S1 Ep 1 - Mike Payne, Sailing from PTC to SolidWorks and Beyond

Mike Payne, the celebrated serial founder behind PTC, SolidWorks and SpaceClaim, shares his journey through the evolution of computer-aided design from punch cards to cloud computing. His candid insights reveal how identifying technological shifts and understanding real engineering needs led to revolutionary CAD innovations that democratized design tools.

• Started career studying physics when computers were rare in undergraduate education
• Learned programming by necessity for a microwave concrete curing project
• First encountered CAD through PDMS at Prime Computer in the early 1980s
• Co-founded PTC with Sam Geisberg, pioneering parametric modeling using "feature recipes"
• Recognized the potential of Windows NT to disrupt workstation-based CAD
• Founded SolidWorks to bring affordable 3D CAD to Windows platforms
• Made key decision to license Parasolid geometry kernel after ACIS proved inadequate
• Developed SpaceClaim for direct modeling, enabling analysts to modify designs
• Created Kenesto to solve cloud-based CAD document management challenges
• Advocates hiring engineers and teaching them programming rather than vice versa
• Believes in evolution rather than revolution in technology development
• Maintains that cloud and desktop software should interoperate seamlessly

If you enjoyed this episode, subscribe to the History of CAD podcast where we explore the untold stories behind the tools that shape modern engineering and design.

Transcript

Introduction to Mike Payne

Roopinder 0:22

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 .

Roopinder 0:58

Our guest today needs no introduction to CAD insiders . Everybody in the design and software industry knows Mike Payne and everybody has a Mike Payne story . Mike Payne is a celebrated serial founder or co-founder in the CAD and PLM space . He founded or co-founded Parametric Technology Corporation , PTC in 1985 , which was a pioneer in parametric associative CAD modeling . In 1993 , he founded SolidWorks , helping bring about a revolution in 3D CAD for millions . He left and co-founded SpaceClaim , literally across the street from SOLIDWORKS . Passionate about cloud-based engineering collaboration , he then founded Kenesto , a cloud document management and workflow app , where he serves as CEO . We join our conversation with Mike as he talks about his love

Early Computing Experiences

Roopinder 1:57

of sailing .

Mike 1:58

I have a sailboat . It's 11 meters , which I'm sure you can translate as 35 feet . It has a five-foot draft and it's a cruiser . Ironically it's a French boat Beneteau .

Roopinder 2:13

Oh , okay , have you ever sailed down the waterway ? There's a waterway .

Mike 2:16

I haven't got that far . I think I've got to spar with the Elizabeth Islands , but there's plenty of places . Everybody calls me Mike .

Roopinder 2:25

Yeah , did the Americans do that ? When you moved here , they started calling you Mike .

Mike 2:29

It just seemed more friendly than Michael .

Roopinder 2:37

My sister calls me Michael , but her husband calls me Mike , so you moved here in 69 . Yeah , how did you know that ? Because of the Dave Weisberg's CAD history Back to days in the UK , you got your start in the physics of semiconductor and vacuum devices which was not just those valves , as they were called in the English vernacular .

Roopinder 3:14

He studied physics . By the way , you have that in common with Elon Musk . He got his degree in physics . He chose to study the first principles and from that he could understand engineering . That first principles is physics . Isn't that true in a way . Yeah , it is .

Mike 3:31

Look , I went to college when there were very few colleges that had computers in undergraduate school .

Roopinder 3:41

That's right . So what were engineers doing ? They had slide rules at that point , right .

Mike 3:45

Oh yeah , we had slide rules , so they're a very complex piece of equipment . Very simple to use , so long as you understood logarithms .

Roopinder 3:54

I think a lot of people say that it gave them a better understanding that numbers than calculators do Were calculators . Yet a thing Did engineers have the Hewlett-Packard calculator ?

Mike 4:06

No , that was long after . I didn't know I'd have to use the computer . I had a senior project about microwaves and seeing how concrete had cured by its dryness and the reflection of microwaves , and so it was these complex equations that needed to be solved , and so the best way was to figure out how to program a computer to do it . It had a time sharing system that was very different . You stood in line with your punch tape behind the other guy . When you punched it roll , it would fail and you'd not compile , and the reams of paper would come out of a line printer and you'd have to go back and stand in line again after you changed the holes in the paper . So you tried to get it right first time .

