Sunday 1 September 2013

PayTV Operators should STOP re-inventing the wheel and should NOT see Google as Evil!

I have been working within the PayTV Software & Systems space for almost fifteen years now, and from the very beginning with Set-Top-Box software, I really wasn't impressed by the software technology. Apart from the hardware being fairly interesting, basically a device for decoding video/audio/data streams based on MPEG / DVB / ATSC / etc protocols and standards, when it came to software, there wasn't really a "wow" factor in it. Of course, we can't forget the really crucial element, the heart of the system, the crown jewels, the revenue-generator, the very interesting & complicated black magic technology called Conditional Access (CA) which is really really cool, the rest of the building blocks was really around the Set-Top-Box user experience / application, which really wasn't all that new: 
  • Essentially the device needed an operating system, a way to draw stuff on-screen, a user navigation interface, and some data source driving the application, traditionally called the Electronic Program Guide (EPG). The STB software thus re-applied knowledge well-known in the computer industry for decades (Model-View-Controller MVC design pattern emerged in the late 1970s), simple operating system and driver / hardware abstraction layers, C-code...
As all hardware devices tend to follow Moore's law at some stage, Set-Top-Boxes (STBs) have evolved to quite powerful machines allowing for migration to newer, modern software implementations (although nowhere near the computing power & rich experience offered in the latest smartphones & tablets - post for another day), the software too, has become more accessible than ever, with more STBs using Free Open Source Software (FOSS), particularly the dominance of the Linux Kernel as the popular operating system of note, displacing historical dominance of VxWorks, STOs, NucleusOS, uCos, etc, etc.

However, there are some components in the STB software stack that remain fundamentally closed: The STB Middleware, EPG, Conditional Access (CA) clients. Okay, ignoring the CA client, which has always been fundamental and will never go away for years to come -- the classic Middleware/EPG components really don't need to be that closed anymore. 

There is also the backend / headend information service data generators that are traditionally closed, although vendors purport "open protocols", the PayTV Operator generally obfuscates the openness by forcing business-specific modifications in the protocols, apparently unique to each PayTV Operator/Broadcaster...

Traditionally & historically though, these software components were provided by third-party vendors that PayTV Operators just accepted as the norm. Highly closed, difficult to integrate with open systems, these vendors capitalised on providing a closed system, to the extent of locking in the PayTV Operator to the entire stack, some vendors reaped the benefits of providing the end-to-end system, one-stop-shop for everything. Later, PayTV operators decided to take more control, diversify the ecosystem by enforcing the use of multiple components, not being locked-in by just one vendor, promoting open standards for integration, and more recently taking more ownership of some of the development and integration...

Yet, the models within which most PayTV operators continue to work - is still pretty much a closed one. There is an aversion to sharing, opening up technology to other parties with a view to extending partnerships as well as creating new strategic relationships. There is an huge element of mistrust, not-invented-here attitude, we-can-implement-this-in-house, etc, etc!! I am really dumbfounded with that approach...

Stop re-inventing the wheel!!! 
PayTV Operators are stuck in this old-school mentality that is beyond comprehension in today's age. The pace and changes in technology are more disruptive than it has ever been...yet PayTV offerings are still years behind modern computing experiences!

I strongly feel that PayTV Operators are playing too slow, and missing on a tremendous opportunity to partner with new players, especially the likes of Google! IMHO, Google really doesn't care about owning the content per se (but yes, Google is all about people discovering content, and this content might be in competition with TV Operators), the PayTV operator can still own the content, protect it, and make tons of money for its premium content including all the highly-sought-after premium sports content. The rest of the services are peripheral: Middleware/EPG experience, Data Services are NOT the core business of the PayTV operator, but grudgingly has to maintain these services. Adding new features really adds pressure and strain to the PayTV operator, yet they are still willing to invest time, and loads of money in re-inventing the wheel, providing their own customized, highly bespoke systems because their third-party vendors had told them so!  

The internet connected world, emergence of advanced technology experiences of smart devices, the browser, social experiences, web applications, video/voice and gaming experiences -- are pushing the PayTV ecosystem on the back-footing. Technology-wise, Google has the lead on almost everything, the need for partnering with Google as a Technology Partner is increasing (IMHO - my gut feeling, no real market research done, I have a nose for this kind of stuff) - I say PayTV CTOs should explore getting in bed with Google sooner than later, business/economics can be worked out later!

