 I think about storage as an archaic term. It's really the big problem out here is the data management problem. In fact, when you think about a client who's running maybe their core applications which could business critical, economics is about continuous operation, they're not thinking about storage, they're thinking about continuous operation. Storage has to be part of a data management architecture. How am I managing multiple copies? How am I doing batch processing so I can take data offline and do analytics on it? How do I do get best resiliency? So, storage is an element of an overall data architecture and that's why if the vocabulary storage disappeared from the industry, then we will focus on data management because any time we have a discussion about storage, people talk to me about dollars per gigabyte rather than talking about the cost of data management and how should we do data management? So, will storage go to cost? I mean, you have a few players building drives and flash and the rest of the software is going to go down to zero price wise from an IPF standpoint. I mean, eventually that kind of is a direction. I'm not saying, you know, not going to zero tomorrow but that's the trend, commoditization. Question is what is the value, right? How are you using data? That's the value and different data will have different economics. All data is not created equal. It is about economics of data not dollars per square inch. And the word metadata is now mainstream. I was made a tweet about that the other day like, you know, all this stuff about the NSA and the prism story about the quote privacy. The word metadata now is now a mainstream term. People now know what metadata means. Talk about metadata because this is data coming in from all over the web and corporations have the same challenge. How is, what is the importance of metadata? You mentioned it about storage and where is that going to come into the data processing world and has it affect your business? You can say, okay metadata, then metadata itself is growing faster than the data itself, right? So you will have metadata, metadata, right? So eventually anything that you're talking about out here is understanding of what you have and what's the economics associated with it. So I don't want to put a label like metadata on it. It is about understanding of what you have, what's its economics, and how you take advantage of it. Different economics and because you and others have perpetuated you being IBM this notion of, okay, different data requires different products. So you have a box with this type of data, about that type of data. You put forth a vision today where that goes away. I like to, I sometimes I say Amazon turned the data center and into it and the API storage into a, into a platform, you know, with an API. And is that the direction that you're headed? This problem of different box for different things. I would rather see a software layer associated with it all open, standard space, and based on the requirements of the application or business use case scenario, you choose the appropriate media or the combination of media to optimize that business scenario. I think that eventually application defines how they want to access the data. There will be gazillions of applications that will say, hey, this is the requirement and I want to access my data based on blocks, or I want to access my data based on file, or I want to access my data based on, based on objects. We can't really create a separate system in a box based on the APIs you are accessing with or the how application wants to access it and what media you attach inside it. That's why we talked about creating the third generation store-wise platform, which is software-defined storage. Underneath that, you can use IBM or non-IBM storage. Hardware can be anything. And our partners are extending it with file, with object, with analytics, with replication, with copy data management, everything. There is so much inefficiency in the industry because we have created separate boxes for every API and for every media. Third generation is all about, as we discussed earlier, you cannot manage all the data on your own. You cannot. You don't even understand metadata. Policy-based data management has come to an end. You're doing an SD, our third generation platform, store-wise platform, is adding code analytics inside so it is self-understanding in terms of what data is coming in. It's creating valuable quote-unquote metadata to repeat your words so that people, the applications can learn it, leverage it, and drive the underneath hardware, which is the storage. We're at a 20-mile stair to that mountain you have to climb called software-defined infrastructure. It's really challenging. It takes a lot of... You can solve the problem, which is the extreme problem, and say, can I embed Watson inside my software-defined storage? But many things you can do with simple and memory analytics. Let's say you have unstructured data, and it is about oil and gas. And there are only 50 or 70 data formats that all the forms get written out. Can you automatically understand what type of data is? It's not that hard. You don't need a Watson. And to be able to deliver the analytics, you can manage the metadata itself and be able to deliver a scenario so applications can say, how I leverage that particular data. So I'm just saying that, yes, it's hard. I'm not belittling it, but there are so many scenarios. Whether you just go into healthcare, you go into oil and gas, you go into, you know, media delivery, you go into high performance computing, you go into risk analytics. It's not that hard. In the history of IBM storage is story driving. Go back decades. This drive, you get the San Jose R&D facility. It's a story of history. But over recent years, some have said it's been this little island out there. And when you looked at coming on, taking over, what did you see? What was your vision? And why now? And what is the next few decades going to look like? I mean, you mentioned storage is going to be commoditized. But what did you see? Was it the opportunity, diamond in the rough? Was it the collapse of different groups within IBM to take the software? Can you share your personal perspective? Can you keep talking for a little bit? Let me say a few words to you. Look, this whole traditional storage industry has been stable for decades. And any time an industry reformed, you have an opportunity to shift the game. Such a fundamental change in our industry, which is $10 to $50 billion in that range, then new players will emerge. You can be resisting that change, or you can be at the front end of it. Out here is to be the leader in both tipping points, make the open player, advance the industry to the semiconductor storage. And when that happens, you change the game for our clients and for ourselves.