[Developers] Semantic TV Guide

[Robert Cook]

Sadly, this appears to be intended for end users (I assume people  
running MythTV, for example).  Here's a key part of the subscription  
agreement:

> Without limiting the generality of Section 1, You shall not, without  
> Company's prior written consent, which consent may be granted or  
> denied in Company's sole discretion, (i) redistribute or republish  
> Licensed Data in any media now known or hereafter developed, in  
> original or in modified form (including, without limitation,  
> translations, changes in formatting, alterations to meta-data  
> content, or variations in the arrangement of Licensed Data); (ii)  
> assign, rent, lease, grant a security interest in, re-license,  
> sublicense, or otherwise transfer any rights to Licensed Data; (iii)  
> remove or alter any trademark, logo, cop...
...and on and on...

The people running this site are certainly doing a service, though --  
they are at least making the data available to a mass audience and  
they are
doing what they can to drive down the costs.

Even if this data were freely available under an open license, it may  
not be the best thing to put in Freebase.  Program listings are  
ephemeral, whereas Freebase data should have continuing value (yes, I  
know one could argue that people may one day want to know how many  
times an episode of "Mad About You" played in syndication in Manitoba  
during 2007, but that's of much lower utility than 97% of the things  
that could be in Freebase.)

This suggests an application where a live data stream can be blended  
with Freebase data on the fly.  Were this data set to have consistent  
keys for television programs (even in the form "Battlestar Galactica  
2003"), then such an application could be developed trivially.

R

On Jan 23, 2008, at 9:51 AM, Toby Segaran wrote:

> Have you taken a look at this?
> http://www.schedulesdirect.org/
>
> Just came across it, so I haven't really dived in, but it might be  
> what you're looking for.
>
> It seems odd to me that the networks would try to protect this  
> information, since having it out there as much as possible could  
> only increase their viewership.
>
> On Jan 22, 2008, at 7:26 PM, Daniel E. Renfer wrote:
>
>> In keeping with the "Anyone Stuck" thread, I would like to suggest  
>> an idea for a freebase-backed application that I have been thinking  
>> about, but have not even attempted to model or build yet.
>>
>> We already have a good deal of information recorded about different  
>> TV series, their episodes, and the cast, crew, and fictional  
>> characters and locations associated with them.
>>
>> We have information about the companies that comprise the studios,  
>> stations, channels, and cable/satellites companies.
>>
>> What we don't have yet is information about these cable companies'  
>> markets, packages, and channel lineups. Once that information is  
>> modeled, we would need information about upcoming programming  
>> information.
>>
>> I did some searching, but was not able to find any sources of  
>> publicly licensed schedule information to be mined. Assuming there  
>> was a place we could pull this info from and dump into freebase, I  
>> think a semantic TV guide application would be a great thing.
>>
>> Imagine being able to specify for instance that I have access to  
>> Comcast's Full Basic package in the Ypsilanti, MI market. Given  
>> those parameters, at what time, and on what channel would I next be  
>> able to find a program in which Tim Allen appears?
>>
>> I don't know how feasible this would be. It would most likely  
>> result in millions of records being produced (CVT's?) that  
>> admittably would be of limited interest even when their dates lie  
>> in the future, and less once they have passed. (except in  
>> aggregated for for statistics)
>>
>> Would the Freebase devs even allow their framework to be used for  
>> such a purpose? Are there any limitations to this that I'm not  
>> considering? Does anyone know of a CC-compatible TV schedule  
>> service that this type of data could be pulled from? Any other  
>> comments/criticisms?
>>
>> Daniel E. Renfer
>> http://kronkltd.net/
>>
>> _______________________________________________
>> Developers mailing list
>> Developers at freebase.com
>> http://lists.freebase.com/mailman/listinfo/developers
>
> _______________________________________________
> Developers mailing list
> Developers at freebase.com
> http://lists.freebase.com/mailman/listinfo/developers