The need for enterprise-level strategic planning to incorporate GIS into existing business systems is becoming increasingly important. This session will outline the fundamentals and considerations in developing a technology strategy as a foundation for building business solutions using ArcGIS technology.
00:01 So this session is creating an effective GIS technology strategy, okay?
00:06 It is correct in the online agenda, and it's correct on the agenda I was given, but on the little one...
00:13 ...that comes with your badge, it appears to be incorrect.
00:18 Okay, so, well, good afternoon; I hope you're having a good conference.
00:21 My name is Andrew Hendrickson; I've spoken to some of you already this week.
00:26 I'm going to be presenting today on a technology strategy with my colleague Andrew Sakowicz.
00:31 He will do some work at the end of my presentation, in actually showing and demonstrating...
00:38 ...what I'm talking about preceding him.
00:40 Okay, so what am I going to be talking about?
00:43 This is kind of part 3 of a couple of presentations that I've done this week.
00:49 So my goal today is to characterize GIS in the enterprise, okay?
00:55 It's a kind of weave a story around a discovery process.
01:01 I want to do that by giving you an architectural vision for the ArcGIS platform.
01:07 ...in the context of some patterns. These are the same patterns we've been discussing in each of the previous sections.
01:13 Don't worry if you weren't in previous sections; I will explain them again.
01:18 There's an approach, okay, there's an approach that's roughly, you know, routed in an enterprise architecture framework.
01:25 And I will describe that today, much deeper than I've described it in previous sessions.
01:31 We will go through what business architecture means. What is information and technical architecture?
01:35 I'll talk about governance. And I'm going to talk about some tools too, alright?
01:39 So we're going to correlate in an information-gathering context, how can you apply information that you're building...
01:47 ...via this discovery process.
01:51 And of those tools, we've got a number of resources, we've got a solution planner tool that we're going to kind of show.
01:59 So the business enterprise, okay, I want to set some context, and another agenda I have today is to really...
02:06 ...to start to push different ways of thinking on you.
02:10 Those of us in a GIS domain, we haven't always interacted at a very high level with IT, although we may serve IT.
02:18 We may be part of IT, okay, but we may not be fully of and within.
02:23 I want to give you a way to drill in and to build some knowledge, to begin to communicate at that level.
02:30 Not denoting that one is of a higher evolution than another; that's not what I mean.
02:35 Alright, so what is the business enterprise? It consists of functional departments.
02:40 People and systems, right?
02:43 Stovepipes, does anybody have stovepipes in their organization where data is stored, right, different areas?
02:48 Do you think data repeats itself across those stovepipes? I'm seeing lots of heads go yes.
02:52 Okay, so we need to recognize this. So successful enterprises have a free flow of information.
02:58 That doesn't mean that you lose, you know, ownership of a particular dataset.
03:04 It doesn't mean that you're not the subject matter expert with regard to the way a certain dataset is utilized.
03:10 It just means that other organizations might need access to it in a different manner.
03:15 So I think this is a very important concept. The word enterprise, to me, is not a proxy for size, okay?
03:24 I don't, when I say enterprise, I don't mean you have to have a minimum of 300 people in an organization, or a minimum of 1,000.
03:30 That's not what I mean. It's more of a train of thought.
03:35 So an enterprise, to me, can be a group of six people serving a business workflow for some sort of organization.
03:44 So GIS, it has an extremely enormous amount of strategic and tactical value when it is embedded deeply, you know...
03:52 ...into systems within organizations.
03:55 Almost to the point where it is abstracted far enough that the end users don't know they're even using GIS.
04:01 In a business organization, okay, we still require a domain supporting this, but the end users...
04:07 ...primarily, they may not even know it's GIS.
04:10 So a lot of benefits occur by treating it in a manner that it can be supported to serve the business back in that way, right?
04:18 So what is it in the enterprise? I argue an architectural approach, it's a workflow-based architecture...
04:24 ...where geographic data and services are integrated, right, and shared access across an organization exists.
04:31 Or at least it's been agreed upon, okay?
04:34 I have to say this, there are some chairs up front, and mixed in.
04:40 Oh, and something else I forget every presentation. There is a survey; please fill it out at the end. Okay.
04:47 So facilities management, land records at...
04:49 You know, I don't really want to force myself on any particular vertical.
04:54 I believe that this type of process is necessary across any industry, okay, or any vertical market, right?
05:02 It's not specific to a utility or specific to, you know, a small government.
05:08 So the enterprise-wide use of a GIS in capability should be governed as part of IT, okay?
05:14 Who's part of their IT department in their organization?
05:18 Okay, I would call that maybe 50 percent of hands that went up.
05:23 That's up from years prior I would argue.
05:26 So there's an organizational concept to this aspect.
05:30 There's an infrastructural aspect to this.
05:33 So what is the infrastructure that we need to enable the ability to support the business enterprise?
05:40 What I will argue is that we have started here in the past.
05:44 We've said, okay, I've got business with x amount of people in it; I need to get this size machine.
05:49 The refrigerator machine, right? I need a big, honking machine, right?
05:55 And we'll build something and the business will come to it.
05:56 I argue that that is not necessarily a good approach in this day and age.
06:01 We need to grow GIS throughout the organization, okay?
06:03 So the infrastructure will grow as well as needed if we do this correctly.
06:08 So managing your GIS as a platform instead of an application. Make sense? Alright, very different approach.
06:16 So tactical growth over time. Okay, what do I mean with this statement?
06:20 So broad access for your business enterprise to the data and to the geoprocessing.
06:25 So the GIS domain remains extremely relevant; you are the subject matter experts that can author geoprocessing...
06:32 ...that gets served out to the bigger business.
06:36 So this common infrastructure, to build out and deploy GIS, what do I mean by common infrastructure?
06:41 I'm really kind of looking at this from a funding perspective, to be honest with you.
06:45 Common in the sense that the business enterprise owns it.
06:50 But it's funded in the same manner across departments.
06:54 This is important, right, so economies of scale through organizational use.
07:00 Often the best technology to integrate disparate systems is GIS; by nature, GIS becomes integration technology.
07:08 I don't mean it's an enterprise service bus, okay, but it is integrating technology by virtue of what it is and what it does.
07:15 Takes datasets from multiple different places and puts them together, spatially enables them.
07:21 So I'm going to promote a lot today the notion of reusability, right, so build once, use many times.
07:29 Do we need multiple basemaps in an organization? No, we probably need one.
07:32 But we need the organizational components to talk to one another and figure out what that basemap should look like.
07:38 Alright, so it's also very important to obtain a business sponsor, okay, at the proper level.
07:43 I will go into this in depth.
07:48 So a successful strategy requires a plan. Remember, our topic today is about strategy.
07:56 So a baseline needs to exist. What do I mean by baseline?
07:59 I don't mean baseline benchmark, per se. I mean a baseline for what exists.
08:04 What do you have today? What are you supporting today?
08:07 Both from an infrastructural standpoint and from a data standpoint and from a business process standpoint.
08:15 So I want you to be able to realize that it's okay to embark on a process of discovery.
08:22 And to do so, you should probably begin to think about promoting a series of events...
08:27 ...with the intent to investigate your current implementation of GIS.
08:31 This is a little different from a GIS grassroots organization, which we have had in the past.
08:36 That's not necessarily what I'm referring to. It could be handled through a center of excellence or something like that.
08:42 They don't really mean a GIS users' community; that's not what I'm referring to here.
08:48 So I want to promote, you know, assessment and requirements gathering sessions for you...
08:53 ...to validate the technology implementation that you have on-site as a baseline to begin with.
08:58 And from there, you can move ahead and begin to do gap analyses.
09:03 Doesn't this sound exciting? [Inaudible audience comment] Painful! Well, I'm going to help you, right?
09:13 And realize it is painful, but it's more painful to commit to it than it is once you embark.
