Drill to Detail - Drill to Detail Ep.37 'How Essbase Won the OLAP Wars' With Special Guest Cameron Lackpour

Episode Date: August 1, 2017

Mark Rittman is joined in this Summer Special episode by none other than Cameron Lackpour, Essbase expert and Oracle ACE Director, to talk about why and how Essbase won the OLAP wars, how Essbase Server works and the role it now plays in Oracle Analytics Cloud and his involvement with user groups over the years. In this specially extended edition he also gives us his reading recommendations for while you're at the pool or, as he will be, out camping, and he also shares his predictions for what we'll hear from Oracle and the analytics industry when he, and Drill to Detail, returns in the autumn after a well-deserved summer break.

🎵 So welcome to the summer special episode of Drill to Detail and I'm your host Mark Rittman. I'm very pleased to be joined today sitting on the virtual beach drinking a virtual cocktail by none other than Cameron Lacpore who many of you will know from the S-Space world and from the Oracle Development Tools user group. So welcome to the show, Cameron. Pull up a virtual deck chair, and why don't you introduce yourself to the listeners? Sure. So as Mark said, I'm Cameron, Cameron Lacpore. I have been in the S-Base space since 1993.

Before that, I used a tool called ComShare System W, which was sort of a multidimensional predecessor to S-Base. Since 93, I have worked for myself with a couple of stints of working for other companies, but back to myself again. Doing S-Base implementations, doing Hyperion planning, planning a budgeting tool that's wrapped around S-Base, as well as some data integration work that, again, feeds S-Base and planning for an awfully long time. Sometimes I say to myself, with grief, can't you get out of this rut? You seem to do the same thing again and again.

I still enjoy it. I'm still at it. with with luck i'll be able to retire with it excellent and so cameron we know each other through the user group scene as well so just tell us a little bit very briefly about what you do with user groups and how we met through the od tug group so uh yeah we met i think in 2010 um i knew of you i you. I'd never had met you. We were both on the board of ODTUG at the time. I was on the board of directors for six years. I term limited out. I'm very happy to say that I was the one who championed term limits.

We didn't used to have them. But now you can't stick around forever. I think it's a good thing. ODTUG is a fantastic organization. It's one of the big Oracle user groups in the States. Really, ODTUG is international, although I guess it's sort of North America focused because of proximity to where most of the developers are. It is a technical user group. It is just fantastic. It's absolutely transformed my life. I mean, meeting you was one of the's, it's absolutely transformed my life.

I mean, meeting you was one of the things, meeting all the people that I'd known online for years and years was another part of the thing. Um, not having it be a vendor-based, um, conference Hyperion was the, the company that was acquired by Oracle. That's where S-Base came from. Their presence, their conferences were fine, but they were sales conferences with some technical stuff mixed in on the side. ODTUG's not like that at all with its Kscope conference. Yeah, I mean, so through Kscope, I met the people that I work with and call colleagues, even though they're literally

all over the world. It's a fantastic group. Yeah, it it's great i mean so so the reason i wanted to get you on there on the show was we had we've had quite a lot of kind of discussion about about olap in the last few episodes and um it would be interesting to get you on to talk about s space because um anyway s space is the olap server that you know won the olap wars and um you've been very closely associated with with s space over over the years that's how i've known you and um so you know back last year we had um my old boss uh graham spicer um on the show i'm talking about oracle express and the o-lap option um which obviously for many years was a competitor to to s space and then both in the end ended up in the oracle world and then we

had more recently we had chris webb who was a bit like you an independent consultant but working with analysis services and then we actually had donald farmer a real kind of a real um treat for me actually had donald farmer on the show you know one of the pms responsible for analysis services which is olap services for microsoft you know back in the days and he went on to uh to click and so on but airspace effectively goes on and on and on and you know and essentially at its core it seems to me like like it's the same core technology, the same classic technology from when Oracle first acquired it back in 10 years ago. So just Cameron, first of all, just explain what Airspace is for the audience who may be a little bit kind of younger than us and maybe haven't worked with OLAP servers. And just tell us, you know, why do you think it's endured so long so what is S-BASE and what does it do and and just at a high level we'll drill into the detail

later but at a high level why do you think we're still working with it and it's still popular now so S-BASE is an OLAP tool OLAP for the young ones amongst us is online analytical processing it's a multi-dimensional. It's not a table. It's not a relational solution at all. It views data as intersections of dimensions, dimensions being descriptions of properties. So the classic one is you are a company that sells products. The concept of sales, the concept of an account or measures is one dimension. That company sells it across a geography account or measures is one dimension.

That company sells it across a geography. Geography is another dimension. It's the counties of England. It's the countries of the EU. It's the states in the United States that roll up into that. I live in Pennsylvania. Pennsylvania is part of the East, part of the United States. East rolls up to US. When I add up all the states in the US, I have total US sales.

Products might be another dimension. So it's this concept of intersections of data that drive S-Base itself. It's a client-server product. It still is a client-server product until its recent webifying. It sat in a data center somewhere in your organization or at a managed service provider. It is static. I mean, I think that's a very fair comment about the tool. It has made me wonder, you know, why is this thing still here? Why am I still able to make a living off of it?

It is because fundamentally it is a very useful product. It is powerful enough that it can handle moderately large data sets. It is simple enough that you don't need to be a data scientist to do it. You don't need to be a hardcore programmer to do it. I would say that in the small group of people globally that do S-Base, a handful of us are what I would call traditional programmers. The rest of us are some mix of business and some mix of technology, probably trending more towards the functional side than the technical side. So I think that's very different, Mark, from the kinds of guests you've had on before. That's why I don't understand most of the podcasts you give, although I do try.

It's interesting that you view it as winning. I suppose when you're on the winning side, it doesn't seem that way. But I think it won the OLAP wars because it was easy to use. As I said before, you don't need to be a technician to make it work. It was sold to power users, not to IT departments. I think that was a big part of it. I am the S-based cliche. I started off with an S-based server under my desk.

It was 1993. The memory was 256 megs. I think I figured out I could have bought a mid midsize Mercedes for the money that that memory cost. Just to show you how things have changed. But that kind of revolutionary approach where IT was sidestepped, where the tool was something more than an access database, that was a big deal. The other thing was that people like me could build applications much, much faster than a product like Express could, as an example. I was still, I worked for Johnson & Johnson.

That is where I was exposed to S-Base. And I was at a subsidiary and I had to build a P&L application, a profit and loss app. And they weren't convinced of S-Base. S-Base was this new and dangerous thing. And so they brought in a consultant to build the same application in Express. I was new to S-Base.

I built it in a week. I had it built in a week and validated. I was done. The Express guy thought I cheated. He thought that I must have come in before him and built it and didn't it and didn't tell anybody that it was, you know, a total setup to him. But that wasn't so. That continued with Oracle, just jumping ahead a little bit. I was on a project, I think in 2008, putting in a Hyperion planning implementation.

