🎧 #149: Change My Mind: The Independent Data Layer

‍

“Talking about it in this way, that the IDL is silver bullet and it will solve all of your problems is simply not true, and it will actually hurt the industry, because owners will spend a lot of time and money expecting a certain business outcome, that may or may not actually be there in terms of reduced costs at the end of the day.”
‍
—Alex Grace

‍

Welcome to Nexus, a newsletter and podcast for smart people applying smart building technology—hosted by James Dice. If you’re new to Nexus, you might want to start here.

The Nexus podcast (Apple | Spotify | YouTube | Other apps) is our chance to explore and learn with the brightest in our industry—together. The project is directly funded by listeners like you who have joined the Nexus Pro membership community.

You can join Nexus Pro to get a weekly-ish deep dive, access to the Nexus Vendor Landscape, and invites to exclusive events with a community of smart buildings nerds.

Episode 149 is the first episode in our new debate style series called Change My Mind. The conversation is led by moderator James Dice, and then flanked by an opposing team with Alex Grace and Gregory Cmar, and a supporting team with Silvia Quaglia and Shaun Cooley. Kenny Trowers is the judge and neutral party that the teams need to convince to one side or the other.
‍

Summary

Episode 149 is the first episode in our new debate style series called Change My Mind. This idea came from Nexus Pro member Pete Swanson. So thank you Pete, and we love that our community continues to inspire us to keep experimenting with the pod. The first topic is the independent data layer, and the debate center’s on if all buildings over a hundred thousand square feet should have an independent data layer in the smart building stack. The conversation is led by moderator James Dice, and then flanked by an opposing team with Alex Grace and Gregory Cmar, and a supporting team with Silvia Quaglia and Shaun Cooley. Kenny Trowers is the judge and neutral party that the teams need to convince to one side or the other. This episode format also features recording with a live virtual audience of our Pro Members that can engage in a chat while the debate is happening, as well as come to the stage to ask questions at the end. Enjoy!
‍

Music credits: There Is A Reality by Common Tiger—licensed under an Music Vine Limited Pro Standard License ID: S441935-15083.

Full transcript

Note: transcript was created using an imperfect machine learning tool and lightly edited by a human (so you can get the gist). Please forgive errors!

[00:00:00] Alex Grace: Talking about it in this way that the IDL is a silver bullet and it will solve all your problems is simply not true, and it will actually hurt the industry because owners will spend a lot of time and money expecting a certain business outcome that may or may not actually be there in terms of reduced costs.

[00:00:15] At the end of the day,

[00:00:16] Shaun Cooley: the problem is that every single one of those defined business outcomes today that's sold to them by some vendor, whether it's their insurance company, the local municipality asking for ESG data, a company doing F D D A company, doing facilities management and maintenance, all of it comes back to an initial integration step, right?

[00:00:36] To implement that business outcome. You start by spending a lot of money and a lot of time. Going into the building and connecting those systems. Uh, and then you do that again over and over and over with every vendor that comes in.

[00:00:50] James Dice: I was trying to figure out how to mute you and I just didn't get there fast enough.

[00:00:53] I really wanna mute someone, so someone next person to go over. I will figure out how to mute you.[00:01:00]

[00:01:02] Hey friends.

[00:01:03] Did you know the best way to continue learning beyond the Nexus podcast is to sign up for our newsletter? The Nexus Newsletter is your one stop shop for staying up to date on the way to smart building trends. Sign up today free of charge, and we'll drop this industry leading resource right into your inbox every Wednesday.

[00:01:18] If you're already signed up, double high five and thank you. But now tell your coworkers and your friends once you're signed up. The best way to continue the learning. Is to join the Nexus Pro membership community or our online course offering. Headlined by our flagship course, nexus Foundations. Diving into these products allows you to connect with our global community of like-minded Changemakers Links are below in the show notes and now on the pod.

[00:01:46] All right, I think they can all see us now. Hello to everyone in the audience. Thanks for coming. Uh, we're doing it live here. So welcome to the Nexus podcast. Today's episode is a new debate style format that we're [00:02:00] gonna be experimenting on here today. It's called Change My Mind, and this idea came from prom member Pete Swanson.

[00:02:07] So thank you, Pete. I think I saw Pete's name. In, in the audience here. So thanks Pete. It's a great idea. So, first of all, I was an athlete in high school, so I was not on the debate team. I'm sure there are people that do both, but I was just purely an athlete. So this is my first formal debate I've ever participated in.

[00:02:25] So thanks everyone for, uh, in advance for the grace. As we feel this process out and as we do with everything, we'll sort of iterate on this after this first session and hope to make future debates the best we can. All that said, here are our rules for the debate. We're gonna target about an hour here.

[00:02:43] We're gonna do q and a at the end of that hour. These folks have been provided with time limits for opening arguments and response arguments. If they don't, if they go over those time limits, well they'll be muted. They're gonna need to be civil and kind, but, We didn't come [00:03:00] here just to be kind. We came here to debate each other.

[00:03:03] So please do be prepared to argue. Argument is encouraged. Are any of you anti confrontational? Like me? You're gonna need to get over that before we start to minimize talking over each other. We're gonna use hand raising to, to sort of get the floor. So if you wanna speak outta turn, you'll, you'll need to do that.

[00:03:23] The resolution that we're debating here, here's the, the, the key statement that we're debating every large building owner and we're defining large by over a hundred thousand square feet, should have an I D L or an independent data layer in the smart building stack and the definition of the IDL that we're gonna use for this.

[00:03:42] We're going to call it a dedicated edge connectivity and middleware layer that is independent of the applications or application layer. And this could be a product, it could be developed by a building owner, it could be an open source project. I don't [00:04:00] think we're here to debate what the best method of providing the I D L is.

[00:04:04] We're here to the debate, sort of the need for the dedicated layer between the devices and applications. Do any of you all before we start, wanna sort of change that definition or, or give us any feedback or briefly add anything to that definition at all? Are we all good?

[00:04:21] Shaun Cooley: Well, I think in the, in the spirit of debate, we, we should have some feedback.

[00:04:25] The only thing I would add is it should also be independent of the devices, not just the application. Right? So it's independent. Above and below.

[00:04:33] James Dice: Above and below. Cool.

[00:04:35] Alex Grace: I'll add some there, James, which is

[00:04:37] James Dice: okay.

[00:04:38] Alex Grace: Are we defining it as raw data only or are we defining it as some level of modeling?

[00:04:44] James Dice: Um, I think what, I think we're defining it as some layer of modeling, but I think when we get to the points that we disagree on, I think that that piece will definitely come up.

[00:04:53] Yeah. Some layer of modeling, um, some layer of data storage, some layer of a p i or some [00:05:00] sort of access for applications. Um, yes. We're defining it as some layer of modeling that's separate from the applications. So Kenny and I will be judges. You guys will compete to change our minds. And then obviously the last thing is no sales or self-promotion.

[00:05:16] We're sort of debating this concept, not really talking about any individual products. And we have our pro members here today, so they're gonna, I, I, I hope they will all weigh in as well. So we're, this is change your mind. So if, if they, I want you guys to pick a team and see if these guys change your mind as well.

[00:05:35] At the end of the discussion, we'll have pro members come up here, like I said earlier, and this, this will be published as a podcast and that's video for the, for the members. So let's do some introductions. Kenny Trauers is senior Associate at Cisco Hennessy Group out of New York. Then we have the advocate team.

[00:05:52] Shaun Cooley is founder and CEO of Mapped Shaun's out of California. Then we have the second member of the [00:06:00] advocate team. We have Silvia Colia is Technical Program Manager of building technology at Dexus out of Sydney, Australia. We're covering a lot of time zones here today. Then we have the skeptic team.