09:19 Because I'll be honest, realization of what's really going on is quite rewarding.
09:25 And one of the concepts that I like to talk about is architecture, okay?
09:30 This quote, I think, is very interesting.
09:31 So "computer architecture, like other architecture, is the art of determining the needs of the user of a structure...
09:37 ...and then designing to meet those needs as effectively as possible within economic and technological constraints."
09:44 Right? Architecture's the glue to me in my approach, okay? Why do I share this quote with you?
09:50 If you could read it, if I was a better PowerPoint slide maker, you would see that here the date says 1962.
09:59 And it comes from a gentleman who worked in a department called the Machine Organization Department.
10:05 This isn't a new concept. We're just going to apply it to the maturity of the technology today.
10:14 So I need to do something kind of interesting for a second here.
10:25 [Audio] Some people like to climb mountains. I like to build planes, in the air.
10:35 I grew up wanting to be on a plane, wanting to be at this height.
10:39 Sometimes the temperature up at altitude will reach 60 below. It's brisk; it's refreshing.
10:45 You never know what you're going to come across up here. Canadian geese, mallards, owls.
10:52 These people back here, that's why I come to work. That's why I build airplanes in the sky.
10:57 We're not just building a plane here; we're building a dream.
11:02 I love this job.
11:06 I don't get a lot of thanks up here, but I look over there and I see that little kid, that look in his eyes...
11:11 ...that's all the thanks I need. [End audio]
11:19 Okay, I had to actually hold that up because I don't have an audio feed coming out.
11:26 I use that, because in a sense, we've been doing that for years, okay?
11:32 Anybody feel like you've ever built a plane while it's flying in the air?
11:36 That's a lot of hands, okay? We don't have to do that, and it is scary.
11:43 So let's embark on a path of discovery and just pretend that you might be able to do it today, okay?
11:48 And I'm going to feed you some steps and some process, alright?
11:50 So the need for discovery, the who, the what, the where, the why, the when.
11:54 The questions and the answers. So now what?
12:02 Execution of any plan requires buy-in, right? Understand where you are, okay?
12:08 Understand where you're going, identify barriers to progress, and develop the strategy.
12:16 Seems, obvious, right? It's not easy to do, but your leadership needs to buy in to this.
12:22 At least your direct leadership, because you need to spend time on this.
12:26 And if we don't see value in it, then you won't be able to achieve it.
12:31 So let's dig in.
12:36 This is a process; these are the steps that I use when I work with clients to develop an architecture.
12:43 Which is ultimately turned into an implementation schedule, which is ultimately implemented.
12:49 I begin with a platform vision for ArcGIS and for GIS in general; I will show that to you.
12:57 Number 2, my second step. Once I've laid out the vision, I start to dig into business architecture.
13:06 Then I dig into information architecture, and technology architecture as well.
13:12 Second, do you see how I'm starting from the top down?
13:17 To the left, you see this little influencer that I have noted here as SLA.
13:23 For those of us who weren't in a previous session with me, you know, service-level agreement...
13:28 ...okay, what is the expected uptime on a particular business process.
13:33 Or also the SLA can exist on what? It might exist on the hardware, okay?
13:40 Or the database, or the SAN, right? Or all the above, right?
13:45 But what happens if the SAN SLA doesn't match the business process workflow expectations?
13:50 We see this a lot, okay, where we're expecting to have a web map available 24/7 but we've only got DBAs...
13:57 ...that can move around table spaces between 8:00 and 5:00.
14:02 I'm seeing genuine laughing out there right now.
14:04 But, you know, this is organic growth; this just happens, okay?
14:09 It's not right or wrong, so, you know, this is this notion of evolution again, okay?
14:14 As we're architecting a new system, let's ask those questions first.
14:18 So we know if I'm told that my web map needs to be up 24/7...
14:22 ...then I'd better have some redundancy in my application server, right?
14:27 And maybe I don't want a direct connection going back to my database; maybe I want to decouple them.
14:31 We'll talk more about this later.
14:33 Now I explode out the technical architecture; why do I explode that one out and not the rest?
14:37 Because by the time I get to this point, and we start talking about platforms, here's where I'm able to go back to the business...
14:44 Okay? ...and confirm what I heard for the business architecture.
14:48 I need to confirm this, right? This is what I think I heard; am I correct?
14:54 Because now I have the baseline; I know what exists.
14:59 Probing on opposing views is really fun; you can cause arguments.
15:07 But at the end we're all friends, really, so, you know, but by probing on opposing views, you get to do just that.
15:14 You know, this projector's in front of me, but I see it differently than you do.
15:17 And you have a different perspective than I do, and you have a different perspective.
15:20 But it's still a projector, right, so that's what I mean by probing on opposing views, right?
15:26 Then from there I create a conceptual architecture and a service abstraction.
15:31 So if I've chosen to integrate with other business systems with web services, which is a contemporary way to do it today...
15:39 ....you know, I want to tell the business, well this is my API.
15:43 And today if you've deployed ArcGIS Server, you have a published API based on an IT standard that's RESTful.
15:50 Or based on SOAP or OGC. You can already communicate with IT folk much better than you could in the past, right?
15:56 This is a standard type of technology.
15:59 Alright, so here, you know, we need to go back and review that conceptual architecture and document the gaps.
16:06 What do I call a solution for my clients?
16:09 I call a solution a combination of core or partner technology from Esri, data, professional services.
16:18 And we have to remember that, in terms of services, there's a big difference between integration, implementation, and configuration.
16:26 They're orders of magnitude different, okay?
16:28 So that's what I mean below that, so when I'm thinking about a solution...
16:32 ...my solution for my customers, internal customers or external customers, you know, is there migrations involved?
16:38 Is there implementation, is there the effect of change management, that just, that can disrupt me, okay?
16:43 And what are the best practices that I have to employ when I'm considering IT best practices and behaviors?
16:49 So I want to influence the planning within the IT domain.
16:53 So building that trusted partnership, I mentioned this earlier.
16:56 I want my production DBAs to trust my SDE DBAs.
17:01 They should realize that they're one and the same person; they're not separate.
17:08 So that promotion of that team approach and discerning the handoff for where and when this...
17:12 ...thing that we're talking about should be built. So here, I'm going to dig in, alright? I told you I'd dig in.
17:18 So use the ArcGIS platform vision when you do this, okay?
17:26 Might argue that this is not an architectural diagram.
17:30 Well, let me show you how it is an architectural diagram, 'cause there's different types of architectural diagrams.
17:36 So all the way to the left, you see asset management, or sometimes I refer to it as data management.
17:42 The collection and organizing of information or spatial assets.
17:45 You're storing them in what we call a geodatabase.
17:48 Or a database, you know, with a spatial data type, right, it's in a database.
17:53 So the ability to store something in one place and reuse it many times.
17:57 As you move to the right, right?
18:04 So, you know, planning and analysis, taking information, data, and transforming it into actionable information.
18:10 Let's correlate this with the concept of geoprocessing, okay?
18:14 I'm going to be very servercentric in this conversation 'cause we don't have a lot of time...
18:18 ...but the geoprocessing framework and building a model and deploying it via ArcGIS Server in this case.
18:23 Field mobility, getting information into and out of the field, whether it's with ArcGIS Mobile, right, or ...
18:30 ...an iPhone or something, but, communicating back and forth from the field.
18:34 And then the last one is operational awareness.
18:37 Taking information and visualizing it and mashing it up together, putting it into a web viewer.
18:42 I think this is architectural. Why?
18:44 Because the technology performs differently, depending upon the pattern of usage.
18:50 We've worked with a lot of clients over the past few years regarding these patterns.
18:55 And in each and every case, our technical workflows that are orchestrated behind the business process...
19:01 ...fall into one of these patterns of usage, okay?
19:04 So by identifying your technical workflows and applying them to each one of these buckets...
