Friday, January 16, 2009

Open Platform VMS. Really?

Yesterday, someone kindly pointed out to me that their video management software, into which they had integrated a number of IP cameras and encoders from different manufacturers, was an 'Open platform'. I agreed that it was a great achievement to work with so many hardware manufacturers and their respective SDKs, to integrate them, test them, support them and continue to keep up with the continuous stream of new and differentiating features that they add every other day. They said that Bosch's encoders were integrated, which made me very happy since I agree that end users should have the option to buy hardware from one provider, and software from another. Choice, I agreed, was in the interest of the end user.

Uncertain about a couple of things, I asked a couple of simple clarifying questions, since not everything is black and white - including integration.

"So, your Video Management System can see our Dinion IP cameras?". Yes he replied. "Great. What about our AutoDome IP camera?". Well, he replied, it should work - it's the same under the hood. "You mean you can control it?". No, but we could if we have a big enough project and prioritize it above the other cameras. "What about playing back our iSCSI recordings, VRM-managed SANs, or edge recordings like CF cards and USB hard drives? What about receiving our video analytics alarms or the precious meta-data that comes with it? Audio? Analog decoders?". No, we don't support those. We don't have time to implement the subtle strengths that make each piece of hardware unique. We don't have developers sitting around twiddling their thumbs.

Is, then, an Open Platform VMS one where its creators have integrated multiple devices so the end user is not tied to one hardware manufacturer? Or, is it a piece of software, where manufacturers of IP camera and encoder (and DVRs for that matter) can, independently of the software creator, make their edge devices compatible with the software, in all its differentiated glory. It reminds me of how every printer in the world works in Microsoft Windows, yet they did not wait for Microsoft to integrate it in, yet every application that runs in Windows can benefit from every subtle feature for the printer they are connected to.

Such 'plug-ins' or 'drivers' are an ancient concept, but I wonder how common it is with Video Management Systems. If not, then end-users are at the mercy of the software vendors, and not the hardware manufacturer, to integrate their hardware of choice. And even if it appears to be integrated, which features have been integrated, how transparently, and how reliably? Good questions to ask when selecting an open platform video management system. Make sure you're getting the most out of your edge devices - there's a lot of intelligence built into them nowadays. It's not just about being able to see a picture as it mainly was in the good old analog days.

Interoperability initiatives, like those proposed by ONVIF ( and PSIA ( address a large number of these issues, especially around device detection, control and streaming. It is clear that a specification is desperately needed, and eventually it will be adopted by a standards body to become a true standard. It is also clear that it will evolve as its sophistication increases. I am optimistic that this is a giant leap to accelerate the rate of adoption of IP video.

No comments: