To BOM or Not to BOM — The Magic Episode | Future of PLM | DemystifyingPLM

Episode Summary

The episode "To BOM or Not to BOM — The Magic Episode" delves into the complexities of Bill of Materials (BOM) management within Product Lifecycle Management (PLM), featuring a panel of experts from various backgrounds in PLM and engineering software. Hosted by Michael Finocchiaro, the podcast is joined by regulars Oleg Shilovitsky, Rob Ferrone, Patrick Hillberg, and special guest Jonathan Scott. Each panelist brings unique insights into BOM management, ranging from development and systems integration to strategic perspectives on data modeling and process optimization.

During the episode, several key technical and strategic insights were shared. First, the panel emphasized the importance of starting with business problems rather than focusing solely on technological solutions when implementing BOMs. Oleg Shilovitsky highlighted that BOMs are essential for enabling communication within businesses and solving operational bottlenecks. Second, Jonathan Scott advised listeners to ensure their logical data models accurately reflect future business goals, suggesting a forward-thinking approach to data management. Lastly, Brion Carroll discussed the flexibility of BOM representations through intelligent linking, allowing for context-specific views that can be tailored to different systems.

The key takeaway for PLM and engineering professionals is to prioritize understanding and addressing core business needs when implementing BOMs, ensuring that these tools support long-term strategic objectives while maintaining adaptability in a rapidly evolving technological landscape.

Full Transcript

Michael Finocchiaro

bye bye bye, ⁓

Oleg Shilovitsky

bum bum bum bum bum.

Rob Ferrone

Thank you.

Jonathan Scott

⁓ yeah,

Patrick Hillberg

Yeah.

Jonathan Scott

careful with that.

Michael Finocchiaro

Bob Rayan here. Hello and welcome to this group of characters. got a herding cats over here. We're herding cats. Michael Finocchiaro on the Future of PLM podcast. I am very happy to be joined by my friends Brian and Rob and Patrick and Oleg are sort of the regulars and we have our special new guest,

Oleg Shilovitsky

Don't say bomb word, not beer.

Jonathan Scott

You

Patrick Hillberg

you

Rob Ferrone

you

Michael Finocchiaro

Mr. Jonathan Scott. Since I've been on two of his webcasts, he decided to.

Jonathan Scott

Hey everybody.

Michael Finocchiaro

Jump on mine. appreciate you being here, Jonathan. It's awesome. Today we're going to have a bomb-versation. can't take credit for that word. I think that was a Rob word, wasn't it? Rob, you came up with the bomb-versation thing. Thank you. was you. ⁓ You there, Rob? Hello? Yeah.

Jonathan Scott

Yeah, thanks for letting me join.

Patrick Hillberg

Thank

Brion Carroll

He doesn't want to accept it.

Rob Ferrone

Yep.

Michael Finocchiaro

He's doing a low profile over there. ⁓

Brion Carroll

He's bleeding autonomy.

Jonathan Scott

You

Rob Ferrone

I've got digital threading shoes in the room

Michael Finocchiaro

All right. Well, digital threat issue. it. So ⁓ let's go around. Like we were going to do these little cards, like with a headshot and with a funny, kind of like a Pokemon card for each person. And we didn't get around to it because the AI kind of freaked out on us for various reasons. what I'm going to do is I'm going to ask each of the panelists to... ⁓ Give a very quick like one minute bio because you we haven't actually done this on the webcast since like the first episode almost a year ago. So we'll just do a quick round table and in the ⁓ quick bio, want ⁓ years of experience with PLM bomb, et cetera. Your perspective, like whether you're coming at it from a ⁓ software developer point of view or from a systems integrator point of view or somewhere in the middle.

Rob Ferrone

Okay.

Michael Finocchiaro

You know, the tip customers have you, which have you been in more retail or discrete or process, you know, which industries, because as we'll talk about the bomb sort of varies, your perception of bomb varies by industry. ⁓ And then sort of ⁓ how close you've been to the solution and anything else you want to throw in about it in a bomb-versation. So let's start with Oleg.

Oleg Shilovitsky

All right, good morning everyone. And again, thank you for having me here. So my experience mostly come from PDM and PLM development, starting in construction industry, then moving to manufacturing, working for the SOAS system, working for Autodesk. And as long as I remember, it was about bombs. So again, if you're at the airport or in the airplane, say BOM, because again, bomb-versation can lead you.

Jonathan Scott

You

Oleg Shilovitsky

Bumper session can lead you to the wrong place and it's a live experience. Okay. So just.

Michael Finocchiaro

Just like, like Trump cards, like Trump cards. Yeah, we know. ⁓ So, so what's your favorite bomb? Hey, how about that? When we say your favorite bomb is your favorite bomb, the E-bomb or the M-bomb or the S-bomb. What's your favorite bomb? Open bomb!

Patrick Hillberg

Yeah.

Oleg Shilovitsky

You promised. So...

Jonathan Scott

His favorite bomb is the open bomb.

Patrick Hillberg

and

Oleg Shilovitsky

I like when someone is doing my work. Thank you very much. I think I started as far as doing bill of materials for electrical boards back in construction work in AutoCAD, but that was in 1990. It was interesting. One of the most challenging ⁓ bombs that I ever touched was back in the 30s when it was for something that was called 7E7. How we say, the people that know, the know.

Patrick Hillberg

Hahaha

Michael Finocchiaro

You

Oleg Shilovitsky

So it's a good name of Dreamliner when it was in the development. it was. Yeah. So the wire harness bombs, it's one of the craziest thing that I've seen and then working across different industries, mostly discrete with a lot of electronics and a lot of BOMs. And yes, open bomb is my favorite these days. So thank you.

Michael Finocchiaro

The seven, eight, yeah. Hahaha! Jonathan, how about you?

Jonathan Scott

Yeah, let's see. So I've been doing this about 25 years or so. My background is really mechanical engineering and systems engineering, but yeah, I've been doing the PLM thing for quite a while. Let's see. trying to remember all the questions on the intro, Michael, but I'll go with, yeah, industries. Yeah. I'll take that one. I'll take that one.

Michael Finocchiaro

⁓ Industries. Industries for five. Industry for five, Bob.

Brion Carroll

Ha ha ha ha ha.

Jonathan Scott

So I'll go with, you know, big stuff, right? Planes, trains, automobiles, ships, that kind of stuff, industrial equipment, energy, those kinds of things. So not as much on the consumer goods and the, you know, I'll say, you know, smaller product side. So that's where I've spent a lot of my time consulting with folks on that. And in terms of proximity to the work, I think it's middle of the road, right? So I don't get my hands down in bills and material a lot these days, right? I spent some of that time when I was doing engineering, but now it's more... consulting on the concepts down to how do you implement it in business processes and Favorite bomb. Let's see. Well, that was a tough one I think I'll go with I'll go with some of the most interesting ones are legacy bombs that I see coming from like aircraft industry right where we still got smart part numbers and And the way that that aerospace does things with installations and and those kinds of things Fascinating to watch those deep level bombs and how people do it and what they still struggle with Yeah

Michael Finocchiaro

Ha ha! Okay. kids and stuff. Great, thanks Jonathan. Hey Brian, how about you? A little view from the fan, the ceiling fan.