So again, the budgeting layer that is wrapped around S-Base. They had had Oracle Financial Analysis. I think that's what the product was called. OFA is what everybody referred to it as. They couldn't maintain it. It was too techie. This was a finance group and it was literally beyond them. They own planning. They've owned it for 10 years. I would love to go back from a monetary perspective. They don't need me. Planning and S-Base is simple enough that they can master it. A big part of mastering that tool in one of the two storage engines, the old storage engine, the legacy one, is called block storage, BSL, block storage option.

The language is like Excel. The language is not the same as Excel, but if you can write an Excel formula, you can write the S-based language, the BSO language into what we call talk scripts. Again, this difference between, you know, well, you can't see me doing the air quotes with my fingers, but real programming and S-based. S-based, it's a text file. You can use it in a text editor and save it to the server. You can use one of the editors that the product provides and write the rule there. Simple to write, simple to test. That's a big deal. That's a big, big deal.

So I think there was product productivity, product functionality that S-Base had. I think other tools probably had it as well. But it was successful. It won the OLAP wars because it stuck around. I think Hyperion bet big on it and didn't give up. Oracle, well, Oracle's so big, I don't know about betting big on it, but they kept the faith and kept the tool around where other companies said, ah, you know, we're really not making any money on it or really this isn't where we want to go.

You could call it inertia. You could call it vision um but espace has abided and it might be the last man standing because it's better it might be the last man standing because oracle is the only company that cares to do that yeah definitely i mean what's interesting is that we're starting to get olap being rediscovered by the the big data generation there are there are there are products out there that are built on i suppose kind of hadoop technology that are starting to think about olap and you've also got i suppose something that's happened with a lot of olap technology or olap use cases it's gone to kind of in memory technology or even just relational i mean a lot of people don't really

understand you mentioned block storage there and what just maybe just explain a little bit what block storage is and and and why i suppose anyway why it was a particularly well suited for i suppose hierarchical aggregations and um and for example planning and budgeting applications this concept of being able to kind of insert data at any level and so on just paint a picture of what block storage technology is and and for you know the audience being people who who probably aren't so aware of that, but are used to in memory or relational or whatever. How would you sell it? How would you sell it to people? How would I sell it? Well, I would tell you that I don't think anybody would create block storage S-Base. And I'll explain the technology in a second today, because it was a workaround around memory limitations and size limitations and speed limitations

that were valid in 1992 when S-Base was created and not today. But it exists today because of inertia, because it's in so many tools. What block storage is, is it looks at valid intersections of those dimensions. So I sell, I don't know, I sell coffee in Pennsylvania, but I don't sell tea because of 1776 and the you know the American Revolution all went wrong there basically when you start

drinking coffee instead of tea that's why you lot are so uh so excitable really I have an electric tea kettle too we can we can get real tea here as well but anyway, a constant criticism of America. So the intersection of tea in the UK and coffee in America, data exists at those two combinations. I don't have tea in the US. So S-Base doesn't need to track that there's this concept that there's a product called tea and this geography called the US. So block storage data only stores intersections where data exists. That makes sense. It's got a pointer list, an index that goes off

and points to it when it's either writing it or reading it. And that's sort of the trick. It doesn't need to think of all the possible combinations. It only needs to think of the combinations that are there. What block storage also does, and again, this is sort of an artifact of its need to put things on disk instead of keeping them in memory because memory was so limited when it was designed,

is upper-level combinations of data are also stored. So there is a longer aggregation time taking those numbers and adding them up, doing whatever the math tells it to do. But when it comes to retrievals, S-base only needs to pull in a small number of blocks to provide the answer if you're looking at an upper level combination of data points. So that's the way BSO works. It is adequate. It's certainly powerful. The language that's around it is quite powerful and is accessible to business users, to power users. Then there's another data

storage model called ASO, aggregate storage option. That's an in-memory database. I mean, it's got all sorts of nuances around it, but I mean, data comes in at level zero, right? The base data is there. And when you, the user, retrieve it or a tool retrieves it, it adds it up in memory. Now, there are some tricks around materializing aggregate views for performance, but that's optional. And it works in a very different way. It uses a bitmap index to get there. It's very different from the way PSO works.

I sure wish they had one engine. It would make life easier. It's transparent to users. A person looking at one database versus another in different technologies, don't know don't care that that's great but it causes a fair amount of angst on the developer side which tool do we use how do we do it here why isn't it the same etc etc so what was what was what was a hybrid then i mean hybrid came along fairly recently or certainly the last few years wasn't that an

attempt to try and combine the two types it is so. So I think it's maybe two years old, maybe a little bit older than that. I don't know. I was so early on the beta, the dates no longer make sense to me or they no longer work for me. So it is, functionally, it is BSO at the bottom level. So it is stored intersections of data

and then this in-memory calculation engine on top of that. The goal, I think, from Oracle's side was twofold. One, keep that traditional block storage option calculation script language going and make it part of ASO, which ASO does not use. So the ASO databases are MDX-based, right? Multidimensional expressions, a totally different, arguably more powerful language, but one that makes people in the finance department lose their mind trying to do it and maintain it and create it. So they tried to bring the two together, keep the calculation power of BSO available, throw on

the in-memory aggregations that ASO does. I have used it in planning applications, and actually I presented on it this year at Kscope with Pete Nitschke from Empower in Australia. When you combine the two of them and do it intelligently, it's magic. A lot of the performance concerns that are around traditional BSO, they're gone uh it is that fast it does it is now closer to what i think a modern architecture would be for an olap engine maybe it's practically there okay okay so so i mean as you say airspace has been through although we said it hadn't essentially hadn't changed obviously it has changed over the years and you mentioned aso which i i got the impression that was a reaction to uhAP services from Microsoft at the time.

I mean, what did you think? I mean, we had Donald Farmer on the show a little while ago and Chris Webb, and they're obviously singing the praises of analysis services. What was your view on analysis services, and how did it kind of affect your world at the time, and what do you think about the way they've gone towards tabular kind of storage, and what's your view on analysis services, really? When I heard of it first, I thought it was going to own the world. I just looked at it and said,

I have to learn this really, really fast or I'm going to be unemployed. And then it didn't happen. I was amazed. I really was. I think architecturally it was a very strong product. It was integrated into every copy of SQL Server there ever was. I think architecturally it was a very strong product. It was integrated into every copy of SQL Server there ever was. At least it was a box you could tick on the install. And so why didn't it succeed? And I think the reason that S-Base has continued is it wasn't tied to applications. When Hyperion merged with Arbor,

and strangely enough, the merger was actually Arbor acquiring Hyperion, but in the event there was no one from Arbor that was actually running the company, it was actually Hyperion. I don't really understand the machinations behind that, but that's how it worked out. Hyperion very quickly said, we're an applications company. We do technology, we do software, but we don't build databases. And so they wrapped Hyperion planning around S-Base.

And then S-Base had a reason to exist outside of a pure database view. Analysis services never seemed to do that. If there were sophisticated applications that used analysis services under the covers, I never saw them. And I think that's why it didn't win the war. I think on its merits, it probably should have. Yeah, it was certainly, I mean, it struck me as a very easy to use technology. And it's very well distributed. And it was very powerful and so on.

