Thursday, 18 February 2010

Proof SeeSaw is better than iPlayer, Blinkbox, ITV Player, MSN Player, 4OD

I have now got scientific objective proof that SeeSaw is better than iPlayer, Hulu, Blinkbox, ITV Player, MSN, 4OD!

How?  you may well ask - the 'proof' is in the screen shots below - look at the volume controls!

iPlayer
iPlayer takes an early lead with the volume going up to 11..


Blinkbox


Blinkbox enter the race with a dull volume control - that doesn't event get over 10! It does have some cute animation effects though!
ITV Player
ITV appear to be a different race - they don't have any numbers! But they do have bars that expand around the volume icon and I count 4!

MSN
MSN have stuck to the slider with 3 bars that appear as the user moves it. It is interesting to notice they are using Flash which is a bit odd for the company behind Silverlight....

SeeSaw

What else is there to say about SeeSaw the volume goes all the way up to 12.
4OD
4OD have joined ITV in not showing a number putting them at an immediate disadvantage. Counting the bars on the icon I get 2 - putting them in last place - sorry guys - but this is an objective review and 4OD deserves better than coming last. A lot of people have forgotten how it BEAT the iPlayer to be the first major VOD service in the UK!

So lets recap

iPlayer - 11
Blinkbox - 10
ITV Player - 4
MSN - 3
SeeSaw - 12
4OD - 2

So there we have it the best Video On Demand service in the UK is SeeSaw as 'the volume goes all the way up to 12 - one more than iPlayer'....

Wednesday, 17 February 2010

SeeSaw is open

SeeSaw.com is now live and open to the public!

As of 6.00am Tuesday 16th February 2010 (yes we waited a bit before announcing it) SeeSaw.com is live and open to the public - no more beta logins! Now it is up to the public to see if they like it!

I have been working on this since July 2007 when I joined ioko and started working on Project Kangaroo. Over the last two and half year we have had a lot of ups and downs and it is good to finally see this site open to the public. For those who are interested a rough timeline is


  • Jan 2007 Project Kangaroo idea conceived and early discussions begin
  • July 2007 Project team assembled
  • Sep 2007 Development team assembled - target completion date Q1 - 2008
  • May 2008 - Change of technology - moved away from Kontiki to Flash streaming
  • Q1 2008 - Venture voluntarily referred themselves to the OFT. Unable to launch until a decision is made so more is developed. 
  • June 2008 - OFT decide to refer venture to competition commission. Development continues refining UI, upgrading from Struts to Tapestry 5. 
  • December 2008 - Closed Alpha begins - project team planning on being ready to start beta the day after Competition Commission judgement.
  • February 2009 - Competition commission declare venture anti-competitve and it all shuts down
  • February - July 2009 - Project Kangaroo's technology was sold to Arqiva
  • September - December 2009 - A UI revamp occurs to the Kangaroo design to make it simpler and remove some 'design by committee' elements
  • December 2009 - SeeSaw.com Alpha begins
  • January 2010 - First public beta invites issued
  • 16 February 2010 - Open beta begins


What's next - well thats for people from the business side of SeeSaw to reveal but I think we are going to be busy delivering at least a some of the ideas that people have suggested!

Thursday, 28 January 2010

Apple iPad and Content

The new Apple iPad has certainly got people attention - whether it will be a success or not is still to be seen. Unfortunately what is undeniable is Apple are not going to let 3rd parties deliver quality content onto it - further locking users into iTunes.

The iTunes story is one where Apple managed to wrest music from the pirates and turn it into a legitimate revenue stream for the music companies. They have managed to make legal downloads easier than piracy and for this they deserve a lot of credit. Their brave move to remove DRM have allowed users to once again 'own' their music - though I doubt they would have done it if other MP3 stores hadn't opened!

With video content however the story is very different. In video the content owners are still (rightly or otherwise) wedded to DRM to protect their content and this current boils down to 3 systems - Microsoft's Playready, Adobe's Flash RTMPE and Apple's Fairplay. There are some smaller players (and subtleties over versions) but these are the ones that are widely accepted by content owners. As it stands on PC and Mac all 3 work on all 3 platforms. However on devices the story is very different. The iPhone and iPad will only support Apple's Fairplay. Other devices will vary but often support limited versions of Microsoft's and Adobe's DRM.

This matters a lot as a consumer as it makes it very hard for me to get (legally) content on to a device. It matters even more as someone who works in the business of delivering VOD solutions. It is right now not possible for us to deliver content to iPod or iPhone and protect it with DRM. It can be argued that we should do it without DRM (like music) but that isn't going to happen quickly (if at all) so for now to get good content we need DRM.

Apple's position as a major video store and controlling the iPod device gives them immense power. They will not license fairplay to anyone else so nobody can deliver DRM content to devices that make a huge proportion of the market.

The iPad makes this position even worse - up til now we have at least had the option of Flash or Playready (via Silverlight) to deliver to every 'PC' type device. Now we have a 'big screen' device no-one but Apple can operate a quality content video store for. This isn't going to be good for the market. It may even be an abuse of a monopoly - they are abusing their strong position in the music and portable device market to lock others out of the video market.

What I actually think will happen is (over time) apple will actually struggle to keep up with the rest of the industry. There are plenty of more open devices (e.g Android, Symbian, Windows Mobile and many more) in the pipeline that will be cheaper than Apple's offerings and begin to get better (and cheaper) content. In the long run the video market will be driven by content, onto cheap commodity technology, but in the short run Apple will make it hard for the rest of us to compete!