Roopinder 4:59

Apparently , so you wouldn't have to make that tape again or punch cards .

Mike 5:04

At that time , only places like Manchester , Errand , MIT had computers as well .

Roopinder 5:11

Yeah , you're bringing back memories . I still remember those punch cards clicking like a deck of cards .

Mike 5:17

That's right . If you were smart , you put sequence numbers in the right-hand column so you could sort them back in order when you draw them .

Roopinder 5:28

At what point were you first introduced to computer graphics that eventually became CAD Graphics ?

Mike 5:35

Didn't come for a long time . That was when I was on the board of CAD Center and we were selling PDMS as Prime Computer . We sold PDMS , which came from a British research center called CAD Center . It's produced on very crude graphics devices by what today's standard is pictures of piping and valves , which was probably the first real aspect of 3D modeling . That was useful . Oh , we

First Encounters with CAD Graphics

Mike 6:07

had some crazy devices . They're expensive and complicated . Tektronix was one of the suppliers and so that couldn't be the future . It was the present , but it couldn't be the future .

Roopinder 6:22

Now , did you realize that at the moment , or is this in retrospect that this was quite not the way to do ? Things Like this would never take off if you thought this was just crap .

Mike 6:32

No , it wasn't crap , because people were building nuclear power stations using this stuff where an oil refiner is , and oil rigs and all the things that needed . Piping graphics was not wonderful , that didn't mean you couldn't work with it . We're going back to the early 80s , when things were beginning to become commercialized People in colleges playing with surface models and things like including designing airplanes to Dassault Aviation . We're talking 40 , 50 years ago and so it all evolved over time , like everything . Which is another lesson , of course , is that there are no revolutions . It's all evolution . You can't just say tomorrow we're all going to do this . It evolved over trials and experimentation and then you moved on . Things got more memory , more speed , better pictures , and that was a total evolution . You can blame the media for giving that name , argue that what they made was a computer which they destroyed afterwards . That was to decode the ciphers . A lot of that stuff is on show in Bletchley Park . Computers have come a long way since they were in a room .

Roopinder 8:15

They were used for important things like bomb trajectories . You used them for curing of concrete . Yeah , that's important , but all the effort that it took to make a little line from point A to point B , somebody had to say we're just making pictures here .

Mike 8:30

But that was the machines like Computerv ision and Applicon, which I used and hated . But it was better than making ICs with tape that you could forget to remove , or the printed circuit board that got photographed and he sipped tape with a Mylar that you photographed and made chips out of .

Roopinder 8:54

At what point do you think somebody said , hey , we should be showing our objects on the computers .

Mike 9:00

One of the things that Prime had was the rights to sell and then had to develop Medusa , which was a 2D drafting which had some elements of 3D Much better drafting system than AutoCAD . But then AutoCAD had the advantage later of being on the IBM PC and everybody could buy cheap equipment and do drafting . And it only shows the volume guy wins in the end . Medusa was a much better system and so were others , but the volume guy won . And today Autodesk gets a huge amount of money every year from AutoCAD it's north of a billion , I think .

Roopinder 9:41

Because it's a democratic thing , it's the numbers All because they put it on the PC .

Mike 9:47

Even though Medusa was better , PTC was better everything was better Not to show about CADM , but that's another story .

Roopinder 9:54

You were a head of development at Prime .

Mike 9:57

Merged with Computer vision at the time when they were both going downhill , it made them go downhill faster .

Roopinder 10:05

I call that two rocks tying themselves together , hoping they'll float that would be like putting Sears and Kmart together . So what was sinking them ? Were they not popular , given that they were just too expensive ?

Mike 10:17

They were expensive , they were 2D , they really didn't have any 3D . They had some facts of wireframe , but wireframe is pretty useless for designing things that evolved before solids . And then there were attempts at solids , but the solids were . I've shown the boss that I can make a solid model , but you couldn't change it . You had to start from scratch . This was the fundamental difference with pro engineer , better known today as prio .

The Birth of PTC

Mike 10:52

It started out as an idea that sam started , the idea he showed me one morning after breakfast . Look , we can make a bracket with a hole in it . Dimension , the hole and the location of the hole . Change the size of the hole , change the location of the hole and , lo and behold , we have models that we could change that was Sam Geisberg , the founder of the PTC Parametric Technology Corporation .