But as you say, there were no applications there. And I think, I mean, looking at, at say another one another for example remember holos remember holos at the time that was a uh around the same i mean i was working in at mars at the time i remember working on a holos system there and uh i mean what what was your did you ever come across that well what was your view on on that technology no no i i did not i did not use it. got out of school at the same time since we're practically the same age and um it was all it was a mainframe product called system w and there was an element of a pure database you if you had an ibm mainframe like maybe it was on other machines but ibm was the only place i saw it you can do analytical databases just like you can do analytical databases in s-based or analysis services

um that isn't where they sold it i I think that was their initial audience, their initial customer audience. But they went to executive information systems, BI, before there was BI. And that was where they made all their money. It was, again, that concept of an application wrapping around a technology and then having a strong message about what that application is. And by the way, that's actually how I got into S-Base. I worked for Johnson & Johnson. J&J was a huge ComShare customer. They tried and could not make the transition off of a mainframe or the mainframe. They licensed Arbor Software, Arbor

being the company that created S-Base. And for a while, they were the largest reseller of S-Base. I thought for sure that ComShare was going to buy Arbor. They were so dependent on it. And I don't know if you know the story about the lawsuit, but the deal with ComShare was it's a read-only database. Your developers, your staff, your IT group, they can feed this application, the S-Base databases that were powering the ComShare EIS tool.

Decision, I think, was the name of the product. But they couldn't use it in read-write fashion. And one of the things that BSL S-based, and again, we're talking about the original engine, does really, really well, and why it's so successful within planning, is it's great at taking users submitting data, and then performing calculations on it, and then viewing it at an aggregated and calculated manner. They weren't allowed to do that. And some sales reps in Europe said, yeah, yeah, yeah, that's what the license says, but you can do anything you want.

And they could. The tool wasn't limited based on the license. Arbor found out and essentially sued them to oblivion. The Comsure, shortly after losing the lawsuit, was acquired several times, the product spun off and off and off, and now they just don't exist. Don't lie. When it comes to product licenses,

I think it's the best thing that I can think out of that one. Yeah, it was a disaster. Definitely. So you must have looked on quite, you know, when you saw Oracle and Oracle OLAP and what they did with oracle olap after you know acquiring um uh express server from ioi you must have looked on and with kind of part sort of schadenfreude and part kind of horror and part amusement at what oracle did

with oracle olap in as much as i think they did a very interesting thing technically in that they incorporated into the database but you know my view was always at the time they built a product that didn't have a customer because they were building an OLAP server for DBAs, and DBAs don't buy OLAP servers. And this was something, we've kind of covered this topic a little bit on the podcast in the past, but that schadenfreude and amusement must have turned to slight horror when you heard that Oracle were acquiring Hyperion. What did you think at the time?

What was your thought when you heard that Oracle, the big Oracle, were coming along and buying S-Space and buying Hyperion? Actually, were you pleased? Actually, I was. I mean, first of all, I thought, well, they could be buying it just to kill it. So wait and see on that. There was no way to know. But I was a partner at the time. I was a Hyperion partner. And Hyperion veered between we love partners to we consulting arm, you know, consulting company that had a software arm. Oracle is very different, right? That's not Oracle's deal. There is an Oracle services group, but at least in the EPM world, at least in the S-based world, they're not a dominant force. Initially, there was no change. I guess Oracle always pauses before it digests whatever company it's acquired. It just didn't change anything.

It didn't go away, so I was happy. And then as it became Oracleized, I saw much... It's really funny. Who would think that a gigantic company like that would be open? But I saw a total mind shift in how the people that owned S-Space treated their customers and their partners. I've been very happy with it. I think amongst people that worked for Hyperion and then were acquired by Oracle, I think they had the greatest culture shock. I think Hyperion was a smallish company, big, but not big like Oracle. What is like that? I think they found becoming Oracle employees difficult. A lot of

them left. A lot of them went to partners. A lot of them went to competitors. I guess that happens with any acquisition. But I think they had a much harder time of it than customers did, a much harder time of it than partners did. A comment about Oracle OLAP. You said that the product was created almost by technologists for technologists. Who were the real people that were going to buy this? What that tool did was, because it was talking about S-Base's success, S-Base was not tied to Oracle. It can read from Oracle. It can still, you know, go against tables and pull data in that way. But it's never been an Oracle-focused tool. I've done probably more implementations using SQL Server than I have using the Oracle DB.

It doesn't care. And that, I think, had a lot to do with the success as well, because companies like SAP couldn't look at the tool and say, well, you'll never work with us because you're from Oracle and it's tied to the Oracle database. We don't use that or they're evil or whatever the reason is for one competitor going after another. S-Base kind of floated off to the side, too small to be crushed and useful enough to stick around. And a tool like Oracle OLAP, for everyone who wasn't an Oracle customer, it was too closely tied to Oracle. I think that's why it didn't succeed. Again, on the merits,

I don't think it was a bad product at all, but that was a case of how it was marketed. Yeah, definitely. I remember going to the Hyperion i think it's the hyperion solutions event in uh in nice uh with mike darren actually from oracle at the time we went to there and remember it was the most expensive taxi driver i've ever had in my life from nice airport to the hotel actually which was interesting and um but it was i remember at the time it was the first event it was the first hyperion event following the acquisition by oracle i remember seeing a few um a few kind of oracle a few hyperion staff including one one or one person who i knew from

oracle who worked on the oracle olap team that had just joined hyperion and and was then facing the thought going back to oracle again and it was it was interesting and but what struck me particularly was two things one was like you say the apprehension from people you know what's going to happen because i think a lot of people saw oracle rightly or wrongly you know driving s driving um driving express into the ground or certainly it was no longer as big a player in the market as s space but also um it was a it was a strange technology as well i remember looking at things like outlines and and so on and the technology you guys used was was i suppose it was it was it was written with the accountant in mind or the financial person in mind rather than the program wasn't it I suppose it was it was it was written with the accountant in mind or the

financial person in mind rather than the program wasn't it I think it was certainly I think S-Base's sort of roots were more from the kind of the spreadsheet side and I think for me it was a product that was built for finance people not for programmers would you agree absolutely absolutely that that was key to success it when yeah that product was new, the number of true believers, the number of evangelists amongst customers, amongst FP&A, finance, planning and administration guys, was huge. They loved the product. Again, it wasn't access. It wasn't waiting for IT to build a table and build a bunch of SQL queries for them so that they could look at numbers. They owned it. They understood spreadsheets. They understood Excel. Actually, initially, it was both Excel and 123. They understood the language, which was similar to Excel formulas. I guess you could argue with all the symbols it uses.

Sometimes it was similar to 123 formulas as well. And that was, if you think about it, who would buy a product like that? An IT department would never buy it. They say, is it relational? No. Is it from Oracle? No. Is it from SAP?

No. Is it from Microsoft? No, at least initially. They didn't want any part of it. Finance groups really didn't want any part of IT if they could help with that. That sort of tension still exists today. And off it went to the races. It is a very different way of thinking about how to develop something than the traditional program review.

That's why I said we don't have people like that. I could think of a handful of people that I personally know that are true computer programmers, but they're absolutely the exception, not the rule. Yeah, yeah. So Oracle then took the products on, and I suppose the core OLAP server has stayed constant, but there's been a lot of administrative and infrastructure additions over the years, and I think some of them look like they came out of actually Hyperion's own thinking, so some of the dimension maintenance things I've seen, and EPMA, and so on, but I suppose a lot's been added into S-Space over the years around infrastructure.

