It’s just data

Exclusive Content

S. Somasegar: We have signed an agreement to partner with NBC Universal to build a Silverlight 2.0 based web broadcast of the 2008 Summer Olympic games.  This agreement also sets MSN as the official home of the 2008 Summer Olympics.  As a part of this, we will provide users with exclusive access to over 3000 hours of live and on-demand video content via Silverlight streaming.  This means that viewers can access every minute of every event.  Additionally, the amount of meta-data attached to each of the streams will be extensive and include links to player bios, medal counts, shortcuts to particular events (i.e. athlete x’s third long-jump attempt), maps of the Olympic facilities, pop-up overlays with real-time event alerts, headlines, video search capabilities, etc. 

Shame on NBC.

Posted by Mark Baker at

w00t! A sport event with some Big Binary Blob (or something like) with lots of pops and stuff as "website"! Now they managed to catch may atention!

Posted by Eduardo Padoan at

@Mark Baker,

Shame on NBC for what?  Doing the “unthinkable” and signing an agreement that allows the “enemy” to profit by providing a useful service that the rest of us can benefit from?  If this was some other company that you were a fan of would the partnership be okay then?  Should NBC just open up the raw streams and let anybody who wants to edit them do just that and rebroadcast the result?  If yes, who’s going to pay for the bandwidth for the raw streams?  NBC, out of the goodness of their heart?

Don’t get me wrong.  I’m all for open content.  But content such as that of the Olympic games costs a significant chunk of change to produce.  Making a profit off the content they create is not evil.  So what exactly should NBC be ashamed of?

@Eduardo Padoan,

If you have no desire to watch it, then don’t.  But we’re well on our way to a world where interactive content-on-demand streamed over the Internet is not only possible, but the norm.  Would it be better if we instead sat on the couch, drank beer, and let the TV producers determine what is interesting content and what is not?

Come on, Man, evolve already!

Posted by M. David Peterson at

providing a useful service that the rest of us can benefit from

Define “us”.  Does it include me?

Posted by Sam Ruby at

@Sam,

After posting my last comment I noticed the Moonlight image and realized what you were inferring.  Looks like I need to research this a bit more, but from the looks of it it seems you are suggesting that this same content will not be available to those running Moonlight.  Is this correct?

/me is digging deeper...

Posted by M. David Peterson at

it seems you are suggesting that this same content will not be available to those running Moonlight.

The announcement doesn’t say...

Posted by Sam Ruby at

The Olympic Games Streaming To A Linux Distro Near You?

Sam Ruby: Exclusive Content As per the above link, Sam Ruby recently pointed to an announcement from S. Somasegar which stated, We have signed an agreement to partner with NBC Universal to build a Silverlight 2.0 based web broadcast of......

Excerpt from O'Reilly Windows DevCenter Blog at

If you have no desire to watch it, then don’t.  But we’re well on our way to a world where interactive content-on-demand streamed over the Internet is not only possible, but the norm.  Would it be better if we instead sat on the couch, drank beer, and let the TV producers determine what is interesting content and what is not?

I never said I’m against “interactive content-on-demand streamed over the Internet”.

Posted by Eduardo Padoan at

Sam,

Moonlight will render your Silverlight 2.0 content (including the Olympics content), every Linux user (provided they run a distro that is not four years old) will be able to use Moonlight and packages will be made available for all distributions. 

What seems to be the problem?

You can test a snapshot from a few weeks ago (it is in heavy flux as we are redoing our media pipeline, so a few things might not work right now):

[link]

And if someone wants to help us improve Moonlight, you can track our QA team’s log:

[link]

Miguel.

Posted by Miguel de Icaza at

What seems to be the problem?

I see potential problems on three axises.

For starters, there is the supported platform issue.  Nobody can predict the future, and there is certainly no guarantee that NBC will see a problem-free rollout on Silverlight/Vista let alone Moonlight/SUSE, other Linux distributions, or Silverlight on IntelMac.  It gets considerably more complicated when discussions expand to include platforms such as the iPhone, which is both popular and has its own set of limitations imposed by the vendor to overlay over this mix.  This part will be interesting to watch.

Second, Silverlight is a vendor controlled standard.  Having two independent and interoperable implementations, one of which is open source, mitigates this somewhat.

Finally, I happen to be fond of the view-source-able and grease-monkey-able and hyperlink-able web.  I view this experiment with the olympics as an exercise in demand generation for RIA, which — outside of YouTube — have made little inroads on the web.

It is the third issue that concerns me the most.

you can track our QA team’s log:

On the subject of QA and standards: this could be improved.

Posted by Sam Ruby at

Having two independent and interoperable implementations, one of which is open source, mitigates this somewhat.

My understanding so far is that the Silverlight video codec is patent-encumbered and only available for Moonlight as a non-free binary blob provided by MSFT.

Some values of two are a lot smaller than others.

Posted by Aristotle Pagaltzis at

Hello Sam,

On the first issue, I agree that this will be an important test of the installer experience;  The user experience bar is high, and we hope that we will be able to match it in terms of usability when it comes to the Linux scenario.

This is what we currently have in mind: [link] for the download experience, am sure it can be improved, and welcome folks' feedback.

On the second issue, I like to think that having two implementations will make for a better standard if such a standard is ever created.  We would be interested in participating on the standardization of Silverlight if other industry players are interested in participating (am speaking for Novell here).

On the third issue, let me break it up in pieces:

On view-source, I love view source, but get the feeling that view-source was very important on the web because layout was hard, the standards and the incompatibilities across the browsers were hard, and plenty of the special effects came from people that had been able to master what was a hostile programming environment (emphasis on was before I get flamed).

This is not to say that View-Source is not useful, just that it was probably more useful when documentation was poor and information scarce or too dense (the O’Reilly DHTML book being a thorough but intimidating book for example). 

That being said, it should be easy to add a “View source” button to Moonlight on right-click, which will just invoke Reflector on the assemblies downloaded.  For stuff written using dynamic languages like IronPython and IronRuby, you do get the source code instead of getting compiled assemblies.

On greasemonkeybility, you can invoke managed code from the browser Javascript already, so this should not be a problem. 

Mashing up stuff will continue to be better if you have a supported interface that is provided by the distributor rather than depending on good luck.

I agree with your assessment on RIAs penetration on the web.  Although I feel that Silverlight is more “webby” than Flash is.  The  user interface is defined in XML and its very easy to generate, maybe adding a JSon interface in addition to XAML is something worth pursuing (Chris Toshok wrote a JSon to XAML script, so you can write your UIs using declarative JSon instead of XAML, but embedding this into SL itself would probably be a step forward).

Miguel.

Posted by Miguel de Icaza at

Add your comment