 Hello and welcome. My name is Shannon Kemp and I'm the Chief Digital Manager for DataVersity. We'd like to thank you for joining the first of the new monthly webinar series, Data Architecture Strategies with Donna Burbank. Today, Donna will be joined by a panel to discuss the emerging trends in data architecture. What's the next big thing? Just a couple of points to get us started. Due to the large number of people that attend these sessions, you will be muted during the webinar. And we very much encourage you to chat with us and with each other throughout the webinar to do so. Please click the chat icon on the top right-hand corner for that feature. For questions, we will be collecting them by the Q&A in the bottom right-hand corner of your screen. Or, if you'd like to tweet, we encourage you to share highlights or questions via Twitter using hashtag DA Strategies. As always, we will send a follow-up email within two business days containing links to the recording of this session and additional information requested throughout the webinar. Now let me introduce our speaker of the series, Donna Burbank. She is a recognized industry expert in information management for 20 years of experience in helping organizations enrich their business opportunities through data and information. She currently is the Managing Director of Global Data Strategy Eliminated, where she assists organizations around the globe in driving value from their data. She has worked with dozens of Fortune 500 companies worldwide in the Americas, Europe, Asia, and Africa, and speaks regularly at industry conferences, including our upcoming Enterprise Data World. And joining Donna today, we have three panelists to help weigh in on today's discussion. Joining us is Ben Neusbaum, the Chief Technology Officer at Adam Rehn. He brings to the table nearly 20 years of software architecture and engineering, server infrastructure, database design, and technology innovation experience with implementation expertise and globally distributed EmitterPrize software solutions for the world's leading enterprises. Also joining us is Danny Sandwell, the Product Marketing Director at Irwin. He is an IT industry veteran with 30 years of experience as a Senior Product Marketing Manager for Irwin. He is responsible for the development and delivery of the company's data, modeling solutions to meet evolving customer needs. And also joining us is Steve Volks, the Chief Technology Officer and Founder at Stream. He is a lifelong technologist, architect, and hands-on development executive. Prior to founding Stream, Steve was the Senior Director of the Advanced Technology Group at Golden Gate Software. And with that, let me give the floor to Donna to get today's webinar and discussion started. Hello and welcome. Hello, it's good to be back with a new year and a new lineup. And I wanted to thank everyone for joining. So a lot of my session from familiar names on the list of attendees and a lot of you joined us for our last year series, which was focused on data modeling. We have now kind of mixed it up a bit partly because of the success of a conference that diversity had last November where, you know, we're more broadly focused on data architecture and got a lot of good feedback as well as a paper we put together. And we'll talk about more about that today. In part of the region, everyone knows I'm a big fan of data modeling and enterprise architecture, but that's only interesting in the context of other things. So you'll see from the lineup this year, data architecture is so exciting nowadays and there's so many things going on from graph databases to artificial intelligence to what's near and near to my heart. How do you make business profit from this? And how do you really help in business data strategy? So you'll see a lot of wide-ranging topics this year. Hope you can join us for all or many of them. And as Shannon may have mentioned, all these are recorded. So if you're not ever able to make them in real time, you can always catch them after the fact. So jumping right in, well, what we're going to color today. So as I mentioned, a lot of the research you'll see today will be based on a research paper we put together last autumn, basically getting some real-world trends on data architecture. We'll walk through that. So basically, again, the most important topic we'll talk about today is how is an architecture going to support your business goals? Because although all of us are technologists in the call and a lot of the people in the audience are as well, that's only interesting when it's doing something, right? So we'll talk a lot about that today. And then probably the hottest thing and why some of you may have joined is what's hot and what's not, right? That's always sort of the interest of a passing trend. So just for your reference, if you want to get a copy of the full paper, we obviously couldn't cover everything today. We'll just get some salient highlights. It's available on Data Diversity's website. It's also available on our Global Data Strategy website under our white papers. And we'll have a reference at the end as well. So jumping right in because we'll get to the meat of the matter, right? So just to start, and this is sort of a framework you may have seen if you've joined any other webinars we've presented. But what I like about it, it really sort of sums up the ecosystem. It sort of aligns with our basic tenet of, you know, a data strategy is only interesting when aligned with a business strategy. So it's a lot of what we do in our practice. And this really is all of the above. So it's the top down. It may be, what are we trying to do as a business? And then how can data support that? But I think even more exciting, and we'll probably talk about a bit on this call, is it's also a bottom up. We have so much more data nowadays, whether it's streaming data or unstructured data or in and of things data. How can we generate new business models from this data? And I think that sort of ecosystem of IT informing business and business informing IT is really what's interesting now in this new, you know, new ecosystem of information technology. And then the rest of it, I think you've all seen this sort of, you know, how do you make sense of that? Whether it's master data or warehousing or analytics or architecture and then the glue that holds that together is really the governance, which is the people and the process and really that culture of data. Once we're all working together towards data, that's where those sort of aha moments and those light bulbs really go off. And if you're sick of seeing the slide, apologize, but people are encouraging us. We actually had a few customers I visited last week that had this printed on their desk with areas highlighted that they wanted to tackle this year. So it's sort of a nice little Zockman framework for data strategy. Right into the findings from this paper, really what we started with was what's driving the business need. Again, you know, and you could argue some of these are technical, some of they're business, but really when it came down to it, I didn't see this as a major surprise, but it was more around how do we get insights from data. So when you see things like data science, data discovery, that shift towards self-service BI, a lot of trends on that. No surprise that a lot of this is now regulatory compliance and data governance, which we're seeing a lot in our practice, whether it's GDPR or any financial regulations. So those are the things I think we expect to see. I was pleased to see the focus on efficiency and cost savings because I definitely see that all the time and say