19:10 ...we can do neat things like realize, hey, wow, I have 2,000 users in operational awareness...
19:15 ...and I only have 12 people editing data.
19:18 Maybe I shouldn't design a system for 12 people that 2,000 people are going to hit, right?
19:25 Because the SLA is different for this pattern of usage versus the data management pattern of usage.
19:33 I'll come back to this. You need to communicate to the business; you need to write. You don't have to write a lot.
19:42 I know you can't read this; it's okay. I'll describe what it is, alright?
19:46 This is just simply a summary of a proposal to do an assessment within an organization...
19:52 ...from an enterprise architectural perspective. So what I'm doing is I'm elaborating to the business...
19:57 ...and I'm saying, "These are my goals; I want to talk to the business stakeholders," okay?
20:02 I have 1 through n goals to accomplish, right? I want to validate a current deployment.
20:08 I want to identify short-term prescriptions so I can just put to sleep, you know, problems that keep bubbling up all the time...
20:15 ...so that we can move on to a proper architecture.
20:17 And there might be other goals that are tailored individually to each of your organizations, okay?
20:21 Deliverables, tell 'em, tell the business what you're going to give 'em.
20:24 You're going to get an architecture diagram, you're going to get a proposal, you're going to get, you know...
20:28 ...some sort of ROI prediction based upon what I'm learning from the business.
20:35 So, inform, invite, interact with your stakeholders. Figure out how they're measured.
20:42 Conduct formal sessions. What's on the screen right now is an actual agenda.
20:49 Some of you folks that I've worked with on-site are seeing this and going, wow, this is what he used with me.
20:53 Yeah, it's the same thing I used with you, right?
20:57 This is an agenda, and it lists out very specifically by task what I want to do.
21:02 And what the required resource is that should be in the room with me.
21:06 And I ask relevant questions for the business, right?
21:09 A lot of times, my business owners or business stakeholders, they don't know what Arc Arc Arc Arc is, right?
21:19 So we need to start thinking more about a solution across your enterprise than a feature/function/product standpoint per se.
21:27 We need you to remain product focused, right, within your domain.
21:32 But when you're communicating with the business, let's just figure out what we need to do to solve the problem.
21:40 So a target landscape for IT. I will read these, 'cause they're...these screens are really small.
21:45 I don't know why I got these mini screens in this room.
21:49 What is this trying to show? Coming across from left to right over here, I've got a timeline.
21:54 And then going down, I've got level of detail increasing, okay?
21:58 On the top, this cube is trying to show an architectural approach...
22:02 ...so the first abstraction is the four patterns, those four patterns that I just showed you.
22:07 We need to align those with your business strategy and with your workflow.
22:12 One step down, specific architectural areas. So a nice example is a utility.
22:18 One utility might have a transmission and distribution going on within the same utility.
22:24 Well, these are two different architectural domains and two different implementations of GIS technology.
22:29 Maybe even two different types of GIS, okay?
22:32 Let's identify it. As I keep going down, here's where I like to think about identifying existing capabilities...
22:37 ...so I know which capabilities I have to build or rebuild to serve the greater good of the business enterprise.
22:45 So you'll start to see things repeating as you do your investigative work.
22:50 A good example is the basemap which I mentioned; another example is maybe geocoding.
22:55 Does geocoding really need to occur in every department with a different soundex and a different streets table?
23:01 Or can we have one and architect it correctly so the business can hit it, right?
23:06 That's kind of what I'm referring to with this.
23:10 Let's talk about the overall timeline of a project very quickly.
23:15 I recognize that this is not easy to do, and this takes some time.
23:19 So what this timeline is trying to show, let's start all the way on the bottom left, and it says "as is."
23:27 On the bottom right, it says "to be."
23:30 And this abstraction here on the bottom, I label it as concrete.
23:34 On the top, I label that abstraction as abstract, okay?
23:38 As I walk across, this is just a different way of showing that process that I laid out earlier.
23:44 I need an as is and to be gap analysis so I could do a detailed diagram. That's the first two boxes.
23:49 As I cross over the timeline, I'm entering into the abstract.
23:54 So let's start to talk about capturing themes for what we want the system to do.
23:59 Do they match those business process workflows that we discussed?
24:04 So this cheesy cloud and this lightning bolt thing here are not meant to denote anything cloud as we've been talking about all week.
24:12 This is just meant to show stormy weather, 'cause this is where it gets a little hairy in organizations sometimes, okay?
24:18 You've got to go back and forth; you've got to synthesize everybody's opposing views.
24:22 So as we come back into reality, or the concrete world, we need a solution design.
24:27 This is where it shakes itself out, then you refine it.
24:30 Your solution architecture should never be final. It's always in draft form.
24:36 So that's why some of the folks that we've worked with, you'll see that our diagrams always say draft, okay?
24:42 Because it should never be finished. And ultimately hitting that to-be state.
24:50 Let's talk about a tool that we use internally, solution maps.
24:55 Solution maps is where we actually lay out a problem.
25:00 As you're in these meetings and you're discussing the refinement of what you need to do, map out what the problems are.
25:07 Map it out. Because then you can map out what is the cause and get agreement from within your organization.
25:14 Getting buy-in and sign-off on what the problem actually is, is important, because there may be a negative business impact.
25:26 And then propose a solution to the greater group, the greater group. The group. The V-team that you've created.
25:34 How do we solve this problem? And what is the positive business outcome?
25:40 This speaks tremendously to business owners and to decision makers, okay?
25:46 The intention, and those of you that are actually photographing this slide...
25:50 ...the intention is not the made-up problem that I put on there.
25:52 It's the process that I'm trying to show you; it's part of the strategy. This is a tactic.
25:58 So let's get back to that.
26:00 Let's dig in a little further and start talking a little bit about the three levels of abstraction that I introduced to you.
26:09 Anybody in here has been through an enterprise architecture training or course or investigation?
26:14 Okay, a couple folks. This will be very familiar to you. This is a value chain.
26:19 This is from something that's called TOGAF, okay, and I've adapted it to the needs for Esri.
26:28 I've adapted that framework to a methodology and strategy that works for us.
26:34 And then I work with clients, and I customize it even further.
26:38 My message is, take your own framework in, use something, and customize it to work for you.
26:46 Some are taxonomic, right, and some are pure framework.
26:49 Zachman's kind of a taxonomy. So a lot of times, if you've heard of Zachman, it works very well with TOGAF, right, together.
26:56 They're not necessarily two separate things. So this is boiled up.
27:00 And let me give you what I've boiled out of this that I think is important in working with the ArcGIS platform.
27:07 I think it's extremely important to define your enterprise value streams.
27:12 So as you're discussing a workflow, ask what are the relationships to external entities.
27:23 When the data is managed here and it leaves this particular organization and it's moved somewhere else...
27:31 ...who is responsible for it and when does ownership begin and end?
27:36 The tough thing here is you've managed spatial data for years in your domain.
27:40 And you're used to being the subject matter experts. That doesn't mean you own the data.
27:45 That 's two different things, okay?
27:47 Another good example's from the utility space, managing an electric network.
27:53 The data needs to be made available to the technology in a different manner for visualization than it does for detailed editing.
28:02 So let's figure out at what point does the ownership begin and end and change the format and optimize it for viewing, okay?
28:09 So the information still sits within the domain for editing, but as it moves over to, say, IT for web maps, it's optimized.
28:18 This allows us to do that. So what are the events that trigger that instantiation?
28:24 So business architecture, like other types of architecture, is composed of processes, functions, workflows, and events.
28:30 Which will be influenced by your corporate strategy. You need to be part of the enterprise.
28:35 As we said on Monday, you know, you're either in or out, right?
28:41 There isn't a gray area, right? You're either in the enterprise or you're not, especially when it comes to funding.
28:51 We'll come back to funding and the stuff as well.
28:54 So you will have this influenced by your corporate strategy; it's developed and managed through organizations...
