Episode Transcript
[00:00:00] Speaker A: Foreign welcome to this episode of our show, True DataOps. I'm your host, Kent Graziano, the data Warrior. Each episode will bring you a podcast covering all things DataOps with the people that are making DataOps what it is today. If you've not done so already, be sure to look up and subscribe to the DataOps Live YouTube channel, because that's where you're going to find all the recordings for our past episodes. So if you missed any of last season's episodes or any of the earlier episodes this season, now's your chance to catch up over the upcoming holidays. Better yet, go to truedataops.org and subscribe to this podcast so you don't miss any of our future episodes.
And my guest today is a guy who's actually interviewed me more times than I can even remember at this point. Eric Kavanaugh is a prolific podcaster, radio host, sometime keynote speaker, and an industry know it all. He's the CEO of Blur Group and an open data advocate for the United nations, and the host of DM Radio, the longest running show in the world of data.
Yeah. Welcome to the show, Eric.
[00:01:13] Speaker B: All right, thanks so much for inviting me. Enjoy being on the hot seat today. So we'll see.
[00:01:19] Speaker A: Flipped it around a little bit. You know, I think, you know, you, you, you've basically pioneered what we're doing now, long before anybody else thought it was a good idea. Yeah.
[00:01:30] Speaker B: You know what's funny is a long, long time ago, I actually pitched this idea to the Data Warehousing Institute when I worked there, and they didn't want it. And I was like, well, I'm going to do it, so I guess I'll do it with someone else. And what I realized is that you cannot have a software company without smart, smart people. And I realized if I could just get three or two smart people in a room with me, virtual room, on a weekly basis, we would have really interesting conversations. And I Knew back in 2001 when I got into the data space that it was going to be big. I had no idea how big it was going to get. I mean, I thought it was going to get big. And it's been like orders of magnitude bigger than I even imagined. And that's just the data side for analytics, primarily operations. And now, of course, AI. AI needs data. It's hungry for good quality data. Don't feed it the bad data. Don't do that. So it's. It's bigger than I thought, and it's a lot of fun.
[00:02:26] Speaker A: So let's keep rocking yeah, yeah, no, I'm with you. I. I started. Luckily, I got started in the mid-90s with my collaboration with Claudia Imhoff and Bill Inman and, you know, just sat down one day. At the time, we called it business intelligence. And it's like, well, logically, yeah, every business needs to understand their data. And, yeah, like you said, who knew it was going to be this big? It's like if I had 100 gig database back then, I was like, wow, that's massive. Right? And I were like, yeah, I can have 20 of those on my phone. Right.
So tell us a little bit about your background in data management and a bit about the Bloor Group and what you all do over there. Sure.
[00:03:12] Speaker B: So the Bloor Group is a hybrid media analyst firm designed to examine the world of data and data management and help folks better understand what the different technologies are, what the methodologies are, what to do, when to do it, how to do it, all that important stuff that business people need to understand.
So, obviously, we've gone long past just the old database environment. Everyone knows what a database is, or at least has some general idea. And now we're into things like data fabric and data mesh. That's an interesting conversation in and of itself, but we basically just try to help people understand what are the tools, what are the technologies, what are the methods? What can you do to improve your own view of the business world, and how can you get that done? So we do what I like to call transparent research, meaning we do live broadcasts and we talk to industry visionaries and practitioners and prospects and customers and all these different people. And really the whole idea is to hash out these concepts in the round and to do so in live broadcasts, because I've always found that when there's a live broadcast, people tend to pay more attention and the guests tend to be on their best behavior because they know it's live and they know that these things are happening in real time. So that. That was the mission and we've been doing it now. We're what, entering year 15? I guess so for a while.
[00:04:37] Speaker A: That's great. Yeah.
[00:04:38] Speaker B: Well.
[00:04:39] Speaker A: And not only is it live, it's being recorded and people are going to be able to replay it. So.
[00:04:44] Speaker B: Right.
[00:04:45] Speaker A: Whatever you said.
[00:04:46] Speaker B: That's right.
[00:04:48] Speaker A: It's now on the Internet and it's there forever.
[00:04:51] Speaker B: What do they say? The Internet never forgets. Right.
[00:04:53] Speaker A: That's right. Yeah. So this open data advocate thing for the un, I didn't know that you did that, and I didn't even know that was a thing. So can you tell us a little bit about what is this thing with the United nations and data?
[00:05:07] Speaker B: Sure. So a really nice guy reached out to me, I guess, almost 10 years ago now, and asked me if I'd be willing to speak at a conference about open data for the un And I said, yeah, sure, I'll do that. I have no problem doing that. So I went over to Abu Dhabi and then later went to Dubai and later went to Kazakhstan, of all places, and talks on open data. And it's funny, Kent, because when I heard that the UAE and these other countries were looking for open data, I was like, open data, huh? Something tells me it's closed data that they want to talk about and security and finding ways to secure their data. And of course, that was a big push. But I will say that some of these Middle Eastern countries are really moving by leaps and bounds to leverage data and to leverage open data for the benefit of their people. And I'm a huge open data advocate. I mean, I've written about this, talked about it for years, and I look at something I like to call outsider trading. I wrote about this a number of years ago, too. And what happened is I just suddenly realized, wait a minute, these investment banks have access to so much real world transactional data that they can normalize trends for specific public companies and then understand with real transactional data who is going to meet or beat market estimates. And I thought to myself, wow, that's. That's kind of an unfair advantage. Like, imagine if we're all playing poker and the rule is I can see all of your cards, but you can't see mine, right? I lose. I'm an idiot. Right? And so that's, that's when I really began pushing forward this whole concept of open data and suggesting that, you know, credit card companies are going to sell their exhaust data, as they call it, to third parties like investment banks. I think there should be some policy or regulation that says they must also publish an anonymized version for the general public to use. So we can all look and see. Soft drinks are spiking in this region, oranges are spiking in that region. You know, think about other ways, too, that open data can help business in general and individual people and individual businesses by understanding, well, how many oranges did come in, how many oranges were in Florida this year, how many widgets were sold in this region or that region, whatever sort of data is not proprietary or would benefit the public at large. I want to see more and more of that in the public domain, such that small and mid sized businesses can benefit from this reality, from this really valuable information and not just investment banks. And frankly, I think that one of the reasons why the stock market has been doing so well despite going through some difficult periods like Covid and the lockdowns and these other things is because insiders know who to bet on. And if they knew who to bet on, well then they're going to keep winning and keep going well until there's some sort of macroeconomic correction, which we haven't seen yet. But yeah, so that it's a bit of a thread on open data and I talked about that for the UN and I talk about that in other channels as well. But I really do think we need a leveling of the playing field in terms of basic information that is verifiable, that is useful for businesses to plan and to understand. So you got your own data. I want people to be able to see what is the big picture of data in the GDP and all that kind of fun stuff. So I get passionate about the open data concept.
[00:08:35] Speaker A: Yeah, I can see that. And it really, it goes back to you, you know, in the very early days of data warehousing where, you know, I think now we call it third party data, is the idea of being able to augment your own data set with external data sets. And that's really what you're talking about is, you know, stuff and putting it in the public domain, which is even better. Right. Because there's, there's data aggregators and folks who sell this day sell data of various kinds. Right. You know, Snowflake Marketplace.
[00:09:05] Speaker B: Right.
[00:09:05] Speaker A: And a pioneer in doing that. And some of Snowflake's customers have gotten some of this data you're talking about like consumer trends and foot traffic and all sorts of things and put it out there actually for free. Some of them monetize it, some of them don't. Just depends on the company and where they're getting the data from. But yeah, there's lots of stuff out there. I'm glad to see some of these initiatives where some governments like New York City publishes a lot of data, but you got to download it and it's spreadsheets and all of that. So it's not completely usable.
[00:09:41] Speaker B: Right.
