Re: Idempotent partial updates

On Mon, Feb 27, 2012 at 8:18 AM, Mike Kelly <mikekelly321@gmail.com> wrote:
> Hi all,
> - In light of the recent proliferation of 'mobile clients' which
> operate on relatively slow and unreliable networks where partial
> idempotent updates would be ideal; is the loss of idempotent partial
> updates considered acceptable?

How would partial PUT work? Client cuts the whole entity into pieces,
and PUT each one individually? That is not atomic - what happens to a
GET request before the whole thing is finished?

I don't think PATCH addresses OP's concern, that is, reducing
retransmission cost in face of network failure (which should really be
TCP's job).

Zhong