[00:06:10] Alex Grace is VP of Business Development at Clockworks Analytics out of Boston, and then his teammate Greg. How do you say your last name? Greg Smar. Smar. So Greg Smar is founder and CTO of Interval Data Systems. He's also outta Boston. So we have like team skeptic Boston folks. Is that, is that just run in the, in the water in Boston?

[00:06:33] Greg Cmar: Absolutely.

[00:06:34] James Dice: Okay.

[00:06:34] Alex Grace: Dirty water.

[00:06:35] James Dice: Yeah. All right.

[00:06:36] Shaun Cooley: Silvia and I can be Team Pacific Ocean then instead of Team Boston.

[00:06:40] James Dice: There you go. So you guys don't know each other, I don't think any of you really know each other. And so I wanted to do something to get our juices flowing a little bit. Um, so we're gonna just debate something stupid to begin with, which is cats versus dogs.

[00:06:53] And so I want to call on Shaun, go first, pick a position, and then we'll, we'll go to the, the [00:07:00] skeptics team to pick the other side. Just, just, just to get the juices flowing.

[00:07:04] Shaun Cooley: I'm gonna go with dogs. I, uh, every cat I've ever had just does not, uh, does not need me. It's, uh, it's nice when the dog. It wants you to be part of its world, needs you a little bit.

[00:07:15] It's, uh, yeah, it needs you a little bit. You leave for two minutes and you come back in and it acts like you've been gone for six months. The, uh, the cats, you can be gone for six months and they just look at you like, eh, get outta my house. I'm you all

[00:07:27] James Dice: right. Skeptics. Argue for cats.

[00:07:32] Greg Cmar: Well, that's a hard one.

[00:07:38] Alex Grace: Okay, so needy. These dogs, you know, I mean Exactly. Exactly. They need you all the time. They're high maintenance. Cats are independent. They clean themselves. They know what they're doing. I don't believe with that at all, but I had to go for it.

[00:07:53] James Dice: Uh, love it. Anything to add, Silvia? To the, to the dog.

[00:07:57] Silvia Quaglia: Dogs are so much better.

[00:07:59] They [00:08:00] love you and, uh, we need love in the world. So that is it. Like a cat doesn't need you, doesn't want you, you are just like a burden for them. Uh, you're just a provider of food. So yeah, no, no, no debate here.

[00:08:14] Greg Cmar: But the nice thing about cats is that you can leave them home alone and not have to worry.

[00:08:20] James Dice: All right?

[00:08:20] Shaun Cooley: But if you do die next to a cat,

[00:08:22] James Dice: Kenny

[00:08:22] Shaun Cooley: will eat you. So, uh, dog, dog will just lay down next to you, which is kinda, kinda nice of it.

[00:08:30] James Dice: That might be the winning, the winning argument.

[00:08:32] Shaun Cooley: That's the mic drop. No rebut

[00:08:34] James Dice: mic drop. So, Kenny did, did they change your mind either way?

[00:08:39] Kenny Trowers: Uh, yeah. I'm team dog.

[00:08:45] James Dice: All right. I love it. Okay, let's start the real debate. Fun, good times to begin with. Thank you for humoring me there. Uh, so we're gonna do the first speaker on the advocate team. The advocate team again is Shaun and Silvia. Silvia, [00:09:00] I think you're gonna go first. You're gonna present your arguments in support of the resolution, and you have timer setting a timer for four minutes right now.

[00:09:09] Go. Okay.

[00:09:11] Silvia Quaglia: So the most important things for a building owner is what is to make profit. And you can only make profit, um, if you rent your space. Correct. So you need to find a tenant that is willing to pay a certain amount of money for your space. And, uh, with anchored tenants, which is what you want, uh, if you want to start to build your property, comes request a lot requests and lease agreement.

[00:09:39] And one of the requests, uh, now that we are seeing more and more frequently, Is full access to building data. And I'm not like saying like that they are specifying the data that they want. They just say full access to building data, which is very vague and it makes it very complicated, obviously. And how do you grant that if you don't have an idl?[00:10:00]

[00:10:00] Do you give them access directly to, uh, your service network? Do you let them plug in whatever device they want? Um, how are they going to interface with a different subsystem? And when you have a tenant, then you need to reduce the opex cost of the building at the minimum. Right? The opex cost are the ongoing cost.

[00:10:21] Um, and a way to do this is to optimize your building or, uh, to have like a data-driven maintenance tool. And again, you need like a third party application to begin their device, uh, to then send the data to their application. So you allowed each contractor to plug in in whatever subsystem network. Um, and even if you have an integrated network, then like it's, it's, it's kind of the same.

[00:10:46] And it can cost a fortune because, uh, like system are complex that like building our complex in general. And then there is this additional complexity related to the, the, the subsystem like access control is very different for [00:11:00] bms. So that's another question. So how you do that, um, your sustainability team needs the energy and the water data to go to a third party analytics platform.

[00:11:13] Again, how you do that, how you normalize the data to them, like how you provide an easy way for whatever stakeholder needs data, um, to access to it. So I can provide like a long list of a stakeholder and different set of data that they need for different proposed. The point is not like I, I can go forever, but the, like, it's literally, it's, it, it's proven very easy that we need to provide a unified data layer for that integration and the, with the CDO system, it's kind of impossible.

[00:11:47] I wish that we could do these things at the edge and maybe we will, but the reality of today is that we cannot. And so that is basically all I have to say [00:12:00] and uh, I really hope that it makes sense to you. Um, I don't want to feel all the time with the different words that are not, I love it.

[00:12:08] James Dice: Yeah. We don't have to fill up all these time limits.

[00:12:11] I just wanted to real quick clarify what you're saying Silvia. So the ability for you to provide the same interface for a bunch of different stakeholders that don't need to know how to basically integrate with the base building systems, they just need to know how to access it in this one place, which you're saying is an api.

[00:12:28] Thank you for starting us out. All right, Alex, we're gonna reset the timer here. Alex, you're the first speaker on the skeptic team. Your job is to basically present arguments in opposition of the resolution. We're not necessarily responding to Silvia quite yet. Why doesn't it make sense from your perspective to do an ideal?

[00:12:50] All right, go. Cool.

[00:12:54] Alex Grace: So I don't think all buildings over a hundred thousand square feet need an I D L because what [00:13:00] buildings need are the defined business outcomes that they're trying to accomplish. So do they have the e s G goals is tenant comfort. What's important to them? Um, what are all the different outcomes and business objectives that they are trying to accomplish?

[00:13:16] That's where we should start from, not from the point of a technology decision that may or may not be applicable to actually solve the business problem, but the owner thinks they're going to solve. So start with the business objective and the outcome you're trying to accomplish. Then figure out what are the applications that are actually gonna deliver those outcomes to me that I have confidence in.

[00:13:37] Then from there, look closely at what the data sources are that need to feed those applications. Then look what the overlap is between those data sources, because talking about it in this way that the I D L is a silver bullet and it will solve all your problems is simply not true, and it will actually hurt the industry because owners will spend a lot of time and money expecting a certain [00:14:00] business outcome that may or may not actually be there in terms of reduced costs at the end of the day.

[00:14:04] So start with the business objectives that you want to achieve. Figure out the applications that will feed that. What are the data sources there? What are the overlaps of those data sources? Then determine, do we have sufficient overlap of data sources here to justify a data layer? I'm saying data layer, not independent data layer, because I, I don't quite understand what is independent, um, about a data layer, uh, independent from the hardware that's underneath and the applications above.

[00:14:33] Okay, that's clear, but it's not independent in terms of, you know, it's another vendor decision for an applica. It's essentially another application that is a middleware layer in your stack. Um, this is not new, right? There's a lot of precedent for this in various industries. Maybe less so in commercial real estate, but for example, pharma always has a, a pie server involved as a middleware layer between their data sources and the applications that they're trying to, to, you know, in regulated [00:15:00] environments, that's extremely common.

