[LINK] Forum on high speed bandwidth in Australia - this Thursday

Paul Brooks pbrooks-link at layer10.com.au
Mon May 4 16:35:51 AEST 2009


Karl Auer wrote:
> On Mon, 2009-05-04 at 12:15 +1000, Paul Brooks wrote:
>   
>> Speed is also misused. Most information travels at the same speed -
>> the speed of light in a vacuum, 300 metres per microsecond in air for
>> radio and microwave links, about 200 metres per microsecond in optical
>> fibre, and somewhat less in metallic cables.
>> What most people look for is capacity, or volume, and the speed is
>> taken for granted these days.
>>     
>
> Well - no, it isn't! I thought I just finished explaining exactly why
> not.
>
> No gamer takes speed for granted (meaning latency, NOT volume). No-one
> browsing the web takes it for granted either. The point I was trying to
> make is that the number of applications where volume is relevant is
> actually quite small.
>   
Karl - I think you missed my point - I was being pedantic about the use 
of the word 'speed'.

I agree with you that for most people the number of applications where 
the data transfer capacity is the limit is quite small. Less when you 
consider that the bottleneck is often not the user's access link 
capacity any more, but some bottleneck in the middle of the cloud the 
user has no control over and for which a change of the user's access 
link technology would make no improvement at all - such is congestion in 
the middle of the Internet, or an explicit rate-limit applied at the 
information source to steady the burstiness of the other end.

I was merely pointing out that information, whether transmitted by 
optical or radio means - regardless of the capacity of a link measured 
in kb/sec or Mb/sec - travels at the same speed  - the speed of light in 
that medium - no slower, no faster.

Gamers do take speed for granted, but I agree with you they don't take 
time for granted - latency is a time measure, and gamers look to 
minimise latency - by choosing topologically closer game servers to 
minimise time delays, not by trying to choose a telecoms network that 
somehow breaks the universal laws of nature by getting information 
across the same distance in a smaller time interval.

> but for most of the population, most of the time,
> it is interactive response that they actually notice and care about.
>   
On this we agree - and probably on much more, if we could agree on 
terminology.

P.




More information about the Link mailing list