#163 What happened at Haystack Connect?
Show notes
If you missed the recent Haystack Connect event, no need to worry. The experts at J2 Innovations, including Scott Muench, VP of Customer Experience, and Gareth Johnson, Chief Software Architect, have got you covered. They're sharing all the highlights and energy from Haystack Connect, considered the most important event in building data modelling. You won't want to miss out on their insider insights!
Transcript
Hello and welcome to Buildings of Tomorrow. My name's John Lester, and in today's episode, we are talking about Project Haystack. We're actually continuing a conversation together with Gareth and Scott around Project Haystack and Haystack Connect. So, as I mentioned, I'm really lucky to be joined today by two special guests, Gareth Johnson, the Chief Software Architect at J2 Innovations, and Scott Mooch, the VP of Customer Experience at J2 Innovations. Gareth, Scott, thanks so much for joining us. Hello, thank you. Thank you very much. We had a great conversation not long ago where we really talked about Project Haystack, and you introduced really clearly some of the, the value, some of the, the secret source behind Project Haystack, what it is, and, and importantly, Scott, what it's not, um, and, and talked a little bit about what that enables us to do. For anyone out there that had the opportunity to listen to that episode, I'm sure they're excited to understand what's new, what's the latest with Project Haystack, because these, these projects, these models, these, these collections of technologies don't happen overnight. It's a community that takes time to really build and grow. Uh, and recently we had, Haystack Connect, if I understand correctly, and I wanted to ask you both, and maybe uh Gareth, we start with you, to give us quickly your, your impressions, your thoughts, how you felt uh during that conference and during the conversation. Oh, I absolutely loved it. I mean, it was a wonderful conference. I attended during the time of COVID, I attended a virtual haystack connect, but there's nothing like being there in person. And I think with like what's really important with an open standard like Haystack is seeing the activity behind it, you know, because you can have open standards, but they fizzle out or there's not much to them or, you know, there's not much activity around them and stuff. And I think what we saw with Haystack Connect this year, it was some some real innovation, um, some real activity and buzz. And you know, a lot of motivation to to improve the standard, to develop new tools around the standard and really move us forward in a very constructive way. And I, you know, even to the extent where I kind of saw, saw different people trying to, you know, there's some various issues. Issues that people are trying to resolve. And they, there was, you know, there were some very, very healthy discussions in perhaps different perspectives as well on how to solve some of those issues too. And I find that debate very engaging and I thought that was a, shows the hallmarks of a very healthy and lively standard um and, and that people are innovating upon. So I was really, really pleased with it. Amazing for you. Yeah, for me it was particularly kind of memorable in the sense that it's 10 years, uh, Project Haystack was founded over 10 years ago and I was at that very first conference and I remember using words like imagine and in the future, uh, we hope for and things like that. And now you fast forward to 10 years later, and one of the messages was haystack in the real world. And session after session after session was use cases and products and services and people saying how they're leveraging it. It was really neat to see that maturity and that penetration into the ecosystem and the creative ideas that people are using. Uh, I just came away with, wow, this has got a lot of really solid traction. There are people that are building entire careers as well as products and companies on it. Um, I, I remember one that was very memorable where there was actually a company, um, almost like a religion, uh, a, a company standard, if you will, that everything they do will be based on data, and that was like the unifying factor and then all the other things around the company were built to support that, that, uh, that central focus, so that was just amazing to see. Um, other things that I saw was some of the energy looking forward, which is some of the topics we can talk about today, um, that. It continues to evolve and it continues to evolve in a really positive way. Perfect, hey, from both of you, I hear a lot of excitement, I, I hear a lot of energy, um, and, and as you mentioned, Gareth, uh the, There are many standards, and there have been many standards, and, and some of them thrive and flourish, and some of them fade and die, let's call it, to, to put it bluntly. And it's great to hear that the, the Project Haystack and the haystack approach is not one of those. Uh, you're really starting to become real, as you, as you mentioned, Scott, and really starting to find traction and people building careers. This is a, this is a quite a, a revelation, let's call it, that something is really becoming. A a powerful force. Now, as you mentioned, Scott, excuse me, uh, something like Project Haystack is never finished. It continues to grow, it looks towards the future, and there's a couple of projects that I know both of you found a lot of energy and excitement in, uh, and that we wanted to talk a little bit about today. There's Building Motif and Zito. Let's start with Zito, I think Scott, uh, something that you particularly connected with during that conference. Yeah, indeed, because kind of my, my background as a system integrator role was about trying to get jobs done in an efficient way, get them done accurately so there's no mistakes, and standards help because it got the devices talking to each other. And in the early days, we, you know, used naming conventions, but boy, standards have taken it to the next level with what we call templates, for instance, in the, in the FI framework, and. As we look at the Project Haystack relative to that focus in Haystack 4.0, there was a concept called protos or prototypes. And so imagine a model of a VAV. is now well defined and it has these points, discharge error, damper, actuator, hot water valve as an example. So we all agree to agree not only to the standard of what metadata gets added to those points, but now how they get assembled into a bigger entity. And this is really significant when it comes to actually creating solutions that are easy to use and not time consuming. And it also opens the door to democratization, which we talked about in the earlier episode, where you have this kind of