29:00 ...and organizational communication and realized through your technical architecture.
29:04 What does it represent? What satisfies your customers, okay?
29:08 How do you compete? How do you sustain operations in some cases? Care for organization and employees.
29:16 We want employees to be happy, right, or they'll leave and go somewhere else.
29:20 Then we have to retrain.
29:24 What are the benefits? It helps with viewing the enterprise through the eyes of the customer.
29:29 Let's get up into the organization and figure out how people are being measured.
29:33 How do they genuinely feel, okay, about it?
29:38 So what are the integrating, connecting components that are necessary? Improving communication.
29:45 And, I think this is such an important concept, accepting the evolution of technology.
29:50 I know ArcIMS was a really good map server, right? I know you guys spent a long time getting it to be fast.
29:59 I know. I loved AXL too. But now, we've got ArcGIS Server; it's better. But it's not a one-to-one move.
30:09 We've got to rearchitecture ArcIMS systems to perform well within ArcGIS Server, and perform better, I should say.
30:18 So accepting the evolution of technology is very important.
30:25 So understanding the processes, functions, and workflows is very important.
30:28 What are the data and information requirements, okay?
30:31 What are the GIS functions that are available via a standard API or maybe something you have to customize to augment a workflow?
30:37 This is the type of thing that business stakeholders would be interested in.
30:41 So where can we spatially enable a workflow that is traditionally managed elsewhere?
30:47 What are the components of a workflow that we can spatially enable?
30:50 This is what you want to feed back; you want to try to probe and figure out where these exist.
30:56 And what are the organizational requirements, wow, that's a big task. I know that, but it's necessary, okay?
31:02 You know, to determine the design patterns. This is very important. This is where we figure out if we need high availability.
31:10 Not in a system architecture diagram. That comes later.
31:14 This where I realize whether I need HA or I need redundancy, okay, or I need, you know, active/passive.
31:20 All that stuff is determined here, right? So some topics, okay?
31:27 The functions and processes. You know, identify key business areas. Try to relate the GIS to a strategic plan.
31:34 Do you need a GIS strategic plan, or do you need GIS to be part of your strategic plan?
31:40 Alright, there's no right or wrong; I'm just giving you something to think about.
31:46 Should you even use the word GIS in your strategic plan, right? Maybe you just use, I don't know, something else.
31:54 Okay, so operations, right? What are the external systems you have to work with?
31:59 What are the procedures and operations? What is the budgetary process?
32:07 Staff and roles of an organization, wow. As you're building something, can you actually support it?
32:15 So as you're working through this business architecture, you need to be thinking, wow, my boss wants this thing up 24/7.
32:22 That means I get five minutes a year of downtime. Can I achieve that with ArcSDE?
32:28 Everyone be quiet. Yes, you can. You absolutely can. But you have to architect it properly, right?
32:34 And then do I have the person that can support that thing when it goes down at two a.m.?
32:39 Okay, that's what I'm referring to.
32:41 Okay, information architecture, okay, helps us map your enterprise business systems. Correlates with workflows.
32:50 So specify which part of the workflows are supported by a particular application today, alright.
32:57 This is a fun one; this is the greatest one ever, is defining who owns the data and who manages the data. It's fun.
33:05 Sometimes I feel like I have a dart, you know, what's that called, a dart board, on my shirt when I do this.
33:13 Eventually, like I said, you become friends again with your colleagues.
33:17 Okay, so like business architecture, your information architecture is composed of processes, functions, and workflows.
33:23 And events too, okay, which are influenced by the corporate strategy, okay...
33:27 ...that are developed and maintained by an organization and realized through a technical architecture.
33:31 So this is another neat one. This represents the fuel that your enterprise must produce to do what?
33:36 To satisfy your customers, to be measured effectively, to get a raise. No. Yeah, actually.
33:43 You know, to retain upward mobility within an organization.
33:49 To integrate with other interfaces and to create actionable information.
33:55 Didn't I say create actionable information on those four patterns? Remember that?
33:59 You see how this stuff aligns with those four patterns time and time again?
34:03 Okay, this is an effective way, as you're talking through this with internal clients or external clients or customers...
34:09 ...whatever you're calling them, stakeholders. If you can begin to lump these things into those four buckets...
34:14 ...and communicate them back to the business enterprise, you're talking about GIS without even mentioning a product.
34:23 Which oftentimes will confuse your funding agent.
34:30 Okay, so it helps with what? Exposing the owner and consumer of data streams.
34:37 You know, promoting and regulating the agreement on interfaces, wow, that's another tough one, right?
34:43 So not only are you, in the previous section, identifying who owns it and when you can move it around.
34:49 But how do you regulate the two different interfaces that exist for exchanging the information?
34:59 This seems simple, but improve documentation and support. I have personal experience with this.
35:05 I mean, as you actually document your information dataflows, you can feed them back to the business...
35:10 ...and show 'em how crazy they are. Can you believe we're doing this? Really? We're doing it that way?
35:18 And why is this eye-opening though, folks? Because they've been supporting business for a long time, and it works, right?
35:26 So again, it's that evolution thing. Accepting the evolution of technology.
35:30 So what I like to see and what I do, alright, and what we all should be doing in this room, is using the Geographic Approach...
35:37 ...as a form of BI, right, business intelligence. This is functionality that we can feed back to an organization...
35:45 ...a business, an entity, an academic institution, a state, a county, municipality, a store, you know, all the above.
35:55 So understanding the processes, functions, and workflows is very important, okay?
36:01 What are the data and information requirements required to solve the challenges that you heard about...
36:06 ....when you were discussing the business architecture?
36:11 What GIS data, what spatial data, maybe is a better way of putting it, is required, you know, to enhance a particular workflow?
36:18 You're the subject matter experts in geospatial, right?
36:21 So as you're listening through this decision maker during the meeting that we're going to have...
36:26 ...'cause remember, we're talking about a strategy; this isn't totally abstract, right?
36:30 What information, you know, do you, or are you aware of, are you a steward of, that you can enhance a particular workflow with?
36:39 'Cause this will allow us to orchestrate those business functions, okay...
36:42 ...and maybe even replace business process and improve it, right?
36:47 We've seen tremendous, you know, improvements on returns on investment by, you know, spatially enabling workflows.
36:54 You heard Jack mention it the other day, right?
36:58 Has anybody actually experienced a return on investment that's pretty nice by spatially enabling a workflow?
37:05 That's very low, right? I would expect many more hands than that, okay?
37:11 So let's continue to think through this, okay, and how can we prove this out?
37:15 And I feel like maybe the response on that question was low because you don't, you just simply don't know.
37:23 Because we've grown organically. You're probably thinking, Wow, I'm sure I do good for the business, right?
37:30 But someday you may get asked to show how good you do for the business. It's a different angle on this.
37:35 Okay, so choose where we can spatially enable information traditionally managed elsewhere.
37:41 And these things are repeating, see, these things are kind of repeating in different viewpoints...
37:45 ...across the different abstractions of architecture here?
37:47 So influences the design patterns again, you know, that the technical architecture will portray.
37:53 Okay, so some topics here; they're the same. These are repeating themselves again.
38:01 Business functions and processes that are key business areas, relationships to the GIS strategic plan, or the strategic plan.
38:09 Operations and organizational. So here, I'm going to talk about technical architecture for a minute.
38:15 So this aligns technologies with business needs. What do I mean by that?
38:21 Well, we might be as high level as some sort of high level denoting, you know, in terms of a tiered architecture...
38:28 ...like an application server or web server, maybe we're talking about an operating system...
38:32 ...or a scripting environment, whatever.
38:35 So let's define the technologies that are required to support that automation environment, okay?
38:39 And identify technologies that are critical to the implementation, okay, for this new applications environment.
38:45 Supporting company-wide connectivity, promoting and creating standardization for the integration of data, right...