And arguably, maybe for a certain point, maybe sort of like five years ago, it wasn't quite so easy to maintain. I mean, would you say that's true? Or what was Oracle's contribution to it over the first few years, really, of things? They made it an enterprise tool. I mean, that changed everything.

No one today, hardly anyone, no one, I think it's probably safe to say, has an S-based server sitting under his desk. I mean, it just really literally doesn't exist. People like me who are crazy enough to do blogging or do things for presentations, do I have a VM that has S-based on it? Absolutely.

Do I run anything other than the smallest databases in the world that I build just for the purposes of talking about it? No. I mean, it's strictly a, really literally, it's a demo box that I can modify. It's sort of a running joke about infrastructure with S-Base and planning and all the other enterprise performance management tools that Oracle has. It's just really, really, really hard, which you may have noticed Oracle has chopped away by going to the cloud, by getting rid of the complexity. I'm positive there's not an Oracle EPM customer anywhere in the world that says we love Oracle EPM infrastructure work it's pretty dire stuff but I guess it's necessary isn't it yeah I mean you need I suppose I think it was it was Donald Farmer actually he was talking about how why he went to go to to click was

because actually taking a product that's successful on the desktop and making it enterprise level is a hard task isn't it I mean it's necessary but making it enterprise level is a hard task, isn't it? I mean, it's necessary, but actually it's quite a hard thing to do, isn't it? Without losing that original, I suppose, kind of spirit of what the product was in the first place. Yes. And that's absolutely, like I said, that's absolutely gone. That kind of skunkworks approach to S-based doesn't exist and can exist. Part of it's because it's mature. Part of it's because there is a body of knowledge and practice around it.

But in part, the other part of it is, A, it's expensive, and B, it would be an extraordinary FP&A person that installed S-Base himself. I guess it's possible, but again, no, not particularly likely. So yes, that kind of excitement is gone. I suppose, though, cloud is actually maybe the reboot of that idea, isn't it?

We'll get on to cloud later on, but certainly I remember sitting in an open world presentation a couple of years ago with Oracle talking about how cloud was almost the reboot or the respin of running S-Space under your desk. I think that's true. I do think that's true. I do think that's true. It takes all that complexity and makes it oracles,

not the customers. That's a very powerful argument because, again, I don't think there's any customer out there that likes it. Why would they? Why would that be something that they want to do? Okay, okay. I hope that it is to revert to S-Pace. We'll get on to cloud in a bit so so just

one other thing i mean just from before we get into kind of i suppose more more recent times um so the hyperion user group scene what was it like and and how did that end up morphing it how did that end up morphing into into being a big part of odys hug what was the kind of story there really it's it's a really interesting uh journey it took so hyperion solutions was a korean uh hyperion's conference that they ran every year kind of a it's kind of analogous to open world there's some technical content there's some business content i thought there was an awful lot of sales content and i i actually got to the point where i would spend most of my conference time sitting in my hotel room working uh because i really didn't want to see a sales pitch.

It was important enough for me to go there and to pay for it, but it wasn't important enough for me to actually attend much of the conference. It was a very frustrating way of doing things. Or Hyperion had regional user groups. They were called Hugs, Hyperion user groups. They were, some of them, you know, like anything, some of them were better run than others.

They were funded in part by Hyperion. I went to Hyperion user group meetings in New Jersey. They were in the North Jersey sales office. That's something that we would never see today with Oracle. So Hyperion kept control and influence over the user groups. And then Hyperion was acquired by Oracle, and all of a sudden, they all went away. I don't know whether Oracle did that out of neglect or whether that was the plan. But all of a sudden, there was a vacuum for user groups that hadn't existed before. The story that I'm told was, or I've heard, was that Oracle had picked OAUG

as the user group for Hyperion people. Mike Riley, who was then president of ODTUG, looked at this, saw an opportunity, reached out to Tim To, I guess he's now an ACE director alumni, and Edward Roski, a fellow ACE director, and said, what can we do about this? And basically stole the Hyperion users and brought them into ODTUG. I was there. I was there at the time. I was on the board at the time. I remember it all. OK, you were on the board. OK, so this part I did not actually see. But what a brilliant coup. He reinvigorated ODTUG. He brought in, I guess, now it's about half.

Maybe it's more than half of the members are EPM people, Hyperion people. It's just utterly transformed what ODTUG does. OAUG still has an EPM and S-base component, but it's small. OAUG is huge to begin with, and I think EPM is sort of lost within their conferences. But I talked before about how ODTUG has really been transformative. It's been great. Again, when I would go to Hyperion Solutions conferences, I would just

roll my eyes and say okay i've gotta work my way through this but you know i'm doing this only because i have to not because i want to od tuck is fun um it fun means a lot fun makes things worthwhile um i've been incredibly happy with them yeah and i i've never done things like writing writing books being with you if it hadn't been for the big kick in the rear end that very first kaleidoscope gave me when I finally was in a room with a bunch of technical

people and said wow I don't know nearly as much as I thought I did I better up my game I remember at the time that I think Edward came to us and it was Edward Roski came to us and said that I suppose the Hyperion users were without a user group. And he brought them to OGTarget. I remember at the time, actually, there was almost concern that it would be too overwhelming and that there was so much energy and so much enthusiasm from the Hyperion kind of users.

That, yeah, it was almost concern that that it would be two groups within one that wouldn't actually i suppose integrate but it's been what's been interesting with with k scope and od tug is how it has integrated how there's a very distinct uh hyperion uh i suppose community there and actually sub communities because there's a space there's planning there's other things as well but it's not not two groups within one. It's one big group. And Edward's been, and yourself, and Tim, Tim Tao, have been kind of really integral parts of the user group going forward. So that's been a real shot in the arm for OG Tug.

Some fantastic conferences. I mean, I'm never one for doing singing and dancing and some of the things I had to do. I think when I was on the board, I had to take part in some kind of quiz. I never felt so uncomfortable as doing that. But I think it's been, I think that was that the Queen Mary? Was that the one where it was in Long Beach?

I'm not sure. But it was definitely, it was very interesting. And it's certainly, I think what's been interesting, I guess, with the Hyperion user community is how distinct and how vibrant it still is, even to this day really um so so

that that's been that's been interesting and but looking at what oracle have done with the product since then so i suppose in some areas things like planning and budgeting are still there and you've got we're going to talk about cloud in a bit there what did you what was your take on the use of within things like oracle bi and in a way looking at what was originally, I suppose, sales applications for S-Base. How well do you think that's gone, really? Well, I've always viewed S-Base as this thing that's been tacked on top of OBI.

I never thought it integrated well. I thought this was something that Oracle decided we have to do this and we're going to do it and off we go. In the field, when I've worked with what I'll call real BI or more pure BI companies, they don't have a clue about S-Base. I mean, I get brought in to do projects where they say, yeah, we need this, but we don't know how to do it. We don't want to know how to do it, so please make it happen. There are all sorts of interesting ways that OBI interacts with S-Base. There are people who do S-Base.

