 Okay, we're back here live inside theCubeSiliconAngle.com and wikibon.org. It's theCUBE, our flagship program. We go out to the events and extract the signal from the noise and share that with you. And SiliconANGLE and Wikibon are proud to be exclusive coverage of IBM Information on Demand as the independent media organization. I'm John Furrier, the founder of SiliconANGLE.com. I'm joined by my co-host. I'm Dave Vellante of wikibon.org and we're here with Randolph Kahn, who's the principal at Kahn Consulting. Now just recently, I think it was last week the week before I interviewed Randy in an extensive interview about his new book called Chucking Daisies. So you can get that on siliconangle.tv and youtube.com slash siliconangle. Check out wikibon.org for all the research. But first of all, welcome, Randy, to the liveCube. Thanks, Dave. We've... Great to be here. You know, I've known you for years through reputation and then I flew out to Chicago that time, you know, we met and really have always appreciated the work that you've done. Let's just review. I don't want to do our whole segment again because that was a really good in-depth interview. But you're writing this book of Chucking Daisies. Let's start there. What is that? Just a brief overview of what that is and then I want to talk about IOD and big data and ECM and everything else that's going on. Yeah, happy to. So Chucking Daisies is about, you know, we've spent the last three decades, big organizations, growing their information, parking lots, with all kinds of content, some stuff that you need, some stuff that you don't need and you wake up one day and you're a big organization. You say, I don't know what the heck I have. I can no longer be an efficient business. And the thing is, I can't get rid of stuff unless I have a legally defensible way to get rid of stuff and how the heck do I do that? So Chucking Daisies is really sort of a Bible for an IT executive to say, hey, I got this big pile of stuff. I have ubiquitously ill-managed information parking lots. How do I methodically go through this stuff and get rid of it? And the idea behind Chucking Daisies is, you know, I think most IT executives don't see information having a life cycle. It comes into existence and it's there, right? So the idea with this daisy, this living creature, it comes into existence, but at some point down the road, its value declines. It starts to die or wither, right? And information is no different except the problem is we're not actually living that reality. So Chucking Daisies helps IT people get rid of stuff in a legally defensible kind of way. Is it an operational playbook or is it strategy and operations? It's both. I think that's a great way to describe it. It's that high-level strategy and the business case and then very tactically, how do you attack the problem? How do you approach the problem? Where do you start? What do you do? So like if it was a book for dummies, fill in the blank, blank for dummies, what would it be called? What would the information management? Cleaning house of dead information for dummies. Blowing stuff away in a legally defensible kind of way for dummies. Cleaning house of your digital debris for dip sticks. I don't know. Yeah, I got a lot of shit need to get rid of it. Right, we can't say that, can we? We're Web TV, we can swear. Oh, but I'm not going to. I have a prohibition. I'd like to though, but I'm not going to do it. I got a big pile of... Crud. Data that I need to figure out. I don't know how to access it on tape. By the way, the first time in my entire speaking career, and I've spoken hundreds of times, I used the S-word a few months ago, and honestly, I got read, I was embarrassed, so I'm not going to use it today, and you're not going to be able to make me. Well, right, I'd be a very professional organization, and even the guy we had on and talked about Oracle, didn't even want to mention the name. He said, we had a competitor, and he basically described Oracle. It's so polite, and so nice. Yeah, so we saw it on the slide. I'm not polite. I'm just not going to use profanity, and you can't make me. Okay, all right, so, okay, then go now to Oracle. We'll go back to Oracle. All right, so people have this parking lots of data, and all kinds of compliance nightmares. I mean, the CIOs back in the old days, yeah, they managed everything. They never got phone calls from the CFO, other than, hey, cut the budget. I want to hire. Now they're getting phone calls from the CFO, and the general counsel. I mean, basically the general counsel has got to pick up the phone saying, hey, what's our exposure? How does that conversation happen? Take us through the life of that first ring, like, hey, this is the general counsel. Yeah, the CIO. What kind of conversation is that? Do you see the first call, and then what happens after that? Yeah, so I think actually it's worthwhile going back in time to understand how you got here, because I think it's pretty interesting. You have this situation where for literally for years and years and years, IT departments bought stuff, and they put that stuff to work in their business, never realizing, of course, that every single application of the piece of information technology to some business process creates information, and the information grows unfettered, and one day they wake up and they say, wow, you know, we have a whole bunch of stuff, and we're no longer just the babysitter of the box. We're no longer the babysitter of the pipe. We actually have to care about the content in the system. Now, how did that come to pass? You know, you don't need many privacy failures to understand that information actually is an asset that people care about. People seek to break into the system to garner information, because it's business, right? In the context of audit litigation investigation, you have all kinds of content that may be responsive to a litigation or to an investigation or to an audit, and it's not very long after that thing is filed, the ID department's contacted saying, hey, get me anything and everything, it's responsive to the ABC lease litigation, for example. Well, in that context, the reality is, huge money is expended from the IT department, huge money is expended from the business. They're basically going to