38:54 ...and implementing integrated patterns ahead of it. What does it help with?
38:59 Meeting business and system requirements and objectives, enabling a flexible partitioning of systems, right...
39:05 ...reducing costs, maintenance, and evolution.
39:08 Increasing reuse and integration across your business enterprise.
39:13 So some topics. What is the landscape, hardware, software, security, standards?
39:18 Okay, the data landscape influences the technical architecture.
39:22 There's an application landscape that exists, an integration landscape.
39:27 Remember, I said I boiled up the TOGAF approach? This is an area where some serious boiling occurred, you know.
39:34 These areas within the framework are very dense, okay?
39:38 And I do want to mention something really quick that I'm very, I don't know, pragmatic maybe, that, you know...
39:44 ...I recognize in my work that sometimes I can't strictly adhere to a framework, 'cause if I did, I'd never get anything done.
39:51 I just wanted to make that comment. Alright, so hopefully this is fairly decent to see from the back.
39:58 But what is this trying to show?
39:59 This is kind of a conceptual, integrated architecture, that I'm describing everything we've talked about to this moment.
40:06 Where on the left-hand side, that's a corporate strategy that's interacting with business functions.
40:10 And business knowledge exists at the bottom.
40:12 There's processes and organizational flow between this, and all the way to the right...
40:17 ...at the top level is technical architecture and security, network, and hardware platforms, okay?
40:24 In there, there's databases, that's kind of irrelevant; what I'm trying to show is that there is a flow, right, amongst organizations...
40:33 ...that interacts both from a business perspective and from a technical perspective.
40:37 This may vary depending upon organizations, but it's always kind of the same.
40:41 At the bottom, you see influencers, budgets, standards, and governance.
40:48 So here, we're going to keep going around the value chain. I'm going to talk about the political landscape and IT governance.
40:57 Why do I consider this so important? One, another thing that I've done with some clients and done for myself...
41:03 ...is the first thing that I do is I build a chart, right, an organizational chart, that's titled Ability to Disrupt, okay?
41:14 And I ask the business, "Is this what I'm hearing?" Number one resource that has the ability to disrupt.
41:20 And you list it across and you rank 'em, and you shuffle 'em around.
41:23 But this is important, right? Why do I talk about this during governance? Because governance is important.
41:29 You can have projects fully halted, or fully promoted; let's look at the glass as half full.
41:36 They could be fully behind it if they feel like they own it, okay?
41:40 So what is governance? It's a subset discipline of corporate governance focused on IT.
41:46 Systems and their performance and risk management.
41:49 Specifies the decision rights and accountability framework, okay, to encourage desirable behaviors.
41:55 It's a fancy of saying that if your apps go down, you're going to get in trouble.
41:59 That's good; I need to know what the business requires of me, okay?
42:02 Evaluation and direction of plans for the use of IT to support the organization and monitoring, right, to achieve the business plan.
42:13 So, putting it all together, this is coming back to the same strategy diagram again.
42:23 Let's put it all together, let's kind of try to summarize it a little bit.
42:29 This is a pretty neat diagram. Think about this as an organizational structure, this being, I guess it does denote lower or higher.
42:37 But not from an evolutionary perspective, right?
42:40 But here, we've been really well as a GIS domain, or a departmental operation.
42:45 Sometimes we used to touch up into here, or maybe we were managed out of this level within an organization.
42:51 I'm going to just tell you that to do some of the things I've talked about today...
42:54 ...you might have to go a notch higher within an organization.
43:00 And that's okay, alright, don't be scared of the IT folks; in the end, IT folks want to end up doing GIS anyway.
43:12 Okay, that's my message on that slide, so presentation back to the business.
43:17 You're going to have to take everything you've learned and assemble some sort of concept or proposal...
43:24 ...and present it back to your business. We want to show that these strategies are driven by business strategies.
43:33 We're not just doing something 'cause we want to, right?
43:36 We're not just virtualizing because we think virtualization is cool
43:40 We're virtualizing it because it's serving the business in some way, shape, or form.
43:45 So value is defined directly as saving, making, or obtaining money; demonstrate that.
43:51 Next year, when we have this session and you all come back, I want to see more hands...
43:56 ...that says you've been able to demonstrate a return on investment with your GIS, okay?
44:00 'Cause it will make you more successful.
44:04 So technology and GIS as partner in business areas, not an adversary or a servant.
44:12 Communicate value proposition in no less or no more than 15 minutes.
44:17 I'm sorry, no less than 15 minutes, no more than 30 minutes.
44:21 Doesn't mean you have to do it like that, but I'm saying if you can't get your message across in a few minutes, that's not good.
44:28 I've been talking more than 30 minutes, right?
44:31 In any case, I mean in terms of presenting this back, you should be able to do it quickly, okay?
44:38 And choose the presenter very carefully; it might not be you.
44:44 Even though you feel like it should be. Maybe you're just not good at convincing, right?
44:50 Maybe you're just not a good presenter. It's okay.
44:54 So understand system needs. Gather information and requirements in the context of those patterns.
44:59 This is awful to read in the back, I'm sorry; I have very small screens, I'll describe them to you.
45:04 What this is showing, starting on the left with asset management, I'm gathering requirements in the context of the pattern.
45:09 So I'm asking the business, or I'm just deriving, I know that I need 30 web editors.
45:17 I've got eight desktop editors, two different types of editing.
45:21 As I move across to planning and analysis, I've got four folk that are doing hard-core geoprocessing.
45:28 I've got about 50 ArcGIS Mobile clients, and then in the end, 30 browser clients, okay?
45:34 Based upon this abstraction and gathering these results, and some internal tests that you should be performing...
45:41 ...or maybe benchmarks that you derive from our website, you can come up with a very simple multiplier...
45:48 ...for figuring out how much horsepower you need to serve each one of these patterns.
45:54 But let's not be mistaken here; without real tests, you're not going to get real results.
45:59 You're going to get approximations based upon tests done elsewhere.
46:04 So information gathering in the context of the patterns is very important, but it's also important to realize...
46:10 ...that you need to correlate this with benchmarks. And this slide is simply showing that.
46:17 Based on certain tests...I'm going to move the slide really quickly; no, I'm just kidding. You get the presentation anyway.
46:23 But based upon some simple tests that we did in Redlands, I can derive a number of cores.
46:29 And this is a conceptual architecture; this is not a system architecture design.
46:33 This is a thumbprint, okay, this is what we think we need, right?
46:37 This is valuable going back to the business, right, and not that expensive.
46:42 I can also promote the notion of business continuity in this manner.
46:47 I heard that this workflow needs to be supported 24/7...
46:51 ...so I'm pretty sure I need redundancy in my production boxes as well as on my staging system.
46:56 And I sure as heck need a development environment.
46:59 So I'm kind of bringing it all together a little bit; I know that this is a lot of information.
47:03 But this is important to gather your requirements in the context of the platform vision, 'cause you can scale the system by pattern.
47:11 But you don't want to hog-tie yourself, you know, by stringing your web viewers to your data management purveyors.
47:23 So we're going to talk about some tools very briefly, okay.
47:27 I consider everything I just talked to you about a bunch of tools.
47:32 So doing discovery in the context of the four patterns of GIS business behavior, to me...
47:37 ...you're talking to the business decision makers at the same time you can go back and talk to your ArcSDE administrator.
47:44 You need to do that, right, when you're executing a strategy.
47:48 Okay, so define the workflows.
47:50 It's very important, you know, and it's of the utmost importance to define your measure of success.
47:55 That's why I list it as a tool. You need to know what you're going for here, at all times.
47:59 Where is the line in the sand, okay? There are tools; there's a capacity planning tool lite, I call it, that's available on our website.
48:06 Andrew can talk about that. You need to have your own test harnesses that are repeatable.
48:11 Every time you release software, I did this earlier and I got a pen thrown at me, but you don't want to go like this and go...
48:16 ...I hope it works, right? You should know, every time I do a software release, I'm going to run this test...
48:24 ...and build a heartbeat over time and say, okay, why did the performance dip here? Wow, it went up here.
48:30 'Cause it goes up every time you get a new version of ArcGIS, right?
48:37 We have a Resource Center, okay? There's a ton of information on it. Andrew will speak a little bit about that too.
48:45 And we've got something that we're definitely excited to talk about...
48:48 ...which is this thing that we're referring to as solution designer.
48:52 This is a new concept. So I'm not sure, I have no idea how I'm doing on time.
48:56 So... You stop now. That's perfect. Okay.
49:01 So we're going to talk a little bit about that, and we'll come back and do Q&A at the end okay?
49:08 No, you're not going to be able to sit over there, dude.
49:19 Thank you, Andrew. You do realize that my name is Andrew too. And we both have beards.
49:25 This is not a requirement for a solution architect.
49:30 Okay, so briefly, so Andy touched on a lot of helpful topics at the strategic level.
49:39 My part of the presentation, a brief presentation, will be how to link this high-level concept into tangible deliverables.
49:48 So for example, what are the resources that we have available to develop [a] strategic plan?
49:53 And then how to use the strategic plan to actually develop the enterprise architecture.
49:58 Then how to use the enterprise architecture to buy the hardware, buy the software, and hire resources, and deploy...
50:03 ...and how to make it happen. So that's the part. So the first, the Resource Center.
50:11 We have an enterprise GIS, and we post information related to enterprise, architecture, security, performance.
50:24 We're working on deployment section; so [there] will be a lot of things like how to deploy NAS and SAN and networks.
50:33 So I encourage you to use this tool as a research tool, to learn about our technology before you commit to any technology...
50:43 ...or before you finalize your strategic plan, it's probably a good place for you to browse to get some information.
50:51 And all the information is at a high level; it's very easy to digest.
50:59 The next thing that I'm going to talk [about] is, I'd like to show you this new tool and where is it coming from...
51:10 ...and what type of resources we are leveraging.
51:16 But the most important thing, how is this tool related to everything that Andrew said.
51:22 So we want to make sure that this tool allows you to capture the current and the target state, just like Andy said...
51:35 ...that allows you to map the technologies to the business processes.
51:40 And it also follows the standard key architecture domains, business architecture, data architecture, technology.
51:48 So what are the benefits of using this tool? What are we trying to accomplish?
51:57 We're going back to the title of our presentation, effective. You can do it without the tool, absolutely.
52:03 But to make it effective, we need to have [a] consistent approach; we need to have ability to share and reuse.
52:10 And if you have 50 Excel spreadsheets and 30 Word documents, it becomes challenging to say, okay, this is our current state...
52:21 ...and this is our target 1, and this is target 2, and this is staging, and this is development.
52:26 And suddenly we need an application to manage our document library.
52:31 So hopefully, you will find that this direction is very helpful.
52:36 We provide consistent diagrams, tables, and including capacity planning.
52:45 So again, this is just a summary of what are the key features.
52:56 How do we use this tool? How do we envision to use this tool?
53:00 This is not going to be the white page where we start from scratch, no.
53:05 The idea is that we will develop templates, the starting points for you that will reflect, match your industry.
53:15 So for example, you might be a global company. So let me just go to the tool, for example.
53:23 So you're a global company, and you say, okay, I need to map all my offices throughout the world.
53:30 So the tool allows you to do, and of course, you can imagine you can drill down into all these points.
53:38 We'll get to this in a second.
53:41 So we will take our best guess on what is the middle-of-the-road pipeline architecture.
53:53 We'll cover perhaps 70 percent, 80 percent, but you're better off with that.
53:57 So you're using this and then you follow all the steps that Andy identified.
54:05 So we need to identify Esri's solution pattern that was defined in our strategy.
54:13 And again, we're going back to the same patterns.
54:16 If you follow these patterns, you probably will easily map these business processes to technology.
54:26 So we allowed this; we need to go a little bit lower and then we need to go into business architecture...
54:33 ...and when Andy was talking about workflows...so how is the replication mapped to business processes?
54:43 Perhaps one of your business processes is, I want to publish our internal data on the public website.
54:51 So we need to have some sort of replication and editing and zooming, and that's how you capture [it].
54:58 Now we're going a little bit, drilling down; we need to actually select an application type.
55:04 So for example, here, we determine that that will be a web application, and because of our standards...
55:12 ...we're tapping into infrastructure architecture a little bit, we were selecting a Silverlight web client.
55:20 That's what we...
55:21 And then below, as you see, we envision these are the core services that will be required to meet these business processes.
55:33 Moving down, now it's time to buy the hardware and configure the hardware.
55:37 So we need to know all the network connectivity that we have.
55:41 We need to select the right hardware, and you can all map that.
55:47 As I said, it might be potentially a lot of data input, but with the templates, they will be quite effective.
55:57 And finally, you can use our predictive models for scalability and capacity.
56:04 So perhaps you have a growth, 20 percent annually, and you want to predict, so that will be available to you as well.
56:12 And at the end, we will produce standard reports. There will be standard Word and standard Visio.
56:20 So let me just very briefly demo what can you do with this tool.
56:30 So you definitely want to define the application, and in these applications, we design each tier.
56:40 So we're really following the application architecture, which as you see...
56:44 ...it's composed of the client browser and we have web services and we have GIS services.
56:49 So let's just click on one and let's see what's behind this; how do we configure.
56:54 So this is where the mapping of the software happens. We'll give you our drop-down.
57:00 This is where your strategic application architecture touches with our technology. You map it. You just map that.
57:11 That pretty much is the same concept throughout. If you need to define the user workflow so it's...
57:22 ...for example, for operational awareness in Amsterdam, we envision that it will be only two map services.
57:34 There will be one map and one image. This is what users will be using. So what is behind?
57:39 Well, now we need to identify how many users, who will be using that?
57:46 Right here, I have a thousand users total, but I only envision 10 percent of them will be concurrent.
57:53 So these are kind of technical concepts, but they're very important to get the capacity right.
57:57 Because if you're a global company and all the information that you have is, well, we have 10,000 of employees...
58:03 ...we have no idea how they will be using this. Well, our capacity could range from one server to a thousand.
58:11 So we need to narrow this down. And you see where I'm going with this.
58:16 There is a lot of detailed information about the networks to calculate the latency.
58:22 So this tool can be used by solution architects but also by engineers who will go and do the installation and tuning.
58:36 So we're very excited about this; it's still very early beta, but I hope it's coming soon.
58:45 Yeah, thanks, I'll take [inaudible]. So thank you, Andrew.
58:52 I don’t know if you guys realize what you just saw, but I don't know, I couldn't be more excited about this.
59:00 What we're effectively going to be able to do is remove the unknown.
59:05 If you do your discovery and you put the input into this tool, which I don't want to say it's a dumb tool, do I?
59:13 If you want to insult me, yes.
59:17 But the tool is dumb unless you give it good input. Does that make sense? Okay?
59:21 So if you give it bad input, is it dumb? No, that's not what I mean.
59:25 I just mean it doesn't presume that you don't know what you're doing and that you haven't done your discovery.
59:29 It actually forces you to realize that you need to go to the business and ask the right questions.
59:35 It's very exciting to me to be able to see right there, through drop-down menus...
59:40 ...that I can actually put in my workflow starting with the patterns.
59:45 Now where do I...I realize in the process of discovery that I need a fully cached...
59:49 ...or half blended, you know, cache web map and half dynamic.
59:54 In the business architecture, right, because you're asking the stakeholder...
59:59 ...How often does that data need to be made available to the end user?
1:00:03 Well, if the answer is once a day, do I need a direct connection to ArcSDE in real time?
1:00:07 No, I can replicate that information across to my application server and store it there for a week at a time.
