Thursday 13 October 2011

Open Source and Cloud, Mobile, Data: what goes where?


After an interesting question from Matt Asay I gave a few answers, then decided that Twitter's not always suited for long conversations (ahem)

I'm working on the theme for OSBC2012. I want to highlight the connection btwn cloud/mobile/data. Can someone help me express that?

My first reaction was that these words are big, and could mean a lot of different things to different people. I don't like to talk about Cloud as it's a blanket term for three different main streams in general: IaaS, PaaS and SaaS (infra, operating system and software).
Mobile? Pretty much the same thing. Mobile has operating systems and software, and is useless without that special infrastructure called Connectivity: a connection to the Web

Data could mean anything as well, but lately Big Data seems to be something to worry about or throw money at. Personally i couldn't care less about data, it's information when stuff becomes interesting: I usually operate at the business logic or application layer where everything comes together and makes sense. below that, on a database level, you might have some structured information but that came through the business logic layer and stopped making total sense when it left that.
So I'm guessing we're just considering all unstructured data here that hasn't passed through the business logic layer yet, or simply can't be classified other than free text

And then tie all that to open Source? I like a challenge

I often talk about business rules versus exceptions, static versus dynamic, infrastructure versus presentation, machines versus humans. My last post touched on that and so do many, many others, including my eBooks. Where do I see Open Source? At the machine level rather than the human level - not that it's unfit for humans, but because humans get pickier when things get closer to them, simple as that

We humans don't care which oil company's gas we use, or who built the roads we drive on. Nor do we give a damn about the people who built our house, or the airports we use to travel.
No, we humans care about the car make and type, and the accessories in it, but especially its colour.We do care about the interior of our house, and many would give an arm or a leg (some would give both) if that were designed by some hot shot interior designer.
We pick our airline with great care, give a buck or so, and take them very personally. We get fussy about overhead luggage and food and drinks, and service provided by flight attendants

That's why Open Source will never come close to the end-user: we want to be treated special, and Open Source is not intended to do so. Open Source is a great initiative aimed at... making the developers feel great about themselves - that's where open touches people

So, I position Open Source at the infrastructure level, and that's where it will stay for decades to comes. More stuff will move down to the infra level, so in the end more stuff will be Open - but it doesn't work the other way around.

Cloud? IaaS and PaaS can, will be and are Open. SaaS? Not SaaS as we know it. Maybe the use of SaaS will drive standardisation and in that way drive functionality down into the fixed, static, boring infrastructural layer, but no, SaaS won't play much of a role there in the coming decades

Mobile? Mobile is perfect to use with Cloud: you get a ubiquitous server for free to which any mobile can connect all the time. Mobile operating systems? Who cares, the hottest news about any mobile is the app store to which it can connect: again, software is the touch point for the end user

Data then? That's interesting to say the least. In my scheme so far, we have software in the Cloud and on mobiles that's still good old 1.0: closed, proprietary, and able to make a buck. Now how can we access that and make our own buck on top of that? By getting its data.
We get that data by interfacing with it, by using its API's. API's will always be semi-open, as they are the property of the one owning the software - no way that's going to be really Open. Handing over your only means to monetise? Look at Twitter and what they've done: make their stream scarcer and scarcer so it automatically becomes worth more and more

So here is how I see Cloud, Mobile and Data connect with regards to Open Source:
  • Cloud will be the sole territory of Open infrastructure and operating systems, enabling old-fashioned software vendors to slowly migrate from their local IT via hybrid Cloud to True Cloud. That will hugely drive standardisation and change the role of system integrators for good. Hardware and storage vendors will get a stiff beating and see their margins drop dead, given the fact that their millions of customers will be replaced by a few thousand at best
  • Mobile will be the sole territory of Open operating systems. Its applications, unlike those from the Cloud, will be directed towards consumers rather than companies / employees, and free (thanks to advertising) or free-ish
  • Data will be linking the two together: application providers in the Cloud as well as on Mobile can make an extra buck by providing API's that will allow their users to retrieve their data. Turning that data into information or even knowledge off-premise before just spitting it all out on-premise will make a fine difference. One way or another, this will be Closed territory
Summing it all up: the farther away from the end-user, the more Open everything will be. The Closer to the end-user... well you can finish that sentence yourself can you?
Matt, the stage is all yours ;-)

0 reacties:

Post a Comment

Thank you for sharing your thoughts! Copy your comment before signing in...