[00:09:41] Speaker A: For the average, for the average small business. But things like data marketplaces, someone else automates that, puts it in and this is going to bring me right to my main topic and basically produces a data product that is easily consumable by smaller businesses. So you want to Talk about data products a little bit here with you. It's become a hot industry buzzword. Kind of started out with the data mesh world, but it's gone beyond that now. I think it's probably one of the concepts of data mesh that people have sort of really attached this idea of data products. So what's your take on that concept and how important do you think it is for this, our ever evolving data landscape?
[00:10:24] Speaker B: Yeah, I think it's very important. I think frankly, that term represents an inflection point in our industry because at the end of the day, the business doesn't care about dynamic data pipelines, ETL versus elt. They don't care necessarily which algorithm is running in the background to surface some business value. What they care about is the business value. They care about understanding what's happening in the business. And so, you know, years ago, a guy I'm sure you know, Dave Wells, very, very smart guy, good friend of mine, he referred to what he called under the hood technology. I was like, what does that mean? He said, well, you don't need to know how carburetors work or how the pistons work in your engine to know that your car works and you can drive it places. And he said, really, we need to move into a direction where most of the inner workings or the nuts and bolts of data management, data warehousing, data analytics, are genuinely under the hood technology. So you don't have to know. Now, there are people who do have to know. Obviously there are teams that have to be able to audit and understand what's going on underneath there to make sure that it's accurate and verifiable and so forth. But for the business people, they just want to know, how many widgets have we sold? What are the possibilities for our next steps? What are some different strategies we can employ to get more business or to optimize our business model? That's where data products come into play. I think that it's a very positive development that we're talking about data products because that's what the business really wants.
[00:11:58] Speaker A: Yeah, people used to think, well, we don't really want black box. But in the reality, from business perspective, perspective, they do. They don't need to know, did you use data vault or dimensional modeling? Are you running on SQL Server, Oracle or Snowflake? Doesn't really matter as long as they can get the data they need and the information they need to run their business and make good business decisions. And so that's led to this. Yeah, I think obviously the productization of Data, enclose it in something that's easily consumable, discoverable, and then of course plays right into your whole open data program is having a catalog of what data is out there. What can I use for my analytics that I can just basically do a couple of clicks, get access to that data and start running the numbers, as it were.
[00:12:52] Speaker B: Transparency is the best disinfectant too.
When you can see things, then there's trust. You want to be able to drill down and understand what's going on, but for the purpose of running the business, you just want to know the data and understand what matters to you. And personalization too. Right. Products that are specific to certain audiences make a lot of sense. We've seen this over the years in many different fashions. Roll ups, hierarchies, for example. The business wants to see this view, the marketers want to see that view. We want to. And you kind of see some of that in the principles of Data Mesh. Though I'll be candid and say I'm still not entirely sure how Data Mesh works. And I think that's because there are lots of different definitions. There are now like 20 different definitions for what Data Mesh means for different groups, for different organizations, which is fine, but you know, I think it's more responsible to focus and talk about the data products and make sure that your IT team and there is some documentation to explain how you got here. You got to have the, the how we calculated this component of the, of the equation. Although that's going to get harder with these large language models, right?
[00:14:04] Speaker A: Yeah. Because then that becomes a little more obfuscated on how do they actually get the answer other than this. LLM calculated it, Right?
Right.
[00:14:15] Speaker B: That's the strange space we're in. But I don't mean to jump tracks on you, but it is something to be interested in and talking about for sure.
[00:14:23] Speaker A: Yeah. So, you know, this is the Data Ops show. So you talk a little bit about Data Ops and how that fits into this and what your perspective is on Data Ops in particular, what it is and where does it fit in this landscape of doing things like having LLMs and AI and producing data products to deliver business value.