1:00:15 Now we give you the ability, once you figure this out, to actually choose these, okay?
1:00:19 And it will spit out a proper architectural diagram and capacity recommendation.
1:00:25 I think this is very significant, and it's almost a paradigm change for us.
1:00:29 We've never had a tool like this. Now, I'm not sure what we're thinking about in terms of deployment yet.
1:00:34 But as you learn the tests that are included within it, you can use the tool to actually navigate through different scenarios...
1:00:48 ...for your deployment options.
1:00:50 So as you're learning from your business what's required to meet the business needs, you can iterate with this.
1:01:00 So I think the application technologically was built in Silverlight, right?
1:01:04 Actually, it's WPF. WPF, okay.
1:01:07 So this is something that we potentially will make available via the web at some point.
1:01:13 So I kind of wanted to review what we covered today, and then we'll take some questions, okay?
1:01:20 So I started today by characterizing GIS in the enterprise, and how to grow it, right?
1:01:27 Gave you an architectural vision with those four patterns. I talked excessively about patterns in practice, time and time again.
1:01:36 From the beginning of our conversation today all the way up through Andrew's demo.
1:01:40 We reviewed at length this discovery process, or approach.
1:01:46 You're going to have a different one; this happens to be the one that we're using, okay?
1:01:50 But I think recognition that you need to do that is quite important.
1:01:55 So what tools are out there? We talked about our resource center; we talked about the solution planner.
1:02:04 So an important thing to leave you with before Q&A is to recognize that the system has changed; the game has changed.
1:02:15 The ArcGIS Server environment today is very different than it was a couple years ago; it's pervasive, right?
1:02:21 It's available in the cloud. You have these other options now to expand out capabilities temporarily and bring them back...
1:02:28 ...to add value to your on-premises deployment.
1:02:32 So it's pervasive whether it's a local install or just somewhere within the enterprise or out in the Amazon cloud today.
1:02:39 So if you've got a massive amount of geoprocessing that you only do once a year, should you really pay for hardware...
1:02:44 ...and provision it and support it on-site to just do that processing that only happens once a year?
1:02:51 Maybe not. Maybe you have to because you're not allowed to do things outside of your DMZ, but...
1:02:58 You know, these are things that we need to think about, but for visualization, for the creation of data, collaboration, discovery...
1:03:05 ...management and analyzing of your spatial data, whether it be on the web, a mobile client, or on the desktop...
1:03:12 ...ArcGIS is pervasive, okay. It's a complete system. It's mature, it's IT class technology, it's enterprise class technology.
1:03:22 We should treat it that way, okay?
1:03:28 So I want to say thank you. We'll take some questions; I think it would be great to get some feedback.
1:03:36 I'll stick around as long as I need to, okay, so have any questions? Yes.
1:03:41 [Audience question] Do you know when the system designer will be available?
1:03:43 See, I knew that would be the first one. Do I know when the system designer will be available?
1:03:52 ASAP, right. No, we don't. Next question, yes, yes, Liza.
1:04:06 [Inaudible audience question]
1:04:18 No, you...Here you go.
1:04:22 No, you only capture, the only dynamic thing is capacity planning. I meant this as a joke.
1:04:30 We have some idea when this tool will be available...
1:04:33 ...and I hope that within a month it will be ready for power user group's evaluation.
1:04:41 And when it passes the test, then we'll make a decision what to do with this.
1:04:45 So I don't see this as a next year thing, no, a month, two or three.
1:04:51 Distribution of that is something that we need to determine, but technically, I think that it will be ready within a couple of months.
1:05:05 ...it's not an operations tool, it's a planning tool, that you can include your own tests in it as part of the templates.
1:05:16 So again, to promote you all doing tests on-site on your own network, the inclusion of these tests would give you results...
1:05:25 ...based on your own environment. This is also a game changer, so to say. Yes.
1:05:34 [Audience question] So, in this designer, you have four patterns.
1:05:38 Yes. [Inaudible audience question]
1:05:47 [Audience question] So, two questions. One, will it combine some results [inaudible]?
1:05:54 And second, you had mentioned about capacity planning. Is that the same as the capacity planning you currently have?
1:05:59 I think an important concept, too, that was a great question, is that although you're not going to manage or...
1:06:03 Okay, so the question; I'll repeat the question first, make sure I get it right.
1:06:07 The question is, If you're choosing by pattern, will the tool be smart enough, I guess...
1:06:11 ...to either combine or not combine them by pattern?
1:06:16 And is this tool a replacement or in addition to current capacity planning recommendations that we do today?
1:06:23 So, do you want to start? Sure.
1:06:27 I envision that we will start with a business process. So the business process, for example, will map to our pattern.
1:06:35 So it will be category of the business process or it will be the category.
1:06:39 And then, this will be mapped to applications, so it's one to many.
1:06:45 Operational awareness can have ArcMap and Silverlight, and you will have an ability to link it.
1:06:55 Yeah, that's, and just to add to that, one of the best practices that we're promoting by use of this tool and the patterns...
1:07:02 ...is that sometimes, or oftentimes, you shouldn't be combining the patterns from a CPU or computational perspective anyway.
1:07:10 As a best practice; doesn't mean that you can't do it by loading up a certain machine.
1:07:15 It's just that the usage patterns are so different, it's not a best practice, okay?
1:07:22 And the second question. To answer the second question, at this point, there are no plans of replacement.
1:07:30 I believe that there is a market for both tools. The other tool comes with a related book and classes.
1:07:37 And many people are familiar with this, so I think that it has a valuable place.
1:07:43 However, the type of, the scale of our users is growing, so if we want to do the global company like Shell or Rio Tinto, BHP...
1:07:55 ...with dozens of offices with hundreds of servers, doing this in a spreadsheet becomes challenging.
1:08:02 So I should say that all the modeling tests are coming from our benchmarks.
1:08:10 And we build models that are consumed by this tool and anybody else from Esri.
1:08:16 So whether you're using Excel to read the results or this tool, the numbers should be the same.
1:08:27 But this tool gives you flexibility to consume your own custom model, so there will be instruction on how to do this.
1:08:39 Because for us, what we're recommending is conduct a benchmark of your system and use this information to predict the scalability.
1:08:52 That would be the ideal. Thanks.
1:08:55 Thanks. That concept of evolution, I think applies here too.
1:09:00 It's just that technology's kind of changed, the customer base has changed.
1:09:03 We need to look at things a little differently. Yeah. There's other hands before, all the way in the back?
1:09:10 [Inaudible audience questions]
1:09:34 Did you catch all that?
1:09:35 Yes. So the first question was, Is it just hard coded for Esri GIS? And we have lookup tables that list our software.
1:09:45 But there will be an admin tool that allows you to insert any type of software you would like to capture.
1:09:54 So it has the flexibility of being an enterprise. And the same with capacity.
1:10:02 These models don't have to be just about the map, zoom and pan.
1:10:07 They could be about SharePoint or e-mails or any non-Esri, just as long as you follow the principle of how we consume these models.
1:10:18 So absolutely, this can grow. I guess Andy and I were making baby steps.
1:10:24 Let us first succeed within [the] Esri domain and perhaps one day we will add the enterprise to the system designer.
1:10:34 I mean, we worked very hard to get it done to show it this year in the state that it's in.
1:10:41 You know, I don't want to be a buzz kill, in a way, but the 45 minutes or an hour of stuff I talked about...
1:10:51 ...you have to do it, or the tool's useless. The tool isn't the silver bullet; it's a tool.
1:10:58 It's part of the strategy that Andrew and I laid out. Did you notice we showed it last?
1:11:04 Why did we show it last? Because it's not useful unless you've gathered all of this information, okay?
1:11:10 You can't start with the tool. So even if we made it available today, how would it work for you?
1:11:18 You've got some discovery to do, right? That's, I think, the message we're also trying to deliver with this tool...
