W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1995

RE: Suggestion: Partial file transfer.

From: Paul Leach <paulle@microsoft.com>
Date: Thu, 24 Aug 95 13:50:09 PDT
Message-Id: <9508242139.AA00662@netmail2.microsoft.com>
To: dror@vocaltec.com, http-wg-request%cuckoo.hpl.hp.com@hplb.hpl.hp.com, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
I like the notion of partial entity transfer. However, I would prefer 
that content-length always refer to the transmitted length of the 
content, for all requests.  Siince you're intorducing new header 
fields, why not add one for the total-length of the entity?

The proposal allows fetch of the tail of a file.  If we go down this 
path, why not generalize it to an arbitrary slice?

Is this makes caches more efficient, why forbid its use by cache servers?

Paul
----------
] From: Dror Tirosh  <dror@vocaltec.com>
] To:  <netmail!http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com>;  
<dror@vocaltec.com>
] Subject: Suggestion: Partial file transfer.
] Date: Saturday, August 24, 1901 10:37PM
]
] A suggestion for a new Header field to support partial file transfer. 
In the normal
] use of HTTP with HTML, many (most?) transfer activities are aborted. 
This makes a
] cache much less efficient, as it have to re-start the file transfer from its
] beginning.

[ excised stuff]
Received on Thursday, 24 August 1995 13:55:18 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:26 EDT