plug and play because those templates enable that. And what this leads up to is the next level of that concept, which is Project Zeto. The idea there is, there's a concept called SECs, and a spec is really kind of the next generation of a prototype. Where not only is it defined in the standard, but it's actually defined in the programming tools, so there's actually a reference design that can be extended through various languages, and so if you're a developer or if you're a user of a of a tool set, you could actually grab one of these specs from a library called VABs, and it would have inheritance all the way back down the tool chain to the standard, so there's no. Um, room for interpretation, it's, it's always going to be right, which is just really cool. And then the other side of that, which is probably the more important rather than the technical, is the business side of it. These, um, these libraries of specifications would be readily available to be used in validation tools. So part of Project Zeto, uh, which is funded by the Department of Uh, Energy here in the United States is to come up with a validation capability for tagged and semantic, semantically tagged data. And so imagine you you're able to. Engineer job using these templates, so generally it's going to be almost 100% right. Uh, if you for some reason mess something up, you would run it through this validation software, and it would say, hey, you forgot this tag, Scott, or this relationship needs to be reinforced. And so, In my mind, it's kind of like where we are going with some of the device standards like BTL verifies that this device talks to this device, but when we get to the data model, what we're talking about is this data model talks to these applications in a standard way, and now we can validate it. And so stay tuned, it's a multi-year project, and we're just kind of getting started with the first kind of iteration of the technology, but I think there's room for excitement there because it's a natural evolution of our industry. Uh, and, and quite an undertaking, I imagine. Yeah, I was, um, Zito is, um, Scott gave a very, very good introduction on what Zito is there. And I think, you know, the two main takeaways from it are it enables you to elegantly define a shape of something, you know. So if we're talking about that VAV, we're talking about all the, you know, the inputs and outputs, everything it's got connected to that VAV and we're talking, we're describing a shape of something. And once we've been able to describe a shape of something, we can, we can validate that data and data validation is. It's critically important. It's the thing when you've, when you've got something like Haystack or or any, any data model, to be honest with you, you, you need to make sure that the, the way those tags have been used, that they are being used in the correct way and there's nothing missing because if there's something missing, it could cause problems if you're running it like a controller. Algorithm against something to solve a particular problem. That data is important. It needs consistency is really important. So forming shapes and validating data that's there so you have a consistent data model that is not full of errors and so forth, not missing anything. It is really important and that's what Zo really is because that, that it is difficult, you know, as, as you know, you need a, you need a good set of tools and technologies, uh, where, you know, even if you're a software developer, you know you need some way to be able to visualize these shapes, so. So Zito kind of helps address that as well. The, the, um, the specs that you were talking about there, Scott, they give an elegant way that you can kind of define what these shapes are. Um, that's something I haven't seen in haystack before, to be honest with you. The Haystack 4 had something, uh, has something called depths. As well. It's a little bit more difficult to use when defining a shape of something. It's, you know, like when J2 when we work with a new customer, customers, they want to be able to model different things in different ways sometimes. So, um, you know, having a way, a better way to be able to do that data model modeling and validation is something that Zito provides. So, again, it's very good. Um, again, you know, hopefully to, you know, your, your SIs and Out there, this will be something which is just the tools that you utilize use. Again, it's a technology that you don't necessarily need to go out and understand yourself, but you will be able to understand the benefits by having really first-class templating, so you can engineer jobs easier and quicker and also having validation tools to make sure that all the data that you've got there is, is valid too. So, I think it offers, um, you know, um, some, some really interesting. Uh, possibilities for our industry. I think there's a, there's one thing that does get missed from Zito as well, it offers simplified querying as well. So once you've got that data model validated, another really nice feature of it is that it really does make querying that data a lot easier too. Yeah, that's a good point because we've talked about different personas and we often say the data scientists or the, the analysts and so forth, and they're really good with queries, you know, they've kind of figured out the syntax, but there's other audiences like end users, system integrator, and uh energy managers, and for them to be able to use kind of plain English, uh, give me all the energy usage caused by my chiller plant feeding my chill water coils. Um. Those are plain English specs that could be pulled out into a query, uh, through the tools, utilizing Zeto, so excellent point. And, and a huge step forward, as, as you say, the, not only do you have this technology that sits in the background, that, that supports people through that engineering process, the commissioning process, the verification means that then what is deployed is really working and connected how it should be, but if you start to also deliver through that operational phase, Uh, you know, the, the, the poor person who, who has to look after the security system and the, the, the lighting system and the building management system and cut keys and, and let people in after hours can also use plain English to find out some information. It's a huge step beyond what you described at the start, Scott, which was naming conventions because often naming conventions are, are actually quite cryptic. Uh, and they don't always help you find the information unless you're inside the, you know, in the in crowd somehow, and you know what these, these, uh, these conventions mean. Um, Gareth, you had some projects also that you're interested in. I think you mentioned one Building Motif to us. Could you give us an introduction there of, of, uh, that project? Sure, um, well, there, well, first of all, you know, there were, there was many, many different projects, um, projects. One of them was Building Motif. I thought that was really, really interesting, which is a building metadata ontology interoperability framework. And, um, yeah, I know it's a big mouthful, isn't it? But the first thing to understand about Zito is, you know, it, it's got some very, very powerful tools for doing data validation. Um, you know, it, it simplifies quite a few things, uh, in order to, to do that data validation, works really well with Haystack. But, you know, it's important to also acknowledge the fact that, you know, outside the world of Haystack there's this thing called the semantic web. You know, where people have already kind of, uh, you know, have other solutions that are kind of data modeling using something like RDF and also validation as well, you know, like the shape the shape constraint language as well enables you to kind of Uh, shackle, shackle, yes, I love these acronyms. I know, but they're quite, they are, they, you know, those, those technologies are, you know, very much for the data scientists and so forth. There's quite a lot to them, and I think, I think Zito as a software developer is a little bit more approachable in my opinion. Haystack is very much a software developer practitioner's toolkit, shall we say. Um, which is great, but you know, building motif, from what I've seen now, it's, it's, you know, it brings in haystack, it brings in brick, it brings in that technology together just to make it, try and make it a bit easier to work with standardized semantic data. It's really trying to bridge the gap between these popular ontologies, so that was very, very interesting. Um, so, you know, it was good to kind of see some different perspectives on data validation and it's very, very important to kind of acknowledge that there are multiple. Technologies out there that do this, and I think especially for the data scientist crowd, you know, that's, that's really important. So, you know that that's something that's evolving and developing all the time which I find interesting. Another thing which which was really cool to see, which I love to see at the Haystack Connect conference are the different working groups because there's always new tags and it's really great for people to be able to stand up and kind of describe their working group, what they've been doing. And the problem they're trying to solve and really make everyone aware of that conversation because not everyone's involved in every working group. So, it's, it's good to kind of hear it, you know, straight from the horse's mouth really on the problem they're trying to solve and why. And I think, you know, those, those contributions as well are always very. Very welcome and really add to a very well rounded conference in terms of the technology, you know, you've got something like Zito, which was the big new shiny thing, but there's there's many, many other things happening in Haystack as well which are really exciting and if not equally as important overall to the overall direction of the standard. Uh, understand, and, uh, the last question I had for both of you, and I already know the, the answer, how, how does the future look? All of these exciting projects, all of these developments, whether they're the, the, the big revolutionary functionality, whether it's the small thing that fills a gap that makes, that makes, uh, the, the overall project more robust. How, how does the, the future feel for you and, and, and, And how does the excitement build after a conference like this for next year and the year after, as you start to see some of these work groups really finalize their activities and bring this into the standard. Scott, let's start with you. I think the biggest word that came into mind for me was collaboration, uh, right after the, the Connect event, the Haystack Connect event, was another major industry uh trade show, and I sat on a, a panel, And I had brick, I had the labs, I had Ashray, all the different standard bodies all agreeing and shaking their head up and down, saying yes, this needs to happen, and so let's all come together, let's find all of our similarities because there's a ton of them. Let's leverage our unique pieces that we have that might be market or audience specific. And um let's ignore the bloggers that use words like protocol wars or data wars or things like that because that's just, that's just hype, that's not the real, that's not the reality. um sitting on this panel, I could feel the cooperation, I could feel the positive energy across all different angles to this problem. So if that's one word, um, cooperation is, is really what the future is about. Thank you, Scott. Same for you, Gareth. Yeah, I mean, I completely agree with Scott. I think cooperation is critically important there. I think, you know, that there is definitely a lot of people motivated in trying to share data and coming, coming to a standard there, especially with everything that's happening with the climate crisis. You know, there's even, there's a A great motivation to be able to do that. I also think there's kind of um what I, you know, what I saw with Zito as well was, um, you know, the idea of building these templates, these shapes of data into a library as well. I think, you know, sharing that and making that openly available, I thought was a really good idea. And that's kind of outside of Zito, I would say. It's kind of more of a, more of a website or websites that would run where you could kind of Access kind of Zeto specs and utilize them in your tools, but the idea again there is sharing and making it easier to engineer and commission these systems, having a better understanding about how they're tagged and so forth and dealing with all that. So I think that's definitely more templating and easier to to commission there using this technology is is is something I see there also. But, um, you know, it was definitely a lot of positivity around kind of moving those standards forward and a greater working together there and everyone being aware of it, far more aware than ever, you know, that that that tree offers a large amount of value and it's critically important that, you know, we're not just going to have haystack, we're not just going to have brick, we have got these different standards and You know, in Siemens we've got like, you know, you know, different ontologies for different Siemens across the whole Siemens business as well. So, you know, there is real motivation to understand how to use those technologies and have them bring together and work together to deliver an enormous amount of value. …