1:11:25 ...is that it requires good input to give you good output, okay? Yeah.
1:11:29 [Inaudible audience question]
1:11:36 It's not public. No. No. If it was public, if this thing was fully released, we would have a whole session on it.
1:11:43 It's kind of just an add-on to the end of this to...
1:11:44 [Inaudible audience question]
1:11:48 Yeah. No, fair enough. Okay, so let me say this then.
1:11:51 So Andrew talked about the resource center and that spot where GIS in the enterprise exists.
1:11:58 We're also actively going to begin blogging after our enterprise SIG this morning.
1:12:03 This will become a community. We will make an announcement on the blog.
1:12:08 And I'm sorry I don't know the URL off the top of my head for the blog, but you can track it down fairly easily, alright?
1:12:14 Is that fair enough? Okay. And Andrew's got something to add too.
1:12:20 I just want to answer your second question; I think you just did.
1:12:25 This tool is not available yet publicly, and I feel that we'll be ready for beta release within a month, perhaps two.
1:12:36 And at that point, we will contact interested users on how to distribute this tool. So...
1:12:46 Yeah, my vision is the blog. We could ask for your card and we'll be glad to take it and derive some sort of list.
1:12:52 But I think the blog is probably the most contemporary way to do this.
1:12:57 And also I would ask, hey, maybe this is a fair thing to bring up now, as we blog about it, blog back.
1:13:03 Let's get going on how this works, how this is working for you, okay.
1:13:07 Let's not create a black box or a practice wrapped around a tool.
1:13:11 Let's create a practice wrapped around best practices of enterprise architecture. Yeah.
1:13:20 [Audience question] Andy, I...well, anyone. I have another question about the planning and analysis usage pattern.
1:13:29 Since, as in management [inaudible] management, and planning and analysis would be, the way I see it...
1:13:36 ...[inaudible] decision methods, decision support tools, geoprocessing stuff.
1:13:42 But the situational awareness would be the web [inaudible]. Typically, I had thought of providing or empowering...
1:13:53 ...my decision makers also through a web enterprise, but the processing is maybe a heavier processing than here...
1:14:02 Correct, yes. [Audience question] ...web. So if we all just separate web stuff and geoprocessing...
1:14:10 ...and that would be separate from [inaudible], would you see that as desktop served or web delivered?
1:14:22 Okay, desktop authored, right? And web delivered as a feature class that's optimized for web delivery.
1:14:30 There may be a case. So the question, I'm sorry.
1:14:32 The question, oh geez, I hope I can repeat this after a long day of talking.
1:14:38 She basically was talking about the four patterns of usage that are brought up or the architecture vision.
1:14:44 And how I talked about separating the use of the technology by pattern.
1:14:50 So her question is, Do I see the planning and analysis pattern as being a place where you serve data back to the business...
1:15:03 ...or for web delivery, right? Right. And the operational awareness pattern purely for web delivery.
1:15:10 I think, if I may just give you my quick interpretation, if I'm right then, I have a sense that your question was...
1:15:21 ...Is operational awareness associated exclusively with web and planning and analysis with the geoprocessing?
1:15:33 I doubt that that's what we're saying.
1:15:34 We're not saying, you can have another web viewer of your geoprocessing results...
1:15:41 ...but this is not [a] core business process.
1:15:45 So it's a supplemental technology that you can use to validate your operational awareness.
1:15:53 So I don't think that there are hard lines between technologies; it's more about...
1:15:59 ...what are the key business processes [that] are served by these patterns.
1:16:04 Yeah, thanks. And I think also, I mean, there may be a case where we need to go from...
1:16:09 ...an operational awareness type application directly back to our server and perform geoprocessing.
1:16:15 What I've seen time and time again, though, is that as I do the investigation back to the business...
1:16:19 ...we really don't need to do that.
1:16:20 We can preprocess the geoprocessing and store it in an optimized web format.
1:16:25 It doesn't truly have to happen in real time.
1:16:27 But if it does, and in some cases it does, then we build the architecture to reflect that.
1:16:34 And we would likely have a separate geoprocessor, geoprocessing server, that exists for the web pattern.
1:16:42 I think that's a summary for what you're asking. There may be a reason that we have to do on-demand geoprocessing.
1:16:50 The Drive Time tool, I mean, something like this, okay, I recognize that.
1:16:53 The lines are still very clear to me in that case, that I have a web mapping application...
1:17:00 ...that if I'm going out to do geoprocessing, I'm not going to do it on top of my data editing environment, right?
1:17:06 I'm going to keep them separate; I want to scale them separately.
1:17:08 What if everybody really loves that geoprocessing on day 2, I quadruple my user base?
1:17:13 This has happened to me, so I need to scale that separately from my data management problem, the actual operational awareness.
1:17:20 [Inaudible audience question]
1:17:23 This presentation will be available on, you know, I've said DVD all week.
1:17:28 We put out a DVD with all the Conference Proceedings on it, and then somebody said...
1:17:32 ...I heard you're going to do online distribution.
1:17:34 I'm going to say DVD, and it ships within a couple months. [Inaudible comment] I'm sorry?
1:17:37 [Audience comment] It's both. Is it both this year?
1:17:39 Okay. So, yeah, you'll just search for this time slot in the Conference Proceedings, and you'll have it.
1:17:48 Other questions? Yes.
1:17:51 [Inaudible audience question] Yes.
1:17:56 [Audience question] You've got a complex business. It could take months before we could actually [inaudible]. Yes.
1:18:05 But there's value being lost. How would I maybe [inaudible]?
1:18:16 Okay, the question is, we've got a complex business, and as we do this investigative work for business architecture...
1:18:22 ...and information architecture, and everything I talked about today, business goes on, right?
1:18:26 And what are the shortfallings that we're encountering by not doing actually on either side of that, right?
1:18:32 By not being able to serve the business because you're doing investigative work, right...
1:18:37 ...or not being able to better serve the business by doing that investigative work.
1:18:43 I think you've got to think about that timeline that I talked about earlier.
1:18:48 Remember the timeline with the storm and the clouds?
1:18:50 And you need to figure out what goes into the beginning of that and what's at the end.
1:18:54 And time and time again, I see people putting too much into the left-hand side.
1:18:58 And they can't get through this before the organization changes or they lose funding or they've just got other things to do.
1:19:03 You need to limit what goes into this, okay? And when you're going back to the business, maybe you're just going for...
1:19:09 I'll go for the geocoder first, and I'll take that geocoder as a project through this process.
1:19:15 Does that make sense? It's really difficult to do that because you're serving the business.
1:19:24 But if you do a good job at the end, that presentation that I talked about...
1:19:27 ...by presenting back to the business what you want to do, they will give you the time.
1:19:32 Sometimes it's in the form of a contractor; sometimes it's in the form of an added resource.
1:19:36 Really, I'm not saying that I'm going to, because of listening to me, you're going to get Accenture, but you might get some help.
1:19:45 [Inaudible audience question] Yes. Sure.
1:20:06 He's asking about what kind of deliverables exist for information architecture specifically.
1:20:10 But there are hundreds of deliverables that can exist as part of this process...
1:20:17 ...because it's part of an enterprise architectural framework.
1:20:20 What I would say to you is do some investigation into the framework itself and just pick the artifacts you're interested in.
1:20:26 Like a data application matrix is really useful to me; I like that.
1:20:29 Maybe it's not useful to you; I'm not sure. That chart that I talked about, ability to disrupt, to me, that's important.
1:20:36 Maybe to you, you don't need to know. Maybe you already know who's going to disrupt you, I don't know.
1:20:40 So there isn't one specific answer for that, alright?
1:20:43 But we can communicate more, I can point you to a website. Just go to TOGAF and take a look at the artifacts.
1:20:48 Scary. Boil it up.
© Esri 2013 http://www.esri.com