Brion Carroll

So. Yeah, so over 40 years in this stuff. ⁓ And when it comes to the industries that I've worked in, ⁓ I'd say, well, it's been a whole slew of automotive, some industrial, but recently like the last, let's say 15 years have been focused more on retail and consumer products where variability. So the clothes that you're wearing, is a product, a colorway and then a size, right? And that could be manufactured in one location or another location or multiple locations where it gets, you know, cut, make, trim, then they fasten it and bag it in another location. So all those bombs, the variability where we found the magic is that you can have just one bomb. And what you do is you make intelligent, we call it overloading the link. You make intelligent the traversal. And so it says traverse to here, the child only if you're looking at this location or only if you're looking at manufacturing or only if you're looking at this ECO and the quantity shifts by two and so on and so forth. So we found that one bomb intelligently traversed would give an infinite number of variabilities as long as you make the link intelligent, right? And so I'd say my favorite mom is all of them.

Michael Finocchiaro

So also worked in retail, right? Which is another animal, isn't it?

Brion Carroll

Yeah, retail, that's the thing where you've got product size, colorway creates like a shirt might be 40 different bombs. And if you have multiple locations, which could have different ⁓ materials from different sources, depending on which location, all being with the same garment, okay, especially outerwear, where you have the outer shell, the inner shell, and then you have things that get connected, and then you have packaging. ⁓ All of these things can have so much diversity.

Patrick Hillberg

Thank

Brion Carroll

that if you can't make intelligent why you traverse from a parent to a child, if you can't make that, which we call overloading the link, if you can't make that smart, then the masses of products, you take an Adidas, three million products a year, well, that's how many bombs? Three million, right? Instead, you should be able to shift that down.

Michael Finocchiaro

Okay.

Brion Carroll

to somewhere near 500,000 because there's variability on all of them. So, yeah, by the way, Mike, do like as maintained bombs. Those I consider like industrial to be really where the heart of evolution is, where it never stops, it just continues to evolve. And that goes back to one bomb for everything, right? Just make intelligent why you go to the child. If you don't go, it's because

Michael Finocchiaro

Thanks Brian.

Brion Carroll

Why you're going doesn't affect that specific child. Go to the next one.

Michael Finocchiaro

Okay, Patrick.

Patrick Hillberg

Oh, let's see. actually, I noticed that Oleg is using what I think is a razor leaf coffee cup because I have a razor leaf coffee cup too. Well, I know you got to do it. You got to do it four times. If you do it four times, they give you a coffee cup and a shirt and a shirt. Right. That's a little way.

Brion Carroll

I don't have one. Mine's just a regular old...

Michael Finocchiaro

I don't have a razorly coffee cup, come on man. ⁓ Uhhh... shit, a shirt! I want the tie! I want the tie!

Jonathan Scott

Yeah, five gets you a jacket.

Oleg Shilovitsky

Hello.

Michael Finocchiaro

I've only done two, man.

Jonathan Scott

Five gets you a jacket. I don't know what a tie. 25 years gets you a tie.

Brion Carroll

Ha ha ha.

Patrick Hillberg

Yeah, that was my deal with them. said, no, there should be Saturday Night Live. Five years, get a jacket, right? Five times you get a jacket. Anyway, and they haven't brought me back for the fifth time. Therefore, apparently, you're holding out on the jacket.

Michael Finocchiaro

Yeah. ⁓

Brion Carroll

Yeah, I see.

Jonathan Scott

Ha ha ha!

Michael Finocchiaro

They want to get me on four times first. ⁓

Patrick Hillberg

Yeah, stop.

Jonathan Scott

No, no, no. In this group, I got to get Brian on next. I haven't had you on Brian, so we got to get you on next.

Brion Carroll

No, we haven't been on.

Patrick Hillberg

So, ⁓ let's see, ⁓ experience wise, I started my career in, ⁓ vision guided robotics and then, ⁓ shifted to PLM about 20 years in. So, 20 years in robotics and 20 years in PLM. And my favorite bomb is probably the BOP, the bill of process. ⁓ cause I really come from the manufacturing background, where by which time really the bomb has kind of been resolved. By the time we get onto the factory four, the configurable bomb has already been configured. So when I moved into PLM world, I was ⁓ an architect for two of the companies. In fact, I was architect at Delmeo when it was still called 7E7. So in any event, you get into change management and you get into bomb management and that sort of thing. And I'll briefly say, and then move on that. The, so I teach a graduate management course called product life cycle management at Oakland university, which is near Detroit. And the last time this group met, we discussed the seven three seven, right? And that's literally the conversation tonight. And what we're talking about today on Tuesday is literally the conversation that we're to have on Thursday. I've never really figured out how to bring the bomb debate into my course, but the conversation we've had leading up to this and then.

Michael Finocchiaro

Patrick Hillberg

what I'm sure will come out of it, has actually given me opportunity to do that. So thank you guys. You're doing ⁓ my class planning for me.

Brion Carroll

Anytime, anytime, what are you going to give us? Maybe a pen or something?

Michael Finocchiaro

I hope it checks in the mail. Yeah, is it checking

Patrick Hillberg

Yeah,

Michael Finocchiaro

them

Patrick Hillberg

right, yeah.

Michael Finocchiaro

out? Mr. Plumber, the data plumber.

Oleg Shilovitsky

Invoices will be coming, I get it.

Patrick Hillberg

Ha ha ha ha.

Rob Ferrone

Yeah, well actually before I was a plumber I was a bomber believe it or not.

Patrick Hillberg

Now, huh, huh.

Brion Carroll

We've recorded that the CIA knows it

Michael Finocchiaro

⁓

Brion Carroll

and you'll be seeing them out south, posted outside your house.

Patrick Hillberg

Michael Finocchiaro

He's not going to any United States for a couple years. think that's a good one.

Oleg Shilovitsky

Now we understand why you prefer to be plumber and not a bummer. That's a real problem.

Michael Finocchiaro

You

Rob Ferrone

So I've been bombing for around 25 years. actually the thing I like about it so much is that the bomb is really the touch point between the people in the organization and the systems and the data and the information. Because for me, it's all about communication and collaboration. And the bomb is where that happens. And so yeah, my background has really been a lot like Jonathan's, know anything kind of big and complicated discrete manufacturing and I think my favorite bomb The the service bomb is a closed contender because it's usually so artistic and beautiful the illustrations of You how you should be taking things apart. But actually I love ⁓ matrices bombs or matrix bombs where you have

Brion Carroll

Yeah.

Rob Ferrone

You know, all of the list of the parts down one side and then the assemblies or the target vehicles or whatever it is across the top. And so which parts are going into which ⁓ assembly. is it for prototype builds or is it something complex like, ⁓ you know, an instrument panel, for example. And so I love that kind of thing. It's like code breaking.

Brion Carroll

Have you seen products that have the Matrix Bomb as a visible representation?

Rob Ferrone

Meaning The the only time No, it tends to happen. It tends to happen in Excel That's what I've seen it. It's like it lends itself perfectly fit. Yeah

Brion Carroll

software products like you know you've got the so and you've got Siemens and you have PTC and you've got open bomb I don't know who who runs that but do you have you ever seen a matrix yeah gee some good have ever seen matrix bomb displayed

Michael Finocchiaro

⁓ Yeah, PLN number four.

Brion Carroll

yeah, well, okay. Yeah, I got you.

Michael Finocchiaro

So ⁓ that's cool. ⁓ I've been doing this about 30 years, ⁓ those last 30 years here in Paris, and I'm from just about every industry and not really from a putting in place the bonds, but making sure the systems that the bombs are running on are performant and the performance problems don't, when you do an expand from the root, you don't blow up the machine.

Brion Carroll

Michael, what about you? Michael, what about you?

Michael Finocchiaro

Which used to be a problem back in the day and my favorite bomb I don't know I guess I can say E-bomb because you know, love manufacturing stuff, but I haven't you know dealing with consumables and glue I shouldn't be anywhere near glue. That's a bad thing. So I'll just gonna stick with the So maybe some people in our audience are not as bomb experts as we are

Brion Carroll

All right.

Patrick Hillberg

Thank Ha ha.

Brion Carroll

As long as you don't have a bag with you, you know, I think it's okay.

Michael Finocchiaro

And Patrick, you being the professor that you are, maybe you can give us a professorial, but not too ⁓ academic definition of what is a bomb. then, well, let me just set it up. I'd like, like Patrick, gives a definition of what is a bomb. And then maybe Jonathan, you can tell us what's inside a bomb, like what is a constituents. And then, ⁓ and then I'll do the next set of questions. Let's start with those two.

Patrick Hillberg

Yeah. ⁓ Sure. ⁓ let me say professors are taught to speak in 90-minute increments, So sit back for a while. So let's see. what is a bomb? Well, what is which bomb, right? So are we talking the engineering bomb? Are we talking the manufacturing bomb? Are we talking the service bomb? Bomb generic, right? ⁓

Jonathan Scott

You

Michael Finocchiaro

So get a coffee, we're gonna be here for a while.

Brion Carroll

You can only inhale four times. you inhale four times, that's the max.

Michael Finocchiaro

Hahaha! Just bomb, generic, without any letter in front.

Patrick Hillberg

It's a breakdown of all the parts in a product. ⁓ I'll actually, I'll go back to my BOP comment. We design our products around the manufacturing facilities we have that can build them. So I think a lot of the BOM originates with the bill of process. What tools does my factory have? Because I can only build the things that my factory can build. If my factory is General Motors, I can't build airplanes because I just don't have the tools for that. But in any event, ⁓ but the engineering process begins with the e-bomb, which is a structure of the high level parts in a bomb. Like I know I've got an engine and I know I've got seats and I know I've got steering wheel. And then within the engine, have pistons and within the pistons, have rings and all those sorts of things. And you guys are probably more experienced than me on that. I'll give one quick anecdote and then we can move on. I was working at a shipyard and ships are full of pumps. This is really, this is the sort of ship that has an airport on top ⁓ and ships are just full of pumps. And in the design process, they can, you know, when the designer chooses a pump, either the pump or the motor or both, it just automatically comes with a coupling between the pump and the motor. Well, from a service perspective, the pumps and the motors last a long time, but the couplings don't. The coupling never showed up on the engineering bomb because in the design process, he didn't have to think about it because he just pulled a pump out of a catalog and the pump automatically came with a coupling and who cares, right? It's not my problem. So I got the pump that I needed to do. So there's these, it's not only different views on the same data, there's data that does not appear such as grease and glues and those sorts of things that Michael was talking about. There's simply data.

Brion Carroll

Mm.

Michael Finocchiaro

But Greece is just underneath the, you know, the Balkans. It's still there. Sorry.

Patrick Hillberg

Say that again? So I'll finish up with that, but I think the ⁓ coupling issue, that was an eye-opener to me. There's data needed by the service people that's simply not ever in the engineering BOM. So there's some additional, the engineering BOM says I need a pump, but then the service people need to come in and say, yeah, but we know that the pump that you ordered out of that catalog does not include the coupling. So we're gonna have come back and add in the fact. There's a couple.

Michael Finocchiaro

Thanks.

Jonathan Scott

I'll

Brion Carroll

That's interesting.

Jonathan Scott

jump in and pick up on what you asked me to talk about, you know, what's in the bomb, right? What are the constituents? And I think one of the challenges talking about that, Patrick led into this perfectly, right, by saying his favorite one is bill of process. Well, the bill of process, that's a BOP, not a BOM. Come on, man. But I get your point. It's when we look at the strict definition, bill of materials, it's about the materials that make up the product. One of the things I think is challenging is that a lot of bombs are multi-level.

Michael Finocchiaro

Jonathan Scott

And so you start to question, how many levels do I go down to and what is actually in that bomb? So your example there with, you know, engineering, not having all the details of what's in that pump, you know, starts to expose some of what's in there. So for sure there are parts, right? Parts and sub assemblies and that kind of thing. But then it gets controversial when you say, well, but did I go down to all of the constituent parts? Like if I bought something, do I get to all the little piece parts? And well, you know, those came from raw materials. Those came from chemicals and compounds. Okay, so depending on who you're talking to, maybe we get our bombs all the way down to compounds and constituent elements and things like that, or maybe we don't. But for sure, we've got those pieces. And the other constituent I'll say is the links, right? Brian was talking about this earlier, right at the very beginning, the relationships between that parent and whatever the children are and grandchildren and great grandchildren, et cetera, et cetera. So you got the relationships and then you got the components. For sure, those are parts of the bomb in my view. But I don't know, maybe somebody takes issue with that. That's what I got for you, if you know.

Michael Finocchiaro

All right, thank you. That was a great answer. That's exactly what I wanted. Rob, maybe you could tell us why is this a religious thing? People are like, why are you doing a podcast about bomb? It's like the most impossible subject. Why is it that it's become sort of these, where's the religious lines being drawn here? Why is it so difficult?

Rob Ferrone

I think a lot of it is entry point is where people come into the world of BOM and PLM and their foundational experiences and you know they learn about or they know what they know because of what they've been doing and so when they ⁓ you know so that forms their religion if you like around BOM and so they have ways of thinking about how things, know, whether like, for example, Patrick talked about starting with the, manufacturing, but actually, ⁓ I would say that, you know, it starts at the end, well, even before the engineering end, you go way back into, you know, marketing and thinking about, okay, what is the product we're going to make? you know, what, what kind of, how many variants does it have? Is it going to be engineered to order configure to order? ⁓ and, so, you know, you go from left to right. But yeah, so and yet people have also system influences. So if you're an SAP person, you might believe that the, you know, the bomb starts with the ERP bomb, and then everything else kind of feeds from that. And yeah, that that's, that's my take.

Oleg Shilovitsky

Okay.

Michael Finocchiaro

⁓ And Oleg, you've obviously built a product based on bombs. ⁓ Why is it that ⁓ you found that knowledge graphs are the best way to represent those? Because I think there's also a debate on how best to put them. Up to now, we've put them in these relational databases and they've been a bit cumbersome. I mean, they still are there and that's what most people are using. You took a different approach. And I'm thinking that that's because of the way people are actually using them and how they're actually structured, is that correct?

Oleg Shilovitsky

Oh, that's a great question. And I actually want to start with what Brian said. He said it's BOM, it's a lot of different dependencies. And in many situations, you want to traverse some of them and you do not want to traverse some others. So I've been thinking about this. The biggest thing that I would say, there is a visible part of BOM. And if you go... I bet money if you ask someone in the university what is BOM, they will say, here is the Excel. I've seen it in so many places. So this is the most visible part of the BOM, like the simple Excel. Now the most invisible part of the BOM is all those things that we discuss and all dependencies. are dependencies in the process, dependencies in materials, there are dependencies in cost, there are dependencies in manners, like all these things together. They're really complex. Now, if you go to all this depth of the complexity and relationships, we understood after the many years of experience, the best way to model it is graph. And also the best way to query this is graph because the needs and the capacity technological are there. But on the other side, we left it as simple as Excel. That's why the user interface is like simple Excel. So those are two things. So answering your question, there is a need for complex data management there. On the other side, everyone to see it as simple as Excel.

Michael Finocchiaro

And so, Brian, what is, can you talk about the, you were talking about already about relationships. So when you're taking ⁓ the bomb and we're looking at all these different views of the bomb, how does that compile interests of relationships as you were trying to say earlier?

Brion Carroll

Well, see, that's the thing. I'm bomb agnostic, right? My belief is a parent to a child relationship is the, just like atoms in the human or in atmosphere, the relationship between a parent and a child is the core of bomb, right? I mean, in actuality, a bill of material is once you go through all that, you list the materials. So a bill of material really is just a list of materials that are in the bomb and quantities associated thereof. But when you go to the bomb as it's structured, like what Oleg says, it's almost like a neuro net, right? Where you go to one and you, ⁓ a parent and you shoot off to the child. And during that, during that move, you should be able to gather content. You should be able to gather information as to why you're going to that child. Or as I speak of before, overloading the links with information that says, go to that child. Why? Because what you're looking for is the M-bomb. And that child doesn't exist because it's a purchase part, meaning the sub-assembly is purchased, which means we're telling you what's in the sub-assembly, but you don't need to go get it because you stop here. So, or you need to get grease or oil or something else. You wouldn't see that in the E-bomb, but you see it in the M-bomb, right? And like with the bill of labor or the bill of process, how that thing gets created, you might find, and I worked with a very large systems company out of Boston that needed to go from E-bomb to M-bomb. And the way we did it was as you went into the E-bomb, as you created the M-bomb, it remembered as you get rid of things, they used to put screws and nuts and bolts into baggies. And the baggies during the bill of labor would be at certain positions so that you could take them out. But from an engineering perspective, they were together because they held the board in place of the chassis and there were, X of them. So we had to allow them to pull all the screws and the nuts and everything and put them in baggies. And what we did was we overloaded the link. We said this child in this baggie used to be in the e-bomb with this parent. So now if somebody went into the e-bomb and did a change, it could find that screw and get rid of it or change it, right? because we remembered where it came from as it went into the baggie. So what I'm saying is bombs are purposeful, contextual. They ⁓ determine based on who's going in, the human, the role, how they should present themselves. And in my mind, a bomb for a product should be one bomb meets all. And how you traverse to it is what makes it contextually accurate.

Michael Finocchiaro

Thank you. That makes me want to wonder because we've mentioned Excel a couple of times and yet there are the big three solutions on the market. Then we have the startup ecosystem. Oleg represents part of that, but there's also Propel, Dura, who was just acquired by Altium. mean, there's a whole ecosystem out there of these systems. Why are we still using Excel, Jonathan? Why is Excel still a thing? it user experience? it because... this other one suck is because we just are too lazy. I don't know. I think that's another open debate, another one of the religious things ⁓ maybe, but what's your take on it? Cause you, do a lot of interviewing and I know you do a lot of consulting work. I'd really like your feedback.

Brion Carroll

haha

Jonathan Scott

Yeah. Absolutely. Yeah, it's I think it's what Oleg mentioned. It's about simplicity, right? So people trying to solve the problem, they've got a problem to solve. They don't wake up thinking, you know, I want a complex software solution to what I do every day. They, you know, are trying to get the right parts to the shop floor or the right parts at the right workstation, whatever it might be. and everybody's got Excel, right? And Word's bad at it and PowerPoint's bad at it. So guess what? You know, it's like, that's how they end up there. It's kind of how, you know, PowerPoint ended up as the best change management tool. Same kind of problem, right? I can take pictures and write some text and they're on individual chunks. So Excel, think is the simplest tool for the job. people, they struggle when they try to progress it to really solve the problem they need to solve. And that's when they run up against it and end up looking at PDM or PLM or something else, or even trying to code it. Because, know, like Brian was describing earlier, these links, these relationships are complex and sometimes they traverse multiple levels and you have trouble just getting it all in your head. How do these parts relate to each other? And then you can't fit it in Excel and you look for something different. But at the start, you're just thinking it's a parent, it's a child, I need a quantity. And usually your CAD drawing, putting it in 2D. That was painful. So Excel was a good representation of what we used to have. And that's where people end up, I think.

Michael Finocchiaro

And ⁓ Rob, do you think, where should people start if they're gonna do this? Should they start with how many bombs do I need or do you start with a system? It's a sort of a chicken and egg problem, right?

Rob Ferrone

You have to start with the business and you have to start with what's the endeavor? are people trying to do? So who's in the business? What information do they need? What's the product? And then from that, you can start to build the data model. So that's the starting point is this master product data model. what information does each group of people need in order to do their job? And then you can start to figure out, okay, how does that information flow? Because I think one of the things that we haven't, we've touched upon it, but you've got information in different bombs. Yeah. So you might have, you know, E-bomb, S-bomb and all the other bombs in between. And how does that information get in there? Is it, is it a cascade from left to right? Or do you have concurrent engineering where someone creates the the service bomb at the same time as the e-bomb, how's that informed each other? And then how are those links maintained? Is there a ⁓ register, graph or something that is saying, hold on a minute, this information is different or how do you know that, for example, there's a part in the end bomb that doesn't need to be in the e-bomb or the other way around? So I think if you can solve all of those ideas first by looking at what master data, master product data you're going to use to run your business. then you can start figuring out how you're going to solve it from a technical perspective.

Michael Finocchiaro

Thanks, Rob. I've got a couple of questions already coming in from the audience. And I was going to throw one at Oleg, because I think this is a good one for you, Oleg. You ready? You sitting down? All right. How tightly do we expect bombs to be linked to system models like MBSC? ⁓ Will the bomb eventually be derived from the system model rather than authored separately? Good question, right?

Oleg Shilovitsky

Absolutely. ⁓ That's the amazing one. Thank you Michael for throwing this at me.

Michael Finocchiaro

⁓ Well, you said you had the most experience other than Brian, so.

Brion Carroll

He did, he did.