[00:15:02] Um, but start with the business case. That's the most important thing. Independent, questionable, plug and play, unfortunately is often not the case. Um, so, you know, just an example here. I've never seen a data layer that will tell me. Uh, that the preheat is before the mixed air temperature or that the deification mode is overriding normal sequence of operations for heating and cooling.

[00:15:30] So there's a lot of details to like how f that's more of a nuanced point about how far the data layer should go and how much modeling should be there versus raw data. But the point is, these are just simple examples to say that the application may actually not benefit from the data layer in terms of the cost, um, reduction that you expect it to, which is why you have to start with the business objectives and the applications that will feed it and go in that direction rather than the other way around.

[00:15:54] And then the last point is really around interoperability and centralization versus decentralization. [00:16:00] So the, the data layer concept, essentially you're centralizing data and then, and then expecting there to be cost benefit by plugging the applications in. Maybe yes, but not necessarily. So we need to look at the advantages of interoperability between applications.

[00:16:14] That's more of a federated. Concept, like federated metadata standards between applications and, and avoiding the need for centralization in the first place.

[00:16:27] James Dice: 20 more seconds if you need it.

[00:16:29] Alex Grace: Um, I think I'll end it there.

[00:16:32] James Dice: You'll end it there. Nice. So, several key points. I'm taking those down. Kenny's taking those down.

[00:16:38] I think we'll, we'll use those when we go to a more open part of the argument. Um, Shaun, over to you. So you have six minutes now. So you're the second speaker on the advocate team. You can present further arguments to add to Silvia's point. You can also begin to respond to some of these arguments that the skeptic team is bringing up.[00:17:00]

[00:17:00] Shaun Cooley: Great. So I, I think I will start by responding to some of the points that Alex made. Um, you know, the, this concept of a divi, a defined business output. Is absolutely what business owners should be thinking about, what building owners and, and other parties involved in the building should be thinking about.

[00:17:17] The problem is that every single one of those defined business outcomes today that's sold to them by some vendor, whether it's their insurance company, the local municipality asking for ESG data, a company doing F D D A company, doing facilities management and maintenance, all of it comes back to an initial integration step, right?

[00:17:37] To implement that business outcome. You start by spending a lot of money and a lot of time going into the building and connecting those systems, uh, and then you do that again over and over and over with every vendor that comes in. Those vendors spend a lot of time and effort. Trying to do those connections.

[00:17:55] They spend a lot of time and effort with what we refer to as the second Dave problem, [00:18:00] right? Somebody changes a backnet id, somebody updates a firmware on a device. Somebody swaps out a piece of equipment, you get to go back in and do all of your integrations again. And so this problem just keeps snowballing for both the vendors and for the, the business owner that's trying to achieve those outcomes.

[00:18:17] And so you're continually throwing money at this problem of integration. The independent data layer steps in and says one party, and whether or not that party is an extra vendor or not, uh, you know, the independent part we just wanna be clear, is independent from the device manufacturers, and it's independent from the application providers.

[00:18:37] For those solution providers, but that one party comes in and owns the entire top to bottom integration. Whether it's manual or automated, sometimes it might require a one time step to, to go through and do that work, but that one vendor is now on the hook for that day two problem for going in and making sure that every time something changes that that's updated.

[00:18:56] Every time a new device is added, it's brought into the independent data layer. Every [00:19:00] time the device is removed, it's removed from the independent data layer, and that allows then the application vendors to decouple themselves from the chaos that's inside of the building and really focus on the application that they're trying to provide.

[00:19:13] And so now they get to spend all their time on the F D D solution or the ESG reporting solution, rather than the headaches of the integration piece that they were previously charging the customer a one-time upfront v4 and putting a lot of time and effort into. Similarly, for the device manufacturers, it allows the device manufacturers to focus on providing the best sensor or the best.

[00:19:33] Device that they possibly can and not worry about all of the solutions that need to be built on top of it. I think that these things really drive forward the entire industry and, and allow all of us to stop thinking about that integration problem and just say it's solved by whatever ideal layer I chose along the way.

[00:19:54] Uh, the other things that I'll, I, I'll, I'll touch on that, um, I don't think have been brought up yet is the [00:20:00] alternatives to this are, uh, likely what Greg will talk about, which is this sort of utopian world where all of the vendors get together, all the device manufacturers get together and agree on a standard, uh, and then decide to move forward with that standard.

[00:20:15] Uh, and then also get to a point where they've actually deployed that standard broadly enough into the world that we can start depending on it. I think that it would be really nice if we can get there. Um, the reality is that making it through a standards organization is gonna take a couple of years.

[00:20:30] Getting vendors to adopt. It's gonna take a couple of more years and then getting it into the field. I think in the, in the response that I gave, uh, on the, on the Nexus Connect forum, uh, you know, I, I mentioned Backnet Secure, uh, which, you know, we're seven years post post standard, and I've yet to see it in a single building, uh, that we've, we've ever integrated with.

[00:20:52] So, you know, I am quite the skeptic that a standards organization is able to make it to the point where this type of thing [00:21:00] happens. Um, you know, in a past life, uh, I was responsible for Cisco's IOT standards efforts, uh, across about 20 different standards bodies. Um, I should send them all to you separately so you can run 'em over the screen.

[00:21:14] But it's everybody from I etf, I E E W, three C three gpp, uh, Laura Alliance, uh, some Ashray work, you know, on and on down the line. And I can tell you that the way that large companies view standards bodies as more about competitive. Advantages more about ways that they can keep their competitors having to play catch up with the things that they were already about to release, and less about actual standardization and interoperability.

[00:21:42] And so I'm quite the skeptic that, that, that mode is gonna solve this. Um, coming back to the independent data layer, I think I still have about a, about a minute. So, you know, the, the other pieces of it is you do really Yes. Really about the, the ease of implementation, right? So again, when you look at the number of [00:22:00] systems that are inside of a building, those systems, speaking protocols like Backnet, K nx, Modbus, Cbus, OICs, you know, on and on and down the line, every one of those protocols has slight differentiation between vendors and how they make use of it.

[00:22:15] And so you end up in this world, especially on older protocols like Modbus or Cbus, uh, where you're really doing a lot of effort to understand how that particular manufacturer implemented that protocol along the way. Giving that to a single party, giving that to a single party to let the single party deal with all of that integration, uh, is a huge advantage of having an independent day layer.

[00:22:37] I think the, the last piece of it is that the cloud connectors, which we didn't talk about earlier, but if you look at companies like Open Path or Matterport or Verge Sense or Butler, all of these companies are purely cloud-based. There is no on-prem protocol to integrate with. Every one of them has different authentication, different rate limiting, different back off algorithms that you have to do different ways of [00:23:00] getting the data and the formats of the data comes in and you really need to be able to integrate with those independently.

[00:23:06] Stop there.

[00:23:07] James Dice: I was trying to figure out how to mute you and I just didn't get there fast enough. I really wanna mute someone so someone next person to go over. I will figure out how to mute you. You, you got me. You got the best of me there.

[00:23:17] Shaun Cooley: You have to switch from chat to people and then it's, uh, mute per person on there extra.

[00:23:26] James Dice: All right. Thank you, Shaun. Greg, last one, we'll give you six minutes as well. Basically adding to Alex's points against the resolution. Also beginning to respond to Shaun and Silvia if you'd like starting now. Okay.

[00:23:42] Greg Cmar: Well, the first point is that every building needs data and you need to have the data to go along with the engineer has specified that these are the, this is the system that you're going to be looking at and these are the points of data that I [00:24:00] want, uh, be to be delivered.

[00:24:04] If we were designing the system from the start, we would say, oh, that's really great. Now where do we want that data so that we can just put the data there. No, no other ever other information. What I'm saying is the independent data layer is trying to figure out. A randomizing of the actual data storage so that you've lost contact with the building information model, with the energy model of the building.

[00:24:30] And you are now just working with what the, uh, contractor has decided that he wants you to see and how he wants it to look at it, rather than what the engineer wants you to see. The difference between the IDL and the, uh, you know, a, a system where the engineer specified it, uh, and was said, the customer will tell you where you store that information.

[00:24:58] And so [00:25:00] my concept is what's called a global key database so that you just have a global key so that the, uh, manufacturer, not every man, the manufacturers don't have to talk to any other manufacturers. They only have to explain their own piece of equipment. They are the ones that will identify and say that.

[00:25:19] When I'm looking at a VA V box with reheat and I installed it in 1970, this is the hardware, this is the make and model of the stuff that I have sent you. Just like the car, it's going to the car and saying, oh, my car's broken and I need something. And if you can only tell him the make and model he'd be able to find, you know, you need, you know, whatever.

[00:25:41] So that's the, the real key. The other side is that we're, as I say, we're already collecting three data points. You have a semantic ID that, uh, basically is defined by the b a s contractor that this is what you're controlling, a [00:26:00] timestamp and a value. All I'm saying is that you need two global points from that manufacturer who sold you the equipment, responding to the spec one for telling you what the system is that he was controlling.

[00:26:14] And the other are what each and every individual points are. At the end, what I'm saying is that when you buy your your system, you're also buying the data and you should be able to manage the data and do what you want with it, rather than, uh, having somebody come in and say, oh, well I'll, I'll pay, you can pay me and I'll show you how you can use the data that you already own.

[00:26:42] There's really no question about what that data is. You have a, a contract, you have a, a design document, you have a schematic that shows you where all the sensors are, and you have, uh, a set of specifications that are telling you [00:27:00] what it is that you were getting. So what I say is that if it's a Honeywell v a v box with reheat, then they will give you, here's what we call that, and they'll give you a number.

[00:27:13] And they'll tell, and these are the 12 points that are gonna be coming across. And they'll give you numbers for each and every one of those so everybody knows what they are. Then now, when you want to use the data, it's available. There's no question about what it is. Everybody knows and everybody can solve the problem after the fact.

[00:27:36] We just need the data attached to the building, not attached to a control system. And that's pretty much all I have to say. I can stop,

[00:27:49] Alex Grace: can I,

[00:27:50] James Dice: all right here,

[00:27:51] Alex Grace: can I, can I add on if there's still time in that six minutes?

[00:27:55] James Dice: Yes, there is. Uh, a minute and a half.

[00:27:58] Alex Grace: Great. So yes.

[00:27:59] James Dice: Great use of, [00:28:00] great use of time.

[00:28:00] Alex Grace: Perfect. Um, I just wanted to respond a little bit earlier to some of the statements, um, for me, Shaun. So I think there was a statement made, which was, you know, A common refrain, I believe, around this conversation, which is, Hey, look, there's all these applications spending all this money on integrations. We should simplify that and integrate once if that business analysis is true.

[00:28:26] Agreed. And my position I'm trying to bring up to the conversation is that in line with what I had mentioned before, starting with the business objectives that you want to achieve and the applications that you actually want to fulfill those objectives, and then determining how many of those integrations need to happen and are they actually the same overlap of data?

[00:28:46] And then what is the cost of that compared to an I D L? I think that's the, that's a reasonable analysis to take place. But starting with the, having that argument, land opre of that process doesn't work. Meaning it's not [00:29:00] true that the, that, that, that implementing an I D L is going to be cheaper than what, 15 seconds.

[00:29:06] So also I think there's a lot of s smushing together of data. It, it, you don't need the same dataset. So I haven't seen another application outside of F T D that needs to understand deification mode. So there's a lot of defining of requirements first and then figuring out the overlap. Second.

[00:29:20] James Dice: All right. Thank you alex. Well done, uh, wait to sneak that in. All right, so Kenny, this is where we, we make all of them mute each other and what we're, do you and I just need to go back and forth to figure out, okay, what are all the key areas that we wanna bring up and let them debate on next? Um, what do you got You.

[00:29:38] You go first.

[00:29:40] Kenny Trowers: Okay. So what came out from Team Pacific basically is that you need the idl, right? So you could have this plug and play to minimize integration and focus on the actual application at hand. I think that was kind of like the. Big point that jumped [00:30:00] out for Team Pacific, um, and for Team Boston, um, it was, uh, more business case driven to figure out what is it that you're trying to achieve and not, you know, implement an ideal just because, right.

[00:30:17] Just kind of like high level and they have some additional points. Um, but those are kind of the two, uh, major points I pulled out of it.

[00:30:26] James Dice: Yep. Yep. All right. I, I have a couple as well. So I think there's a, there's a question around, um, do, are the use cases that we're trying to enable here, do they require a another layer here or can they just be enabled with a sort of a full stack solution?

[00:30:45] It seems like what Alex is saying is, We might not need an ideal because it might be simpler just to hire one vendor to do this and one vendor to do that. And maybe that's all we need to accomplish here. So I think maybe we'll hit that one. I think that'd be good to sort of [00:31:00] clarify that piece. And then there's Greg's point.

[00:31:03] What Greg's saying is if, if maybe this doesn't need to exist at all because maybe this will get solved with a standard. Right? I think Shaun hit that a little bit, but I think maybe there's some more to explore there. And then I think, Alex, you, you, you brought up this point around the, the humidification mode, right?

[00:31:20] And the, this, um, you called it a nuanced piece, but I, I think it's a good point. Uh, we can't hide the fact that you're coming from Clockworks Analytics and you guys. Expect a lot of the idl, you're expecting the IDL to be very sophisticated for your application to sit on top. And I think there are other people in the Nexus community, if you look at the discussion on the forum, um, Josh from brainbox AI was bringing sort of the same point, which is owner, for me to be able to sit on top of the idl, I need a certain sophistication.

[00:31:54] I need this point to be exposed, I need to know which control points I can override, et cetera. [00:32:00] So I think that's another, that's the third sort of piece that I think we, we sort of should discuss here. Those are the three points for me. Kenny, does that sound good to you? Yep. Yep. All right. Um, and remind me what your two points, you had two points, didn't you?

[00:32:17] Kenny Trowers: Yeah, it was, um, ideal is needed for simple integration, um, for, you know, tenant spaces, plug and play, you know, approach. And then Team Boston was arguing, you know, You need the business case to sort of make your decision and not just implement an ideal.

[00:32:34] James Dice: Totally. All right. Let's start with with those two. Um. So I, I don't know that I heard the, and I kind of just want to have this be opened up so all you guys can unmute. When you're ready to jump in. Just please just raise your hand. Yes, let's do it. Sure. Good. Shaun, thank you for, for demonstrating when you're ready to jump in. Um, so the first thing I want to, to open up the floor on [00:33:00] is Silvia's point.

[00:33:01] So I think Silvia is saying, um, let me just try to restate what she's saying. A bunch of different stakeholders need the same data and I want to provide one user interface. Really the only way to do that is the independent data layer. Um, unless you just tell all these different stakeholders that, oh, you need to go grab that from Backnet and that from Modbus, good luck.

[00:33:23] Or you can say, I implemented this technology over here. They maybe have an API or they can do some sort of export for you, uh, there. So let's open it up. Can this be done? What Silvia is saying without the IDL is I think the main question,

[00:33:43] Alex Grace: can we first define, let's go a level deeper. So this is where the conversation often starts, which is like, everyone needs the same data. Let's go a level deeper and say, what data are we actually talking about here specifically? Let's also recognize that there are different sectors with very different needs.

[00:33:58] James Dice: Okay, Silvia, go [00:34:00] ahead. What data are we talking about here?

[00:34:02] Silvia Quaglia: Right, so when you go to the leasing agreement with the tenant, they don't say to you which data they want because they are not sure. They just wanna access to data, and you are signing on that. So the access come with something that is easy and therefore, because they don't really know what they're gonna have, you need to give a certain type of access and, and then you sign an agreement that say, Hey, you're gonna have this type of data for the H V C, this type of data, for the access control, et cetera, et cetera.

[00:34:34] Right? So of course you cannot do this by saying, Hey, you need to plug in your device, in our network to get this data from the, to get this data right.

[00:34:47] James Dice: Alex, go ahead. Response. Oh, sorry, Shaun, Alex wasn't raising his hand.

[00:34:52] Shaun Cooley: Uh, yeah, I, and I, look, I'll just add to that. When we say what points of data are we getting, it's, it's the, the goal is all of them, [00:35:00] right?

[00:35:00] We want all the points from all the systems, we want all the point names, we want all the configurations, and we want those normalized into a form that everybody can understand. And if you can take that a step further and model it so that, you know, the, the system from Honeywell looks the same as the system from jci, which looks the same as the system from Carrier, you're in a much better place to start making decisions about these.

[00:35:22] I think, you know, the, the, this, this question about which data should we collect? I think the inverse of that is when an F D D company shows up, or an energy optimizer shows up, their first step is always collecting data for a month or so before they come back and make suggestions about what you need to change inside of the building.

[00:35:42] You can eliminate that entire month if you've been collecting the data in advance and have it available.

[00:35:48] James Dice: Alex, you wanna respond before we move on?

[00:35:51] Alex Grace: Yeah. I would say when you drill a level down to say, well help, help me understand what data you really need and what are the reporting requirements around it, it ends up being a lot simpler than we're making out [00:36:00] to be in the first place.

[00:36:01] So when we're talking about E s G objectives, what are we talking about? We're talking about operational avoidable carbon related to E S G. We're talking about energy consumption. We're actually talking about a handful of data points. We're not talking about the position of the zone dampers. We're not talking about, you know, 5,000 of the points.

[00:36:21] We're talking about five or 50 at most of the, of the 5,000. So I think like we have to get into the specifics here, because without the specifics, you don't, you, you can't really justify the business case and you might be buying something that you think is unlocking value, but yet, if you haven't def, if you haven't defined first and foremost what it is you need to really get out of it, then you may waste a lot of money.

[00:36:45] On an abstraction layer that has not actually delivered value to the vendors that you think there's overlap with, of which there's not.

[00:36:51] James Dice: This is actually the point number two. So, yes, this is a perfect segue into our second point here. Um, Shaun, go ahead.

[00:36:57] Shaun Cooley: Yeah, I, I think I'll, I'll respond to that by [00:37:00] saying, you know, if you are the building owner and you have tenants asking you for data, uh, you've got one of two ways that that can go.

[00:37:07] Either you're going to put in engineering effort to collect and provide that data to them, uh, or you're going to allow that solution to unfettered access to plug random devices into your B m S network. I don't think that either one of those are a good, biting your tenants with access to data. The third one is you have an independent data layer that you've already moved the data into.

[00:37:30] It moves in real time into a cloud solution. And you can grant access to those tenants, that data with a single click, rather than allowing them to plug something in or putting your own engineering effort into it. So I think that, you know, if you're looking at it from Dexus point of view or any other building owner's point of view, having that data in a centralized repository where you can start to push it off to those tenants without them interfering with your operations, becomes a pretty big deal.[00:38:00]

[00:38:01] James Dice: Greg?

[00:38:03] Greg Cmar: Uh, one of the other things that I, I thought you should know that data is important to your tenants, but it's also important to a lot of other people. If we had all the data that was available that you could access, uh, You basically have a complete energy audit.

[00:38:20] James Dice: So she's saying I want to provide it to my tenants, but that's not the only stakeholder.

[00:38:24] I also wanna provide it to the sustainability folks and to whoever else. Exactly. I feel like you're arguing for her there.

[00:38:31] I

[00:38:31] Shaun Cooley: do, I do actually wanna,

[00:38:32] Greg Cmar: although I'm saying that the data should be there,

[00:38:34] Shaun Cooley: I do just want to take a quick stab and say, I don't think Greg is too far off of Team Idl here. Alex, Alex is pretty far off team idl, but

[00:38:42] James Dice: Yeah,

[00:38:42] Shaun Cooley: but I think I'm,

[00:38:44] Greg Cmar: to tell you the truth, I, I have built,

[00:38:45] Shaun Cooley: yeah. I think that the, you know, when, when I hear Greg talk, I think we're in, we're in like violent agreement that there should be a centralized repository of data and that all the data should be there. You know, as much as you can collect for as long as possible. I think the big difference is Greg says there [00:39:00] should be a standard that makes the manufacturers provide it in an understandable way.

[00:39:05] And you know, as an I D L manufacturer, I would love that future. I, I think that would be fantastic if that exists because. It takes a lot of the work off of companies like ours, but you know, in the meantime, the IDL is the way to get there now versus waiting for a standard. But otherwise, I think we're roughly saying the, the same things.

[00:39:27] James Dice: Silvia, did you have your hand raised?

[00:39:29] Silvia Quaglia: Yeah, yeah. Sorry, my point was actually that the same things of Shaun, uh, like I would love to not have to build an idea because to be fair, it's a lot of money for a building owner, right? Like you have to have the standard right approach. You have to like keep like doing it.

[00:39:46] You have to decide at the points that you're gonna to integrate, et cetera, et cetera. So it's an additional cost for the building owner, but the standard that you are talking about, Greg is not there. Like, I'd have to solve a problem now, not in 10 years when everybody will agree on [00:40:00] something.

[00:40:00] Right?

[00:40:01] James Dice: Uh, you also had your hand raised, Alex, go ahead.

[00:40:04] Alex Grace: So if your use case is I need to give all my tenants all the data, I'm not sure why, but that's what they want and it's contractually obligated to do so that is a great use case to have some kind of layer that provides that. There's zero debate there. That is a very small subset of the market. There's a lot of building portfolios that do not have that use case.

[00:40:26] Um, and starting with the idea that before here, here's why I care about the topic. If you go ahead and implement an IDL and spent years or a long time and a lot of money trying to accomplish that, to meet a use case that is unclear, whether it's really necessary, it can hurt the overall smart buildings effort of your organization.

[00:40:45] Smart Buildings programs needs to produce business value within portfolios. To not be canceled and have those budgets completely cut to not revisit smart billings applications in the future because that bridge has been burned internally. [00:41:00] So it's very important to make sure that we're trying to accomplish those objectives.

[00:41:03] Not everyone has that. And then secondly, you know, within f D D, someone can have access to the data very easily. Like once you have to, you have to restore the data again anyway on the application layer. So there's not really a redundancy there that you're solving and users can have access in that way.

[00:41:19] There's a third point, which is around authentication, which is making sure,

[00:41:23] James Dice: which is, hold, save the third point.

[00:41:25] Alex Grace: Okay.

[00:41:25] James Dice: Um, you guys wanna respond to those first two go?

[00:41:29] Silvia Quaglia: Uh, point, the point that I'm making is not that just only the tenants need that there, there is a lot of other different stakeholder, right?

[00:41:36] So you are just, uh, like providing an easy way for different stakeholder, which are not just only tenant. Therefore, for each building there will be a lot of different stakeholder that needs stay, different data. So it's just an easy way for them to get it. Without spending a lot of money for different third party application to go there and plug in their device and get the data that they want.

[00:41:59] Shaun Cooley: Yeah. And, and I'll, [00:42:00] I'll just, um, since I know Alex can't hear, uh, I, the, the point was that you don't know in advance what these tenants are asking for, right? So collecting all of the data, making available as needed can save you a lot of time and effort down the line. You know, the, the other thing that I'll add is you, you, you've touched on a point that it takes years to implement an I D L, uh, and substantial cost.

