 So, I have this friend who's a building architect and I go out, you know, to have a beer with him now and again and we talk about the lines of business that we're in and the parallels between them. The purpose of this talk is to tell you a little bit about what he and I discussed and try to bump you into a new way of thinking because conferences are great for sort of getting you out of your standard head space and think about novel issues. So, can I have the clicker, by the way? Is this this one? Okay. So, both fields involve a lot of engineering paired with a lot of aesthetics and in both fields tastes change over time and the technology tends to change out from underneath you. Architects get new building materials and methods and we get new frameworks, technologies and infrastructure. So it's not really surprising that if you have constant change, a lot of different camps arise and it's not that anybody is really right or wrong, it's more that they have different problem approaches and they have different advantages and disadvantages. So, what do you pick? The question that I was asking him was how do you ensure that the form of the solution follows its function? And he actually surprised me and he said that it really shouldn't. This notion of form follows function, it's actually really bogus. In fact, form should follow purpose. You see, the normal approach that we use is that we build a system by breaking it down into a number of functions and then we connect those functions up. In the architectural sense, think of a school and you think about classrooms and hallways and so the purpose, the function of the building is to break it up into a number of classrooms and then to shuttle people back and forth between those classrooms. But really the purpose of a school is just education and this idea of classrooms and hallways between them is only one way to accomplish that purpose but certainly not the only one. So we're trying to get to this thing where form follows purpose. Now we're great at central planning and personally I've seen a whole lot of top-down enterprise architecture work being in government. And central planning is great in theory but in practice it's very ugly and it's difficult to keep clean and pure. Often it's not even used in the first place. The other interesting thing about central planning is that it introduces an artificial separation of functions which is then inefficient because it then buys you the problem of getting people and data back and forth between the functions. So how do you get it right? A couple of key principles to think about. The first is to try to keep things human-centric and not tech-centric. And I think Apple provides a good example of this in the tech field. They hide complexity and they make it simple. The second is to try to design for open spaces with a lot of common reference points. You don't try to design for separate functions with passageways in between them. You know this conference room that we're in right now is just one gigantic space that they have the ability to subdivide as they see fit. And so it's this fluid space that can be used for a lot of different purposes. The web is like this too. It looks like a lot of individual compartments but really it's a big open space with links that allow you to jump around from here to there. So this is one of my favorite pictures. It's an architectural picture of reality. And as we all know, reality is a mess. When things get complicated, the problem, the solution gets messy. And basically when the going gets complicated, the smart start to specialize in the architecture world, they actually now, there's this idea of a design architect and a project architect and they do very different things. You know, similarly for us in IT, there ought to be a difference between doing the architecture work of figuring out what the system does and doing the architecture work of figuring out how it ought to do those things. We need really strong reductionist thinkers to figure out how we're going to implement it, but you need a creative and expansive thinker to help you work through issues like, how do you design for open and repurposeable data? Finally, last thought is that we need a lot of viewpoint diversity when we do IT work. In some ways, talking about the need for diversity is a cliche, but it's really important in architecture work. There's a type of person out there that thinks very clearly and very deeply and immediately jumps to a great technical insight. And then there's the kind of person who knows how to think around corners and think over walls. And whether you're building something new or revamping something that already exists, you're going to need both. So just finally, give a little bit of thought to these principles. They're not really meant to be solutions for you. They're more meant to be riddles. It's in the figuring out how this applies to you and the specifics of your situation that it might help you gain some insight into what you ought to do next. Thanks. Terrific.