[LINK] ACS new? submission on ISP filtering

[Roger Clarke]

>>  At 9:42 +1030 19/3/10, Glen Turner wrote:
>>>  ISPs can save significant bandwidth by hosting or peering
>>>  with content distribution networks for those huge static
>>>  items which do consume a lot of bandwidth.

>On 19/03/2010, at 11:15 AM, Roger Clarke wrote:
>>  So the http HEAD method doesn't work for such things?
>>  (We're talking specifically of 'static' items).

At 11:23 +1100 19/3/10, Stilgherrian wrote:
>I'd say that doing an HTTP HEAD to see if the content has changed, 
>then a GET to fetch the content if it has, triples the latency time 
>before the content actually gets fetched. I don't think I'd like to 
>do that.

*You* (or I) might like to do that.  But the decision is the ISP's, 
and they could well choose to trade off latency against cost.

(And the HEAD and the response should each fit into a single 
IP-datagram, and the overheads of TCP are the same either way - 
assuming the keep-open option is set on, which I'd very much expect 
to be the case;  so I suspect that "triples the latency" is a serious 
over-estimate.  No, I don't have a simulator handy to experiment 
with, but presumably someone has).


-- 
Roger Clarke                                 http://www.rogerclarke.com/

Xamax Consultancy Pty Ltd      78 Sidaway St, Chapman ACT 2611 AUSTRALIA
                    Tel: +61 2 6288 1472, and 6288 6916
mailto:Roger.Clarke at xamax.com.au                http://www.xamax.com.au/

Visiting Professor in the Cyberspace Law & Policy Centre      Uni of NSW
Visiting Professor in Computer Science    Australian National University