I've highlighted some new killer features highly sought after by TV Operators as the "next wave of innovation" to remain "current and competitive", that I think can be really short-circuited reducing the time-to-market drastically, if these old behemoth TV Operators opt to partner with Google instead! This is what I would do if I headed up a technology department for a large PayTV Operator:
  • Advanced Search
    • This should be a no-brainer but apparently not. Google is the best in search. It's all about data formatting, indexing and really intelligent algorithms. Why are broadcasters re-inventing the wheel with creating arbitrary metadata formats, creating own crude search systems and trying to be like Google? 
    • It would be  wiser to partner with Google's Search team, use Google APIs for data mining and leverage of the world's prominent search provider.
    • How hard can searching through a month's worth of schedule data, with metadata on content tags like genre (action, movie, comedy, etc.), title, director, actor, parental rating, etc - Google could give you the design within a couple of days or less, along with rich data sets for testing...
  • Personalised Recommendations
    • Okay, this might not come as a natural choice for Google, one could easily lie in bed with Amazon, since Amazon by far has had one of the most influential recommendation systems to date. 
    • TiVo have had over a decade of testing and evolution of their recommendations, so would also be a worthy partner.
    • Although Amazon/Tivo do pose a threat since they're directly involved in TV (as if Google isn't - but GoogleTV is heading in different direction altogether) - so Google could be a partner here. Coupled with its enormous knowledge-base of Search, Google Now could be a good fit for a recommendations partner.
    • With Chromecast, recommendations might be more targeted as well (Is the default view a competitor, i.e. Google will feed traffic to its own Play Store?? Maybe because PayTV Operators treat Google as Evil, but perhaps partnering could offer Google different incentives, like profit-sharing?)
  • Targeted Advertising
    • Similar to search, the king of advertising is definitely Google again.
    • So the TV data format might be a little different to web profiles, but the concepts are still the same: How can you target specific advertisers to specific audiences (based on behaviour, past search requests, demographic / social profile, etc, etc.)
    • Google can offer technology services and expert consultants in this area - Partnering with them would definitely shorten the architecture / design stage.
    • I know of Targeted Advertising systems for TV that have been in development for more than five years, and are only just going live with first release this year - surely this is too slow an approach??
  • Home Networking - Sharing of Content
    • PayTV Operators would like to have a gateway STB in the home, acting as the central hub to share content within multiple devices in the same local home network. 
    • They also would like to enable off-screen, on-the-go sharing on portable media devices. 
    • Due to their traditionally closed systems, PayTV operators extend their partnerships with existing Middleware vendors to support Home Networking - which is usually built on UPnP stack, or other standard for home networking (including MOCA, etc).
    • Google is your friend - partner with Chromecast!!!
      • All you need is to partner with Google, extend Chromecast to the STB domain, use it as a model for sharing content within the home, over Wi-Fi
      • This is sure to disrupt traditional models for Home Networking!
        • Embed chromecast engines in STBs within the home
        • Content can be discovered and served from one STB to another using the chromecast protocols
  • Advanced Operating Systems
    • Android - need I say more??!!
    • Traditional closed STB-middleware systems can be displaced and disrupted in pretty much the same way as the mobile phone disruption, knocking the likes of Nokia/SymbianOS in the dust
    • Although Digital TV penetration does not have the same numbers as Mobile Phone penetration, it is still big-enough of a pie to beg the question of why hasn't Android/Google pushed out a completely open stack/design for Set-Top-Boxes (matter of time I say)
    • Related to operating systems is really the Chrome-OS experience
      • Whilst broadband benetration / IP-TV isn't as easy and ubiquitous as satellite dishes installed almost in every home (Apart from cable providers), a cloud-based STB is surely the next big thing, and who else better suited to partner with than Google??
  • Application Stores
    • Related to Android & Chrome - the chrome app store is an compelling experience from a browser - surely this experience can be brought to STBs??
    • If PayTV operators want to share the whole application store experience, instead of re-inventing the wheel with yet-another-app-store, partner with Google to see if something can be re-used or tailored for the TV experience
    • A PayTV operator venturing into its own independent app store providing APIs to third party application developers is quite non-sensical

2 comments:

  1. This comment has been removed by a blog administrator.

    ReplyDelete
  2. PAY TV Operators have been offering these services for some time now (many years) so I am bewildered by this article...Apps on TV - Yes even Apps on TV - They used to be called Widgets remember Yahoo Widgets.

    ReplyDelete