that all the time. What is that saying? If you don't have time to do it right, you have to time do it again. And I think my personal opinion is now that we have more business people sort of being involved in this, they sort of see the logic of that. Can we design it and really understand the architecture before we get into the hottest new technology? Because they're probably not as excited about the new technology or they want to get stuff done. The other one that I saw that was hot, but not as hot as I would have thought given my particular customer base and a lot of what we're doing is that idea of digital business transformation. I sort of expected that to be hired. But maybe, you know, in the whole ecosystem of companies, that's not the biggest thing. But I'm seeing more and more of that. And I'm seeing that's why people are focusing more on data because to do digital business transformation, of course, you need the data piece. So those are my thoughts. I kind of wanted to pass it around the table, so to speak. Steve, I was wondering what are your thoughts on this? Any surprises here? Is this what you expected to see? Steve from Stream? Hi. I needed to unmute. Technology issue there. It's kind of in keeping with what we're seeing. There is definitely an emphasis on getting value out of data and becoming a data-driven organization. And so having reporting a business intelligence, and that is a very kind of loose term, right? It can mean kind of old school, you run the end-of-day report and see what's happening, or it could be the new world of real-time insights into what's happening into your business. It kind of covers both of those things. And yeah, as organizations get more data-driven, you would expect them to be focusing on getting value out of their data. The digital transformation is a long slog. So it's something that people may also be doing it, but they don't even know that it's called that. I mean, we hear people refer to data modernization or just architecture. Architecture is an ongoing thing. It's always evolving. So it may be that they're using different terms for the same thing, but almost everyone that we're talking to is under some form of transformation and obtaining new sources of data that they can integrate with their old. So I think it is very much what we would expect. Cloud, I thought, would be a lot higher though. That was one area that we are seeing so many different cloud use cases. I expected more people to be talking about it. Maybe migration wasn't the right word. There may be adoption would have been a better word. Yeah, that's fair. And I like your point on the digital transformation. So it reminds me of the dot-com boom or the e-commerce. You might not think you're doing that. I'm just selling stuff on the web, right? I'm not using that fancy term for it, but I'm doing it. Exactly. Danny, what are your thoughts on this? Any surprises or is what you expected to see? No, I think agreement on the cloud side because I do see a fair amount of requirements from that side. And as Steve said, digital business transformation, I think that's business today. I think whether people are recognizing it or even tying back to it, it's kind of what we do. The part that wasn't surprising and really I think comes to the core of it is reporting business intelligence and probably throw in analytics somewhere in and around there along with self-service. Because I think that one of the biggest changes I see in terms of how people are approaching data architecture is that we talk about wanting a business architecture and business drivers, but then what part of the business is driving it? I think historically it's been the operational business. We've built things to make the business run. And then analytics were kind of what's the best we could get with what we have and an interesting way to put it together. But I think people are really starting to look at at least their 2B data architecture as driving out of the other side of business, which is business strategy. So really looking at not just data to run the business, but data to drive the business, data to drive the next big thing. And then having a data architecture that's actually tilted that way that drives people to results that are bigger than just this year's sales or processing an order, but really understanding what are the markets that are out there, what are the opportunities. And that's the big thing that I hear people talking about is return on opportunity versus return on investment and measuring it versus what they think data could mean to their business in terms of steering it going forward. Yeah, I would agree with that. And I guess we could all say whether transformation was the right term, but I'm seeing that same shift. It's not just about reporting about what happened last year. It's using data for what can we do next year. What are the new innovations we can do? Yeah, I agree with that. Ben, what are your thoughts on this? So I think I was a little surprised to see the reporting in VI so high just because of some of the transformation that's been taking place over the last five years around getting more from just big data and just doing reports and kind of reporting on it to how do we actually use this data to make our business operate better in this area? You know, more that understanding and that intelligence and applying, you know, new tools and techniques like machine learning, artificial intelligence or creating knowledge representation that's shared across the enterprise. So you can break down those department walls and kind of start operating as a holistic organization, speaking the same language as humans and as well as machines so that you can all interact together. So that was kind of the one thing that surprised me that that was still so high. Yeah, no, I mean, I think a lot of this, I think some of us were in the industry was always about the new hot stuff and we have to realize that maybe a lot of folks are still, you know, using some of the tried and true things that don't go away when this new stuff still comes. You know, I don't think it's an either or probably is an and. So, but yeah, definitely agreed. Actually, that's a good segue into our next slide because I personally had some surprises and it's sort of just the car lottery of what we just talked about, but to specifically call out what was not hot. And I think, and I have my thoughts and I'll pass it over to the gentleman to get theirs. But you know, some of these, I think we expected to be bigger. So big data. I mean, when we talk about later in the presentation, we'll say a lot of folks are using big data. I think certainly a lot of people are going to the cloud. And with any survey, is it that this wasn't the major business driver or the major driver, which was the question of how it was answered. And it's a means to an end. You know, maybe cloud is just so prevalent now they don't think it as a, you know, it's just what we do. You've gotten into the previous comments. It's not necessarily our main driver. That was my thought there on the bench comment on the artificial intelligence. That's top of mind for me. It's something several of my clients are using at least tipping their toes into it. But again, I think it's probably is sort of an emerging thing now that maybe isn't embedded in every organization across across the country in the globe. The one that maybe my two cents, because I was also surprised that software development isn't as high as I know a lot of people are. I mean, what I see, unfortunately, and I hope someday we'll get better as organizations, there still seems to be this divide. You sort of do software or you do data as if they're separate things when they should be integrated. So the teams, different parts of the organization, folks are trained differently of personally taking several software development classes where data is never mentioned. It's sort of a thing you use or it's easier to code around data. And I think until we get that divide, right, we're going to have a lot of issues. I think they should be two sides of the same coin. So that's my thought. So I'm going to, since you sort of have the good segue, Ben, I'm going to pass this back to you. Are these the same surprises you're seeing or any other thoughts you had here? I think that's right. I think at this point, you know, big data cloud and cloud are more enablers. They're not the driving objective of the enterprise. They're more kind of assisting and tools to get to a broader objective that's really focused on the value from the data and making data a business asset that can be, you know, leverage invested in and then show a return. Right. So I think that's where we see the kind of trend and thinking and how to think about your data as an enterprise. Yeah. Steve, your thoughts on that? I think this is incredibly interesting, especially when you start to think about what a data architecture is, right? A data architecture basically means that you've thought about your data. You've thought about how you collect it, where you store it, what form it needs to be in, and what questions you want to ask of it. And, you know, as one of the attendees just pointed out in the chat. Yeah, thank you, Gail, I think it was. I remember years and years ago, people would build applications and create a database perfect for the applications and then someone would have to try and write reports and you had to build these horrendous queries to actually get back anything useful in the reports. And you'd hope that after many, many years that this situation has changed. But when you look at what happened with big data, oh, we'll just throw all of our logs, all of our raw data into Hadoop and then hope by some magic we can get some value out of it afterwards, which is why a lot of the big data initiatives failed. That was obviously just, you know, Calcukulant, right? You're not going to get value out of things unless you've thought about it beforehand. So I think using the term data architecture here, it kind of implies to me that these pieces haven't been considered when it comes to how do I arrange my data. And if you think about how AI often happens, you know, AI requires data scientists, right? They are given some data by the people that are responsible for the data infrastructure. You're moving data around, storing it somewhere. And they may not be the technologists involved in, say, a data lake, which means that they get given typically raw, unprepared data that they then have to massage and get into the right form, extract the feature sets, and then do the machine learning on. And they're kind of okay with that. They've got the tool set to do that. But that means that the formats and the features that were required for machine learning weren't part of the data architecture. And so if you start to think, well, now how do I operationalize what I've learned in my machine learning? How do I turn my models into something that is able to make real-time predictions, real-time classification, assist my customers in real-time, aid my business in real-time. Then you run into a huge hurdle because none of the real-time data pipes are in the right format to actually execute a machine learning model. So I think this will change over time as a lot of these use cases, and I agree with software development as well, as a lot of these use cases will end up influencing the data architecture. And so maybe people will think ahead of time that, you know, what should I put in a data lake? What format should it be in? Well, what do I put in my data warehouse? How does that need to look? And where do I put the data? What do I put in the cloud? What do I put in the lake? What do I have running out live over a Kafka bus? You know, all these different options that you have, you need to think about how they all fit together and work out the best technology for the questions you're going to ask. You know, I think it makes a lot of sense. And I noticed there's a lot of chat going back and forth, which is always happening and good on these. I think because this is a panel, I think we'll maybe address some of them sooner. And yeah, I think Gail and Tony are having quite the back and forth. I think you hit on as well, Steve. And actually, I need to scold you a bit because you're a product developer. You haven't developed that big red magic button that just automatically takes everything from the lake and organizes it. But I do think there is some frustration between these teams and these silos that we're sort of hitting on. And one of the questions I think is, you know, we sort of touched on that either one does data or one does software or the folks come to us after the fact to say how come the data doesn't match what we wanted. And then sort of how do we address that? So, you know, one way to address that is to really have some of these data designs and sort of a canonical format or can we design the data that goes into an API? Or, you know, Danny, I know design is sort of near and dear to your heart. So what are your thoughts on how to sort of break down some of these silos that people are sort of bringing up on the chat? Any thoughts there? I think that, you know, that breaking down the silos, you've nailed it on the head, you know, in terms of, you know, do you make many data sources and then try to fit them together? Or do you try to figure out the right data and then map everything to it because it makes sense for the business? You know, obviously, or maybe not obviously, but in my mind, you know, you always want to, you know, get out the redundancy and the rework and the rationalization aspects of it and have that sort of single source of the truth that everybody's working on. I think Wendy nailed it in the chat, you know, on your previous question in terms of, you know, emerging trends and big data. I think that it's all about data governance first because, you know, we've seen so many things in the data world that have, you know, charged forward and then, you know, had a pause and a reflection based on the fact that, you know, that the governance models aren't in place. So maybe, you know, a lot of this is looking at governance models first, making sure that, you know, the data is fit for purpose, that it's understood, that it's under control because, you know, going out and investing in a lot of A.I. and machine learning, if your data is crap, it's garbage in and garbage out, you're training something based on bad information. You know, and then as Steve said, do you have the wherewithal to keep, you know, to make that machine ascension learning being through the process, you know, ongoing? How do you operationalize it? So, you know, I think that, you know, people are recognizing that you can't let people go off the reservation. And the more you do that, the more time you take, you know, bringing them back and going through the, you know, the gyrations and, you know, gymnastics involved in doing that. So maybe this is just a sign that people are starting to think that, let's give some of the basics in place, you know, in terms of where we want to go. And then leverage that to start bringing in some of these other things, as Benjamin said, that aren't necessarily drivers, they're enablers that are going to make it happen. And that's a great sign because, you know, we have a long history in this business of running to the next best thing. And it's hot, it's new, so you've got to have it, put it in and see what it can do. I think if this is a sign that people are starting to think before they, you know, before they go and make those decisions. Crazy talk. I know, I know. It's only taken 30 years in my life to see, you know, some Brent clubs and hints that this is actually happening, but it's very heartening if it is. Yeah, and I think that's going to be a theme throughout this discussion of what are those tried and true just fundamentals that don't go to a no matter what. We do. What are these new, really game-changing innovations that are truly innovative? And what's really a business driver? And it's almost the intersection between those three. And on that note, I mean, we sort of had some of these what's hot, what's not. And what can maybe make our brains explode sometimes when you read what's hot in the media versus what are people actually doing? You know, because I think one is always ahead or not necessarily in sync with the other. One of the topics that unless you're living under a rock, you could not have missed is this idea of blockchain. And that's hot in the news. I think a lot of us are trying to get our brains around is this a fad? Is it a new way of doing finance? Is crypto currency the new best thing? Or is blockchain one of these, you know, this is almost summarizing that what I sort of just mentioned. That part of blockchain or a foundational technology, which is a new way of doing things that we could leverage. How much of is it a fad? And how much of it could be used for finance? You know, when you look here at the numbers of the survey, it's a fairly small number right now actively pursuing blockchain technology. I don't think it's a surprise. I think a lot of us are getting our brains around it. What I did find interesting was the different types of use cases. It's not, of course, just cryptocurrency. I think the underlying technology of blockchain can and may be used for a lot of different things. Again, I want to pass this back to you, Ben, because I know this is near and dear to your heart. What do you thought? Is it just cryptocurrency? Is it something else we could use for other use cases? Or is it a fad? So, blockchain is super hyped right now on, you know, kind of this panacea that can do all the things. And, you know, it can't do all the things. We definitely see the hype from cryptocurrencies, but that's not necessarily that interesting to the enterprise. I think what you've got listed here, you know, kind of the underlying blockchain technology does provide some valuable, notional concepts for the enterprise. And, you know, the nature of, you know, distributed consensus, being able to, you know, have a network that can agree on certain things that happen, and that can retain that order and that transparency are interesting when it comes to, you know, tracking supply chain management, or, you know, looking at financial transactions across different, maybe separate financial organizations. So, it does have merit in terms of the blockchain technology. I think, you know, what we'll see is there's a lot of experimentation and curiosity around how is it actually useful, and, you know, kind of what, you know, Daniel was saying, not just jumping and chasing that next trend for the sake of, you know, doing blockchain for the sake of blockchain, right, but actually finding where it fits into the enterprise. So there's a long adoption curve ahead here. And I think what I'm seeing is, you know, as there, as different blockchains emerge and organizations adopt and try them, I think, you know, we're going to go through some growing pains and some performance pains and figure out, you know, which ones actually get implemented well and where does it actually fit as part of a broader data architecture. Yeah, and I think that's, I think we're all the juries out and I, you know, if and when we do a similar study next year, I'd be curious if this changes in either direction, which is sort of a good segue into the next topic. It really was, in terms of platforms, we talked a bit about strategy and, you know, what are the main business, the techie drivers, I found this interesting of, you know, there's a lot of talk of different systems. Right now, what are people actually using? And don't everybody drop off the call when you'll see that the main data source we listed was spreadsheets? I know that's not an emerging database technology, but we have to be real, right? They are ubiquitous, they're not going away, and they're really handy for what they are. You know, the other, which is not necessarily a surprise, relational databases are still clearly that workhorse, it's one of the call-outs, you know, mentioned. We will see that it's switching, I think, from on-prem to the cloud. I'm not surprised there. And still, when we're thinking of what is still currently in use, motoring away, running the company, you'll see there's actually more legacy than big data, right? So, you know, close in the running. But there's some mainframes still running some big banks out there, right? And whoever coded them must have done it well because they're still working, right? So, not necessarily that's going to be the new emerging technology, but you'll see some of the tried and true, are still the tried and true because, of course, when there's a new technology, actual adoption is going to lag. So I don't see any huge surprises there. When we go into sort of what's emerging, I think I found this one really interesting because this really is the what's hot, what's next. Again, some, no surprises, big data. Everybody is either looking at it, seeing if there's a use case, at least testing the waters movement to the cloud, as I mentioned. I think what's also not unsurprising is a lot of folks don't know. And when you see that there's a few big leaders, cloud and big data, no surprise. But a lot of other horses there are pretty close in the running. And I think given that there are some very valid choices out there, graph database, real-time streaming, can we make use of Internet of Things? I mean, some of my clients are very traditional clients doing a lot with Internet of Things and real-time data streaming and some very interesting new technologies when they put on their really creative hat. But how do you choose? So I think there is some valid confusion of what is the right, a lot of people hit on it in the comments. You have a business strategy. What's the right technology for that? It isn't always a relational database. It's still a big fan. They do certain transactional data very well. I wouldn't necessarily use that for my IoT streaming data, right? So the only other color commentary I'll mention before I pass it over to the team, just basic statistics, right? So I remember when I took a statistics class, one of my marketing colleagues mentioned that no matter what the survey, everyone will get at least 5%. So you might do a new product launch and say, well, at least 5% of the people like my product, but usually that's to the people who didn't read the question or there was some confusion. So you'll see here that 5%, just about that magic 5% are saying that legacy systems will be their new technology. I think that might have been a badly worded question. I personally don't see a lot of people moving to COBOL. It probably is maybe hiring folks to maintain those systems that probably was an outlier as we all do. We're all data people, right? Yeah, I guess someone had a comment about not using spread system plan to start, right? So never know. You see, I actually had one client and then I will stop my little rant. It was actually a water company and they did a lot of acquisitions of some smaller water companies and we said, can we see the customer master data from this acquisition and they handed us a paper notebook. Literally the guy that ran the water company wrote down all his customers in the big paper notebook. So never be surprised at some of these that spreadsheets in that case would have been a step above their current master data implementation. So, you know, Danny, what do you think? And again, given your experience in the market, is this a surprise to you? Anything you would have expected to either see people using now more of or things that people would be looking at in the future? It doesn't because I'm sure the sampling, you know, there's a lot of different companies in a lot of different stages of their life cycle that have answered this. So, you know, the answers that you'll get from a startup, you know, will be very different from, you know, the answers that you'll get from a company that has, you know, a long history and a, you know, that history involving a fair amount of infrastructure supporting their business. The dimension that I'd love to see on this is not just, you know, which, but when. Because I think that's the big piece. I think that, you know, you've got different types of folks out there. And as I said, they work for different types of companies. You talked about, you know, your sort of more traditional customers that are there. You know, they will get there, but they'll get there, you know, when it becomes more, you know, more accepted, when there's more experience, when they can get there, when they're not trying it out. So, you know, I think that there's a lot of things that people may not necessarily think are on the roadmap today, but they're on their longer term roadmap. And that roadmap is an agile roadmap, which means, you know, the justification from a business perspective and the capability to actually make it work are going to be big, you know, big sort of decision makers in terms of when that road might map item actually, you know, hits the hopper and goes into the next scrum. So, you know, I think from that perspective, you know, and, you know, I think back to the previous slide, blockchain, you know, very similar. You know, as I went out to understand and look at what blockchain is, you know, when it started to get hot, you know, the first thing I felt, you know, I'm a child of the late 60s, 70s here, you know, this sounds like hippie technology, right? There's a lot of people all getting together and doing the right things, and we don't need any of those traditional, you know, checks and balances because we're all going to do the right thing. I think that scares a lot of people off, but Steve actually, you know, spurred a thought in my mind on our pre-call yesterday or Tuesday, whenever it was, talking about, you know, Walmart. And, you know, when a Walmart decides that blockchain is a technology that's viable in their world, trust me, there'll be a lot of other folks because, you know, I worked out in industry for companies that fed. You know, and supplied Walmart and trust me, their supply chain management is your supply chain management. I went through the same thing, you know, when I was in a different area of software, you know, looking at, you know, eco software and, you know, corporate governance. And, you know, when a giant like that says this is what it takes to do business with us, that's when you're going to start to see a lot of adoption outside of that. So I think that, you know, none of this really surprises me because it's a wide range of people. Their sort of visibility and timeframe may not be reflected in what we called, you know, what do you plan to use. And I think that there's some natural things that go on out in the business world that will accelerate these things at the right time. Yeah, and I think you had some good points, although hippie technology, you're going to be quoted on that. And you should have hit on a lot of the questions. I mean, this for nerds like me, this survey was fascinating to do because just seeing a lot of the direct comments were interesting. You could slice and dice this a million ways. And of course, there's a lot of data people on the call. So a lot of the questions are asking about sample size and region and types of companies. In the download report, we do put that information in. So it was several hundred respondents across. Yes, it was a diversity survey. So in that sense, we sort of picked the audience. We didn't ask cab drivers in Jersey City, you know, so slightly different because I would segment your answer by region, by type of industry. You know, and this was a fairly good broad section of large companies, small companies, etc. But, you know, there's actually drivers in Silicon Valley accepting Bitcoin, right? So, you know, it really depends on so many areas. So Steve, I'm passing it back to you because I know these a lot of new technologies have neared your heart. So what are your thoughts on these findings? Oh, I love these findings because as a company whose platform supports streaming integration, I just see endless integration opportunities. But the key thing here is that we absolutely see this. We see companies that have legacy systems, but they recognize that building analytics or other types of applications to get real-time insights out of those legacy systems will be prohibitively expensive. Maybe this part on this chart on this slide is people that are thinking about hiring developers to actually build analytics applications on legacy systems. But in reality, we're seeing people that are doing real-time data movement from say a tandem, the HP non-stop system into Hadoop. Some of it going into Hadoop, some of it into Kafka, some of it into the cloud and doing the analytics or whatever they want to do that data there. So, again, it comes back to purpose, right? And it looks like people have, given the breadth that we see here, have bought into that purpose idea somewhat. Graph databases are very, very useful for asking questions of connected data. So it's no surprise since we live in an increasingly more connected world that you're seeing arising graph databases there. Doing real-time analytics allows you to get insights immediately rather than kind of end-of-day batch jobs. And as the world gets faster and faster and businesses need to react more quickly, it's no surprise that we're seeing a move towards real-time analytics away from the old-school analytics. IoT, I agree, is very, very interesting. And what interested me, I tend a lot of IoT conferences. There's a lot of presentations from the insurance companies. And insurance is, you know, it's one of the oldest industries we have. They use hundreds of years' worth of data just to work out where to build the new headquarters, just because it's like the place needs likely to be hit by a natural disaster. And yet these same companies are investing in things like smart homes. If you can have a home with IoT sensors that spots a water leak before it causes thousands of dollars' worth of damage, that's saving the insurance company money. So IoT has so many different applications. I think we're really only touching the tip of the iceberg when it comes to what will actually happen there. And it's just, you know, recognition that it doesn't just have to be consumer IoT. It isn't just fancy gadgets. There can be things here that can help almost every industry, especially things like agriculture and logistics, right? So I'm not surprised at the