[LINK] Broadband for a Broad Land

[Richard Chirgwin]

On 3/01/11 12:54 PM, Tom Worthington wrote:
> Frank O'Connor wrote:
>> At 10:59 AM +1100 28/12/10, Tom Worthington wrote:
>>> I started writing a submission for the NBN Inquiry:
>>> <http://blog.tomw.net.au/2010/12/broadband-for-broad-land.html>.
>> Mmm ... many Internet apps are Web based, and don't benefit from the
>> increase in bandwidth beyond the initial 12 Mbs ... The problem on the Web is latency ... and depending how the NBN
>> implements this the Web may or may not be a better experience. ...
> I assume that the NBN's optical network will cut out many hops between
> data source and the customer at home, reducing latency. What happens
> with the wireless sevice is anyone guess. Are there any specifications
> from NBN Co. about latency?
None about the wireless. I don't think wireless has been specified yet.

Most of the latency of Internet applications comes from router 
processing, doesn't it? The reason I mention this is that the NBN is a 
layer 2 network - it's not providing the router layer. I'm happy to be 
corrected on this by those more knowledegable, but: if (say) iiNet has 
ten hops between its Sydney data centre and San Francisco, the NBN won't 
affect this. It will deliver customer traffic to the iiNet data centre, 
which is where the traffic will enter iiNet's routing and pick up the 
same ten hops.

RC
>> I'd argue that education needs a new application/standard to supplement
>> the Web. Something built from the ground up to be interactive with the
>> student, responsive to their inputs and needs, and malleable enough and
>> easy enough to use for the teacher to add content and make the necessary
>> tweaks and changes to meet their students' needs. ...
> Building new approaches from the ground up is not cheap or easy. As an
> example, the Australian National University is spending $3.8M on its
> Engineering 'Hubs And Spokes' Project:
> <http://www.deewr.gov.au/Ministers/Gillard/Media/Releases/Pages/Article_081211_140416.aspx>.
>
> This is starting to produce useful results:
> <http://engnet.anu.edu.au/DEpeople/Kim.Blackmore/papers/Ascilite2010.pdf>.
>
> My approach has been to adapt existing educational approaches and
> technologies. That way I am more likely to get courses approved and have
> the resources to build them. With this I make the e-learning look as
> much like a traditional course as possible.
>
>> If the changes happen at the application level the amount of capital and
>> other investment costs by schools should be minimal. ...
> Schools have been designed on the assumption that students spend most of
> the time in classes. The schools will need fewer small classrooms and
> more large spaces. Universities have been built with more large lecture
> theatres than will be needed. More small rooms will be needed as well as
> "leanring commons". Remodelling these campuses will not be easy or cheap.
>
> The Victorian government had produced a useful set of publications on
>                               the process of mapping from pedagogical
> approaches to the design of learning spaces: "Victorian School Design":
> <http://www.education.vic.gov.au/management/infrastructure/schooldesign.htm>.
>
> Take the example of a school with 500 students, which is a bit more than
> the average of about 380 students per school in Australia:
> <http://www.abs.gov.au/AUSSTATS/abs@.nsf/Latestproducts/4221.0Main%20Features22009?opendocument&tabname=Summary&prodno=4221.0&issue=2009&num=&view=>.
>
>
> Assume an average class size of 25, to make the maths easier (the
> average is about 24
> <http://austcolled.com.au/notepad/article/class-size-vexed-question-or-huge-distraction>).
>
> So a school would require 20 classrooms each for 25 students, assuming
> the students spent most of their time in class. If the students now
> spend one fifth of their time in class, then only 4 classrooms will be
> needed. What will then be needed are other spaces for 400 students,
> perhaps a commons for 200, and other smaller spaces for 200.
>
>