Is that interesting the way that I say interesting, or is that interesting as in interesting? Interesting like as in the mind recoils in horror. That's how the product works. I guess that's the way you would use it right yes yes it is a very strange melding of the two products and i don't think ultimately a very successful one um what again what i've noticed about people who are bi practitioners so they just try to sidestep it and use it as little as possible. So that's one side, by the way, of how S-Base has been integrated into Oracle's products. I don't know if you know this, it's the reporting engine for Fusion GL. Okay. Yeah, I mean, people who run reports may

not realize this, but there is a headless ASO S-Base database that reports everything. That's how you do analytical reports. That's how you do standard reports out of the GL. There is Gary Adeshek, a new workplace, has blogged about connecting to his company's Fusion GL using SmartView. SmartView being the add-in that

S-Base uses within Excel and the Fusion GL uses and OBI uses and a whole bunch of other products. I think he was as shocked as anybody else. He had no idea it was there. And all of a sudden, ta-da, he had S-base against the GL directly. So depending on the, you know, Oracle's not a stupid company. They're certainly capable of doing creative things. I think in the case of Fusion, it worked really well and has been seamless. And in the course of the BI product, not so much. As to why they weren't able to do that, I don't know.

I don't have any insight into it. It's a tough thing to do. It's a tough thing to do, isn't it? I mean, I think to try and fundamentally, I mean, I must have presented for so many years on S-Base and OBI integration. integration and everyone got a video on youtube with my son uh who was about i don't know eight at the time introducing me at monterey on a on a talk at kscope on s space and obi integration and i think fundamentally there's two different technologies one's multi-dimensional one's relational and they're quite hard at the kind of the point at which you try to integrate them to bring together in a meaningful way and there's been things over the years i mean recently the last thing i saw was using s space in a headless way using um i think it's hybrid storage

as an acceleration layer but then arguably maybe an in-memory database is better for that i think it was i think it's been an interesting story but i i think agree with you that i'm not necessarily sure it's been a success really um but um i mean what about okay so obviously planning is still there hfm and so on all the core things are still there what about things like things like epma and odi and so on i mean i suppose the the development side of things and loading data in and maintaining dimensions and so on how's that gone over the last few years i always found that a bit fragmented and a bit kind of half done really what's your view on that oh i space studio and so on right right i mean i would absolutely agree that they have bit fragmented and a bit kind of half done really what's your view on that oh i space studio

and so on right right i mean i would absolutely agree that they have a fragmented approach to this so epma enterprise performance management oh my goodness i've forgotten what the a is what is a administration i think that's what it is was this it was a very good to have a single place to do dimension management for all of the tools. The execution has been troublesome. It really has. It's worked very well for HFM, HFM being the financial consolidation tool that doesn't actually use S-Base underneath it. There's a whole story there. It works well for that product. But when other products, when planning applications and S-Base applications were into it, I've been on project where I just wanted to cry.

It was so incredibly hard to use. And the crazy thing is these other products, to the point of fragmentation, these other products have their own ways of loading data and building dimensions and managing all of that. So if you are a company that owns HFM and planning and S-Base and other products like strategic finance, you may try to make it work in EPMA. You may give up in despair and start using it across all three tools and have to have three skill sets and do all the things that go around

managing metadata in that way. And then you go off and buy Oracle's data relationship management, DRM. I mean, that's nuts, right? That's hard. The answer now is DRM. The answer is buy our very powerful and very expensive product, and it will master and manage all of your metadata. And it actually does a pretty good job of that. But it is not a standard part of the product and that fragmentation and those individual ways of still doing things

persists. Studio, by the way, is an example of just a brilliant idea that it's just kind of stalled. It is a way to build S-based databases off of Datamart star schemas. That's it. That's it in one line.

But that's what it should do. I mean, that's how you should build these things. Unless you're building the most hand-built, hand-designed thing that isn't standard in any way, sort of the S-based server under the desk. But for your sales analysis applications, why are you managing metadata? Why are you building in S-based?

They're called load rules, but they're mini-DI to get a table or get a file and put it into a dimension. Why would you do that separately in an S-base administration tool? It sits in a table somewhere, doesn't it? Grab the table, build the application. It has some traction, but I don't i don't see it very often i think only bigger customers get it that product as far as i know will never make it to the cloud there will not be a cloud instance of of sb studio that's too bad because it really really has its place so yeah it's it's an interesting it's an interesting situation they find themselves in okay okay so we get into cloud in a second one last thing was exolytics exolytics was was was i i famously i

bought an exolytic server and i i think you did i i was too well but my my strategy at the time was was it was it was to basically to it the competition either think we're either insane or incredibly successful either way would be kind Either way, it would worry them about competing with us. So either we were insane to spend $130,000 or pounds at the time to buy one of these things or doing very well. And I think in practice, it was more of a marketing thing for my old company. We had one, we used it for development, we used it, we won some work during it and so on.

But it was a terabyte of memory, really. And really, it was actually Exalytics. So it was actually S-Space in my mind that was the product that did best on Exalytics. It was a product of all of them. It was, although it was designed, Exalytics was designed to be used with Oracle BI, it was actually S-Space in my mind

that seemed to get the traction with customers and get success. You know, what was your, did you find that as well? So, you know, again, interesting take take interesting the way you want to do i should i should first mention by the way to the audience mark you i was on i was doing a case scope presentation with a couple of other people and we needed to do some benchmarking against exolytics and we cast about going how would we do this you offered it up um you and robin moffat just i i can't tell you how how impressed i was by your generosity of course in the event

we found out that actually we could totally beat xlytics with the commodity commodity hardware and then you know cast about like how do we how do we say you know your baby's ugly we beat it with fourteen thousand dollars worth of hardware but that's that's then that's this is now um exolytics the the problem exolytics had and has with customers is okay now space and its tools are um enterprise class products there are skeptical it departments that say really you're going to make me pay all that money for a server? I can build a Linux box for a quarter of the price. Tell me why it's not as fast. And there are certainly use cases where Exalytics is faster, where it has functions, features to S-Base that are only on Exalytics, not on commodity S-Base,

commodity server S-Base. But I don't think it's been the success that it could have been. Part of it was price. Part of it was the complexity of bringing into a customer. Part of it was the big customer only focus that the Exalytics team has or had and has. Again, I thought that was going to be the future, and I was surprised when it didn't happen. Maybe I'm too optimistic when I see technology announcements like this, but I thought for sure I'm going to be on a bunch of Xalytics projects. I never have been. It's not happened once. Okay.

I'm a small company, but I do get around, and I just have not seen it. So tell us about cloud then. So cloud is another rebirth of S-Space really. And I noticed that there was S-Space within Oracle Analytics Cloud. So tell us a bit about S-Space's moving to the cloud. What is OAC and what's S-Space's role in this? So, I mean, S-Space had to move to the cloud.

