Open Systems, Proprietary Cloud

Sure there are increasing numbers of edge devices, cameras mostly, that are open platform, but they’re not entirely open. Too often there are elements of programming that are available only to administrators using proprietary software platforms – systems designed to work with ‘own’ devices. For all their faults, things are improving in this area as manufacturers sort themselves out into alliances and the work of ONVIF and PSIA filters through the product stack. 

But just as our hardware becomes more interoperable, a major new trend is threatening to cart the heart of networked security systems back to a darker age of proprietary networks. What is this trend? It’s the movement of management and monitoring of security solutions – commercial and increasingly, domestic – onto mobile devices running Apple iOS and various incarnations of Google’s Android. And from the wider perspective it’s fierce competition for ownership of internet ecosystems and ultimately, proprietary ownership of cloud. 

The difficulty with proprietary ownership of cloud is this. Third party makers building applications that manage security systems via mobile devices must bond their apps to cloud – they must integrate with cloud as intimately as possible so as to ensure the best reliability and performance. And the result is that relationships with underlying cloud providers are genetically embedded into code and are necessarily for keeps. 

At the heart of the issue is commercial competition, fierce and deep. Companies like Apple, Samsung, Google, Amazon, Microsoft and Facebook are engaged in a war of epic proportions in order to establish the dominance of their proprietary internet ecosystems. They’ll stop at nothing – including hamstringing competitor devices at network borders – to promote their individual corporate hegemony. 

To highlight the conflict, we see patent wars being fought as the owners of these sprawling commercial platforms seek to disable or slow down competitors. Any really clever app builder further down the food chain is promptly swallowed by bigger fish looking to shore up their business models and guarantee a profitable future.  

If this war of the platforms sounds unrealistic to you, consider the workstation you use. Whose proprietary web browser won the battle in the 1990s? It wasn’t Netscape, was it? The company that created the Secure Sockets Layer and Java was run under by market forces and its shadow exists only through Mozilla’s Gecko-based Firefox. Now expand and apply that tussle to mobile operating systems and to the fast growing cloud. Yes – it really is a battlefield out there and there won’t be a crowd of winners. 

So, what’s the point of these observations? In short, if you’re going to take your electronic security system outside the traditional boundaries of your corporate network, you need to understand the play. All the old rules still apply. Cost. Security. Uptime. Inter-network performance. Accessibility. Future-proofing. Interoperability. And all this against a paradigm of no guarantees. 

Oh, sure, you want the flexibility to jump providers but you’ll only keep it by weakening the integration layer between your app and the underlying service layer. And what if you commit to a cloud service that cannot scale or that doesn’t exist in vital areas of your local or global market?  

There are challenges when you start talking about cloud services and about mobile applications and they relate to a definition of terms. Simply put, the term ‘cloud’ applies to any software or hardware services delivered over a network. Pretty easy, eh? And old, too. 

By this definition, the cloud has been around since the mid-1990s and it infiltrates every aspect of WAN communication. This means that the war for dominance of cloud – a thing that we can clearly see is pervasive – is really a war to own internet services. And that’s something security integrators and security managers looking to lock part of their systems into cloud need to pay serious attention to. 

The 4 biggest players are Apple, Google, Amazon and Facebook. Each of these companies is immensely rich in assets and this gives them the raw power to invest in serious amounts of infrastructure, including global networks of data centres. There are other players in the mix, too, including Microsoft.  

While much of the competition between these players will happen in social networking and ecommerce, it’s really the area of software, tablets, smartphones and data centres that should be of most interest to security people. In this area the central conflict is that between Apple’s iOS which is used by Apple in iPhones and iPads and Google’s Linux-based Android operating system used in smartphones and tablets by everybody else. 

Given the hype surrounding Apple, you could be excused for thinking it was the market leader but Android holds a 75 per cent share even though its latest version, Jelly Bean, is somewhat clunky. Android is open source and Google works with third party hardware manufacturers and their software development teams to push the technology forward, while keeping its own supporting apps closed source. 

So far it’s a model that’s worked well, though there have been a couple of misfires along the way. And just for reference, Google activates about 1.4 million third party Android devices per day. Will Google stay out of the mobile device market? The company released Android-powered Nexus tablets built by ASUS in 2012 and its $US12.5 billion purchase of smart phone maker Motorola Mobility last year argues definitely not. 

Apple’s response to Google’s Android has been the voice-activated Siri and the hasty decision to chop Google Maps from iOS. From a security integrator or end user’s point of view, these moves should be seen as outlining the nature of the competition. 

Apple and Amazon are also contesting another front – digital music and digital books. Meanwhile, Amazon rules eCommerce, including books. About one third of online purchases take place through Amazon, which is a vast number. And Amazon gives away Kindle tablets at break-even to give its customers a mobile interface to its online storefronts. Strengthening its act is Amazon Web Services – AWS Cloud – and the company has global infrastructure capable of handling vast traffic in real time.  