subpoena, I want all the emails related to Joe, Mary, and Mike related to the case, they got to go dig in through a bunch of stuff. That's kind of a random example, right? It's not a random, it's exactly what it is, so the reason that the law department now generally has to communicate with the ID department, it's not only, reactively, we need stuff for the ABC lease lawsuit, but even more fundamentally, when I buy IT, if I buy information technology applications, hardware, software, some appliance, whatever it is, I care up front now in a kind of way about my governance issues, my compliance issues, my regulatory and legal drivers, like I never did before, right? And that's just the reality today. So that's absolutely a conversation that now happens on a regular basis. No, I'm sorry, no, you go ahead. So how is big data changing this? Because I talked last week or whenever it was about information as a liability, but increasingly people looking at it as an asset, and not that they didn't before, but now with big data, does that outweigh the general counsel's concern or is he or she still the tail wagging the IT dog? Yeah, so let me just tell you something, lawyers should never drive the business conversation. If you're doing that, you're going to be out of business rapidly. But it happens, right? Let me just tell you something, they should not be driving the conversation. They're a part of that conversation, but they can't be driving the conversation because in the end you have a business to run it. If you're not running your business, you're not focusing on business stuff, you've missed the point. So that's the first thing. As it relates to the lawyers or the law department, just like the IT department, they're a service bureau. They're a service bureau within the organization for risk mitigation and risk avoidance and litigation response or whatever. But let me go back to the heart of your question, because I think it's really interesting. And there's a rub here. So if you go back to big data, the idea with big data is if we could take all this mass of stuff across various information parking lots, and I can get into that information and connect dots and map out the future and see connections that I didn't. Using technology, there's some business advantage to that. That drives the conversation, but wait. But wait, is that the end of the conversation? No, if I keep that stuff around and it no longer has value, is it a liability? Absolutely. If I keep that stuff around, it's no longer being used. Can it be access in the context of audit litigation investigation? Absolutely. So as the next phase of that conversation, big data says, hey, I may have some value, connect my dots, and then there's also a conversation that says, if I'm a liability, I'm no longer utilized, I'm no longer a value to the enterprise, there's got to be some weight in a legally defensible kind of way without creating liability, without causing us heartburn to get rid of it in a methodical way. And I have to say, it's every big business's challenge today to address that. Okay, so, and we talked about this earlier, but I want to dig into it a little bit. So how do you do that? Do you say, okay, everything that's a work-in process that's where the project's completed goes? Every sort of development prototype that we don't need anymore is gone? Or is it on a case-by-case basis? I mean, how do I scale that decision point within my organization? How do I accommodate the massive scale of data that's coming in and actually do a good job? Great question. So I started a business last year called DELV, and DELV's pure reason for being is to help organizations legally, defensively get rid of huge chunks of data. To be able to do major house cleaning in an IT infrastructure today for a big organization, it's a really complex undertaking, right? Exactly as you suggest. Two things have to happen. Number one, I need to make sure that I've taken care of my record-keeping responsibilities. I have this huge policy, I don't know what it is. Well, blowing it away without regard to the laws that say, thou shall retain that stuff, you can't know if you've met your legal requirements, your legal retention requirements, unless you know what that stuff is. So there's gotta be some diligence around what the heck is that stuff? So that's the first level of analysis. The second level analysis is, even if I'm not a record, even if there's no laws that say, thou shall retain me in accordance with A, B, and C regulation or laws, if it nonetheless is potential evidence, it also has to be preserved. So again, you look at this gargantuan mass of stuff, the only way that I can blow it away in a legally defensible kind of way and sleep at night, right? And not where a potential liability is two levels of diligence. One, am I a record? Not, okay, great. Second of all, am I evidence? If I'm evidence, I nonetheless have to be preserved, right? And to do that again, in today's world of hundreds of terabytes or petabytes of content for a big organization, you need to harness technology to be able to do that. Now when you say are you evidence, you're saying are you active evidence, or what if you have the potential to be evidence down the road? Yeah, so, well, down the road is a more complicated question. Let's try it this way. The idea under the rules of civil procedure and the rules of evidence is that it's not just evidence that I know of, it's that something could potentially become or lead to evidence. And so it's a broader definition that what you need to actually preserve is broader than what is evidence today. But the point you make is actually slightly different than that point, which is, do I need to predict the future of what is needed in the end of that, is no. Now, do I, if I'm on the throes of an investigation, if a federal regulator is knocking at my door, clearly I'm not going to turn around and say, fellas, hurry up, get the fire a-burning. So that's clearly not the reality. But yeah, is there this sense that you need to look down the road and possibly know a kind of hostage that you have or complaints that have come in, seeing where that's going so that you don't proactively destroy stuff that you know or should have known you needed to preserve? Okay, well here's another corner