[00:14:45] Speaker B: Right. So I think it's a fantastic concept. And I've been watching this really from the beginning. The folks at Data Kitchen, obviously the folks at DataOps Live, there are a bunch of different companies that are getting into the DataOps space. Basically, it's a separation of concerns. From my perspective, DataOps refers to the operations around data management that must be transparent that must be auditable, that are named in appropriate ways such that you don't muddy the waters. Right. You want to be able to manage data in a marshalling area that is transparent and understandable and has an audit trail. And then you connect it to your analytics and then you connect it to your operations and other places like that. But to have the discrete marshaling area for data management I think makes a ton of sense. And it's what we learned in DevOps. Right. You figure I've always thought and talked about this business, it divide and Dave Wells years business wants to get things done. It says not so fast. We have to maintain all this stuff. I think DevOps came along and really helped mitigate that conflict because you had developers working directly with the business. And then of course what happened is they did all kind of great stuff and maybe didn't document it as well as they should have, but there's always one missing piece somewhere. But now you get automation. You know, I remember Wescape coming along and automating the documentation of data warehousing. This is one of the most exciting parts about where we're going in this whole data journey is that in the cloud. And I think the cloud really absorbed a lot of the principles of soa. Remember we used to talk about service oriented architecture. Even though it's gone away, service orientation has not gone away. And I think the best practices and principles of SOA just got baked into the cloud. And now you see this automated testing, automated development, automating. Basically any part of the process that you can, you want to automate. We had a guy in the DM radio probably 10 years ago, I was like, well, you don't want to automate everything. He goes, oh yes you do.
Yeah, really, as much as possible. But where I'm kind of going with this is if you are responsible in your data management, if you do have a formalized data ops methodology and discipline, and discipline is really important stuff, then you are going to solve yourself so many other problems. You're not going to get problems that you would get otherwise of muddying the waters and not being able to unwind things. You know, that's the danger of the business moving too fast with like a DevOps is that you can't then unwind that stuff very easily and you want to have those audit trails. So to me, data ops is a wonderful, very important discipline, especially in the age of AI, because these AI engines are going to need highly trustworthy data. You're not going to want to point your data at or Your AI, just any data. Because that's what Microsoft did with Tay with the unveiled.
[00:17:47] Speaker A: Hey.
[00:17:47] Speaker B: On Twitter, like 10 years ago, and it just started insulting people like, whoa, timeout. Because they trained it on nonsense. You got to train these things on corporate, trusted, governed data. And that's where DataOps comes into play.
[00:18:00] Speaker A: Yeah. And do you think it's possible for people to even deliver value at the scale they're trying to do today if they're not adopting some sort of a data ops, an automation approach?
[00:18:11] Speaker B: No, no, I sure don't. You know, because you, you. You sort of quickly realize that if you haven't separated out the management of information assets that are going into operations, that are going into analytics, if you can't control that and understand it, then you've lost control is really what the bottom line. So it's important to know what's going. It's like the recipe, right. It's like managing the different products that go into your recipes. You want to make sure you know where they came from. They have, you know, expiration, expiration dates, etc. So if you run a restaurant, you have to be very careful about what the sous chef does. Right. The sous chef prepares everything such that the chefs and everything can do their job and you get good food at the other side. But data ops is almost like that sous chef job. It's like making sure all the products are in line, in order, are appropriate. They're what we wanted. And that's a really important aspect of running a good kitchen, of running a good restaurant, is having a good sous chef. And I think that's what data ops is.
[00:19:17] Speaker A: Oh, yeah. It's interesting because one of the seven pillars talks about basically component design. And, you know, back in the day, and I know you remember this, we talked about modular programming. Right?
[00:19:27] Speaker B: Right.
[00:19:27] Speaker A: Same kind of thing. Getting things down to the smallest possible component, which kind of gets to that SOA concept in microservices again, but in from a data perspective. So we can put together whatever data products we need based on what the consumers of the data require.
[00:19:46] Speaker B: Right, yeah, yeah, I agree with that.
[00:19:48] Speaker A: I mean, I like, I like the sous chef thing. I'm, I'm a. I'm a huge foodie, so that resonates really well with me. It's like, oh, yeah, I just came.
[00:19:57] Speaker B: Up with that and then round. So you inspired me. There you go.
[00:19:59] Speaker A: All right, awesome. That's good. We'll have to. Have to hang on to that analogy for a while here, being this could be a new role. Instead of data Wrangler, You're a data sous chef.
[00:20:10] Speaker B: Yeah.
All right. Well, because they, you know, one of the greater, more interesting and I think appropriate mantras that I heard or just descriptions years ago was that seat belts don't make your car go more slowly.
[00:20:27] Speaker A: Right.
[00:20:27] Speaker B: Seat belts protect your car. And that's governance. And so one of the principles of data ops is governance and change management. And governance is very important because it's like when you're mentally free to explore ideas, which you are if you have a data Ops program. It's just like the chef, if the chef has a good sous chef, the chef can play around with all kind of stuff and doesn't have to worry about something being past its expiration date or something being inappropriate. They know they have that trust in their team and so they can explore ideas. And that's the beauty of analytics, is being able to come up with ideas and have that so called conversation with your data. And when you have good governance, then the business knows they can play around, they can try stuff, they can use this, you know, this new data because they have that trust, because they have that belief in the quality. And if you don't have that, I mean, we all know the stories where people lose trust in data, they just don't use it and they do something else.
[00:21:28] Speaker A: Exactly.
[00:21:29] Speaker B: And so, you know, there's a real psychology to this stuff. And to me, the data ops discipline provides enough governance and enough auditability and certitude to enable the conversation with data, to enable the fun stuff, which is the analytics.
[00:21:47] Speaker A: Yeah. And I think you just answered one of the questions I was going to ask you is how critical do you think data governance is, especially in the world of AI and LLMs and things like that?
[00:21:59] Speaker B: Yep, it's huge. I mean, it's absolutely huge. And you have to have it. I mean, in data warehousing you want to make sure that this is trusted information. It's your certified report capable information. It's what you would tell the auditors, for example. So it's really important, you have to know that whole lineage, what the heck happened. And luckily we have tools that can do that these days. You know, years ago it was very difficult to do that. It was possible, but it would take a lot of time. But you know, the, the component about automation is really important too. Automated regression testing and things of this nature, you know, the more you can automate, the better off you're going to be. So, yeah, I mean, governance is absolutely crucial and automation helps you get there by tackling tedious Things. I mean, goodness gracious. The, you know, the amount of time you can spend manually doing stuff that can be automated, it could be all day long. It could be your entire job. And that's no fun. You know, it's just, it's not a fun job to manually check 10,000 records to make sure that they all have the appropriate fields in them. That's not fun at all.
[00:23:05] Speaker A: Yeah, yeah. One of my earlier jobs in the very early 90s, working for a little software company that was a startup that was building stuff on Oracle, is I somehow ended up being the guy in charge of version control of Oracle forms and reports and DDL scripts on Unix and having to manually create the branches and create, if you remember, tarballs, to send off to the customers for, you know, to be unrolled on their machine or to install the software. I don't know how I ended up with that particular job because I was also the chief architect. Oh, yeah, that was. Yeah, you want. Yeah. Talk about tedious. Yeah, that, that, that was like, oh, my gosh, that was. If I never have to do that again, which I don't expect I ever will. Yeah, that, that's, that's, that's awesome. But yeah, being able to automate that sort of stuff. So that's the getting into things like CICD and version control and all that. Being able to automate that. Yeah. Leaves you a lot more time to actually write good code, figure out what you're supposed to be doing, find the right data, do the right transformations if you need transformations, build the data pipelines. But being able to have something to help you keep track of all of that, so you're not writing over code that did work and now it doesn't work and you've got no way to roll back because you didn't version it, you didn't do a branch.
That's all critically important.
[00:24:37] Speaker B: Yeah, that's right. I love the manufacturing discipline that gets woven into your view of the data ops world. Right. Is like, well, let's think about this, guys. If we have all these different steps in the process, we want to discreetly manage each one such that we can branch something, fix it, and it doesn't take the whole system down. Right. What was the other analogy I came up with a while ago for those folks who are really kind of getting up there in age, you might recall that the Christmas lights you used to buy, like in the 1970s, if one of those lights went out, the whole thing went down. And then someone very clever figured out, hey, if we just Route the circuit twice, then we don't have to have that problem and just the one light that goes out will go out and you can fix it. Because before you would have to go, like, find the light that's burned out.