As an aside I often hear - but iPlayer does it. They do deliver to PS3, Wii, iPhone and dozens of other devices. The way they do this is by cheating and not using DRM. Due to 'the unique way it is funded' the BBC can force rights holders to release content without requiring DRM. This means there are tonnes (just google it) of methods of downloading iPlayer content without DRM. This is good for the consumer in the short run - but it is bad in the long run as it is a huge disincentive for anyone to invest money to compete with this stream of free, DRM free content!

Wednesday, 27 January 2010

SeeSaw and Technology

The SeeSaw beta is now open so I thought it would be interesting to mention what powers it.

There are a few things I would single out
Tapestry 5
The web application is written in the Tapestry 5 framework - this is a relatively rare choice as it is quite new - it was in beta when we started developing (we actually migrated from Struts mid way).
I think this is one of the choices that has paid off for us substantially. That's not to say there hasn't been pain along the way (as I sure some of the team who had to learn it will agree) but I think overall it has paid off.

The key win features in my mind are

  • Components - it is hard to underestimate how powerful component based pages are - we can actually re-use large chunks of code with ease across the site
  • HTML templates - having templates that are basically HTML makes it so much easier to bridge the gap between the code world and the web development world
  • Javascript - now this would be the most debatable one - as I know some of our team don't like the way T5 handles Javascript - but the effortless way we handle AJAX via zones I think is a real killer feature

I won't say much more about this - as I'm saving it for 'In the brain of Ben Gidley - Tapestry In Action for real' a talk I'm doing on the subject of Tapestry 5 on March 23rd 2010 at Skillsmatter.

Jetty
The SeeSaw application runs entirely on Jetty. We don't need (or want) one of the more fully featured application servers. Jetty is stable, easy to run, manageable, lightweight, cheap and most of all just works. We develop on it, test on it and run on it. Lots of shops develop on lightweight servers like Jetty/Tomcat and then deploy on something heavier like Websphere or Weblogic. We just don't need it.

We also get top class production support for our production Jetty's from Webtide. When we have an issue (which is rare) we know we will get immediate support. They are the only company I know that operates support on a de-escalation basis - you get the senior people first who then hand off to junior people if it doesn't need their expertise!

Hibernate
Hibernate is like an old friend. I remember back when Hibernate was just reaching version 2 debating with people if this new fangled Object Relational stuff would ever take off. Now it seems to be the obvious choice for accessing a database from Java.
Hibernate has made it easy for our developers to focus on writing business code (e.g pages/components) and not worry too much about the database. It also performs well, is very easy to configure to cache our content and generally 'just works'.

Zeus ZXTM
Caching is a BIG issue when building a site that was designed to scale like SeeSaw is - when it was Project Kangaroo it was aiming to be the no. 2 destination in the UK - the new owners are also quite ambitious!
The site therefore relies on a range of caching to perform under load. The most valuable of these is the Zeus ZXTM cache. The best thing to say about it is - we can't break it. Seriously we max out our network links before it breaks. It doesn't degrade, it just keeps going.
The ZXTM doesn't just cache - it also handles SSL, GZIP and load balancing across our application servers. I think without it (or something similar) we would need a lot more application servers that we have.

Tuesday, 26 January 2010

SeeSaw beta is open

So the beta invites are finally out for SeeSaw.com (well the first batch anyhow) - so people can now see what I have spent the last 2.5 years doing!

The SeeSaw (then Project Kangaroo) journey began in July 2007 when I left Macrovision and joined ioko to work on this new joint venture between BBC, ITV and C4. It has been a long road with the lowlight being last February when the competition commission banned the venture from proceeding. We had been ready to start a beta the day it was announced - as the whole team were expecting to be allowed to proceed!

Several months later the code was picked up by Arqiva to be remade into the SeeSaw that opens today. This involved quite a few changes including a complete UI redesign (I must say I like this one better!).

So finally nearly 3 years later we are inviting people in to see the 'future of TV' - now we just have to see how popular it will be. I hope all the team who have help make SeeSaw a reality (not forgetting all the Kangaroo people) enjoy the site and take some credit!

Wednesday, 11 November 2009

IE 7 and dom:loaded

We recently hit a very annoying bug/feature that we could only see in IE7. The page works fine normally but in IE 7 we got javascript errors saying objects were not defined.

The page looks something like

<head>
<script src="ourcorescripts.js/>
</head>
<body>
// Some stuff
<script>
document.observe("dom:loaded") = function(){
// do stuff
}
</script>
</body>


What happens is in IE7 (sometimes) the event handler fires and objects defined in 'ourframework.js' are not ready yet. A similar problem was noted by Atlassian - their solution was to use window:load. This didn't really work for us as it is a bit too late as we want some of our logic to load sooner.

What seems to be happening is in IE 7 it is firing dom:loaded before it has finished parsing all of the JavaScript files. This seems odd as it doesn't happen in any other browser. Looking into this further it comes to light dom:loaded is not actually an IE 7 feature but is simulated by prototype. Prototype uses
code that registered a 'deferred' script to fire the event. For some reason in IE 7 this doesn't seem to always work.

Our solution was more low tech. We simply fired another event my:dom:loaded from the footer and listened to that instead. This solved the problem and seems to fire nearly as soon as dom:loaded.

Thursday, 5 November 2009

Social Network network....


I have been slowly linking together all the social sites that I use and then surprised myself by how linked up they are. So i drew myself a diagram
So far it seems the best place to post new stuff is either Brizzly or my Blog. That way it gets everywhere!

I think I will see how many more services I can get linked in!