Learning about streaming, with Klip as a first case study

I’m currently trying to learn more about streaming in its various forms. Streaming is all about the delivery method of media such as audio, film or TV. There are basically two forms of streaming: live streaming and archived streaming.

With live streaming, you take the content and broadcast it live over the internet. It involves having a camera to capture the content, an encoder to digitise the content, a media publisher where the streams are made available to users and a Content Delivery Network to distribute and deliver the content. This way, the user is able to view an event live. Good examples of live streaming are live coverage on YouTube of an Arcade Fire concert at Madison Square Garden or people being able to view the first match of the FA Cup live on Facebook.

Archived streaming is more about ‘on-demand’ streaming (think BBC’s iPlayer and Lovefilm’s Player) or Marks & Spencer broadcasting a vodcast on food you can order for Christmas. The technology used to deliver this is fairly similar to live content, apart from the fact that the ‘content source’ for archived streaming can be as varied as a tape or an audio file.

I don’t think one has to be a great visionary to predict a great future for streaming, with a multitude of devices to stream video content to. A good, recent example of these bright prospects is Klip, a mobile video app that lets you share video content with your friends. These are the reasons this service caught my attention:

  1. It’s the highest ranked social video app in the App store – Klip is an iOs (as in ‘Apple only’) app that lets users view, capture, share and discover video on their iPhone. Other examples of social video apps are Rounds, Vlix and Socialcam.
  2. High quality video streaming – A key aim for Klip is to provide the highest quality video streaming around for mobile device. You simply shoot a new ‘Klip’ or grab one from your Klip Camera Roll and share it with the Klip community, your friends on Facebook, Twitter, on your YouTube channel, or by email.
  3. Underlying technology – Even though I’m not a techie, it will be interesting to find out more about the underlying Klip technology that enables users to swipe a video for a preview. The video will then play at the accelerated speed at which you move your finger across the video. If you shake the phone, all the videos will begin playing on the page.
  4. Adaptive streaming – I learnt that because mobile bandwidth can be unreliable, Klip includes adaptive video streaming which means that it automatically adjusts the quality of the video streaming based on a user’s mobile network conditions.
  5. Cloud component – The sharing on Klip is done around hashtags so that users can easily surface and find content by topic or event. Big on Klip’s product roadmap is the cloud component in its technology, enabling users to search and index hashtags in realtime.

Main learning point: Good to learn more about streaming. The term is often used very loosely and distinguishing between live and archived streaming feels like a good starting point. There will be more examples to follow, but Klip is an interesting one since its aim is to achieve the highest quality streaming as well as encouraging its users to really interact with its content.

Related links for further learning:







Lessons learnt from Marc Andreesen: making your own cake and eating it

It’s always interesting to read Marc Andreesen’s thoughts on things. As a co-founder of browser firm Netscape and investor in Internet success stories like Zynga, Twitter, LinkedIn and Groupon I find that Andreesen tends to have a good overall perspective on things, being familiar with the world of startups as well as with that of established companies like Hewlett Packard (where Andreesen is a Board Member).

For instance, when Andreesen writes about Why Software Is Eating The World I pay attention and take note. If Andreesen then builds on this ‘eating’ theme  by comparing building startups to Baking a Cake in 3 Minutes I really want to understand where he’s coming from.   :

  1. Software companies are taking over the world – Andreesen’s main point is that we are in the middle of a big technological and economic shift in which software companies are getting ready to take over large parts of the economy. He believes that “all of the technology required to transform through software finally works and can be delivered at a global scale.”
  2. Any good examples then? Andreesen cites well known examples such as Amazon, Netflix, Google, Zynga, Flickr, Skype and Spotify to strengthen his argument. The two common denominators between these success stories are (1) that they are all software based and (2) that they all have used their software to overtake their more traditional competitors and transform their sectors in the process.

That’s clear then: software companies are here to stay and will continue using their technologies to play an instrumental part in society and commerce. In a Y Combinator Startup School event at Stanford University, Andreesen then spoke on his lessons learnt from starting up Netscape (and how you can bake your cake too quickly):

  1. Hyper-growth – Andreesen highlights the risks of true hyper-growth, based on his experiences at Netscape: “The problem with with true hyper-growth… It’s the problem like baking a cake in three minutes. You’re in the kitchen and you have sugar, flour, egg on the ceiling. What the hell are you doing?” His main point being that a sudden flux of new employees (particularly when they all come from a specific company) can be really disruptive to an existing company culture.
  2. Management training – Management training is another thing which Andreesen thinks is important for fast-growing startups. “Many times, engineers are promoted to managers, but they have never been trained. Someone has to teach them how to do it” he feels.
  3. User centric – When Andreesen was asked how Netscape had managed to pull ahead of its competitors, he explained how he always prioritised his users. “When Netscape started there was no web analytics. Essentially all our feedback came from emails from people. Whenever push came to shove I’d always be on the side of the user,” he says.
  4. Don’t let your cake explode! I guess Andreesen startup and cake analogy all centres on his experience that startups can grow too quickly and not necessarily in the right direction or in an appropriate manner.
Main learning point: people like Marc Andreesen have been there and done it. Its’ interesting to hear their thoughts on current industry trends and to get an honest view of their lessons learnt. With Andreesen having been on both sides of the fence, founding a startup as well as investing in them, he knows what to do and what not do when baking a cake.

Cadillac and the promise of its new infotainment system

Not that I ever would be able to afford one (or be able to properly reverse park it), I got very excited about driving a Cadillac this week. Not because of its great looks, slick interior or amazing rims, but purely since Cadillac just announced Cadillac User Experience (CUE).

These are the main things worth looking out for in CUE:

  1. Voice recognition – Using technology developed by Nuance  who are also behind the Siri voice recognition of the new iPhone 4s, enabling users to interact with their phone and car stereo without ever having to take their eyes off the road.
  2. Backed by Linux and ARM – CUE is based on a Linux operating system and a chip by ARM who seem the chip processor of choice for most smartphones these days.
  3. Connected applications – Utilising the technical setup as mentioned in the previous point, many types of connected applications like those on mobile phones will be integrated into CUE to some degree. For example, CUE will be able to read out text messages sent to the driver and it will enable the useof Internet radio and other online services through its system.
  4. Give drivers a choice –  In addition to its work on the back-end operating system, Cadillac is looking to redo the instrument panel of its cars by giving drivers more choice in what they see. They can make it look like today’s cars or can customise it with any number of features to give them just the information they seek.

Main learning point: it will be interesting to see how Cadillac will go about implementing CUE in its cars and to find how drivers will adapt to the new touch screen, voice recognition and interconnecting technologies that come with this new system. I can easily see myself peaking through the shop window of a Cadillac dealership in the future, trying to catch a glimpse of CUE in action!

Related links for further learning: