Mythbusting the NBN - Peter Ferris

Discussion in 'Networking, Telephony & Internet' started by HyRax1, Aug 26, 2011.

  1. HyRax1

    HyRax1 ¡Viva la Resolutión!

    Joined:
    Jun 28, 2001
    Messages:
    7,812
    Location:
    At a desk
    I went to this information night about Mythbusting the NBN as part of Macquarie University's Engineering Information Week a month ago held at Cochlear Limited's new corporate headquarters in Sydney NSW. I was skeptical thinking that this would be a bit of a sales pitch but it ended up being one of the single-most informative presentations about the NBN I have ever seen.

    They have finally posted up the video (OK, they posted it up two weeks ago and I've only just checked for it again tonight).

    Prepare to have a large chunk of your questions answered and myths surrounding the NBN busted. There is some good general information as well as good, proper technical information. Peter Ferris is an excellent presenter and provides an brilliant and straight-to-the-point presentation without any hint of spin.

    The video is accompanied by an excellent slide presentation that has been edited into the video. They provide a concise overview of what the NBN is, how it works and how it is being deployed as well as addressing those "myths".



    Attendees were primarily engineering students and university faculty.

    Total viewing time is just under 57 minutes. Also available in 1080p from YouTube as well.
     
  2. Blinky

    Blinky Member

    Joined:
    Jul 4, 2001
    Messages:
    2,530
    Location:
    South TAS.
    First and foremost I think the NBN is a good thing, hopefully fixing a flaw which has existed for a long time.
    Now I'm only 12mins in to the video, but I think that by not forcing a RSPs to offer services across all access types is fundamentally flawed.
    All it does is perpetuate the same business models and issues we see in the current market. RSPs will pick the eyes out of the high value markets and abandon the rest. The other end of the market will survive by exploiting the government financial incentives, which in turn, (as they are only marginal markets) creates monopolised businesses; i.e. the current state of the retail satellite market.
    The proposed technology model already foreshadows an extension of the 'great technology divide' by not mandating a common access definition. They could have made a positive change for marginal users by (at least) defining an minimum exposure level to non-metro zones within a retailers SLA.

    So what's changed? The tech has changed, and as such, the numbers involved have changed too - shiny new bigger, faster, more (capacity, speed, la la la). The only other thing which it is designed to do, is right the wrong the Liberal Government caused by half privatising Telstra; yet only to the extent that it gets the government's finger back into the revenue pie.

    <possible future edit here> Before I get into the USO or lack there of, I had better watch more of the vid. I hope it gets better, I won't be able to cope if my $&*#*! Liberal voting Dad is right about the NBN. :Paranoid:
     
  3. Duideka

    Duideka Member

    Joined:
    Jun 21, 2009
    Messages:
    309
    Location:
    Perth
    But what about new ISP's who are starting? you can't seriously expect them to have national coverage from day 1.

    iiNet started in a kids garage ;)
     
  4. Blinky

    Blinky Member

    Joined:
    Jul 4, 2001
    Messages:
    2,530
    Location:
    South TAS.
    Fair point. I guess that why we have the tiering system eh :)
     
  5. caspian

    caspian Member

    Joined:
    Mar 11, 2002
    Messages:
    9,216
    Location:
    Melbourne
    yep, that's precisely what will happen - except this time there won't be Telstra there with a nice guaranteed wholesale available product when the RSPs can't afford to do it themselves.

    I predict Telstra will service 100% of the POIs, but there may be some where no other RSP does, but this time around Telstra doesn't have to resell their access - so you're stuck with them as a retailer too.
     
  6. Devils

    Devils (Taking a Break)

    Joined:
    Nov 10, 2009
    Messages:
    185
    Location:
    Brisbane
    Verizon in the US only puts fibre in markets where it will make money & has millions of connections.
    http://en.wikipedia.org/wiki/Verizon_FiOS
    "Verizon announced in March 2010 they were winding down their FiOS expansion, concentrating on completing their network in areas that already had FiOS franchises".

    They are winding down because you need the population density to make money.

