could have been enterprise software or something else. And I'll only speak now to obviously because my my my view was from within the design and manufacturing

Modernizing Siemens And Closing Reflections

Grant 27:44

group. Nothing to do with AEC. Okay, just to be clear.

Roopinder 27:54

Like I I think I'm an engineer. I think engineers are the most important people in the world, right? I think if you're an engineer and you go into software, that's good because you know how the software should behave. It should serve you, right? I think if you're going from software into a software company, you've got to learn engineering and design. And that's a much harder thing to learn. I don't even know if people can do it. I know Autodesk did it once. They were lucky to make AutoCAD because they were serving the needs of architects mostly. They managed to get it right, but I think that was good timing more than make it a good product. They happened to create a drafting program, but not because they were architects. But that window closed, I think. After that, you had to know the business. SolidWorks, for example, created its company from engineers. All the people from low to high were engineers. Later, they let in marketing people and software developers.

Grant 28:46

My observations were when new roles were created or people left for whatever reasons, there was a real focus on recruiting externally, and not and that's not to say that they weren't cultivating talent from within, but there were certainly a lot of new people, a lot of new people, and that was valued, like different perspectives, different inputs, etc. For me personally, the culture of the company had changed to a point where I didn't feel like it really resonated for me and I wanted to go do something different. Conversely, Siemens, actually, and the other thing I should say is Autodesk had completed its cloud and subscription transition. Particularly, let's call it the subscription transition. By 2018, it was That was a huge sea change. We'd spent three to four years going through the pain of that, come out the other side, and you could see the trajectory that the company was on. Whatever was going to support the rest of that trajectory, there was a desire to bring a lot of people from the outside that might support whatever this next future part held. At Siemens, Brenda had gone over, I went over. There was a few other, it wasn't like it was a huge exodus. When I got there, the culture was very different. I was immediately immersed in a world of working with people who had worked together for a long period of time. It was really refreshing. Under Tony's leadership, there was a really close bond between lots of people. They valued other people that were there, their expertise, their experience, but they were also recognizing that they also needed some sort of external, some different inputs to take them wherever that next part was going to be. Brenda, me, and one or two of the other people, we'd been obviously exposed a lot to the cloud and subscription transition. At that point in 2018, with few exceptions, nothing was running in the cloud at Siemens at DISW, digital industry software, outside of the former mentor business, so the IC and PCB businesses, there was no subscriptions. So I went over there and took on a role that was in pricing and packaging. That was that that was a big part of what I was responsible for together with my team. And then the other bit was MA. Basically, at Siemens, it was just a different mix. They were bringing in some external inputs, but part of that was, yeah, hey, we need to go and figure out some of these new things too. And so that was what I got involved in.

Roopinder 30:50

Bringing Siemens into this century, making them more modern. And Siemens seemed to value that. They were an old company, I think really old. They went to their 175th anniversary, right? 175, 175 years, and only the last bit of it had computers in it. They had discovered cloud. And here was Autodesk already recognizing cloud and database cat, and it was getting there. And also, I gotta say, looking back on it, a Looking at it from the outside, I've got to say they probably valued a lot of the marketing that Autodesk had. Autodesk was known as a marketing juggernaut, more than it was known for its technology. It was known as a marketing company. Brenda had that, yeah, company had that, and they were probably benefiting from that as well because they're an engineering manufacturing company predominantly.

Grant 31:38

And yeah, Brenda was in the process of retooling and re-upping what was going on from an outbound marketing perspective. So that was good. Within fairly short order, we were working on doing the research and putting proposals forward about how to bring about a broad subscription model. We worked in partnership with some great people inside of Siemens to get that done on the finance, legal, and other sides, almost mirroring the kind of path it had gone. I wasn't closely involved in it at Autodesk, but very aware of it at Siemens. There was a bunch of us working on that. Once the green light came from Siemens Corporate down this path, it happened really fast. It was a little weird, given it's Siemens and things typically don't happen quickly. There'd also been a leadership change at the Siemens Corporate level and a new leader for the digital industries group, which I think really helped because they were not cast from the same mold as some of the previous leaders. They were much more dynamic, very engaging. That stuff all got turned around in six to twelve months from first concept to we're announcing it, we're doing it. And it was done in increments because Tony and the leadership team were again very aware about not creating seismic shock in the installed base.

Roopinder 32:44

Yes. Tony is an engineer as well. And so that that's very good. A lot of things that did happen lately that I thought were quite uncharacteristic of Siemens, that they are become very modern, responsive to today's engineer. They got out of their dark suits and ties, for example. Oh my gosh, when did they stop wearing ties?

Roopinder 33:03

Great talking to you. Talk to you later. Bye-bye. This concludes this episode of the History of CAD podcast. Please join me in our next episode as we add to the rich history of design software. If you'd like to be a guest on the show or know someone who has stories to share, please let me know via email at roopinder@engtechnica.com.

Closing, Sponsor, And How To Get Involved

Roopinder 33:24

Companies can sponsor an episode of the History of CAD. Get the details on our website, www.ENGtechnica.com/ sponsorship.