Roopinder 11:23

Absolutely , did he recruit you for ?

Mike 11:26

Effectively . His brother had worked with me for me at Prime running the mechanical section where we had to improve Medusa . I had left Prime and I get this call from a guy with a very heavy Russian accent saying my name is Sam . You don't know me , but I'm doing this thing in Billerica , massachusetts . My brother said I should talk with you . Would you like to have breakfast ? So I went and had breakfast with Sam . That was his brother , Vladimir . Right , vladimir was the problem .

Roopinder 12:02

Yes , Okay , Sam doesn't sound so Russian , but Vladimir . The problem yes , Sam doesn't sound so Russian , but Vladimir , you had a clue .

Mike 12:08

Yes , but Vladimir didn't have such a thick accent . Anyway , he shows me this and I look at that . That's what we should have been doing at Prime , maybe we can do something with this . And he said would you join me ? But I don't have any money , but we're trying to raise money . And there was another guy , Danny Dean . That was the group that started Parametric Technology . The three of us .

Roopinder 12:32

This idea of jumping into a company , even though they were just startups and I don't even know if they were paying themselves at this point .

Mike 12:42

This happened again when you joined SolidWorks . It was really hard to raise money because there were venture guys , experts , and we would run into the experts . The experts weren't expert at all , they just thought they were experts and oh , that'll never work . Well , the fundamental difference between what everybody else was doing at that time we didn't change the geometry . We made a recipe to make the geometry and you replayed the recipe and that was the list of features that you wanted .

Mike 13:16

The computers weren't fast enough to do what HP was trying to do at the time with direct modeling , because that takes a lot of computer effort . The computers weren't that fast . We were there in the era of mini computers . What was that hp's program co-create ? They were doing that in germany and you didn't have the horsepower to do it . So we used mini computers . That was the first embodiment . There were two . We had it on a VAX , which you've never heard of , and we had it on an IBM PC . But it couldn't stay on the IBM PC because it was 16 bits which it soon ran out of address space , memory , space in general and everything .

Roopinder 14:03

I remember DEC because that was Ken Olson's company . Why didn't you go to Prime ? Didn't you run on Prime ?

Mike 14:09

We had access to it at Adage . As we started to develop , we decided we were not going to play anymore with remote graphics , but rather these emerging workstations like Sun and Silicon Graphics , the Micro VAX . We focused on those where the graphics was included with the CPU , which was a lot different than having a remote tectonics display . The cost was better , so it became affordable . I think the thing PTC did at the very beginning was to go to those workstations exclusively . The next 10 years was that way . Okay , from the time we started . So it wasn't the way . But there were 10 years that these workstations flourished , got faster , and that forced us into how we stored the data . Because it had to work on different architectures with different byte orders and bit orders , and the only way to really do that was to make it textual and let the operating system figure that out . So the data was stored in text form because there had to be different computer formats and one of the problems we were determined to avoid was things that computer vision had to grade . You had to move your data to the next version with some conversion , which was ridiculous for the user . So we tried to make the data work with the next version . Pro engineer . We had to do everything . We had to write double byte language stuff , but wasn't any . If we wanted to support Japanese characters , we had to come up with our own storage . We had to do our own graphics . There was no hoops or something .

Mike 16:07

Until later , after we did graphics , Carl Bass showed up and said I got this lovely graphic software . I said that's very nice . What do you want us to do with it ? And he said it's wonderful and you should buy it . And I said maybe I should show you what we've got . And he says where did you get that ? I said we do 3D modeling . We have to show it on the screen . We couldn't buy it from anybody , so we wrote it , oh , and it was pretty good . So , Carl , in a way , that was probably his first failure to get a sale .

Mike 16:41

This is Carl Bass , eventually CEO of Autodesk that was it Figure Software , which later became Autodesk , which later became TechSoft . On the way , Carl Bass stayed at one of the steps , the one called Autodesk . We were fortunate in a way that we were able to hire somebody that had worked at Prime with Vladimir and that Sam knew called Leonid Rice and it created the geometry in ProEngineer . The geometry in Creo today is one of the two best geometric modelers .

Roopinder 17:20

This is a code that became known as Granite .

Mike 17:23

The modeling inside is what Leonid created and the other best one is what's in GM . That Granite is the modeler inside . They never really sold it . They put that label on it , but they've never really sold it .