[00:42:23] Um, I know I'm not supposed to pitch my own, so I'll talk about Nove Buildings, iot, and Code Labs. Um, it doesn't take years. It, it doesn't even take months and it doesn't cost a lot of money. They're all very, very inexpensive and very easy ways to get in between, uh, you know, the, the device manufacturers that you have today and the system integrators, if you wanna talk about expensive.

[00:42:48] Wait until, you know, Siemens or Schneider sends you a bill for adding two extra backnet points to, uh, to a backnet system that you want access to. Uh, you know, when you get that bill for $400,000, [00:43:00] uh, that, that's expensive, right? So getting in there in advance of that and collecting all of that data is a very attainable goal these days with the ideals that exist out there.

[00:43:12] James Dice: Um, Alex, do you have a third point? I wanna move on if you don't.

[00:43:17] Alex Grace: Uh, the third point is just around authentication of just really thinking through how you make sure that the users or the tenants in your building actually have access to only their, their data and not others, which is not a trivial point that has sufficient amount of complexity involve.

[00:43:32] James Dice: I'm gonna set that aside due to time. I think that's a product specific thing. Different products can solve that in different ways. I really want to hit something that we haven't gotten to yet, which is Alex, your point around, um, these different sort of, you said nuanced, you said, um, humidification mode, these basically these higher levels of technical detail that certain applications need.

[00:43:55] And I think if I sort of summarize what you said earlier, Alex, [00:44:00] you're saying, um, we've come across buildings that have had IDLs implemented and again, big like diversity on what certain IDLs are capable of. Um, you've come across buildings where what was in the idl and mostly talking about data modeling here.

[00:44:15] So context around the data and the I D L, that wasn't enough context for you guys to do what you need to do. Um, again, brain box. Josh from Brainbox has brought up a, a similar point, not enough data in this i d l for us to, you know, get anything useful out of it. So we basically need to start over to enable our solution at the application layer.

[00:44:37] I'd love for the I D L team to respond to that. Right. What have you seen, what is, what is your response to that sort of skeptic argument?

[00:44:50] Shaun Cooley: I did take that one. Go ahead, Shaun. I think you know, the, some, somebody probably mentioned it in the chat already, but the xkcd, you know, the standards organization doesn't do what I need.

[00:44:59] I'm gonna [00:45:00] go start my own standards. Organization is the, the same sort of concept as, as what you just described. You know, when you come in and there's an I D L layer that does 90% of what you need, but it doesn't have that last 10% deciding you're gonna throw the whole thing out and start from scratch again, just doesn't make any sense.

[00:45:18] Right. It, it would make a lot more sense to go in and. Try to extend and work with the ideal layer to say, Hey, to put f d d on top of this, here are these extra things that you guys aren't collecting or you're not modeling properly that we need from this layer. And again, now you're enabling this layer to do more and to be more robust along the way.

[00:45:38] But throwing it all out and saying, I'm gonna go spend three months doing my own integration, uh, just, just seems like a, like a really bad approach.

[00:45:46] Alex Grace: The, the issue is I'm not, I guess I'm not clear on where the solving of that integration is in the first place. Like the applications still need to integrate to the I D L, which is going to have particular nuances, [00:46:00] and I'm not sure where the efficiency really is gained from what you're describing.

[00:46:05] Shaun Cooley: Yeah. So, ha, happy, happy to respond to that. So the IDLs are including data normalization. They're not just data lakes. That are taking all the data out of the building and shoving it into a, a big, you know, I, I joke Data lakes are actually data swamps, but they're not just shoving it into that, that one spot in the process of normalizing the data.

[00:46:27] It means that the f d D vendor only integrates into the a p I once. They don't have to customize that integration for every single building that they go into. It's the data layer that's taking on the effort of taking each of those unique snowflake civil building and turning them into the normalized form that the data layer is producing on the API side.

[00:46:47] Alex Grace: Yeah, so I think, I think part of the challenge of this conversation is, is, you know, we're obviously not here representing the, um, individual applications that we both are a part of. Um, there's a bit of a chicken or the egg situation, right? So, [00:47:00] you know, certainly I've seen if, if someone has an independent data layer, if, if.

[00:47:06] Tens of thousands of buildings had independent data layers with one point of API integration to perfectly normalized data. Obviously that would be fantastic. Right. Um, on the flip side, I've seen a lot of owners that have thought they were b, that they were buying something that would create efficiencies for them with lots of applications that ended up not doing that because each application still needs to store the data, needs to have their own data models to be able for their application specific.

[00:47:36] And there often isn't that much overlap. Like what security data or a work, a workplace planning space planning application needs versus what F D D needs is very different data and it's different types of velocity of data. It's different types of normalization that is needed. It's not the same thing. So there ends up being a lot of duplication anyway.

[00:47:57] So it's a bitter chicken of the egg that if someone has already gone down this [00:48:00] direction, that of course is great, but I do think it does a disservice to the overall industry by leading people to do this without thinking through all of the trade-offs.

[00:48:07] James Dice: Silvia, go ahead.

[00:48:10] Silvia Quaglia: So just, I just wanna understand, so how you gonna to, um, like so you are a solution is that I plug in a data integrator for each third party application that needs their data in a building.

[00:48:23] How it can be, uh, like a, a, a great, uh, intelligent approach for a big portfolio.

[00:48:30] James Dice: And Shaun, go ahead. If you wanna add onto it.

[00:48:32] Shaun Cooley: Yeah, I, I was just gonna also respond to Alex in, in saying that, you know, this, this, um, our view of the world is that each of these companies has chosen a select few number of points that help them to deliver their business today.

[00:48:47] And I think that it is a very short-sighted view to say that there are no other points in the building that would add value to that solution. And so, you know, when you are looking at, uh, you know, just the, you know, [00:49:00] that the certain components of the HVAC system and not other ones, or you're looking at the components of the HVAC system without also looking at the access readers or the occupancy counters or the, the footfall traffic monitoring, I think that you are artificially.

[00:49:14] Narrowing the scope of the solution that you deliver. And it's not to say that it doesn't provide value to customers in that, in that, uh, you know, who is you here as Species or Royal U? It's F D, D, yeah. FTD solutions. Any, any solution. ESG reporting, Ft. D um, you know, energy optimization, all of these, you know, the, the businesses are built on, you know, we've found this particular set of data that we can pull outta the building and use to produce our, our output.

[00:49:39] And we've found customers who are willing to pay for that output. But I think that once you settle in on those, uh, uh, sorry. Again, once these vendors settle in on that narrow set of points, it becomes very hard for them to open up their scope. And so this argument that you don't need those other points, I don't fully buy because I, I think that you can get better data and better [00:50:00] visibility by seeing everything.

[00:50:01] James Dice: All right. Last, last thing, then we're gonna kind of head towards conclusion slash opening it up for the audience.

[00:50:08] Alex Grace: I, I think, I think what individual f d D applications need is a separate discussion outside of the IDL debate. Uh, I think that occupancy counting point should be built into the b a s and not its separate siloed data point anyway, but that's a more nuanced point.

[00:50:22] In general. I just want to course correct on a couple things I've mentioned so far that I think are some false assumptions. You don't need to necessarily have different integrators involved. You don't need to necessarily have hardware involved. Um, F D D doesn't require hardware, uh, data to come. And then defining what is the data specifically.

[00:50:40] So if we're talking about H V A C data, I don't, I don't see why there would need to be an integrator involved at all.

[00:50:46] James Dice: Okay. All right. So we've hit the points that I wanted to hit. I think we'll call a winner here. Kenny and I are gonna decide the winner here, and then we're gonna let people come in from the audience and ask [00:51:00] questions.

[00:51:00] So we have about 20 minutes left. Um, Kenny, I'm gonna open up the floor to you. What are your reflections here? What have you learned today? You were the judge. I've, I've already declared myself as biased. Um, I'll say my piece, but you go ahead. You're the neutral judge. What, what have you learned? What, who do you think won the debate here?

