Re: AW: AW: URL: Background and Requirements

From: Warner ten Kate (
Date: Mon, Nov 23 1998

Message-Id: <>
Date: Mon, 23 Nov 1998 18:33:14 +0100
From: Warner ten Kate <>
To: Henning Timcke <>
Cc: www-tv <>
Subject: Re: AW: AW: URL: Background and Requirements

Henning Timcke wrote:
> What is your answer to the URL's on tape ?

I have some difficulty in understanding your questions. 
I also wonder where URIs come into play. If you have 
proposals, please, come forward.

> Okay, please tell me more:
> store locally on tape ?
> How will the extra-information be stored that tells at what time on the tape the URL is ?
> Do you intend to use diffrent methods as in HTTP ?

When saying 'what time on the tape' are you referring 
to the physical position on that tape ?

I suppose the same concepts and mechanisms can be used as 
when referencing content on a tape or in a Broadcast stream, 
Particularly, when the tape uses the broadcast format as 
"file system" (e.g., MPEG TS).

It is up to the tape device design how to arrange (fast) access 
across a tape. Possibly, the tape is arranged with additional 
information to assist the driver. Anyhow, that seems to me 
outside the scope of the URI. To keep this focus I think it 
helps if we talk in terms of storage devices.

> I propose to use XML to handle the hooks neccessary.

I am not sure I understand what you suggest here.
What hooks ? And how would you use XML ?

I assume you mean to use XML for specifying the application 
language. That also seems outside our current scope of 
discussing URIs.

> Time Code could be used as is and another *Time* Code should be developed:
> topology relevant time for download.

I do not understand this phrase:
What 'Time code' are you referring to?
For what and how could it be used 'as is' ?
What is the other ' *Time* Code '
What do you mean by 'topology relevant time' ?
What topology are you referring to ? (position on tape ?)

Before answering all this, can you think of what the consequences
are for URIs: are we missing a requirement ?