Roopinder 17:38

I was reading CAD History by Dave Weisberg . He mentions Steve Wolfe . I don't know if you remember Steve Wolfe .

Mike 17:44

course I know Steve Wolf .

Mike 17:46

He was the curmudgeon who called us the dark side . Later , when SolidWorks became part of the Dassault Systemes , you went to the dark side and never worked . Is that the fellow you mean ?

Roopinder 17:58

Yes , that's the fellow . He took PTC to task for giving demos that highlighted the benefits or the advantages of PTC without showing any of its deficiencies , and it was a blender , I believe .

Mike 18:14

I would only say I'd never , ever heard of a salesperson who would show their disadvantages they don't do that . Shortages , they don't do that . I remember a trip I made to , say , petersburg in russia and met some people that did design work and they said I said , look , I can't tell you about that because I'm not a salesman . I said that's why we want to talk to you , because in the soviet era we were used to talking to the people who made it . We didn't have salesmen . Engineers are very good sometimes at highlighting the flaws rather than the benefits , and the sales guy is supposed to sell the benefits . Steve Wolf was possibly right . It was better than anything else out there . He could have acknowledged that instead of saying what was wrong with it . Every piece of software has bugs , whenever Microsoft tells you . That's why they shut your computer down twice a week at night .

Roopinder 19:09

Yes , every software does have bugs , correct ? Weisberg also mentions one thing I wanted to ask you about . He said you left PTC in a fit of pique and that's not his words , that's mine but he had a management dispute . Those are his

Leaving PTC and Founding SolidWorks

Roopinder 19:24

words .

Mike 19:24

He never asked me . The companies got bigger and it was less tutorial and ruled by marketing and it was time to move along .

Roopinder 19:37

There wasn't a product company at that point .

Mike 19:40

Steve Wolske he was a businessman , Dick Harrison was the sales guy . But there's this embryo system from Microsoft called Windows NT , which they build as a 32-bit operating system . They gave a presentation about it . I realized the world was going to change . But at that time there was marketing deals with all of these workstation vendors and the end of the workstation vendors could be foretold simply by hearing about Microsoft and 32 bits .

Mike 20:19

And so I got a copy of that . I sent it to a single guy in Israel and he made . This was before I left PTC . He created the Pro ENGINEER running on Windows NT Before Microsoft had formally released it . It was full of bugs , the operating system , but nevertheless we made it work . And just one guy , it's just your Skunk Works , nobody knew about it . And there was this show called Autofact and I was asked could we do a sample of something on Windows NT ? And I said no , I got a whole product and so we would show it at Autofact and people got interested . And then the marketing team come along and say , no , you've got to show it on workstations that we sponsor . And that helped me realize that getting into Windows would not be easy , as it later turned out . So that was probably a realization that you don't get to innovate anymore .

Roopinder 21:27

It reminds me of what Steve Jobs once said if you don't cannibalize your own product , somebody else will .

Mike 21:34

I did some work with Barry Feidelman looking at companies for him . One of them was a group of five people who didn't want to talk to me , but Barry said they had to because they thought I was the opposition . In fact they wanted to do 3D modeling on Windows . I talked to Barry later and said I like the idea , but there's not enough to make it a minimally viable product . They're not asking for enough money , they're asking for too long a time and they don't understand what they got to do . Barry said that's why I won't fund them . This was called Winchester Design . Barry said I'd fund it if you joined it . I said I wasn't planning on that . Then I go away and I think about it . Then I called barry and said okay , I'll think about that .

Roopinder 22:30

I might do it this was before the days . You could just look them up on the on the Internet .

Mike 22:35

You had to find out who are these guys I had to go and see them and I had a prototype . But prototype had holes in it , to put it mildly . One of the things that had to work right in 3D modeling was to be able to do rounds , and it didn't do it very well . The original modeling in it was called ACIS and there were other component things . This was where it started getting interesting . We didn't make everything at SolidWorks . Things that we could buy , we bought . And again people said we were stupid because you'd never make them all work together . I made them all work together .

Mike 23:12

There was Bruce Morgan . He was in charge of Spatial and Bruce came and he said I want my money for ACIS . I said we haven't paid anybody else till we ship the product and we're not paying you till we ship the product . ACIS doesn't work for us . And he says what do you mean ? It doesn't work . I said you can't do half the things we need to do and we give you bug reports . Or Scott Harris gave him bug reports and you ignore us . Okay , so it doesn't work . And he says but you've still got to pay me . I said we're not going to pay you .