Oleg Shilovitsky

Yeah. So I think the question here, you're trying to understand the relationships are going to two places. The first is the business is what Rob mentioned, because the way it related is dependent on your business process. So if your business process starts from the requirements, it is how they will be connected from the business process. If they are like, I don't remember who said the favorite is the maintenance. If your business goes from maintenance perspective, you eventually will have to interlink it with additional pieces of information. But this way or another, system requirements and connections to these tools that help you to identify and create and manage those requirements better, will be tools that will interlink or intertwine all this network together and will be connected. So I don't think there is one straightforward way how to connect them. What is important is the flexibility. think Rob mentioned that we need to build a data model as we understand the requirements. So as you're building this data model, you will understand how interlink those pieces together. So the right data will show up in the right place, like we like to say, at the right time.

Michael Finocchiaro

I think also Jonathan, you touched on it and I think that Brian and Patrick probably have experienced that too, having ⁓ looked at different industries. But it's also interesting to look at, process industries where they have a bomb, except they call it a recipe or a formula, right? And that is where Jonathan was saying you could go all the way down to the actual chemical reactions. That's what process industries do, which is interesting. And then you've got fashion with its own thing. ⁓ The struggle that... the big three have had is trying to get one size fits all, And I don't think we have, nobody's been really, really successful at that because there's always been a flex PLM or an info for process or whatever. Do you guys think that that, mean, Patrick, do you think that there is a way of doing an all in solution does it all? Or do you think that these things are so specific that we're always going to have to be looking for point solutions in a lot of cases?

Patrick Hillberg

Yeah, I think it's, I think it's, it would be nice to have it all. should have, we should have one, one ring to rule them all, but it's not realistic. There's a aphorism or a law, Conway's law that says that the, the, ⁓ the product that you build replicates the organizational structure, which built it. So if, if you look at, at a Toyota versus a General Motors, you'll see that they are built in very different ways. And that's the difference in culture in the organization at Toyota versus the culture and organization at GM or pick one, right? and then the BOM represents the product. mean, the BOM is a textual representation of the product infrastructure, of the product structure. The product structure is a representation of the organizational structure. So without... collapsing onto a single organizational structure as a academic professor, I'd sit here and say that, I don't think we can have one bomb to rule them all. I think it's going to be ⁓ very individualized based on how the company was created.

Brion Carroll

So, so Patrick, I want to jump in for a second. So I think. I think that, yeah, yeah, sure. I'll just sit over here on the balcony and go, Hey, what are you guys doing? having been at both consumer products, automotive and aerospace, and then, retail fashion. ⁓ the area that I think complicates the delivery is.

Michael Finocchiaro

was actually going to go to Brian. was my next one. So it's all good. ⁓

Patrick Hillberg

Thanks.

Jonathan Scott

You

Brion Carroll

And I'll go back to the clothes that you're wearing, right? Each one of you are wearing a certain style, which is the product, a certain colorway, which of course is the color, and a certain size. And so how you model a thing that you're going to deliver to market, the SKU, the stock keeping unit, right? It could be an airplane, 737 model, whatever serial number acts, right? Or it could be an automobile, which has a VIN, which represents a certain structure. So that's similar to what you classify as a skew or stock keeping unit in retail. if, but the bombs, I challenge anyone listening to show me how overloading the link won't work. Period. I don't care who you are. I don't care what industry you're in. I don't care if you're doing lesser bees of batches.

Patrick Hillberg

My comment on that is not, I don't see it as a technical issue. It's an organizational issue. I mean, you would need to convince General Motors that they should operate like Boeing. How do you do that?

Brion Carroll

Right, right, right. Right, right. Or somebody that they respect. How would we do that? Right. Whoever it is that has succeeded, right?

Patrick Hillberg

Yeah. Yeah. How do you convince

Michael Finocchiaro

⁓

Patrick Hillberg

any company that they're wrong and somebody else is right? Because to get to one, you're going to have to convince everybody else that they are wrong and adopt what is right.

Brion Carroll

But you could be slightly wrong or you could be slightly inefficient. Not wrong. You don't necessarily have to say the word wrong, but less, less efficient than company X, right? Whatever that is.

Michael Finocchiaro

Yeah, you gotta talk money. There's gotta be money.

Patrick Hillberg

Although change is, yeah, I mean, I'm a little bit going off onto a topic, but change is inefficient. It's difficult to change. Once you're through, you're better off, but the process of change is inefficient and seems inefficient. Sure.

Brion Carroll

Of course.

Jonathan Scott

No, I want to separate something, I think, Brian, that you're saying just to make sure I'm tracking it. Because I hear what you're saying about the overloading the link part, right? But I think that's a technical answer to still getting the same result for the end user, which is somebody sees that bomb one way and somebody else sees it a different way, right? So there's still an E-bomb and an M-bomb. It's just technically it's one bomb. So that part, I think, is interesting that it comes back to, Patrick, what you were saying about

Brion Carroll

Right. Correct.

Jonathan Scott

organizational differences. You know what, if your organization wasn't broken down into engineering and manufacturing, would you have an E-bomb and an M-bomb? I don't know, right? But maybe you would because you still had those functions even if you didn't have those organizations or departments in an org chart.

Brion Carroll

You still may need to add grease or fixtures. And if you're talking about recipes, you have boil time or mix time, right? And those things become operational in the manufacturing floor, which are purely ⁓ defined without action in the engineering world, right? Because your recipe has not only chemicals, but also ⁓ things that do things with the chemicals. Sorry, Rob.

Michael Finocchiaro

I Rob had his hand up.

Rob Ferrone

Yeah, you know, posted the other day about this and we talked about, you know, we are talking about e-bombs, M-bombs, all the other bombs, but actually, you know, imagine if you had a Michael bomb or an Oleg bomb or a Jonathan bomb where that was just the information you needed as an employee of the company to do your role. And you could see...

Oleg Shilovitsky

and

Rob Ferrone

the things that it took you to do your day job. So that's one way of thinking about things. And so maybe you won't have a department bomb, the E-bomb. If someone needs everything and needs to see everything, great, they've got it and they can roll it up and see the weight of the product or the total cost of the product. other thing is, Michael, you mentioned having, can one system do it all in most companies unless they're relatively Unless there's simple companies, you've got multiple systems in play and multiple ways of connecting those systems. that, you know, when you add in the stuff that Patrick's been talking about, the organisational differences, cultural differences, manufacturing differences, you know, it's exponential in terms of the complexity. that then comes back to a post I did the other day about what are the universal things that you can learn around the idea of having visibility of the product data that you're going to be using for your business. versus all the way to the kind of unique way that you then execute it as an organization.

Michael Finocchiaro

Thanks, Rob. I'll look at his hand up and then I've got another segue.

Oleg Shilovitsky

Yes, I think what we are trying to resolve here is what I see in many organizations. I'm not sure it should be resolved because there are four pieces here and they're all different and we should not connect them together. They impacted each other. But if I will go from organization on the top level, organization is thinking how to operate and they need to send them and they call it with need to send this bomb and we need to synchronize that bomb. So that's something that can be very different in any organization. This is what I think, Patrick, you said how to teach, how to tell to Boeing to work like GM or vice versa. I mean, each organization, even five people, we see them, they have a very specific opinion how they want to work. You will not go in to change them. And this is what they will be trying to do. And then there is a specific industry aspects because a bomb for ⁓ consumer goods or the bomb for processing industry, in other words, a recipe of a cake will be different than the bill of materials for industrial machinery, and you want it or not, it will be different. That's something that we will need to acknowledge, and we will not be able to marry people that are doing cakes with the people that are doing machines. They just speak different language. And then we go with the some...