[00:51:21] Kenny Trowers: So, I think, you know, great, great point from both ends, right? Um, but I would say that, you know, team Pacific kinda came out punching, uh, with the first argument. And, and stayed afloat. And, and I also got, you know, Alex's point, right? But I, I looked at it in two different categories. I think implementing an ideal is probably great for a developer tenant space, but from a owner occupied, um, maybe you need to understand what the business case is or the objectives, [00:52:00] right?

[00:52:00] Because then, you know, but from a developer standpoint, you know, how do you plan for the next generation of tenants, right? Where it, everything's gonna be driven by data. Right? And just an example, you know, today I'm hearing clients saying, Hey, I want to be able. To provide data to my clients, to my tenants in the future when their lease is up.

[00:52:24] Right. So how do I do that? Right? Um, so, uh, I would say a lot of the points, uh, arguments that came out of the Team Pacific, um, kind of pushed me over the edge to say, Hey, you know, we, we may need an ideal sort of as your plug and play. Um, so that's, that's kind of where I stand.

[00:52:46] James Dice: Yeah. And, and I, I think as Andrew just said in the chat, I think it's a little bit difficult to, uh, Call the, the skeptic team losers.

[00:52:56] I think having Silvia on the team allows a, you know, a certain [00:53:00] sense of truth to be happening when a building owner says they want something. It's pretty difficult to argue with, uh, you know, you don't want that thing Mrs. Byer in this case, right? So it's a little bit difficult to argue with that. So maybe in the future we'll have, try to make it a little bit more balanced.

[00:53:16] Uh, yeah. And Alex did a great, you guys did a great job of coming up against that obstacle, uh, with flying colors. So, well, well done to everyone. I'm not changing my position based on what we've talked about today. Just for the record, uh, my t-shirt that I'm wearing is still, still, uh, still gonna keep wearing it.

[00:53:35] Shaun Cooley: I was expecting you to rip it off Holk Hogan style and have another shirt underneath it,

[00:53:39] James Dice: another shirt that says IDL was a winner or something. All right. So thank you to all of you. It's been, this has been an awesome first debate. Uh, it's really fun. A lot of, lot of comments in the chat about how great this is.

[00:53:53] Let's keep it going for a little bit longer. I would love for an audience member to attempt to come on stage with us. [00:54:00] So I guess whoever can figure that out first. I don't see anyone yet. Oh, Andrew Rogers. I could have predicted that first person. All right. We know whose team this guy's on.

[00:54:15] Andrew Rodgers: I'm just doing this cuz Rosie said I had to.

[00:54:18] Um, so I think one, one topic that didn't get covered that I'm curious, uh, everyone's perspective on is, uh, I think we could argue a lot about data modeling and whether or not you can build a rich enough model for any particular application, whether A F D D or ASO or whatever, tenant experience, et cetera.

[00:54:39] Um, but ultimately, If you do do point-based, like point system-based integrations, there's cost to maintaining those integrations. And I much prefer the idea of like an IDL in a building times the independent systems in that building versus every [00:55:00] independent system that needs to connect with another independent system.

[00:55:03] James Dice: This is a great question.

[00:55:03] Andrew Rodgers: Time in the building.

[00:55:05] James Dice: Yep. This is a great question. Anybody wanna respond to that? Go ahead, Shaun. Yeah,

[00:55:11] Shaun Cooley: I, I think it's a really good point and you know, it is it, it goes beyond that where when you start looking at who's updating those individual connections, who's maintaining the security for them, who has access to them, where is your data going?

[00:55:25] Who's storing that data long term? Where are they sharing it? You know, what country is the data stored in? You get into a lot of questions that are really, really hard to answer and I think you can go. You know, back to like Target getting hacked, you know, that came in through an HVAC vendor having full access to A B M S network and they were able to, to leap off and do, do something else with it.

[00:55:46] You know, you, you start to put yourself in a weird spot when you have all these vendors. And the last, the last part of it is when you terminate the contracts with those vendors, uh, somebody's gotta go in and find any boxes that they put in or any [00:56:00] additional logging they turned on, uh, trend logging, they turned on in your BMS network and turn that stuff back off and take those boxes out.

[00:56:07] And a lot of times they just get left behind. Um, you know, not, not to ever be seen again until somebody bumps into it years later.

[00:56:16] Alex, go ahead.

[00:56:17] Alex Grace: Can I just wanted to ask if, you know, sometimes these conversations aren't, I d l are very theoretical, so many to many, what, what are we talking about?

[00:56:26] James Dice: Many to many different, I think Andrew used the word point solution so. FTD over here. Energy management over here, tenant app over here. Um, what else do you guys have? Silvia, uh, name your other apps over cmms.

[00:56:40] Alex Grace: My core question though is again, what percent of data between those applications is actually shared?

[00:56:47] Shaun Cooley: No, I,

[00:56:47] Silvia Quaglia: between this application,

[00:56:49] Shaun Cooley: but I, I don't think that was Andrew's point. Andrew's point many to many between the applications and the devices.

[00:56:54] James Dice: Devices. Also,

[00:56:55] Shaun Cooley: each application is, is talking to a bunch of devices along the way.

[00:56:59] Alex Grace: What, what [00:57:00] devices? I mean, F D D talks to the BAS.

[00:57:04] James Dice: Yeah. And, and what else Talks to the BS supervisor? Control talks to the bs. If there's meters on the bas, somebody, somebody else is talking to the bas. So he's talking about just with hvac you could be, the HVAC system could be talking to 5, 6, 7 different applications, tenant SUBMETERING data.

[00:57:22] So that's just one to many on hvac. Silvia's talking about,

[00:57:26] Alex Grace: right. But the amount of data you're talking about on meters, you're talking about a handful, a subset of points that meters overlap with FTD and a meter analytics. If those are two different solutions, they're probably not, they're probably the same solution.

[00:57:39] Automated supervisory control is typically writing back to a handful of set points.

[00:57:43] James Dice: Yeah. And Silvia talk about how many different applications you're, you're talking about here, because this is, I feel like this gets narrowed in to only a few applications, but you guys are trying to enable much more it seems like, and I think a lot of building owners that [00:58:00] go down this road, Are, are not such,

[00:58:02] Silvia Quaglia: so, but I cannot really talk directly about, uh, like, uh, who I'm working for.

[00:58:08] But I can tell you that definitely there are not just one user cases, like, uh, we are talking about the asset. Uh, like also, uh, we have like a hazard register to be maintained and that like, you know, that part, the, the MAC address and all this stuff, like the device, um, uh, device details need to be sent out in a certain way.

[00:58:29] We're talking about the, like the network, for example, we need to maintain the network in a certain way and we want to see, for example, if the camera, the CT camera are active or not, et cetera, et cetera. So there are application for each different type of things. Um, we are having, obviously having analytics, we have ftd.

[00:58:47] Uh, there are like the, as I say, the tenants that needs access to data for ESG reason or whatever other reason. So it's just like a enable also for the future. It's not just only right now what we [00:59:00] have. It's, uh, what we are going to have in five years time.

[00:59:04] James Dice: Um, okay. Jim Meacham is coming on to ask our final question.

[00:59:09] Jim Meacham: Awesome debate. Thank you guys. I guess, like I put this, uh, alluded to this in the chat earlier, which the chat was super fun by the way too. Um, you guys that are up there debating may have missed all the nice quips. Uh, but, uh, what, you know, what really makes it independent, right? Like I think that's a key element of it because is it, is it only if you don't add any application value that you are truly an independent data layer and the moment you're trying to actually do something with the data, you're no longer independent? I'm curious about that designation of independent data layer, cuz there's lots of data layers.

[00:59:53] James Dice: Shaun?

[00:59:53] Alex Grace: Yeah.