Mike 23:48

I had already got a DLL for the geometry , but could replace one with another and the other was Parasolid and I showed Bruce . I said let me show you what you can't do . So I showed him doing rounds but it still worked perfect but was quite good in Parasolid and some other geometry but he couldn't do . Bruce says which version of ACIS is that ? I said I never said it was a version of ACIS . He said I still want my money . I said you ask for it one more time , I'll give you your software back .

Roopinder 24:23

He never shipped SolidWorks with ASIS in it , though , correct .

Mike 24:26

Oh , never shipped . So then I gave him his software back and said Bruce , have a good day . And then we worked out a deal with Siemens . It was Unigraphics at the time with John Stevenson , and that part is history .

Roopinder 24:41

Go back a little bit . You were a really big fish for Jon Hirschtick and group to land . You were the programming development at PTC . Here was SolidWorks just a couple of ragtag developers if I can call them that , people with a dream . Not much , no means .

Mike 25:01

Actually it turned out that Tommy Lee did the modeling part . He was very good . Dave Corcoran did the assembly stuff . I got Linda to be in charge of drafting and the system stuff and they hired good people and eventually which isn't in his book we figured we couldn't find enough people in Boston area . So we started this office in Cambridge , england , with a lot of people who had worked on various CAD things in the past and wanted to work on real products and some of those people were really very good .

Roopinder 25:43

Isn't Cambridge the place where D-cubed ?

Mike 25:47

Yes , d-cubed in Cambridge and so was Parasolid development , so that was convenient . That really helped with Parasolid , because we just go there and wait , you've got all these bugs to fix . Why don't you fix them ?

Roopinder 26:02

Was that a surprise to you that you were even able to license something ?

Mike 26:06

Graphics , especially Siemens . They have this component business . I found them to be very cooperative and they live up to their contract . My experience with them was always very good . I got criticized . I had to keep explaining to the board how do we know about Unigraphics ? And then Siemens will support it . I said because they have in the past and I have no reason to believe that they won't in the future . And then I had to listen to Bernard saying that how many years ago is that ? Nearly 30 years ago , bernard was telling me that . And they still behave properly . You can't be in the compartment business and I ran that for DS and Spatial and you had to treat customers as customers and you never told the people in Paris what they were doing .

Roopinder 27:01

if they told you , you kept it completely confidential didn't your title change after SolidWorks was acquired by Dassault Systèmes ? But you actually worked for Dassault systems then , officially .

Mike 27:13

We had a handshake agreement on that , not a contract . I stayed for eight years and five of them were making spatial work . I must say that it did not go the way of acquisitions in some other company . Solidworks prospered with Dassault as the owner , as you can tell .

Roopinder 27:36

You were in a position where you went into Spatial which sold ACIS , so you could actually correct the mistakes they made in the past . Their modeling wasn't solid .

Mike 27:46

I knew about some of the

The SpaceClaim Era

Mike 27:48

mistakes from the experience at SolidWorks and what one heard from the experience at SolidWorks . And what one heard . Okay , there was an interregnum period . There was three months between announcing the purchase and actually closing the deal . In that three months I learned that it was even worse than I thought it was . The place needed reorganizing . There was no quality assurance . So I took the SolidWorks test and 70% of them ran with underneath . So I had battery of computers and I would run these tests and with SOLIDWORKS on ACIS to make sure that ACIS behaved itself and continued working , that ASES behaved itself and continued working . I had one of the founders of DS in charge of development . His name was Didier Boursier , who had recently been in charge of geometry at CGM and he had moved on from that and came out for two years to help implement some engineering judgment . Of course , there wasn't much .

Roopinder 28:56

You left . And then what ? Sailed around Nantucket Bay for a little while .

Mike 29:00

Oh , then somehow I got together with David Taylor and Blake and Danny Dean and that was a space plane . I did that for a while and then I didn't do that for a while . Then eventually a friend of mine had a VC who had a failing company . They said look , we're funded if you run this and make something of it . So what we made of it was a cloud storage product that worked for CAD assemblers . How do you get an assembly with its relationships into the cloud ? You can't exactly zip it up and take it there If you have library components etc . Because then you'd make copies of them . So we came up with this simple idea why don't I treat the cloud as a remote disk from your computer ? But it's not in your office , it's all the cloud .