Now Microsoft is muscling in on the act. It has a new tablet, called Surface, and its own mobile operating system and the big American also controls mobile phone manufacturer, Nokia. Microsoft has huge potential thanks to the pervasive nature of its software – a proprietary OS that became the global operating system. And Microsoft has a presence in mobile, media centres and the car industry. Interestingly, Microsoft is now talking about itself as a software and service company – that could be a sign of things to come.

The key thing for me though, and for all or you, is that this contest between Amazon, Apple and Google and now Microsoft, is vertical. Each of these huge players is setting up its own ecosystem comprising alliances, firmware, software, hardware devices, eCommerce portals, data centres and, increasingly, cloud services. 

Google’s App Engine is a Python app cloud, while Amazon’s EC2 and S3 provide generic compute and storage clouds. Meanwhile, Platform-as-a-Service provider EngineYard is a cloud for Ruby-on-Rails – an open source web framework which is unsurprisingly backed by Amazon and Oracle. There’s also Apple’s MobileMe cloud, Google Gears and Microsoft Live Mesh. Sure, these are limited for the moment but they tell us where the infrastructure investment is going. 

If there’s a weakness with cloud it’s the way vital alliances will lock service providers into code-deep relationships with the proprietary solutions of monolithic corporate players whose capabilities may not be suitable enough, or flexible enough, in the future. 

If a third party provider with its applications symbiotically melded to a provider wants to change clouds that’s not likely to be easy to do. And there’s no way for apps not to be melded to a chosen cloud. The tightest possible integration is required if apps are to be stable, fast and scalable and to offer the the global performance they need in real time. 

Then there are proprietary apps that third party organisations link their own apps to in a proprietary cloud. These lock client apps to provider apps. And thanks to the closed ecosystems we talked about earlier, these proprietary apps won’t work outside the original cloud, throwing the continuity of the service into disarray. To say nothing of re-encoding issues that are almost inevitable.  

Affordable cloud is here. Google Drive, AWS and smaller competitors like iCloud, DropBox, SkyDrive, MelbourneIT, RackSpace and Box offer consumers and small businesses storage in the cloud and other services at very low cost. And there are more and more remote management apps being released. These open up a range of applications that have been vapourware for 10 years or so. 

I argue remote management apps combined with edge storage using affordable 128GB microSD servers and cloud backup of motion and/or alarm events really is the future of many applications in video surveillance. This change will take place in an affordable pervasive online cloud environment. The question is not whether this will happen but what cloud will look like when it arrives – will it be proprietary? WIll it be open source? 

As some readers will already have discerned, the current situation is thoroughly fluid. While cloud – software and hardware services over a network, remember – has been around a long time, the establishment of large cloud provider ecosystems between which fierce competition forces down prices is still in its infancy. 

It’s not certain who will dominate proprietary cloud services but it is likely that the company that does dominate in the proprietary market segment will not be a startup but instead an organisation already investing heavily in a revenue generating vertical platform of which scalable cloud services are a necessary extension. The market leader will be a company that can handle huge transaction rates, high data throughputs, the best interface to users and which have the most pervasive vertical presence to multiple markets. 

Also vital is cost. This latter is tied to numbers of data centres and their proximity to commercial, industrial and increasingly, residential, concentrations. Proximity is the key for affordable cloud services when you start taking high bandwidth. Such a demand relates to economies of scale and this too, suggests the winner will be a big player. 

When thought of in this way, potentially dominant players other than Google become clearer – eBay and Amazon – the latter in particular. And if Microsoft it flexes its software development muscle, it will make a serious impact. But there are rumblings – and a push against proprietary cloud. Recently, monster cloud user Netflix, rolled out a private content delivery network (CDN) and it did so to ensure reliability, ownership of business, low cost on bandwidth and greater control of its cost model. 

Could similar developments take place in smaller applications with non-proprietary cloud owned by end users, or by manufacturers/developers and integrators? Yes – open cloud solutions do exist. The most notable is OpenStack, which was developed by Rackspace and NASA under the Apache Licence and is supported by such venerables as IBM, HP, Cisco, Dell, Linux, Intel, AMD and VMWare and 150 others. 

OpenStack is managed by the OpenStack Foundation and is free open source software for building infrastructure-as-a-service solutions. I think it’s fair to argue that while late to the party, open source cloud will win the battle for the cloud-based platforms of the future. Once a particular weight of companies throws weight behind OpenStack the result will not be in doubt. It’s my feeling that level of support has already been reached. 

At the same time I think open source will not own the entire market.  

For those extreme sports people wanting to get in on the bleeding edge of cloud and multicloud, the short term play is deep integration of portable applications with a weather eye on open source. For me, low cost, no-lock open source is the best route to the networked electronic security product we’ve dreamed about so long, that magical unicorn, the open platform solution.