Brion Carroll

Okay.

Oleg Shilovitsky

specific foundational principles because we said there is an item, an instance, an assembly, and you kind of can call them all the same and they function in the bombs with the very similar principles. And this is where the core technology, if you want to talk about core technology for bomb is coming. And then the fourth level is the technology, which becomes very interesting because in the past we managed one technology to use one technology to manage those bombs. And then we send relational database are amazing and start using them. And then we said SQL and the relational database sucks because it's a performance is crazy. And then we moved to graphs. This is what open bomb does. And now we are saying, you know what, if we can represent everything as a language, because we can represent the bomb as a language, and this is the experiments that we do, then a language model can work to describe any bomb. So think about this for. they will not displace each other. We will still need to do something for specific company and specific industry and specific product and implementation. Yes, it will be simple. People would like to have it. And then we will use some technology. Most of our customers don't care about technology, but wait, if it works so fast, it's important.

Michael Finocchiaro

Thanks, Sardeg. That was a really great way of framing it. Actually, I was gonna come back to you, Brian. I know you raised your hand. I wanted to ask, I think another thing that makes BOMB such a touch point and so complicated is that they kind of concentrate all these things around configuration and change, which are very, very complex and very, very widely different depending whether you're building an airplane or you're making Coca-Cola. So I don't know if that's where you were going to go with it. That's where I wanted to go with it, right.

Brion Carroll

Well, I was going to go two places. Well, now that you asked that question, I still think that why you traverse to the child can be overloaded with content that says the effectivity date of this, right? So this gets into the change, you know, effective because it's ⁓ something that will create. ⁓ harm to humans, right? You may say that we need to do it now. I don't care how much you got left of that other thing, we're replacing it with this. So the effectivity date is now. Yes, yeah, exactly. But I liked what Oleg was saying regarding the language. When I originally designed or led the humans that designed the FlexPLM product, ⁓ my original intent before PTC bought the company was to actually have different

Michael Finocchiaro

It goes back to our 737 debate.

Brion Carroll

profiles for different industries. And one of the industries was RX, right? And if you read a something of pharmaceutical, it can be a language which actually has intelligence within the language used, right? You've got verbs, nouns, the nouns are actual things. Like if you say, I'm going to take this ingredient, this chemical, that chemical points to a chemical, right? And my process is to do the following with it. and then enjoying it with something else and then do something else. And that becomes a recipe, right? A pharmaceutical recipe. And we would have done that had we stayed autonomous. We would have done that as a language, as Oleg is saying, because it'd be great. It's almost like a children's book, right? You're reading through and you want to click on the dinosaur and, my God, quick, shut the door. Don't scare the children. I'm telling you right now, don't do it. But that language really creates kind of like this ability to open doors into things if you choose to go deeper, right? ⁓ So I'm just, he sent me off in that direction. I apologize for getting going down that cul-de-sac off our main drive. But I still think the effectivity and change is something that you can put in the link because if an ECO is the reason that this part is being swapped out, then it's original pointing to a

Michael Finocchiaro

It's okay.

Brion Carroll

a thing with a quantity of zero is because of that ECO. And this other part, which is a quantity of one, is being brought in because of that same ECO. And that's how you swap a part, right? Still one bomb, right? In my mind. Okay.

Patrick Hillberg

Michael Finocchiaro

Right. So that's going to lead into another ⁓ thing, which is one of the questions in the chat. But just before that, I had a question for Rob because Rob, okay, well, Patrick, you can jump in just after Rob.

Patrick Hillberg

Yeah, I'm sorry. just want to add one thing I know what Brian says. And I think that's an important point of maintaining reasoning. And I know, you know, in the DMS that this group has and we, you know, there's some other, you know, likely suspects who are all in that conversation, right? Maintaining reasoning is a hugely valuable thing that is not well done currently. It could be, but, you know, maintaining a link back to the reason why the change occurred and what was all the What was all the thinking and such like? I think that's a valuable thing. OK, I just wanted to emphasize that point that Brian made.

Brion Carroll

Yeah.

Michael Finocchiaro

That's okay. That does actually tie into my question because ⁓ Rob and I have been at a couple of conferences together and you recall Dr. Youssef Houshmand's thing about multiple sources of truth, single source of change. Do you want to talk about that for a second, Rob? Because I know you were in answer to those conversations.

Rob Ferrone

⁓ Yeah, mean change changes is a great topic because as we know products static it depends where you're where you're you know, which perspective you've got on the bomb, but let's say that the most simple form of change is Um, someone makes a change to the design and then at some point that cascades down through the different departments. And then that change, you know, goes into manufacturing and, you know, then, you know, who needs to be aware of it, et cetera, et cetera. But then when you think about if you're making lots of, um, uh, discrete, uh, products and you start making changes, then you need to be able to control which changes have been on which products. Um, you know, especially where maintenance is critical. And I know that I gave a talk recently about change control and defense. And there was a famous example with the the Harrier aircraft where I think it was a Harrier, sorry, during the Gulf War that they they made so many, well, they control of the changes that they're making to the aircraft. So that actually of the of the fleet, only 10 % could fly at any one time because the upgrades that they wanted to do weren't compatible. And so they had to then re-baseline everything in order to then do the next generation of upgrades. ⁓ yeah. Sorry, Michael, I forgot the original question now.

Brion Carroll

I thought that was bad, bro.

Oleg Shilovitsky

That's the problem with bombing.

Michael Finocchiaro

Nice. ⁓

Rob Ferrone

Okay.

Michael Finocchiaro

Well, we're heading into the last quarter of an hour. had a question from Moose out. I hope I said that right. ⁓ I think all of us should throw in our comment because it's sort of about, mean, Brian started and Oleg also sort of lead us that way because we were talking about PLM as a language. I've already written in another article about PLM as code, kind of like we've at this transformation of infrastructure from all these crazy ⁓ diagrams of rational rows to just Kubernetes and YAML files. ⁓ So how do we see the role of AI and automation handling traditional bomb stuff, right? How in the next three to five years, we're going to see it change. It absolutely cannot stay exactly the same, right? Something's going to boot. So what kind of shift will there be? I would also throw in ⁓ Sadie's comment last year from Microsoft that SaaS is dead, which is true. ⁓ It's problematic too, because everybody took 10 years to get there, right? From 2010 to now, everybody was like, got to be on the cloud. And suddenly like, ⁓ maybe the whole model has changed. ⁓ So how do you guys feel about that? We've all have a certain gravitas, having done this same thing for so long. How do you guys see us moving? I want to start with Ole, go for it.

Oleg Shilovitsky

All right, so I think what is the important thing to move forward is to start with openness, because ⁓ that's something that ⁓ solves the human... We originated from a place that everyone wants to hold the data as close as possible. But as Rob said, in the organization, the human factor and the openness coming together. So I think if we resolve this and we will be able to deliver, I mean, not speaking about technology for the moment and speaking mostly about behavior, we will be able to solution and the business model behind this. that allows the companies to work on the shared data. That would be an amazing thing because back for the last 30 years, I'm in this business, you you sell enterprise solution, you hold the data, sell applications. Now that's becoming a problem today, especially when people start hiring agents, AI agents and not people. How do you sell seeds? Did you think about this? that's a problem. You can't sell more seeds because people, companies are hiring agents. Something is broken. A lot of people are now concerned about this. And if we will not solve a problem of data communication, and we will not solve a business problem, that's the most important thing. Because otherwise, if there is no business, everything collapses. And my best prediction on what can happen here is that we will learn as the telecom companies to stop charging for the long distance calls and start charging for data. I'm not saying it will be exactly like this, but there is a lot of conversations that we have now with customers and prospects about how to make it in the way so we allow the companies to grow and to connect and not to be restricted by the business models. think that's, I think that is the most important thing because this business allows to go and to unlock many things. And then I think the second thing is to discover how to use all these amazing technologies. that ⁓ we experienced for the last 20 years and 30 years and to use it for the bond management because keep in mind the four or five top companies that managing bonds and enterprise today, I will quote Martin Eigner that said, those are technologies from 90s. So it's all amazing things with technologies for 90s. That's my take.

Michael Finocchiaro

Thanks, Tilly. So Patrick, do you think we're gonna end up with outcome based pricing?

Patrick Hillberg

I, well, first, let me say, okay, so first, before I come down negative on this, let me say I'm no Luddite. My PhD dissertation from 2004, I developed what I call a digital reliability twin and I use neural networks to improve throughput in this digital reliability twin. ⁓ that's, and that was based on objective data. I think with, When we talk about AI, I think we need to start breaking up AI into specific use cases. So I see some value in AI in terms of drug development and a variety of other things, but I don't think there's a viable business model for LLMs. ChatGPT had $11 billion in revenue on $34 billion of expense. committed to $1 trillion of expense and yet they're losing way. I mean, they're losing much more money. So, and we're not currently paying. We're not currently paying for this. so in any event, I'm, I'm definitely keeping my powder dry on the LLM, ⁓ standpoint, I think, and I'll add one more thing and then we can move on, but, and in PLM we use. PLM to design dangerous products. I mean, they're planes, they're trains, they're automobiles, they're guns. I've got three members of the military in my classes here. And what we see in things like the 737 MAX, which we talked about, and the GM ignition switch, and VW diesel gate, and, and, and, and, and, and all these scandalous failures is there are weak signals within the culture who recognize the fact that these catastrophes are likely, but it's difficult for the culture to raise those issues up. And when we move into ⁓ stochastic mimicry, which emphasizes the most likely response rather than a weak signal response, we're simply building ⁓ an IT infrastructure, which will increase the likelihood ⁓ of catastrophic failures. So first, I don't think there's a business model for LLMs. and in general. And I think it's a particularly bad use of LLMs in the design of products which can kill people.

Michael Finocchiaro

Thank you very much, Patrick. That was depressing. I just don't remember who had the hand up first, Rob or Brian. Okay. Because I did want to, well, okay, go ahead, Rob. then I did want to kind of modify the question before giving it to Brian and then Jonathan can close. Go ahead.

Patrick Hillberg

Yeah. You should take my class.

Jonathan Scott

You

Brion Carroll

I rubbed it, rubbed it.

Rob Ferrone

Quick one for me. I think change and change impact is going to be a great place to apply AI because I think ⁓ AI can figure out the potential impacts ⁓ better than humans can, especially across an enterprise where many, different things could be affected. Then you've also got the really good examples of people like Colab where they are giving the engineers information at the time they need it. For example, they're developing a part and you give them the requirements, you give them any lessons learned and all that type of information is when they're doing the work so that they can make the smartest decisions. So I think there some really great examples in engineering that are actually happening today that will make a difference.

Brion Carroll

you

Michael Finocchiaro

Thank you Rob. ⁓ Brian, you already talking about ⁓ engineering and bombs in terms of nouns and verbs. And so if, you think if we have the right guard rails in place, we have, you my amenities agent that's always doubting what the other ones are saying. Do you think we could get to the point where PLM could become a language where your dream of one bomb to rule them all with these powerful relationships, which are all driven by agents in order to. keep the bounds and keep the thing moving. You think that's where we're going?

Brion Carroll

Yeah. Well, I think let's go back to how bombs get created from CAD, right? I believe that in the CAD world, if you can define a screw, characteristically, like Rob was saying, the requirements of it, it should be able to generate the 3D from the screws that are already in its learning tables, right? And that learnedness becomes something that manifests a screw that's longer than any of the other ones or has a different thread count based on a spec. So that CAD geometry, which could help people doing the fixtures and the mold design or whatever else they do when they create a screw in manufacturing. So that, think, is something that AI will have. Another thing, and this goes to Patrick's point of the 737 and the issues of validating is everything here that is necessary and is there something we should be aware of and continue to check, right? Like if you look at the ⁓ suspension of a plane landing, the shocks have, you know, unbelievable pressure on them. So ⁓ the logic in AI should be able to say, check the shocks three times a year, right? Because you're going to have so many landings, you want to make sure that the grommets and whatever else is in there. So that I think is something that AI will do, is design validation. But the third thing I think is ⁓ being able to in the bomb world say, I want to templatize a wide range of products that have similarity, whether it's housing. Look at the fan above my head, you've got blades, you've got housing, you've got lights on, ⁓ that you should be able to auto-gen. 95 % of what a design should be or what a bomb should be, ⁓ bringing in all of the components and the engineer would then finish it off. And then you click a button and it validates that I forget something. And then it goes out and does whatever. So I think AI has position within this world. I also think to your point or question Mike is that you could design a language. This goes back to the ingredients of a recipe, but I don't think you're ever going to take the fan above my head and describe it semantically, but you can take a recipe and semantically describe it so that the batches follow that operational step and you can log the batches. This was back at the pharmaceutical or, you know, consumer packaged goods or whatever, right? So in that area, I think that language and semantic bomb development could exist. AI could help them. But I don't think in CP or retail or whatever else, there's a way to do that. Cause I can't even think of the logic of the language. ⁓ that's better than the bomb in its presentation, right? I'm done.

Michael Finocchiaro

Jonathan, we've got a couple of minutes to wrap up. ⁓ There's actually a question in the chat about ⁓ causal sources of truth, but I think all of us can look at the chat afterwards. thank you, by the way, for everybody listening and for the folks that did comment. We really appreciate that. So go ahead, Jonathan, I'll let you make your...

Jonathan Scott

Yeah. Yeah, I want to come back to the question that we're all answering here. And I also want to throw in one more topic before we wrap up today. So coming back to that question, think that AI is a real challenge in the PLM space. And I'm going to bring it back to the bill of materials explicitly and say, think one of the challenges that companies are going to see is that the big companies are going to have, A, the budgets and B, the data to do what they want to do with AI on their product definition.

Michael Finocchiaro

Thank you.