Roopinder 29:55

This was before Onshape . Did you say that's the original Onshape ?

Mike 29:59

No , I didn't do anything with the inside of the devices . I only took their files and put them there and the version . So you've got all the versions of everything and you think you've got everything on your desktop . We have so much data from some companies that the metadata takes up too much space on your computer , so we've got that virtual as well . So you think you've got this vast file system , but really isn't there at all .

Roopinder 30:31

Can I spend a little bit more time on Space Claim , because that's like your trilogy , as it were . So you started that along with Blake and the others , and then it got sold to ANSYS .

Mike 30:43

That's interesting because it is direct modeling . Now the computers were good enough to do that . It's not easy . What happened was that very early on , in large companies they have people who design and they use things like Cotillard , and the people to analysis use Ansys , and so if you want to change your model , you can't do it because the designers won't let you . So what we originally thought of was that Ansys would be a beautiful partner for this . So at a very early Needham conference we showed Jim Cashman of Ansys this thing , which he got interested in , and five years later later they bought it and , because it allowed them to go into analysis without the regimentation of large companies , to let the analyst change the model and then , when they finally made the structure right , give it back to the person who would then put it back into the CATIA model .

Roopinder 31:53

All these FEA programs had a very limited modeling capability .

Mike 31:58

Well , this was a way they could change the models . It was about change not really create , but change . They actually did some dynamic stuff with changing the model and the simulation was almost in real time updated . It was very impressive .

Roopinder 32:16

Let me ask you something about . You said direct modeling and space and a space claim was one of the that was their whole strength was direct modeling . I always thought it must have been somewhat of a change of religion for you , because you did all the things that were not direct modeling , history-based . So here you are now saying was that , like I'll throw this all out the window , it should have been direct modeling . No , not at all .

Mike 32:40

The direct modeling is more about change rather than create and in this case , with answers , it was more changed by the people who didn't create it . If you like fish , does that mean you won't like steak ? It's that and that's how we did this document management stuff , which is , it's all related . Now we evolve things because we keep all these documents , especially for some builders , that which are PDFs mostly . We can index them for lack of a better word All of those documents , the RFIs associated , the specs and the rest of it , and find any word or phrase in those documents , which is a natural evolution from saving the documents . Now we save the indexing of documents . That's a good way to describe it and , yes , we can read all the words inside the PDF drawings . So you take steps , you evolve and the technology changes . What we can do now we couldn't do a few years ago .

Roopinder 33:51

You realize you did all that without ever mentioning the name of the company , so I'll do that for you . It's Kenesto .

Mike 33:56

Correct , you didn't mention it . You're right , but that was an evolution , and the idea of virtualizing everything sounded

Conesto and Cloud Solutions

Mike 34:06

crazy . But the computers now we've got enough storage and enough bandwidth to be able to do that . So if you click on a folder , it takes a fraction of a second for it to appear . Then it will stay during a session .

Roopinder 34:21

Offers us unlimited computing , unlimited space . When did you convert to believing that was a good thing ? Because that became an evolution for all of us , didn't it ?

Mike 34:30

That goes back to . I remember at PTC , when we went to double byte characters , there was an argument to be made that if it will double the amount of storage for characters , okay , so we should only do it for people who have a language that needed double-by characters Chinese or Japanese . However , I looked at it from , say , a . That will make unreliable software because there'll be too many if statements and then you'll get it wrong . So we should make them all double-by and if there isn't enough memory this year , there'll be enough memory next year .

Mike 35:15

Moore's law has almost been followed for decades . One of the things that we did in very early on was that same discussion about storing the geometry that you last had . That took space , but the disks got bigger . In solid works it's pretty well known that the picture of the assembly is stored in the assembly . So when you bring it up you immediately have a picture of your assembly and can manipulate the picture while the rest of the work comes in another thread . And that took more space on a disk . You couldn't have done that in pro-engineering back then . They could do it now if they chose to . I don't know they may have . So you evolve if you are dependent on the underlying platform that you're using .

Roopinder 36:08

Would you say software is always ahead of the hardware in terms of demand .