    Here is Australia most of the network will be loss making as its rual. The projected return is 7% which is pathethic given the risk profile of the project. People with the NBN may well decide on lower cost plans. Its an engineering masterpiece but financial white elephant.
     
    Last edited: Aug 28, 2011
  7. martino

    martino Member

    Joined:
    Mar 8, 2005
    Messages:
    1,225
    It's not being built to generate profits - its being built so everyone can have access to decent reliable broadband.
     
  8. Blinky

    Blinky Member

    Joined:
    Jul 4, 2001
    Messages:
    2,530
    Location:
    South TAS.
    I still haven't got through downloading and watching it. Sigh, the joys of rural internet... If I could rewind I think I had some issue with myth four.

    Anyway, I'd like to take small issue with the comment above. I'm content that it doesn't make a profit, even though it must survive for a good many years before it gets it's head above water...
    It's the existence reasoning, in the second part I'm concerned with. If it's "being built so everyone can have access to decent reliable broadband" why is there such a mixed level of service?
    In plainer terms, why are the taxes of the people in the 7% coverage area only worth (up to) 12Mb/s yet the Metro folk's tax dollar is worth (up to) 100Mb/s ?
     
  9. martino

    martino Member

    Joined:
    Mar 8, 2005
    Messages:
    1,225
    It's that last regional/rural/remote 7% that would cost the most to service. Running fibre to individual premises that are kilometers apart would run into tens of thousands per dwelling.

    I think the rule of thumb being applied is towns of 1000+ will see FTTP. Those near a fibre service area but just out of reach may be able to put money in to be included also.

    It would be rough missing out, but the cut off has to be somewhere - it's just not feasible to fibre 100% of premises.
     
  10. caspian

    caspian Member

    Joined:
    Mar 11, 2002
    Messages:
    9,216
    Location:
    Melbourne
    that's what it comes down to.

    Telstra wanted 18% for a similar solution and the howls of indignation were heard far and wide.

    7% covers what the money could earn invested otherwise. risk? there are literally hundreds of very, very smart people projecting and sampling what the uptake rate and profiles will be across both NBNCo and government departments, I think they know what to expect.
     
  11. Auriga

    Auriga Member

    Joined:
    Jun 28, 2001
    Messages:
    1,196
    Location:
    Warnbro, WA
    Very informative and non political. As a communications technician myself, it's very obviously a hierarchical theory based presentation for an engineering audience.

    Not a lot of nitty gritty on installation and maintenance side of things. Or CPE or possible setups across business and residential. Although there was talk about locating the aerial fiber close to LV power corridors on poles and not requiring internal bearers in the cable. As well as mentioning not installing any poles. I find this hard to believe as there is a large number of Telstra aerial poles out there separate from power utilities.
     
  12. caspian

    caspian Member

    Joined:
    Mar 11, 2002
    Messages:
    9,216
    Location:
    Melbourne
    at the moment there is only one installation setup - nothing discrete for business use. that will come later, with the release of B-ONTs with more ethernet interfaces and no voice etc.

    there is no internal cable caternary bearer for the aerial deployment. it is being done with ADSS cabling that is light enough to be self supporting, along with the obvious benefit of having to metallics.

    NBNCo will install zero poles. the deployment follows Telstra's CAN (apart from the first release sites which had to be built pre-Telstra agreement). if Telstra's CAN is aerial then so will be the NBN, otherwise it will be underground pit and pipe.
     
  13. Blinky

    Blinky Member

    Joined:
    Jul 4, 2001
    Messages:
    2,530
    Location:
    South TAS.
    Yes agreed and / or I know, to all points.

    But you've haven't explained why the divide is so great.
    I don't mean from a technical POV, I'm referring to a social one.
    If any bias was to be given, it should be the other way around. A location with poor access to services should have better telecommunication access as a physical ones are lacking.
    The question is, how do you justify that you are carving up a per capita unit of public funds but returning a service which is disproportionate. It would be fairer (in a public money sense) to deploy a lesser class of service with a corresponding lower cost, to a larger volume of users.
     