From a marketing perspective alone, it had to do it. It's a platform as a service, so it is a server that you can touch. Bits of it you actually can poke about in, unlike the EPM, the Oracle EPM world. It's all applications, and it's all software as a service. So everything's locked down. You can do what the application does. You typically get one application per SaaS instance, and that's it. S-based cloud, it's a platform, as many databases

as your CPUs and disk space memory can manage. It gets rid of that complexity that a true enterprise product has, like the on-premises S-base. For the longest time, I thought many people thought it was going to be a tool all on its own. It is actually part of OAC, Oracle Analytics Cloud. Depending on the version of OAC, it is either one half of the tool, where it's S-base on one side as the database engine and data visualizer on the other.

I believe that's the standard product OAC standard. And then there's OAC enterprise. It throws in Bix. You know, when you talk about a rebirth, the data visualizer is kind of philosophically a user-driven tool. It isn't just philosophically, it is a user-driven tool. That is kind of the ethos that S-Base had in the beginning. I think that's a large part of what Oracle hopes to capture with S-Base in the cloud.

They have added features to it. There are, just like Axalytics, has functions and features that are not in commodity S-based in the cloud. They have added features to it. There are, just like Xalytics has functions and features that are not in commodity S-based, so too does S-based cloud beyond what's in Xalytics. I would argue that cloud, just like with the other Oracle products, cloud is where all of the development's going to happen in the future. It's happening there today, and I don't see that changing in the near future. In S-based cloud, I'll call it a space cloud, but the product really is OAC.

Oracle have tried to simplify the building of applications and databases, so you can take what they call unstructured data, but it's a workbook with columns that represent regions of the country, regions of the world, products. And it can intelligently figure this all out and build a database. In essence, it is like Power BI. It strikes me as very Power BI-ish. I don't know if it does as good of a job as Power BI, but it is certainly their attempt to make it happen. You can build S-based databases directly from Excel itself. So putting aside the unstructured data concept, the thing, the low rules, the definitions of dimensions, the definition of the application,

which has historically been done through the S-based administration services console console can be done directly in Excel and just upload it into the cloud and ta-da, you now have a dimension, you now have data, etc. I think that's powerful. I think it's a powerful idea. I think in execution, it means that developers have to be able to sort of build a dimension in their head because that's how it works. The interactivity is all around creating these workbooks for upload to S-based cloud, but it isn't actually about managing the workbooks themselves, about building the actual content

of the workbooks. We'll see where that goes. If Oracle can put that dimension editor into Excel itself to make it a Windows application again, I think that'll be huge. I think that will merge the power of Excel and all the things that you can do in Excel to data and metadata and allow developers and power users to see what that looks like, and then build it for real in S-Base. That kind of flexibility, again, comes back to the rebirth of S-Base and a return to its roots. Currently, the way the product works as of the 26th of July 2017, there are no good dimension editors.

I know that that is their number one priority to get that in there. Today, I would say that building a complicated application in S-based cloud is going to be a challenge. It doesn't handle dimension building well today because there really is no way to view it. There's no way to view it in Excel where there's a very simple tree view of it.

And there's really no way to see it in the cloud. Yeah, you can drill up and down dimensions in a very cumbersome way but the traditional tree view that we're used to, tree view controls that we're used to seeing dimensions in doesn't exist in the cloud. So it's something they are aware

of in terms of a weakness and I'm anxious to see them fix the problem. They they will they have to there really is no choice okay okay so who who who is the the user persona this is aimed at then do you think and what is the problem it solves that couldn't be solved by something else in your view i mean that's a classic kind of product management kind of question, but who is this really aimed at, really, do you think? I think really, really, it's aimed at existing S-based customers. I think it's, you know, I said before that Oracle had to put this into the cloud. Why do they have to pull it into the cloud? Because the rest of their products are there. So,

you know, inevitably, S-based must follow along like everything else that Oracle does. I think truly net new customers for S-Base will only be driven by data visualizer. I think they'll have a very hard time saying why S-Base when there are so many other options out there in the world today. It's not that S-Base is a bad product. It's not that it is diminished in its value. There are lots of other options today that weren't there when S-Base was born. And I think their job of selling it to customers that do not already have S-Base on-premises,

I think it's going to be a very hard sell. They certainly have their work before them to make that happen. From an existing on-premises customer perspective, if the whole industry is being pushed to the cloud, if development in on-premises S-PACE is static, there is no real development except patch upgrades, what other choice does a customer have um they're they're sort of forced into that role okay okay what about what

about the more what about things like epm and and planning and budgeting and and and those tools i mean i i presume they've been taken care of in a separate line of products really in the cloud they have been yeah i mean not only do they have this this service uh versus platform um philosophy but the development for s base has forked um there is the bi owned space that's a space cloud that is actually if were you to buy were you persistent enough to buy on-premises S-Base beyond the fact that on-premises software and the EPM and S-Base is not actually being sold? You have to go out and buy it yourself on shop.oracle.com or whatever the name of that website is. That's owned by the BI team. That's not owned by EPM. And so this stuff that all these new functions full hybrid

in S-Base Cloud sandboxing we're going to have a real debugger. We almost have it. We actually do have it in S-Base Cloud. A real honest to goodness debugger that lets us look at numbers that we don't have to split out onto a thousand sheets to trace

the lineage of data through a calculation. That's cloud. That's S-based cloud. That's not EPM cloud. What Oracle will bring across from the S-based side into the S-based that's underneath EPM and when they'll do it, I think that's to be seen.

I can certainly see some things not coming across because they're direct competitors. Hyperion Planning, PBCS, Planning, Budgeting, Cloud Service has this concept of sandboxing. So does S-Base Cloud. They're not the same. They're not handled the same way. The EPM guys had a problem.

S-Base wasn't going to solve it, so they've done all sorts of wonderful stuff in planning to make that occur. The S-Base team came along later and they said, yeah, we don't, I assume this is what they said. They said, yeah, that's wonderful for them, but that's not what we need. So we're going to do it our own way. Will the two of them ever exist? Will applications that get migrated from S-Base to planning have to be partially rewritten? I mean, I think the answer is no and yes. That's never going to come across. And yeah, it's going to cause a bit of pain

when it comes to a conversion, if that conversion happens. Okay, okay. So what's it like being an S-based developer now? What's it like being a partner, an S-based developer, and an S-based expert? Certainly, from my experience with the move to the cloud, it's brought its own challenges, for example. What's it like now trying to continue to be an S-based developer and expert in the world of the cloud?

To say that the cloud is disruptive, particularly on the developer and partner side, is to vastly understate what's happened. Projects have become short, probably incorrectly so. Particularly, I'm thinking primarily around planning. The development time in cloud and the development time in on-premises is roughly the same. That's fine the infrastructure

work that again nobody no customer ever ever in the world ever wanted to own that's gone that's good that that bit of pain is is out of the way that's oracle's um problem and expertise now but the way the products have been sold is i think the sales reps heard six weeks to develop, and that comes across as six weeks to implement. The design issues, the requirements issues, the integration, the data integration bits, that's all the same. And it doesn't happen in six weeks when you throw in the actual development of the databases themselves. So there has been a lot of difficulty for partners in particular to adapt to this. There's been a lot of, how do you tell a customer who's bought the product and thinks that it's going to happen in six weeks that, yeah, it's actually more like three months.