Mike 36:12

It's not just hardware . One of the things we use is a structure , an S3 , which is storage as a service which spans much more than the address space on a computer . That's how you can store huge amounts of data in what looks like a file system that it isn't . Things evolve and then you need to use them and you can't just fix a few bugs every release . You need to advance the product . I'm absolutely amazed that none of these CAD companies have managed to figure out how to have a desktop product and a cloud product interoperate and be the same .

Roopinder 36:53

Autodesk says Fusion is like that a desktop product and a cloud product interoperate and be the same ? Autodesk says Fusion is like that .

Mike 36:58

No it's not . It's on your desktop and you've stored the files in the cloud . That's not . We haven't made a Fusion on the web .

Roopinder 37:06

I've called that the Achilles heel of Onshape . That it can't . It just doesn't work , oh no .

Mike 37:15

They could have stored it in JSON files . JSON is unbelievably fast to read . We're doing that for our little building project where you take IFC but a million lines and in six seconds you can turn it all into JSON and then read that JSON very fast , like in a second .

Roopinder 37:33

Have you mentioned Jon Hirschtick ?

Mike 37:35

No , he would tell me all the reasons why he can't do it . But think about it . Why do I care that something's on the cloud or on my desktop ? I choose which is the best environment for what I'm doing . How can I use Onshape when my internet goes down ? Because I have Comcast Xfinity .

Roopinder 37:56

Mike , we have that in Jon Hirschtick . doesn't listen to me either .

Mike 38:00

The point is he never asked me . If he asked me , I'd tell him the latest version of Outlook on the cloud on your desktop . Who the hell cares ? If I don't have a cloud connection , I can still read my email . Why the hell I can't do that ? Why I don't have a cloud interface which is optimized for the cloud and one for the desktop , but the files are the same .

Roopinder 38:26

We ought to team up on John . Ten years I've said this is your Achilles heel . You can't work when I'm offline . This is not good . You need to do something about that .

Mike 38:34

And they've

Reflections on Industry Evolution

Mike 38:35

resisted yeah , because maybe they don't know how , or it's just supposed to be a tinker toy . It's got to be pride . I don't understand pride . Customers are what counts , but the people who built the software are never going to use it . It was built for a developer , not for a user .

Roopinder 38:55

That's true . I was so amazed by a CAD company which I won't mention their name . I worked for them as a consultant . I worked there for months . Not once did I see the program itself on anyone's screen . They were using email , they were using PowerPoint . I was in a marketing group , but you would think somebody there would actually use the product somehow somewhere . Um , yeah , it .

Mike 39:19

No , that that's the whole point . One of the things I can tell you about PTC we mostly hired mechanical engineers , no programming . I would give them C books and say learn it tonight , especially if they were the Russians that arrived and wanted jobs . You don't know C , that's fine . Here's a book .

Roopinder 39:41

Much easier than the converse , which is trying to teach a programmer engineering .

Mike 39:47

I told the American Society of Mechanical Engineers they gave me an award . I don't know why , but you professors want to teach mechanical engineers programming . They're smart enough to figure out how software works . When we did SolidWorks I had different books C+ and then , because nobody knew C+ , I give them those books and say learn it . There were a lot of people with a mechanical engineering background who were doing the programming so they knew what dimensions , linear dimensions were , and radial dimensions and feature control symbols and all those lovely things . I am a great believer in people , unless you're doing pure math stuff , but then you should be a mathematician .

Roopinder 40:33

I got to call you on that because you learned electrical engineering right , so you had to learn on the job yourself how to be all these things like feature control blocks and everything . The parts list .

Mike 40:43

You forget that I went to a college where in the first year you had to do all forms of engineering , so I had to do something I really hated , which was engineering drawing . Oh yeah , you had to use the right pencil 2H or HB or whatever it was to have the right thickness on the lines with a drawing machine and the squares . It was horrible . You couldn't smudge it to make it look right .

Roopinder 41:12

And I had to make all your letters so perfect .

Mike 41:15

I hated it . Computers were a much better way of doing draw things .

Roopinder 41:19

I remember the electric erasers , the little things that had the eraser to remove a line . Those were the lucky guys that had the mechanical erasers . I only had a manual eraser .

Mike 41:30

We had to do things like strength of materials , civil engineering , mechanical engineering . That was the first year .

Roopinder 41:39

You really had a rounded engineering education .