case that I want to run by you. So in the instance where, okay, I don't have to predict it, so great, let's get rid of it. Now, if in fact down the road, that record that I deleted becomes evidence in a case, and I've deleted it, okay, well, so it's gone, but I don't know who else has it outside of my organization. Yeah, that's that's. If I'm the GC, I'm saying, well, don't delete it because at some point we might need it and a plaintiff's attorney is going to find it on some client's laptop. Yeah, so let me just tell you something from whether it's Dell or my other business con consulting. We're never counseling people to keep stuff and retain stuff because of the eventuality of maybe needing it years down the road. That's not the way records management happens. That's not the way record policies are built. That's not the way records retention rules are developed. Get rid of stuff. This is the deal. When you develop the stuff, here's my business need for that stuff, here's my legal requirements, and I smush that together. I come up with a rule, it says thou shalt live three days, three years, three decades, whatever the heck it is. At the end of that period of time, absent a loss or investigation, threatened, imminent, anticipated, you get rid of that stuff. If it doesn't exist thereafter, it doesn't exist. Now, the point you made is a different point. Let's just say there's a random copy out there. I mean, again, one of the reasons that Delve is busy as heck, helping lots of big organizations defensively dispose of stuff is, this is a reality for everybody. I have all these systems, I have all these copies, I have all these systems across all these copies. How in the heck do I make sure that that stuff isn't around? So again, I think that it's not even some third party having it. Imagine most organizations having so many copies and so many versions of the exact same thing across their enterprise, trying to manage that stuff. It's a gargantuan headache. Yeah, so what's the answer? How do you know if you delete something that it's actually gone? Yeah, so I'll give you a perfect example. We're doing a project right now for a big financial services company. The first phase of that project is doing nothing but finding mathematically, numerically precise duplicates, not near duplicates, not close to this thing. Not a change period. No, like exacts. And even that exercise, I mean just taking away exact duplicates, multiple, in some cases, dozens of instances of the exact same thing replicated over and over and over again, just taking that out of the mix. It's a gargantuan savings in and of itself. By the way, it's not that simple. You'd think it'd be a really simple exercise, but when you're talking about petabytes of data for big institutions, being able to execute upon just that simple task in and of itself is not simple. I mean, I want to ask you, on that point, what is missing technically? It's for the entrepreneurs out there. I know guys working on some big data cool tech that need a solution for their tech and want to navigate. Is there anything that you could share? It's supposed to say, hey, super geeks. Alpha geeks. Solve this problem. If you could build X, Y, and Z, talk to me. Yeah. What is that? Yeah, so what I will tell you is there's some incredibly wonderful technology out there and there's some wonderful technology that take care of a whole bunch of complicated IT issues. In the end, to take on a defensible disposition, it's not a simple task. It's not a task for those that don't want to invest some resources and some money. And if you can make that process simpler, make the analysis cheaper. What's the mix of manual automation right now? When you go into these gigs, is it like all manual? Yeah, in the Delve side of the house, all the heavy lifting is done by technology. Delve has scores of really smart people that do stuff, but in the end, without good technology, nothing's getting done. Got it. So I'm looking at the Delve website. It's Delve, by the way. One of our guys listened to our interview the other day. Yeah, because I started. And he said, I loved it. And he said, was he saying Delve? I'm like, yeah, Delve. He wanted to rate it up. So I'm looking at one of the things is auto classification, right? I mean, that's like a critical barrier now enabler to be able to defensively dispose at scale. So what do you use technology to do that? You use whatever technology works for your clients. There's a variety of technologies in the marketplace. And what Delve teams do is they go in and they actually train your retention rules with your content to a system. So once I train your rules, simplify. First of all, we come in and simplify the rules and then train those rules to the system. And then it crawls your content. It learns your content and it tries to make business classification decisions on your content across, again, hundreds of terabytes of stuff, taking all kinds of text files to classify those. So then again, in the ordinary course of business with that diligence process, they can go away. But you're a consulting firm, Delve, right? Yeah, it's absolutely. Are you inventing IP? We do not have any of our own technology. We're using technology in the marketplace. And in many organizations, they've already bought technology and they want to harness this value. They don't know how the heck to use it in the context of water classification. So we're helping them. So you use whatever the client wants to use? Well, assuming that it works, assuming that it's going to do a sufficiently good job, the answer's yes. Will you make technology recommendations to that client? We do. We do. If they have a void and they need to fill that void, we certainly will help them make that decision. Sure? Awesome. Okay, we got to run. Okay. Chuck your daisies, people. Okay. Get rid of that crap. I'm not going to say it either. I've never sworn on the cube. You know, they got to watch your data. It's parking lots of data. I mean, we have great food for our bloggers on this segment. So guys, great content. It's a services angle. It's a business angle. The tech angle. Chucking daisies, Delves. Great, great consulting firm. Randy Kahn, thanks for coming to the cube. Okay, we're right back with our next guest here for this short break.