[00:59:55] Shaun Cooley: Yeah. So, you know, our, our view on this is that the BSG dashboards are [01:00:00] reporting. It's no longer an independent data layer, an independent data layer. You cannot be directly associated with any of the device manufacturers, and you cannot be directly associated with the solutions that are built on top of it.

[01:00:11] I think the moment an independent data layer starts providing f d D or starts providing, um, you know, ES or now it's just a solution that has APIs that allow others to access that same data. And so to truly be independent in there, it is, it is literally just the data layer. Um, with, you know, normalization is okay, some enrichment of data is okay, but you can't provide, you know, analytics, dashboards and, uh, and, you know, changes back into the system.

[01:00:40] James Dice: Before we go to the other side, I'd love to hear Silvia's take and then we'll go to Alex and Greg and this will be our sort of last question

[01:00:47] Silvia Quaglia: here. Well, it dependent for us means that it's obviously independent from, uh, the device level and the application level, but also it also means that we are not stuck with the one vendor solution.

[01:00:59] [01:01:00] So, um, as you might or might not know, but like a building owner hated to be stuck like just with the one provider for this, for one things. So what they tried to do is obviously to use, uh, the most open source solution possible, uh, and also a solution that doesn't allow them to sign for 10 years or whatever long it is, uh, to one provider.

[01:01:25] So independence means both that you are independent from different contractor, that they are, like BMS contractor, access contractor, whatever, independent from the application. So not an FTP solution that is providing the I D L. And also that it's, uh, independent in a way that is open source and that you are not stuck with a solution that it's from a certain just type of provider.

[01:01:49] So it can be supported by just that provider.

[01:01:51] Jim Meacham: Yeah. So if it was a software that was multi-vendor, right? Like, so if it was a piece of [01:02:00] software where multiple vendors could support it, then that would qualify, I think, based on what Silvia was saying. But if it's a, if it's the software's only supported by the vendor, cuz then you have this issue of the, the integration is the hard part and all the labor associated with integration and you can't just rip and replace that, right?

[01:02:19] Like you'd have to redo all of that integration work to move between vendors with most solutions. Unless you had a like open source being kind of like multi-vendor kind of maybe, uh, the analogy is like Niagara of IDLs, right? Where anyone can work inside of it. Um, and you could, you could own it and move between, I don't know if there's anything like that that exists. It'd be interesting to know.

[01:02:44] James Dice: Kenny, were you, did you have your hand rights?

[01:02:45] Kenny Trowers: Yeah, I mean, he, he covered the point. I was gonna say, you know, you kind of have it in your, uh, newsletter, right. Is the ideal, the new vendor lock-in. Right. And, and maybe there is, is there a Niagara version [01:03:00] in the ideal space? I'd be curious to, to know that.

[01:03:04] Cuz that was always a concern, right? In their lockdown.

[01:03:06] Jim Meacham: Yeah. More software. Yep.

[01:03:08] James Dice: Okay. Because of the time, we're gonna do closing remarks from everybody on the debate team. Thanks Jim. Um, Greg response and closing remarks.

[01:03:20] Greg Cmar: Uh, the independent data layer from my point is that the provider doesn't, the provider of the independent layer doesn't cross silos.

[01:03:32] So if he un uncovers a, a problem that he finds, The I D L vendor can't go fix the problem because that will, uh, put them, uh, in conflict with the people that are providing the hardware and the application. So that's what I think of as independent. It's you're only providing information, you're not providing any of the services necessary to correct any of your problems.

[01:03:59] James Dice: [01:04:00] Cool. Shaun, you're next?

[01:04:02] Shaun Cooley: Sure. Yeah. Look, I, I think just. Briefly touching back on the independent part, uh, you know, one, one of the keys of an independent data layer is being able to swap out devices under, underneath without breaking the applications that are above it. So, you know, if five years from now an even higher resolution occupancy tracking system comes into play, or, you know, an even better air quality monitoring system comes into play, you should be able to plug those pieces in without, uh, without breaking the, the setup.

[01:04:29] Uh, and that was just, uh, you know, Jim's last point about, uh, the, the vendor and, and, uh, ideal, uh, integration. Look, I, I think this was a, a, a very good, uh, first debate despite the technical issues. I, uh, I'm gonna, I'm gonna leave saying Alex is, is still not, uh, a believer, but, uh, Greg is, is not too far off.

[01:04:51] Um, you know, and, and I will, I will concede that if, uh, Greg's, uh, future comes to be where we have standards that define all these things and they [01:05:00] self-describe and show their data somewhere, we're all in a, in a much, much better place. Um, so, you know, no, no argument from me on that. I just think that there's a, a, a many decade gap between where we are now and that actually being widely deployed.

[01:05:15] James Dice: Silvia next

[01:05:16] Silvia Quaglia: for me, like what you would say It's true, it's correct that probably Greg is more on our side. Uh, Alex it's is I think that Alex is also on our side, but it's just the saying that. To enable an ideal, you need to have some good business cases instead of enabler everywhere. Uh, which is kind of correct, like, uh, it just solves some of the issue that we are having to share to enable data to be available for everyone.

[01:05:44] But it's not that it's worth it everywhere. Uh, sometimes like, uh, you don't need it and it's just a cost for the building. So it's true that you need a business case to do that because it's a cost, an additional cost. So I feel that we are all agree that the ideal is [01:06:00] solving an issue right now of our industry.

[01:06:02] And, uh, it, it's not gonna be forever. Hopefully, uh, hopefully we will find the standard that Greg is saying, but at the moment, that's what we got and that's what we have to play with.

[01:06:12] James Dice: Thank you, Silvia. Alex, what is your, your closing statement?

[01:06:16] Alex Grace: I think I'd like to reframe, right? I think this is not a Holy War.

[01:06:19] So belief, not belief doesn't make a lot of sense. It's about business case. And does the business case actually drive the cost benefit that we think it does? Needs to be looked at very carefully on a case by case basis. It's not always the same. It depends on the objectives you're trying to accomplish.

[01:06:34] It depends on the applications that are gonna serve those. And a lot of this discussion is focused around commercial, multi-tenant office space. There are a lot of buildings over a hundred thousand square feet that are not multi-tenant commercial offices. And I wanna represent that side of the market in this discussion because a lot of the, the use cases that we've defined or talked about needing so far are simply not relevant for a university or a pharmaceutical manufacturing portfolio.

[01:06:59] Um, so there's [01:07:00] real differences here. And the point is not to say good, bad belief, not belief, but simply to say, here are the practical considerations. Start with that business value. Work your way backwards, not the other way around.

[01:07:10] James Dice: Absolutely. Yeah. I think what we've done here is we've, we've come up with a bunch of nuance around.

[01:07:15] It's not a, it's not a yes or no or a Holy War, like you said. It's a, um, this is a tool that we can apply in the stack. And just because James wears a t-shirt to make a joke doesn't mean that everybody out there needs to go do this thing right now. Um, any closing remarks from you, Kenny?

[01:07:32] Kenny Trowers: No, I mean, great discussion. Um, and at the end of this, it all seems like we're kind of going towards the ideal, uh, story, but just from different angles. So it was great. It was fun.

[01:07:46] James Dice: Yes. Thank you all and, and thank you all for bearing with us. We'll try again next month and, uh, thanks everyone to the audience. I've been distracted by the chat most this whole time, but the chat's been hilarious.

[01:07:59] [01:08:00] Hopefully there's a way we can download it and, and revisit it all later. But thanks everyone, and uh, see you all later.

[01:08:10] Rosy Khalife: Okay friends, as we're trying these new formats, please let us know what you think in your podcast player right now, or on the episode page on our website. There's a link to a survey for this specific episode. We'd love to hear from you and we wanna hear your feedback. Also, don't forget to sign up for the Nexus newsletter or invite your coworkers and friends with a link below.

[01:08:31] Catch you next time.

‍

‍