Mike 41:41

I didn't have to learn what feature control symbols were . We had to draw the bloody things , did you ?

Roopinder 41:47

ever have to make splines oh yeah , those slate-looking things .

Mike 41:52

Nobody in our present generation knows what that is .

Roopinder 41:55

Yes , Bezier curves with the little pins they used to stick on in places . The control points .

Mike 42:01

Second year was all different forms of electrical engineering and electronics synchronizing machines . I remember somebody did it in the wrong direction on a three-phase motor or generator and that went bang .

Roopinder 42:17

When you were studying electrical engineering , you were on the cusp of vacuum tubes to semiconductors .

Mike 42:23

I didn't learn anything about vacuum tubes , but there were semiconductors . One of the professors had written a book semiconductor devices . I got into trouble in high school because I didn't want to go to Oxford or Cambridge and study classics or pure science . I wanted to do engineering . So I went to Southampton and King's College and then King's College .

Roopinder 42:48

You're back at Southampton , aren't you ? Or you still have some affiliation with Southampton , aren't you , or you'd ?

Mike 42:52

Still have some affiliation with Southampton . I advise them . Sometimes there's students at Future Worlds which they're encouraging entrepreneurs , and this was not the Europe of the past or present . This is the England of today . There's an entrepreneurial spirit . Now I would happen to think which Martyn would not so that it was coincident with Brexit . We've had this conversation .

Mike 43:21

Martin Day knows how to run other countries rather well , doesn't he ? I'm not knocking Martyn Day . On the contrary , he has found the formula for conferences in the mechanical space and in the AEC space . No one else has succeeded They've all gone by the wayside in getting a huge number of people from North America to go to his conferences .

Roopinder 43:46

That brings me back to the show you mentioned . Do you remember it ? Yeah , yeah , it's all gone . They have user group meetings . COFES is gone . Of course , ptc shows is no longer a thing .

Mike 43:56

He's to be applauded . He's found a formula that works . People interchange views . He should allow more time for that , because you have chats with people in related industries , even competitors . You have competitors that are friends in other companies all the time . It's a small world we're in . You're not out to annihilate the competitor . They've got to exist because then you do better too .

Roopinder 44:23

Absolutely . Competition brings out the best in you .

Mike 44:26

That's a brief history of my version of the world . I mentioned PDMS . That research company was privatized that's how I was on the board of it . It became known as Aviva , which continued to have success with PDMS

Engineering vs Programming Mindset

Mike 44:43

, which is still around . When it was privatized , they had more products and employees . Part of sorting that out was to have fewer products and employees . They focused on PDMS and made a damn good business out of it . They got bought by Hexagon . It's part of Hexagon now , which is probably a natural home for it . A missing S-train was acquired by .

Mike 45:10

Hexagon . That is something that goes back to the late 70s PDMS . That's how long this stuff has been around .

Mike 45:20

3D modeling they were the original solid modeler in essence , yeah it was really lines that were turned into tubes , but on the display there's lots of graphics tricks you can play . Yeah , it's been around that long and it's all about evolving . PTC did not evolve into SolidWorks . The difference was a more friendly user interface . It was sold through researchers at a lower price . It was on Windows . Now AutoCAD has grown in revenue from the early beginnings on the IBM PC , which shows that when said we won't have to do drafting , it didn't quite turn out way . We quickly made two products when somebody would say they would buy Pro/ENGINEER if we had drawings and it will no more go away in building because we have BIM 2.0 , according to Martyn Day , where we don't even have BIM 1.0 because Revit is really not a BIM model . But even if we did have complete models of buildings , drawings won't go away for a long time , if ever they haven't gone away in mechanical . Autodesk continues to sell more AutoCAD and more AutoCAD , not so much Fusion and not so much .

Roopinder 46:47

Inventor , not so much Fusion . I never hear numbers about Fusion .

Mike 46:52

You can easily do the math . They get this from Revit , this from AutoCAD . That doesn't lead much to anything else .

Roopinder 46:59

Thank you for your time , a gracious use of your time , because we've gone way over Mike , we've gone way over our time and I really appreciate that You've got better things to do than talk to a journalist . I'm sure I appreciate that You've got better things to do than talk to a journalist . I'm sure at repinder , at engtechnicacom , companies can sponsor an episode of the History of CAD . Get the details on our website engtechnica . com/ sponsorship . Thank you .