W3C home > Mailing lists > Public > public-sysapps@w3.org > May 2013

RE: [sysapps/raw socket api]:Proposal for resolution of remaining issues.

From: Nilsson, Claes1 <Claes1.Nilsson@sonymobile.com>
Date: Wed, 8 May 2013 15:12:34 +0200
To: 'Jonas Sicking' <jonas@sicking.cc>
CC: "public-sysapps@w3.org" <public-sysapps@w3.org>, "Isberg, Anders" <Anders.Isberg@sonymobile.com>, "Edenbrandt, Anders" <Anders.Edenbrandt@sonymobile.com>, "Isaksson, Björn" <Bjorn.Isaksson@sonymobile.com>
Message-ID: <6DFA1B20D858A14488A66D6EEDF26AA3010F4E5ACD17@seldmbx03.corpusers.net>
A quick answer listing some use cases.

·        A mobile device is acting as wifi hotspot. This means that both wifi and the mobile network is active. If an app wants to communicate through UDP or TCP with a device in the wifi network or send a UDP multicast on wifi then the socket has to be bound to the wifi address. This has to be explicitly stated as the mobile network is the "default interface".

·        Similar as above for a desktop computer having a LAN connection but wants to communicate with TCP or UDP to a device (e.g. a media device) in the local wifi network.

·        An application running on a desktop computer that has two network cards wants to create a TCPServerSocket using one of the interfaces (which may not be the "default" one).

From: Jonas Sicking [mailto:jonas@sicking.cc]
Sent: den 7 maj 2013 18:45
To: Nilsson, Claes1
Cc: public-sysapps@w3.org<mailto:public-sysapps@w3.org>; Isberg, Anders; Edenbrandt, Anders; Isaksson, Björn
Subject: Re: [sysapps/raw socket api]:Proposal for resolution of remaining issues.

On Tue, May 7, 2013 at 8:08 AM, Nilsson, Claes1 <Claes1.Nilsson@sonymobile.com<mailto:Claes1.Nilsson@sonymobile.com>> wrote:
https://github.com/sysapps/raw-sockets/issues/11: Devices have more than one network interface. However, the issue is whether web applications should be able to select a specific local network interface to use for a socket or if always the "default interface"/the configured interface should be used. My view is that we should provide this possibility by an optional field in the constructor's options attribute. I must admit that I have difficulties in motivating this by tangible use cases but I haven't seen any existing TCP or UDP socket API that  does not provide the possibility to bind a socket to a local address. So there must be use cases and I propose that we keep this possibility in the specification. Objections?

I definitely think we need use cases for all features of the spec. The fact that all existing libraries support an outgoing IP number might be an effect of that they are commonly used on servers rather than just clients.
/ Jonas
Received on Wednesday, 8 May 2013 13:13:08 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 1 July 2021 16:04:42 UTC