[00:25:26] Speaker A: Yes.
[00:25:27] Speaker B: To get the whole thing to work again. That was an engineering issue. Like someone sat down and thought, hey, wait a minute, if we just run the circuit twice here, we don't have to have the whole thing of the single point of failure that was the literal single point of failure in the old design of Christmas lights is that one light going out would kill the whole thing. And you don't want that. That's what DataOps has figured out. Especially what you folks are doing at DataOps Live is you figured out how to branch these little bits and pieces such that you can work on something. It's like a mechanic being able to work on a part of your car without having to. Your car have to be in the shop. Right. I mean, that's, that's some pretty interesting stuff that you can address these issues. Well, you also see it with, you know, with the Teslas and some other things where they can just update the software. You didn't have to bring it into the shop at all. They just updated the software. And, you know, that's when the user experience really hums and that's when people are happy and that's when morale goes up.
[00:26:24] Speaker A: Yeah, exactly. And it's like the. Some of the innovations at Snowflake helped with this a lot. Like, Zero Copy Clone is my absolute favorite feature that Snowflake came up with, because now you can not only branch your code, but you branch your data, which saves so much work and so much headaches because you're not going to go in, run a new. We'll just say, for lack of a better example, an ETL script that updates all the data in the database and go, oh, crap, that was wrong. How do I roll back? Oh, now I got to go find a backup tape. I got to do a restore of the database. You don't have to do that in Snowflake. You do a zero copy clone first and run all your tests on the clone and if it works, great. If not, you blow it away and do another clone and do the next iteration.
That was a huge game changer, especially in my mind when I was originally working at Snowflake. It's like, wow, we can do some agile DevOps stuff now with data. And so that got backed, baked into what DataOps Live does on the Snowflake platform.
Yeah, it's very exciting because as you know, how many customers did you ever work with that had enough disk space to actually make a full copy of their production data warehouse in order to test the new queries and get the indexes right before they moved it into production?
[00:27:44] Speaker B: That's brutal, right?
[00:27:46] Speaker A: Yeah. Nobody could do it. You could. And nowadays you're talking hundreds of terabytes. There's no way anybody could afford to do that without something like this. Zero copy clown.
[00:27:55] Speaker B: Yeah.
[00:27:56] Speaker A: So go ahead. Looking at the seven pillars, a true data ops, you know, if you are there any there that jump out to you as, as being, you know, I guess, critically important. I think you already said governance, which is one of the pillars there.
[00:28:10] Speaker B: Yeah, it's a good, I mean they're all good, they're all important. You know, I think collaboration and self service is really important. And you know, self service is hard to pull off. You really have to think about workflows and things of this nature. But you know, collaboration is always important. Self service is important. You don't want people waiting on other people to make decisions about things. You know, that's one of the keys, I think, to success with, especially with analytics. Because if you realize you need a new data source and you have to go file some IT ticket to get someone to come come in, provision the source for you, and that takes, you know, a week or 10 days or something, that's going to kill whatever creative juice you just had flowing.
[00:28:53] Speaker A: Yeah.
[00:28:53] Speaker B: So you want to be able to, to give people their own little sandboxes, their own marshaling areas, if you will, to play around with their data. And this is something I learned about, gosh, almost 20 years ago in the OLAP space with a company called Raza. There was a client of mine that got bought by Hyperion, that got bought by Oracle. It's now the Oracle DRM tool. But they were talking about hierarchy management. And the guy who came up with the idea, Doug Cosby, very, very clever guy, super nice guy too. He said it was for a big bank. And he goes, well, you guys have a hierarchy problem. And they're like, what do you mean by that? He goes, you need to roll up hierarchies by different divisions, by different departments, and give people their own view of the same data. Right. And that's the key. So you don't want to mess with the data layer itself. That's, that's the raw data. But what you want to be able to do is analyze it from this perspective, from that perspective, with different dimensions and that's what we're now able to do much more effectively. But it's very important to have that governance in place, to have the audit trails and then to let people do what they want. You know, years and years ago I wrote my thesis on deconstruction as a, as a concept. And I remember studying Jacques Derrida and he talked all about deconstruction and it was interesting. It's called Structure, sign and play in the discourse of the human sciences is this paper that he wrote. And basically what he did is he said, look, you have this concept of a center, but it's not really a center, but it is central to the integrity of this thing. And the best way I could describe it is to say you have a number line that goes from negative 10 to 10. The zero in the middle is the foundation. It is the center upon which you can even have 10 versus negative 10. But zero is an abstraction, but at least it's a concept that helps you kind of wrap the data around to be able to do some analysis. And I get excited about this stuff because it challenges you to really think through what is it that we're trying to understand? What are the dependencies here? What is the implication for our business? These are analytical questions that are enabled by having some discipline, by having some formation with the star schema, for example, the Kimball versus Inman, all these different ways of thinking. And of course a lot of that stuff is now baked into guess what, the large language models, the LLMs. Because that in my opinion, that really is where everything is going to go. So it makes the data ops storyline much more important because you're going to want these anchors of truth, these embeddings that are your trusted corporate data. And you're going to have a real good chance of getting that right if you have DataOps as a central discipline for managing your environment. So basically it's almost like now or never. People get it right now because if you don't and like a few years go down, you can't untrain models very well, you know, you'll have to start over again and that's going to be very unpleasant.
[00:31:52] Speaker A: Yeah, no kidding. Well, unfortunately we're got to wrap up Eric, and you and I could go on for literally hours about pretty much all these topics that we're just hit on. What's next for you? Do you have any conferences or meetups or anything that you're going to be speaking at in the next couple of months?
[00:32:10] Speaker B: Well, we're always doing DM radio Thursdays at 3 Eastern. I will, I'll promote a couple interesting things. We do have a stealth project we're working on right now with humans, which are unique marketing identified names. So we're trying to solve for the loss of third party cookies. We've got a project, the, the code name is cookie cutter. So we're working on that. I'm also working with a really wonderful group of people in Pittsburgh from University of Pittsburgh, andy Hannah and 1486 Labs. Back to that conversation around alternative data. They're building out a group, an organization that will serve as liaisons for purchasing and selling data at scale. So alternative data, third party data. They're trying to be sherpas, if you will, to help the the business world understand what goes into buying data, what goes into selling data. How can we grease the tracks and facilitate this conversation in this practice? So watch for some some updates about 1486 labs and blue Street Data and otherwise send me an email. Info dmradio Biz I'm always curious to get new folks on the show and and to share more insights about what to do, how to do it, when to do it, where to do it, who to do it with, all that fun stuff.
[00:33:24] Speaker A: Okay. And where can folks find and listen to DM radio?
[00:33:28] Speaker B: DMradio biz we're coast to coast. We're in Los Angeles, San Francisco, Chicago, Atlanta, DC. Our TV show Future Proof now has its own time slot in Washington, D.C. so we're pretty excited about that. And you could be on all these shows by just reaching out and let me know you want to come on and talk about data.
[00:33:47] Speaker A: Awesome. All right, well, thanks for being my guest today, Eric, and you know, thanks for everyone else for joining and hope you enjoyed this episode of the True Data Ops podcast. We're going to be taking our holiday break for the next month or so, actually, and we won't be back until January 24th of 2024. And it's here already. Yes. Hard to believe. So, kicking off the new year, I'm going to be talking with my buddy, the snowflake, data superhero and evangelist for Thought spot, Sonny Rivera. So be sure to tune in for that and as always, be sure to like and repost the replays from today's show and tell your friends about the true DataOps pod. Don't forget to go to TrueDataOps.org and subscribe to the podcast so you don't miss any future episodes. So until next time and next year, this is Kent Graziano, the Data Warrior, wishing you Merry Christmas, happy Holidays, and happy New Year. Bye for now.
[00:34:46] Speaker B: Take care.