breadth. I am heartened by the breadth. Yeah. Yeah. And what I used to have hit on, and I won't go into a full rant because we've limited time, and then Shannon starts texting me this stuff. But that's what I find so fascinating about being in this business right now, because you hit a nail on the head that very, very traditional businesses are trying very non-traditional things like drones and IoT. I mean, some of my clients were sort of the oldest of the old, you know, water companies or oil companies or, you know, consumer electronics that they're just trying brand new things. I have a small Head Start company I'm working with, organization nonprofit, doing some really next generation things with predictive analytics. You know, so who would have thought, but now with this democratization of all these technologies, it isn't just the big guys doing it or the people on the leading edge, it's everybody. And I think that's what's really fun rant ended. And I'll pass it over to Ben. What are your thoughts on this? Similar thoughts or anything you saw differently? Yeah, I think mostly similar thoughts. The one thing I would have been curious to see on here would be around the AI and machine learning from the previous ones to see where that was interest was lying. Yeah. Yeah. So many of the different slices on this. Who of this was using one today and then going to the next one tomorrow? You know, so fascinating with this data that we should do a data lake just on this. But, you know, and some of the comments are interesting. And I think we're all, again, we're sort of preaching to the converted on this particular webinar, right? We're all some sort of related to data architecture. But, you know, some organizations are looking at some of these new things like blockchain, but unless you have that right governance and data quality, you know, that's not as helpful. And what I find interesting, you know, one of my clients late last year wanted to do a big new artificial intelligence program, and they called us in to start with the data model. Like, who would have thought that's almost the old and new converging right there, but they understood that if we don't understand the basic rules and don't have basic quality, our AI is going to be really crazy, right? So, which leads me into my second rant of the day that I think I'll agree with, but I find fascinating. And Danny hit it on earlier. I put this in the category of the what's old is new again section. I really like this cartoon when I found it. It looks similar, but this one's powered by Hadoop. And it's sort of what in this world are just tried and true fundamentals that don't go away and what our old school and should go away. Are we going to be building brand new things on mainframe? Maybe not. But a lot of the, and I think, Steve, you hit on this too, a lot of the fundamentals of mainstream are seen in other technologies, like the cloud. You know, so it's again using what works and keeping it, whether it's around for 20 years might show you something that is good. And one of the reasons I like this is we still use wheels, right? Tesla, the most innovative new car, still has wheels. Literally, why reinvent the wheel if it works? So, particularly like this one because I actually spoke to some of that Hadoop ecosystem. You know, in some of these findings, so I'm not bitter that they said data scientists are the sexiest job of the 21st century. Because I think a lot of us in other areas of data think we're off these sexy two. And I think this one comes down to it in this survey. And again, this is sort of these were diversity listeners or readers, but over 96% were engaged in data modeling. And I'm just going to pause there because if we think back on the previous slide where the majority of the organizations are looking at one or more of the above of artificial intelligence and machine learning and blockchain and, you know, graph databases. You still need a data model because that's really the core. When we're talking about aligning business and IT, especially at the logical conceptual level, it really is the data model that does that. It should be sort of technology neutral. And I think the next slide sort of touches on this is that if you look at types of data models or types of models that are being used, a lot of them are very business centric. So, of course, you'll have a physical data model, especially for a relational database, but others as well. But you'll see the high percentage of conceptual models, logical models, near and near to my heart, business process models. For example, if I'm going to do a master data management implementation, I cannot do that without looking at business process. And how does your data flow across the organization? You'll see another high percentage. I see them as sisters to each other. Data flow diagrams and business process models are two sort of sides of the technical and business coin. So to me, and we've talked a lot on this call already about kind of that convergence of business and IT. And I think this role of the data model is kind of that bridging the gap a bit. And one more slide before I open it up to the gentleman as well. What I also found interesting is when we look at, okay, so we can model, these are almost classic to data architecture, some of these my data models. And it was well, things like component diagrams and data lineage, you know, it's the gamut. But when we came down to say who is now building the data architecture? Well, there was the well done answer, you know, data architecture, building a data architecture. But I think just like we saw the breadth of technologies, the breadth of stakeholders. So people like the business architect, the data governance officer, not necessarily building it themselves. This was a multiple choice question, but a lot more people involved, which I think is a good thing. But I think that's also where we're seeing that larger prevalence of some of these business-centric models, because that's really where you're doing the collaboration. The relational database developer might be looking at the physical data model, but everybody should be looking at more of the business-centric model. So, Danny, I know this is near and dear to your heart. What are your thoughts on this? Surprising? What you expected? Thoughts? Well, a couple of things. First, I think you are better about the data scientists, but that's just my... You can't do three slides and ask me to remember what was on the first two. No, seriously, I think this is great. I've seen this over the last number of years in terms of the different types of people that are at what we would call data shows. You know, diversity, enterprise, day to world, absolutely. Starting to see a higher percentage of business people walking around. One of the questions they asked me is, I know that there's a group that has all these great graphical pictures of my data, but how do I talk to them and let me have access to it? So it is good to see these folks, I think it's absolutely real. Back to the models that people are using and as a data modeling provider for how long, we've seen that the reason people bought it when I started was because they wanted to build databases faster. But now the value of the logical model, the conceptual model, the usage of it, not just to talk to business people, but also for architecture purposes, design purposes downstream, whether it's building your warehouse, trying to figure out what you have in your lakes, or even a data virtualization model. So it's the one place where you can sort of start to get some apples to apples sense and remember. And any relationship was not, believe it or not, didn't come out of relational databases. It's a human way of looking at information and how those things relate to each other and how we relate to it. So it's not surprising. Yes, business process. One of the pieces that really isn't listed here, and maybe it's because of the audience that was pinged, but we're also finding that there's a lot more interest in data for the traditional enterprise architects. So people are doing enterprise architecture, going, you know, goals, strategies for the business, business capabilities down into the technology and infrastructure that supports that. The information model and integration of information and data into that is a much higher priority and quite frankly is starting to drive a lot of opportunities and a lot of reason why people are looking at enterprise architecture because it is another enterprise architecture use case. It's an enterprise data architecture, right? So how better to put that into the context than using those technologies? So I think it's, you know, this all aligns with what I've seen both on the data modeling side, the process modeling side, but also as we move out into, you know, other architectural disciplines, data just continues to drive, you know, and become the raison d'etre and really become a key component of those things as they move them forward where they might not have been before. Yeah, and I would agree. I mean, we do that so much. We're even stepping above process, you're right. Is it business capability? Folks who joined some of my webinars last year might have seen we even do something like a business motivation model, you know, actually mapping out where your goals are and why we're doing this with hearts and minds. So definitely agree on that. Just in the interest of the clock moving ahead, you know, sort of the sister of these data models is the other sexiest job. Yes, I am bitter, Danny. The other sexiest job in the 21st century is metadata. Who would have thought? So in this was a survey we did last year, which is also available out on the website, is over 80% said that metadata is as important, if not more important than in the past, that this is definitely a growing trend, not something that, you know, as we're looking at all those different types of technologies, the glues that holds us all together is in the metadata. And so when we looked, and this was sort of a graph of, not just really again with data, we could always be, we could keep talking about how we sliced and diced it, but not necessarily the same people answering each year, but you'll see the trends of what the trends were last year in the gray or this year in the red. But you got some similar trends, data governance, data quality, analytics, MDM, master data management. I think 2017, the difference, we saw a little bit more on the regulation than some of the master data. And I want to ask you, Steve, in terms of either metadata or the modeling, which are kind of together, what are your thoughts on any of this? Is this what you expected or any insights here? It would have been nice to have actually seen, again, a kind of cross question, which was what types of technologies were you applying the best practices in modeling and metadata management? I would guess, without having done that survey, that big data would fall short on that. And that would kind of help explain issues people have been having, again, throwing stuff into a lake, right? It kind of shows that having an idea of what your data looks like, or at least planning for it, and planning for it in the context of a process, in the context of your data flow, and what manipulation happens to that data in order to actually achieve something and to be able to ask questions a bit later is really important. And it's important from a architecture perspective and ensuring that the right things are done to the data in the right places so you can actually ask the right questions of it either in real time or after it's landed somewhere. But also in regulations, you look at something like GDPR, which is kind of going to hit everyone. The first step people have to do is actually understand what data they have, actually do a data inventory. If they don't have models and metadata management or something like that in place, then they're going to have to start doing that. They're going to have to start categorizing their data, and there you're looking for things like personal identifiable information, anything that could possibly relate to a customer. Because if your customer says, please forget about me, you have to show that you've forgotten about them. And if you don't know where all your customer data is, then you're in trouble. If you've thrown your web log files or application log files into Hadoop, so you could do some analysis on them afterwards, are you sure that there's no customer data hidden around in those logs? So I think it is something that is and always will be important for people to understand and plan what their data should look like, and then to manage it going forward. So the one thing that was a little bit saddened in the previous slide is how, and on this one as well, the previous graph slide, that software developers and programmers aren't really involved in the process at all. So I think that was part of the comments I was seeing, whizzing by as well, that everyone that has a play with data, everyone that works with data, and if you think about software developers creating logs, even web logs that are going to end up as part of your data structure, especially with big data, then they kind of have to be part of that, and they have to understand the data governance, data security, data protection, and data management as well. Yeah, I would agree with that, and that sort of resonates with my comment right in the beginning. I think we've gotten a lot better with business, whatever, I mean, business is so broad, and IT working together, but I think that programmer data gap is one that we still haven't quite gotten right yet. There's probably a lot more there. Again, looking at the clock, just quickly, another one of the tried-and-truel what's old and is new again is really this idea of master data management. And so, you know, I personally would have thought that this was higher, so much, and one of the questions, and we can talk about it a bit, if we have time to Q&A, someone asked, you know, does data governance, doesn't that include data quality? And if you remember back to that sort of graph we had in the beginning of the overlapping, there's governance and quality and architecture. I mean, so many of these are so intertwined. I find it hard to do governance without a master data set. I can't do master data without governance in place. I can't do master data without data architecture in place, or data models, or data lineage. So in my experience, I've seen this number be higher. But I'd be curious what the others think, Ben, are you involved in that master data management, or what's your thoughts there? Yeah, our approach and our perspective on master data, and most data within the enterprise is it needs to come into a singular knowledge representation of that data across all the sources and be mastered into a single set so that everyone knows which data is the right data to be using, because if you have many sources, you know, what we've seen is, you know, if you have one source of truth that it's hard to interact with and work with, that'll get copied across the enterprise sometimes. And then that gets mutated in different ways by different departments. And nobody really knows which data can really be trusted. And so a lot of enterprises are fairly messy at this point. And, you know, I think mastering the data and getting it to a single source of truth from all of those sources and doing that reconciliation process is a first step to really accelerating the value you can get out of your data. And so we do put a high importance on that. Yeah. And what order, Steve, do you think this is an outdated term with the technologies we could virtualize it or something else? Or do you think at its core this is still a discipline that has its place? Well, there's always the question of, or should you have a single view? Should that single view be in one place? Or is it distributed in virtual, right? And the answer to that kind of depends a lot on what data sets you're talking about. We definitely see data flying all over the place, right? It's being moved from one place to another, and it's being moved to a system. Or analytics are happening in the place that kind of most makes sense for the questions you're trying to ask. It's hard to keep data in one place. And you can have something that is your system of record, but that system may not always be suited for analytics. And the types of information that you may want to add to it can also vary. You may want to enrich your data in some way in order to make more sense of it. So the actual record set expands based on the analytics you're trying to do. And for AI, the feature sets you're trying to extract. So I don't think it's simple. And again, I think maybe not everyone understands everything that's included in mass data management as well. So some people may be doing it and not knowing they're doing it. But I would say, again, unfortunately with big data, that's probably the area where we're not seeing so much of that. Interesting. So given that we're getting close to the clock, and there are a lot of questions, I'm going to move on and then there's at least one question I want to cover. But just a couple things I think we should have touched on. Data is a strategic asset. Companies are getting that. A lot of folks are getting that business focus. More business people are involved, which I think most of us agree is a great thing. Maybe the developers are second. Next year we'll talk about the rise in application developers looking at data. Some of the tried and true stuff is still tried and true. The relational data that it faces probably are not going anywhere anytime soon, but there's just a lot more options. But even with these options, I think some of these core things like metadata, like data models, which is really that glue between this collaboration that we're talking about really fits. So we probably have time just for one question and I'm going to pick it. I think it ties into what we were just talking about. There's a question about what is the framework or the best practices for a big data governance? And I'll just take a quick shot of it and I'll give a call out because the other questions we saw are these webinars recorded. And yes, they all are. As well as the ones last year, the one we did in December actually touched a lot on this and I kind of have is that bimodal data governance where I think even big data has a component in things like master data and reference data. So some data needs to be very, that sort of, I guess, old fastener traditional governance where we have standardized reference that a small group approves it and publishes it out and do not touch. That doesn't go away. You still sort of have state codes that are state codes or drug codes at a hospital. Those are pretty important. Those need to be locked down. But more and more of the discovery and the data lake where it's all about look through the data and come up with insights. I see that more as sort of the Wikipedia approach. If the previous one was encyclopedia where a few people define it, the next one was Wikipedia where we really discover some of the governing by is this the right query? Or maybe it wasn't. Is this the algorithm we should use? It isn't just the data. It's the way you govern even some of the algorithms. So that's my quick two cents on that. And I'll pass it over to, I guess, the three of us, or maybe give your parting thoughts on that. So, Danny, your thoughts on governance and the big data world. You know, governance comes down to trust. Right? And so the question is, what are you using your big data for? And is it a big bet? Is it a risky bet? You trust it. So, you know, is there a governing model? You know, I would say, you know, data governance overarching, there's a model if it's important. And if it includes opportunity and includes risk, then it needs to be incorporated in and treated, you know, like all the rest of the data. If it's, you know, if it's not, then again, it sits on the priority and the roadmap for, you know, for when we're going to get it there. So, you know, that's my thought on it. It is, you know, we don't govern because it's the, you know, because we do it because there's, you know, outcomes. Either we avoid going to jail or, you know, or we, you know, increase strategic data usage in our organization. And that's the questions that you should be asking as you look at your big data environment and how you incorporate governance into it. But absolutely, there'll be aspects of it that have to be. And, you know, it's really got to be driven out of the business. Shannon, do we have time for the two others to comment or do we need to cut it at the hour? Go for it. Thanks, Ben. What are your thoughts on this? Yeah, so just real quick. I think governance is and will only continue to become more and more important for organizations. You know, GDPR, a great example of that. I think as we go, we'll continue to see governance, even trumping big data implementations, you know, if it comes down to choosing between being able to put the compliance and governance processes in place to adhere to those requirements and doing something that's, you know, the latest big data technology that's hip and cool and sexy and everybody loves it, governance will win. So I think as data architects and those that are working to form and shape the strategy around data within organizations, we have to be responsible enough to build and design with governance in mind from the start so that those solutions don't get, you know, cut short because they don't meet those requirements. Right. And Steve, your parting thoughts. I think the trust aspect is really, really essential. If a future organization doesn't trust that their cloud-based blockchain technology running smart contracts using artificial intelligence is using the right data and using the right algorithms and you can't prove that that is something to be trusted, then the business is never going to run on that. So you need to not only trust the data, but also trust the pipelines and the analytics and the machine learning that is done on that data if we're ever going to get to a point where they are making decisions for the business with little or just, you know, non-human intervention. Great. We will have to wrap it at that. Folks, if you want to join next month, it will not be a panel. It will be more prescriptive on maybe how we actually build some of the stuff we talked about and we hope to see you next month. Thanks, everyone, for joining. Thanks, everyone. And just a reminder, I was going to follow up email by end of day and Monday with links to the slides, links to the recording and anything else requested throughout. Thanks to all our attendees for being so engaged in everything we do and thanks to our panelists for joining us today in our first data architecture strategies webinar. Thanks, all. Thank you. Thank you.