  14. caspian

    caspian Member

    Joined:
    Mar 11, 2002
    Messages:
    9,216
    Location:
    Melbourne
    a social one? how would you define such a cutoff?

    should they also have half a dozen takeaway food stores, a major hospital, an international airport and an extensive public transport system?

    of course not. infrastructure gets provided to serve need, and that's a function of population, uptake and cost. in the case of the NBN, there's a cost level to provide fibre access to a locality, which needs an acceptable uptake rate to justify it without the price/service going to unjustifiable levels. the cutoff point chosen was 1000 inhabitants. there has to be a cutoff point sooner or later, and it has to be able to defined.

    so we should service the majority of users to a lesser extent, and the minority to a greater extent? how is that fair, equitable, or even sensible? or even possible, for that matter?
     
  15. martino

    martino Member

    Joined:
    Mar 8, 2005
    Messages:
    1,225
    So by that standard, it wouldn't be fibre to the premises for anyone then - because if the last 7% miss out, so should everyone else.

    I have no idea what a "lesser class of service" applicable to 100% of the population would be - a good chunk of that 7% wouldn't have copper currently. Radio phones and wireless-based POTS are deployed alternatives for example. Satellite is another.

    Regardless, the problem of city/country divide you bring up is not one constrained to broadband delivery . It's the same issue across all services.

    Lastly, the taxpayer is not footing the bill for the bulk of the NBN build, they're simply acting as guarantor on the debt.
     
  16. Blinky

    Blinky Member

    Joined:
    Jul 4, 2001
    Messages:
    2,530
    Location:
    South TAS.
    Yes and no, everybody to the same level!

    If it means that the bar is dropped for the city folk then lower it, make the NBN standard 12MB/s for everybody. They sold the concept of the NBN on (in part) that it would remedy the poor rural service, but they are not achieving the goal they in part set out to, all they have done is moved the goal posts.

    If the 'bush' won't be disadvantaged by the lower speeds then the city won't need 100MB/s or even a bare basic 25MB/s, cuz all the apps and services will be designed to target 12MB/s the lowest common denominator right? BS, it will be the same situation as we have now wrapped up with a different flavour coating.
    You stop and consider the negative ROI in at least the first half of the project's lifetime, any other project wouldn't proceed in the currently proposed form. Something like FTTN would be implemented, which is more affordable in the short term and yet doesn't hamper the upgrade path should FTTH become more viable in the future. It was an interesting video nevertheless, even if I had to use a download manager to get it because it wouldn't stream without issue from youtube during peak times. :(

    Look, I don't wish to take away your faster internet, god knows I'd sell my granny if she were still alive to get back on to HFC again, but it's not clean water we are taking about and it's not fair for everybody concerned; which is why it has got as far as it has anyway. You could after all, do more than survive on 12MB/s - really.
     
  17. kabab

    kabab Member

    Joined:
    Dec 26, 2001
    Messages:
    2,466
    Location:
    Sydney
    I dunno i think its reasonable it's just not financially viable to give everyone the same connection..

    It just comes with the territory, its like living in the city and demanding you have the same air quality as someone out in the sticks..
     
  18. caspian

    caspian Member

    Joined:
    Mar 11, 2002
    Messages:
    9,216
    Location:
    Melbourne
    two points:

    (a) so we should artifically disadvantage some people to no benefit or reduced costs just because it's not viable to provide the same service to rural areas?
    (b) the standard is already 12Mbps for everybody. they're just doing better than that for some areas.

    the standard has always been 12Mbps minimum and that hasn't changed. that will so improve rural services beyond current levels that it's a quantum leap.

    your point seems to be based on the assumption that delivery of 100Mbps services to 93% of the population is more expensive than delivering 12Mbps. it's not, the technology is the same - in fact it can do gigabit right now. the only additional costs are in bandwidth and that's ultimately borne by the end user.

    it would cost substantially more to deliver those sorts of speeds to rural areas because fibre can't be used due to the geographical distances and lack of population density involved. the minimum speed of service in those areas is defined by the technology that can deliver services cost effectively.