Jonathan Scott

Right. And it's not going to be LLMs. It's going to be a different kind of model, but an intelligence that's trained on their data to understand how to operate on bombs, product definitions, all these things. Yeah. So it's, think that's coming, but it's going to be tough for the smaller companies for a couple of reasons, budgets and data again. Right. So a specific example on bombs. I know tons of companies that when they create a bill of material, they've got alternate parts. And I've seen at least five different ways to represent an alternate part compared with the primary part.

Brion Carroll

All right. Yeah. Right.

Jonathan Scott

Okay, so what is the model that's going to get trained eventually to work on bombs, not an LLM, but this bomb model? How is it going to interpret that? Our data is not semantically rich. One company is different from another and we're going to misinterpret and people are going to put the wrong part in and et cetera, et cetera. So I think we have to watch for that and see what does that mean about the difference between big companies and little companies and how they can use AI in their product development. Okay, so that's my answer. The topic I wanted to get into, but we won't have time today. Let's not forget about builds and material. Another thing that I think is really important, variance, right? We talked about MBSC earlier a little bit, and I think that ties into the whole picture, but all of these structures that we have, eBOM, MBOM, whatever, comes back to Brian's point about one overloaded structure. When you try to have variation, but you've got multiple copies of your definition of your product, now it's really tricky, right? And it's something important for people to think about when they're trying to decide how to represent bombs in their company.

Brion Carroll

Right. Yeah, and I think that overloading the link solves that problem without duplicating content, you know, religiously. By the way, Oleg, anything that's an agent should have a ID to get in, and that would be licensed. You can't have agents floating around penetrating your product without having license to penetrate your product. So regardless of whether it's a human or it's an agent, it should have a user ID that enables it to go in. And there should not be seven agents look like one. It should be seven agents each having a license to do what it is they're going to do. Period. Oleg Shilovitsky (1:00:57) Congratulations, Jonathan Scott (1:00:57) But the trouble is what? Oleg Shilovitsky (1:00:58) congratulations. You are the group of people on the software vendors now that's saying that we need to start selling seats per AI agent. I I'm not saying good or bad, but this is the group, okay? But then, yeah. Brion Carroll (1:01:06) Yeah, you got it. You got it. Well, I did that with PT. I did that at PTC where flex connect is the way you connect systems and each one has to have a license. Each system you want to Right. Yes. Well, you got it. Oleg Shilovitsky (1:01:17) then we have thousand agents and we need to sell to each of them licenses it's a great model i just want to see who will buy it Michael Finocchiaro (1:01:22) Rob, Rob had his hand up. Rob Ferrone (1:01:30) I think that's a legacy hand Michael. Let's wrap it up. What do you want? 60 sec or less than 30 second bomb conclusion from all of us Michael? How do you want to Brion Carroll (1:01:32) That's a legacy hand. It's not part of his current anatomy. Jonathan Scott (1:01:33) You Michael Finocchiaro (1:01:37) That's a legacy, Ann. Okay. ⁓ Bomb inclusion, yeah. Okay, go. ⁓ Okay, let's go ahead. You start. We'll go ⁓ Rob, Patrick, Oleg, Jonathan, Brian. I'll say it's, we're out. Go. 30 seconds. Oleg Shilovitsky (1:01:50) That's it. Rob Ferrone (1:02:01) Okay, so my advice to companies is when you're thinking about the bomb, don't start with the bomb type, but start with your business, start with the people in the business, and then go to the information that needs to flow around the business, the right information at the right time to get the right stuff done, and then build your bomb solution from there. Michael Finocchiaro (1:02:23) Boom, Brian, Patrick. Patrick Hillberg (1:02:26) ⁓ I'm really pushing the idea of we need a digital thread as the foundation of the information model and then the various X-bombs simply become views into that thread, purpose built views into that thread. Rob Ferrone (1:02:33) Thank Michael Finocchiaro (1:02:40) Thanks, Brian. sorry, Patrick. Oleg? Oleg? I saw the strings. I saw you pulling the string and I his mouth open. Oleg? ⁓ Brion Carroll (1:02:42) How'd you know I was doing that? How'd you know? Hahaha Oleg Shilovitsky (1:02:53) Yes, I would, I'm always saying to all companies, the bombs is the lifeblood of your business. It's how you operate your business and what problems you want to say. So if you have a bottleneck, you need to solve it. So we always start and come into companies and saying what problems you want to solve instead of what bombs you want to build. So start with the problem and then what model and what process and what tools will fit it. And yeah, that's the, that's the bomb. in the manner, but it's something that enables your business and enables your communication between people. Michael Finocchiaro (1:03:28) Boom. Mr. Jonathan. Jonathan Scott (1:03:31) Yeah, I think my advice is get your data model right. Right. And I don't mean technical because your technology might change under the hood over time and it should, but get your logical data model right. Think about the business you're in, how to support what you want to do, like everybody else is saying, but also think about where you want to be in five, 10, 15 years. Right. Nobody thought about product as a service 20 years ago or few people did. But now if you think about it, well, your data model might look a little different to the things you want to track to get into that. So. Try to think about where you want to go and get your data model. Michael Finocchiaro (1:04:02) Boom ⁓ or bomb and now Brian closing comment 30 seconds. Brion Carroll (1:04:08) So, ⁓ yep. So when I say that you can, you should be able to represent any bomb by overloading the link, that doesn't mean that you're not going to be trimming that bomb based on the, the bomb traversal and pushing it to ERP. Okay. You will be pushing it out to different systems for them to do what they do. And Mike, Michael brought it up as, know, each system has its own purpose and you can't forget the fact that they exist. But I believe that bombs can be made. by intelligently creating a link reference that has contextual applicability. Should I travel to the child? Depending on how I'm looking at the bomb, I may or I may not. So 27.6 seconds. Michael Finocchiaro (1:04:52) Thank you, Brian. ⁓ Wow, that's amazing, man. Thank you everybody. We have not come up with another topic and I'm sure we will. We will take suggestions in the chat. Thank you once again for everybody joining. Please look out for next week. I have a AI Across the Product Lifecycle podcast with a couple of awesome startups. And by the the startups are also not sleeping on the bomb topic. I have at least three or four that are very uniquely using that. And I think, Jonathan, you've talked to a lot of them too. Thank you for attending this episode. We had a great time. think we took a very boring topic and made it rather interesting. I appreciate that. And again, thanks to Jonathan for joining our group. think we might welcome him back. What do you think guys? Thumbs up if he can come back. ⁓ Yeah, get the mugs, bribes, all these bribes. Geez, you guys. right. All right, take care and we'll see you on the next episode, everybody. Brion Carroll (1:05:39) Yeah, sounds like a possibility. Oleg Shilovitsky (1:05:40) It all depends on this, Jonathan. Jonathan Scott (1:05:43) Yeah, he's... Brion Carroll (1:05:43) Well, you gotta get something to pens or cups, coffee cups. Oleg Shilovitsky (1:05:46) Ha ha ha ha. Jonathan Scott (1:05:50) Thanks for having me though guys, it was a lot of fun. Brion Carroll (1:05:55) Take care. Oleg Shilovitsky (1:05:57) Thank you. Michael Finocchiaro (1:05:59) Okay, I killed the recording.