That's a very difficult message for the person that owns the implementation on the customer side to take back to his management and say, I goofed, please triple the amount of money that we're gonna put into this, or quadruple the amount of money we're gonna put into this project from a consulting side. So it's been a challenge. It's been a real, real challenge.

And I'm not gonna harp anymore on partners, but one other thing, partners are going out of business, partners are selling themselves, partners are doing massive layoffs. I can think of some companies that I know that have had 50% layoffs amongst application developers. So cloud makes things hard. Cloud has not been a panacea from being somebody like me. We'll see how it goes in the future I hope that Oracle with Oracle's announcement that there will be a future

for on-premises S-Base and planning and HFM even though it comes out in a year and a half I hope that brings some stability to what we do but again we'll see that's all on the business side of it that's all on the S-Base developer side let's talk about the product in the EPM space they are

throwing I always say millions but they're they're throwing a lot of money at those tools they are moving by leaps and bounds they're they bring in this is this is the promise of cloud right they've got one image to develop for. They've got this Linux VM that they build and they just throw it out there on whatever the enterprise version of VirtualBox is. Sometimes I'm overwhelmed. This is a product that moved very, very slowly and then there were big, big releases

and it took a year before everyone was on that release. Now the thing changes every other month. I think from a development perspective, the product management development teams, the shackles are off. And they're doing great stuff. So there's this really weird dynamic going on here. The product's literally getting better

and better, faster than it ever has before. You can do, as a developer, you can do things that were never possible before. They've integrated Groovy into the planning, the PBCS side of things, the ePBCS side of things. That's not anywhere on the on-premises

world. Will it ever come to on-premises? I don't even know if it'll come in what I imagine is the final release of on-premises planning in S-PACE. Colossal. Fantastic. Great. We're going to do really cool stuff. And yet, we have to do this in six weeks so i don't know

where that's going to go um it is certainly tumultuous okay may you live in interesting times is the cliche right yeah so both you and i are oracle aces i mean how have you found how have you found i suppose the ace program recently and how have you found things like actual actually actually access to software as well I mean one of the things I found quite hard with keeping abreast of things like kind of BICS and RAC and so on is getting access to software but I've not been so active in that in that space recently I mean is it is you know how have you found I suppose the advocacy side and getting access to software and so on.

Yeah. So the ACE program as a concept is brilliant. I mean, what did Napoleon say with such bubbles, men are led? I, I very much. So that sounds like I took a dig at the program.

I'm not, I mean, the, the concept of having an honorific like that being something that many, many people want very, very badly and we'll do what it takes to happen, that's brilliant. Whoever thought that up, that was a fantastic idea. I hope you got a promotion out of it. It's changed. of the changes that have happened to this move to a point-based system where you have to, you the ace,

have to perform certain acts and get to a certain score, almost like a video game, to retain your title. That put me off a little bit, actually. I agree. That put me off, and also this is the first year

I'm not going to Open World because I'm not speaking this year, and so therefore you don't get to go to the briefing, and I get that. I get the budgets are cut and so on, but that was an interesting change. Yeah, I think it went from being an advocacy

and evangelism model and focus to one being marketing-based. And I put me off as one way of putting it. I found it really annoying and disappointing that that was the direction the program is taking. It's their program, Oracle can do with it as they like but

they've lost by doing so they've lost some of the power of the program and I think there are now people that are very purposely doing things because that's, it's exactly of a mixed blessing and not going. That's a crazy place, right? That's just like insanity for a week.

I think I came out of those between the two days of the ex-director briefing beforehand, the weekend, and then the week itself. I think it's worse than, from a recovery perspective, I think it's worse than Kscope. And I have, I get about two hours a night sleep at Kscope. So you can imagine what open world is like. So yes, I'll miss the excitement. No, I won't miss the madness, but you know, maybe that is the sign of a good conference. But in terms of, so for the audience that doesn't know this, there was a program, there may still be a program where Oracle said, yep, you're ACE directors. We want you to have access to these tools. So they set up these trial programs, which was great if it was something that you did, but EPM S-Base is so tiny,

there was no way for us to get pods. The pods were not available. And so I looked at it and said, that's fantastic for other people, but I don't care because they're not getting the access to anything. It is a challenge for developers. It's certainly a challenge for independent developers. You basically beg, borrow, and steal access on other people's servers. Sometimes Oracle, you know, this magical hand comes out of the sky and Cameron, you have it for the next 45 days.

Do what you want. And off I go and try to blog about it and learn about it. But that's very much the exception and not the rule. Yeah, definitely. Definitely. I mean, yeah. Definitely. about it and learn about it but that's very much the exception and not the rule yeah definitely definitely i mean yeah definitely i mean it's certainly for me i've never had problems getting access to things and even now even i'm not so involved in in that world at this moment but i can always ring up somebody i can always email somebody and i'll get access to anything really um but i think that it's it's i think it's hard being an OLAP or S-based developer and a BI developer in the cloud world.

And I think unless you work for a partner or unless you work for a company implementing this, it's not quite as easy as it used to be. I mean, something I put forward as an idea a while ago was some kind of ACE director cloud program where we paid some money. We paid $100, $200 a year to get access to these this software on a kind of trial basis to cover the actual real costs that oracle would have in in setting this up and providing it but you know

made it possible to access things and but i think i guess one of the kind of fundamental things is is that oracle are a cloud company now and they want they want advocates and they want experts to be focusing on cloud and you know if if if if that means that people who are more comfortable on premise which was me a while ago and and you maybe at some point and so on that you know they don't want that as advocates now and it's not necessarily a bad thing it's about using the budget in a way that they can use effectively and so on and it's i guess the thing i'd most want is access to this software access to to bix to oac and so on without that that's the major kind of issue for me really but but i think you know the fact is the ace program is still there the people behind it we know their intentions are good and so on as well

but it's been hard to maintain that level of expertise i think for me over the last couple of years with it being so hard to access the cloud software. Totally agree. I mean, historically, you went to OTN, downloaded the on-premises software onto your server. And so long as you were using it for non-commercial use, do whatever you like. And I mean, you know, I would never be able to write the things that I write and do the things that I do if I didn't have access to software that way. I completely understand your frustration with how do I get access to OAC if I am not a company that can afford even the discounted version of OAC. It's pretty expensive. It's certainly a heck of a lot more expensive than a laptop with a lot of memory and a VM, which is the way I've done my S-based stuff in the past

for when I write a blog or teach myself how to do something. So it is different. And it would be fantastic if there was some way for us to get access to it, maybe just the cost. I mean, I don't know. Maybe there would be some sort of group way of doing it. There has to be something. What happens in the partner world is they spend a lot of money. I mean, it's just as simple as that.

They get a discounted price. I don't really know what the discounted price is because I don't buy this stuff. But in talking to my fellow consultants at other companies, they have sign-up sheets. You know, hi, my name is Cameron. I'm working on a planning application or I need to write a presentation. I'm booking our PBCS instance for two days. That's a very strange way of doing things compared to the way it used to work with on-premises. I think it's a temporary.