    sure it would. every privately funded toll road in the country, for starters. major infrastructure runs a capex recovery period over a number of years, that's no different to your getting a private loan to buy a car. sure you could save up for 5 years then buy it for cash, or would you rather have the benefits of the car now and pay it back over those 5 years?

    mmm... I was the lead level 2 project team on the Telstra FTTN project, so I happen to know a little about that one, and my opinion is a little different on the subject. the technology worked OK but suffered steep performance roll-off within about 250 metres from the port, which if you do the simple cross-sectional maths based on the proposed pillar migration plan, meant that substantially less than 20% of users would have seen a performance improvement over ADSL2+. and that's only after spending $10-12 billion on CAN maintenance, which was something that everyone conveniently swept under the rug.

    oh yeah - and you would still need your house rewired with CAT5 to the modem socket and a central filter to maximise performance, so there's no saving there.

    I literally burst out laughing when I saw that, funniest post of the day. a perfect example of precisely why the NBN is needed.
     
  19. Duideka

    Duideka Member

    Joined:
    Jun 21, 2009
    Messages:
    309
    Location:
    Perth
    Have you had a look at the coverage maps?

    All of these dots are areas getting FTTH, 1Gbit/s promised for now:
    [​IMG]

    That really is a huge area, you can't expect much more... even places in the middle of nowhere are getting FTTH.

    NBNco has already promised that people who are currently on copper, who are outside of the FTTH coverage, can opt to keep the copper if they prefer it - but consider how far most people in rural Australia are from the exchange, even if they can get ADSL2+ it's probably very slow.

    Plus there is always the option to expand it further with revenue generated after the first 10 year period, 93% is a massive initial goal - the current copper network that covers probably 99% of the country for voice (less for data obv) likely started with a very small coverage aim.

    This is a good point also, statistically from day 1 NBNco is delivering a 1:1 78.125Mbit/s service, this assumes everyone is on a 100Mbps or faster plan and downloading as fast as they can, we all know this is never the case and the average monthly usage is measured at the lower end of 10's of gigabytes.

    Theoretically NBNco could deliver a 2Gbps service tomorrow using the GPON's and not run into problems, haven't even touched the fact every single piece of networking equipment for the NBN has been estimated and mostly tendered already for just $1,5bn - how much do you think upgrading the 2.5Gb GPON's to 10G in areas that need it would cost? I can't imagine more than $250m for a full replace, and that would deliver 312Mbit/s 1:1 - you could more than likely provide 8Gbps burst speeds on that without much issues.

    I am of course talking about the NBNco infrastructure, ISP's networks area another situation, many ISP's networks would grind to a halt if their customers got upgraded to 100Mbps connections, but if anything is going to kick ISP's into gear, it's the NBN - I have seen more upgrades to ISP's networks in the past few months than any other time, ISP's are already chopping and changing their networks to support the onslaught of traffic to come.
     
    Last edited by a moderator: Aug 29, 2011
  20. caspian

    caspian Member

    Joined:
    Mar 11, 2002
    Messages:
    9,216
    Location:
    Melbourne
    sure, but my point was to illustrate that it costs no more to deliver gigabit capability than it does lower speeds in terms of the existing network equipment. the only variable then becomes usage based backhaul, and the users pay for that depending on demand.

    actual demand will obviously be well sub this, based on a range of service speeds, and the fact that not everyone tries to download the internet each month.

    not just yet, the ONT UNI-D ports are only 1Gbps capable.

    it's just a matter of replacing the GPON SFPs with 10Gbps XFPs. the real question is demand, which is a function of what people can do with it. personally I would be hard pressed to use a 100Mbps service just now, although I can see the applications within a couple of years - particularly around HD streaming video. beyond that, WDM-PON would provide further capacity, but I suspect that more extensive hardware replacement would be required to make that leap. again, that's something that will be driven by demand.

    hopefully the ISPs would have the good sense to realise that if they're going to sell that sort of capacity, they need to look at their capacity provisioning. not only are 100Mbps services a new game, but if people are going to be paying the premium price for the flagship service, they are going to expect it to perform as advertised.
     

Share This Page