In practice, I think this is something that's more, it says more about the kind of, I suppose, the stage Oracle app with these products, excuse me, in that I think there is limited availability in time it's going to get it's going to better but certainly is a challenge I think now let's kind of move on a bit I'm conscious that it's the summer special and we're sitting on the beach at the moment or actually in the campsite for yourself and yeah just just for laughs but I think be interesting to kind of to you know have a couple of kind of end the show with a couple of summer and fall themes really and and one of the

things i thought would be interesting would be to um to think about what you'll be reading over the summer i mean typically for me i tend to take away a few articles a few books a few white papers and surreptitiously read those by the uh by the poolside and pretend I'm reading something kind of not interesting and not technical and actually read this instead. For you, Cameron, I understand you like going camping. Is that correct? That's true. I'm sort of a bit fanatical about it.

Camping and hiking, at least currently, are my two big hobbies. Fantastic. So I wouldn't be by the beach. I wouldn't be by the poolside, but it would be in a camp somewhere um so what would you be reading so what would you once you put the tent up what would your three things you've been reading then over the summer be then give us give us an idea so so i'm a i'm a history buff um i could have been a history professor except i didn't have the nerve to actually do it but i I still read the books. I have an old book called Popsky's Private Army. It is the memoir of a guy in the British Army that was sort of the analog of the Long Range Desert Group, but different, much more active than them.

So I guess the kind of current mania for Special Forces, He was one of the first groups like that. It's interesting because it's a compelling story, but what I find most interesting about this particular author, Vladimir Penikoff, I'm sure I've pronounced that incorrectly, he fights with his brain. And I guess as someone who likes to pretend that I'm sort of smart, I find that much more interesting, a much more interesting way of looking at life. I'm also reading a book. I'm actually currently in the middle of this one by Richard Feynman called Surely You're Joking, Mr. Feynman. Yeah, good.

Yeah, you read it. Yeah. Yeah. Brilliant. Again, you know, this is a person of intellect and of mind. And if I'm not as smart as that, at least I can read about it and try to follow what they do. And then the last book is sort of a guide. It's called 60 Hikes Within 60 Miles, the Philadelphia region edition.

So I love planning for these things. I love kind of anticipating and mapping out what I'm going to do in a hike. Hiking for me, camping for me is where I, maybe unlike you, this is where I completely turn on. And I don't bring a laptop with me. And with luck, my cell phone doesn't have a signal. My mobile's inoperable. And I just relax. And I exercise.

The two of them combined is just vital to my mental well-being. Which is tough because it's very difficult to do that in the northeast of the u.s it's cold it's you know rainy and snowy and icy but yep yep with luck i'll get through all three of these books fantastic so you're so you're not someone then who takes you know your phone with you you don't take kind of wi-fi with you and so on really you tend to go back to basics really when you're when you're camping oh yeah rustic camping primitive camping whatever whatever you like to call it water i mean even water i i've got enough fancy filters to drinks right out of streams without any you know issues about health so yeah it's sort of the real deal yep fantastic fantastic okay so another similar sort of topic really you know looking forward so you got the summer which is the summer, which is nice, and I think, you know, we're all looking forward to and I'm not necessarily working in that area at the moment my son had his exams this year so we said

to him actually rather than take him take him out of school we would actually go to the JW Marriott in San Antonio in August so apparently be quite hot though which would be interesting but we're actually then going up to... then we're going over to Seattle and getting an RV and going on the west coast and spending two weeks actually traveling around there in an RV. And not necessarily back to basics like you are, but certainly out in the country and doing that. So I'll be over there actually at some point as well, which would be nice. Very nice. You're doing the classic American road vacation.

Exactly, exactly, exactly. So I'm hoping to get an RV the size of a tour bus as bus as well should be kind of good but but looking forward to that so past that so we've got the summer then we've got open world we've got the conference season and and and you know what you guys call the fall but actually correctly is autumn um what do you think if you had to cast your mind forward to maybe a couple of predictions what's gonna what do you think will be announced at open world what do you think will be the next major things that will be announced in our world that, you know, if you had to kind of be a betting man, what would you predict will be some interesting announcements and product developments over the next few months? The march to cloud will continue. I mean, there's nothing extraordinary about that statement.

I think it's intensifying. I think it will continue to disrupt the market, both on the, you know, again, selfishly on the partner the market both on the you know again selfishly on the partner side but on the customer side as well uh i i sometimes wonder about the traction that oracle's getting with these products uh there are the numbers and i i don't have any reason to disbelieve those numbers but then there is the reality where there are a lot of customers that have not switched um the it's. Let's just put it this way. Again, this is one of my, I guess, sort of a series of, I thought it would be this way, but it's not.

It's occurring again. However, I think that Oracle's drive to the cloud, the way they're bringing functionality to cloud products that they're not doing at all for on-premises, absolutely going to see that again and again and again. If I were to think particularly within the, I'll call it the BI space and the EPM space, the buzzword right now is machine learning.

Machine learning this, machine learning that. When I ask people, and how does that relate to what we do, I always receive a different answer. So yes, I suppose there's an element of hype to that. Having said that, it is a compelling concept, and I can definitely see where this programmatic analysis of data to truly drive decisions and to make the decisions, I can absolutely see that impacting what we do.

Will machine learning need to use S-PACE? I don't know. I don't know that it will. Maybe. It will be fascinating to watch how that plays out, whether it really, really is just a flash in the pan or truly the future. I tend to think it truly is the future. Again, we'll see. I won't be going to open world so i won't know but i'll i'll keep tabs on it virtually at least okay okay and one last thing really is where's case scope going to be next year i've heard it's going to be in florida it is going to be in florida it's going to be in the same places that the hype korean solutions conferences were every other year it's a disney world um yes yes so for people who bring their children to conferences, so long as it doesn't interfere with school exams or, yeah, that sort of thing, I think there'll be lots and lots of families there. As there were this year in San Antonio. Yeah.

With, you know, the pool resort that's there. But, yes, it's going to be there. I will tell you that I was on the board when the decision was made to go with that. I'd been there so many years as part of Hyperion, part of Hyperion Solutions Conferences. I was a little nonplussed, but I have been assured that things have gotten better. It's a much nicer hotel than it was 10 plus years ago.

And then Kscope will do its, ODTUG will do its normal fantastic job and make it just a special conference. I'm very much looking forward to it. Yeah, yeah, fantastic. I remember going to the, I remember going to the IOUG, the Collaborate conference that was there

and they hired out, they had the, Universal Studios was hired out for the evening and we could go there and get a go on all the rides with no queues there. I mean, that was, that was interesting. But just particularly going to Disneyland

is fantastic really. And yeah, I mean, I think actually, to be be honest i'll be there next year i think it'll be really interesting to go so i'm looking forward to it i'm looking forward to it my kids are just about just about young enough to go and and still enjoy it i think really you know i think i think this is probably the last year we could do it really so it's fantastic timing great great i'm glad to hear that fantastic well look cameron it's been really really good speaking to you fantastic speaking to you thank you very much for for coming on and taking part also in the summer special bit at the end there um thank you very much and um have a good evening

and it's been it's been great to chat to you again it's it's very much been my pleasure mark thank you for for having me. Cheers. Thanks, Cameron.