[AusNOG] Aust Govt will build National Broadband Network, no company will be awarded the tender.
technical at halenet.com.au
Sat Apr 11 17:45:45 EST 2009
----- Original Message -----
From: Matthew Moyle-Croft
Cc: ausnog at ausnog.net
Sent: Saturday, April 11, 2009 11:35 AM
Subject: Re: [AusNOG] Aust Govt will build National Broadband Network, no company will be awarded the tender.
Your emails seem to be based around the idea you don't like the idea of a fibre network and you're highlighting any risk regarding them.
Not at all. i am just sharing some of my experience in deploying FTTH. I am very much in favour of FTTH. In fact I have deployed it. My focus is that it must be based on a business plan and the infrastructure should be installed with good design. Because fibre is fragile and the equipment to locate and repair faults is jolly expensive the design needs to address those risks. (Prior planning prevents piss poor performance). I have identified a number of ways to get it underground without massive expense or using Telstra ducts. I am about to test another method in the next month.
My other issue is that governments are not good at achieving asset cost minimisation. For every dollar that is spent a return needs to be recovered from the consumer, also duplicating Telstras existing fibre network without a business case is dumb. My other concern is that if this was such a good idea why hasn't optus, iinet internode and others done it. From what I have worked out, it is because the business case does not stack up and because the applications do not yet exist in commercial form and or are available to consumers other than through 2or 3 providers. ie IPTV and on my second point foxtel. There is one telco begining with t that has exclusive access to content which only 2 other companies can sell Optus Austar (and make a margin). There is no reason that many of the existing offerings cannot be done over ADSL2, so why strand commercially deployed networks with a taxpayer funded one before it is needed. When it is needed and a business case exists it will happen.
Many countries around the world have significant aerial infrastructure. US and Japan especially as markets I'm familiar with. They have aerial Cu, Fibre and Coax. None of which, in my discussions with various O/S telcos seem to worry them significantly in terms of maintenance, risk etc.
Speak to someone that maintains them. Just because they are not worried, doesn't make it right. From my experience in T half of the managers wouldn't know ship from clay. You needed to talk to the 8 or so methods and practices fellows. They do the research and make the process changes. Do they have the same environmental concerns as we experience in Oz. Of all the issues we face the issue that we humans cannot control is weather. We can legislate and regulate against cowboys and idiots, we can even take security steps to prevent such. In many areas aerial may be a solution, but I very much doubt 70% of oz is in that category. A robust FTTH network needs to provision all the end points at the time of construction seal it and leave it alone, other than periodic testing of reference fibres to identify degragation and schedule repairs. By provisioning I am referring to connectorised entry points to allow for customer connection by low skilled staff.
Verizon have said that the maintenance costs of their fibre network (FiOS) (some above, some underground) are significantly less than Cu and more reliable.
So it should be the fibre network is how many years old and the copper is how old? All networks maintenance costs increase with age, then there is the issue of fibre not being affected by the issues that affect copper. If you read what I wrote I am referring to storm, motor vehicles and servere events that will cause wide spread damage on an infrequent basis. But they do happen frequently especially in the summer months.
They have developed significant technology around FTTP install which makes it easy to do and not require as skilled people.
They have developed connectorised methods to reduce the need for skilled staff. Trees falling on cables does not use connectorised methods for repair.
In SA there is significant amounts of aerial fibre from one telco.
That is IEN cable not FTTH, installed with few slice points and on major power routes that do not have trees growing up through them like much of the distribution FTTH network will.
The outages caused by fibre physical failure have been few and far between. Certainly no more than I'd expect from underground cable, maybe less due to the fact that people can see above electricity cables vs can't see underground cables!
My experience would suggest that tree damage would outnumber cut cables 10 to 1.
If there are certain areas where the design of the electricity poles is so poor that there are outages more than every month, then I'd hope the percentage underground would include those areas. Most of the "backbone" will be underground as it's going to be purchased from existing Telco installs or swapped for
equitiy in the new company.
That will depend on the business plan. No one will invest if there is no return
At the moment I don't see much wrong with what the government has agreed to do. The only significant change would be to have a time machine and undo the selling of Telstra in the manner it was in 1997.
The suggestion you make regarding repurchasing TLS and then using it's profits to do fibre I think are ultimately fruitless as the cost is ultimately going to be higher ($43b plus the $6b/year), take longer and not deliver an additional network.
I really do not know how you work that out. Telstra already has the duct and pipe space and an extensive fibre network. It will take many years to duplicate that and cost more than $43 Bn. Most if not all RIM cabinets are within 4 Km of customers houses. To convert RIM areas which cover upto 240 customers to FTTH will take little to do. Yes some RIM cabinets do go upto 480 customers but they are in high density areas.
Have you considered what will happen if Telstra decides to build out their existing network? If so how will NBN mk2 ever be viable and who will ba paying for the losses. NBN will strand many of the ADSL deployments and the existing ADSL 2 providers will have to come back to the pack and operate on minimal margins. I don't see how that will be good for encouraging investment over the next 8 years. If someone can show me a business plan that says it will be viable that also stands up to scrutiny then I will willingly eat me words backwards forwards
It's certainly a valid point to ask about what other things could be done for the $43b. It's the role of government to weigh these things up. I guess the decision point is that GDP is likely to get a 0.5% boost from FTTP but none from water projects.
The bradfield scheme has already shown that there will be a massive return. Some say 200,000 forrestry jobs in Queenland alone. Let alone the 1000's of jobs that are being lost or are at loss becasue water buy backs etc. This decision has nothing to do with return. It is about saving face.
Water is primarily a state issue,
Not any more the states have agreed for the feds to manage the murray darling water. The Bradfield scheme would go a long way to solving the water problems of 4 states. Tell me that would return more than 0.5 GDP. There are many millions of dollars of investment that are stranded at present because of water issues. More importantly many of the people that have water licenses and are paying for allocations are not allowed to use them.
so I guess they're doing what all Federal Govts do and leave the ugly stuff to the states.
Yes unless there is a photo shot in in for them.
I would be interested to hear from some of the major shareholders of some of the larger ISP's about what their thoughs are. But at the end of the day we need to see the devil in the detail in the business plan which doesn't exist for all of this. In the meantime it may discourage investment.
----- Original Message -----
From: Matthew Moyle-Croft
Last night I saw reference to 70% of the network being on power poles and only 30% underground. emm I hope they don't get bush fires, cyclones, cars running into poles, garbage trucks pulling the cables down etc etc etc.
Does your electricity on power poles go out much because of this? Mine doesn't. I think you're overblowing the risk here.
I don't, based on my experience as a field manager for X and research I undertook as part of a operations research degree I did into X's network performance fault analysis as well as work I did as a state analyst. Are you saying you don't have power black outs. If that is the case you are very fortunate. I have 10 - 20 or more a year
Does your Optus Cable/Foxtel go out?
I don't live in an area that has cable. If it did exist here there would be 10 or 20 faults per year. Believe it or not some areas are more prone to storm damage than others. Hence why the design needs to reflect the risk.
I'll point out that undergrounding cable didn't stop the San Jose vandalism last night! I've had more issues with water getting into Telstra's Cu cables going to my house than overhead power issues.
You really should compare apples with apples not oranges. The issue with the cable to your house is not poor design, it is poor maintenance. My comments all relate to design and the business plan
Heck, the water pipes in the street here crack three times as often as the electricity has gone out due to someone doing external aggression on power poles!
Different issue. Water pipes break because of old age (rust), tree roots, soil cracking, back hoe fade etc. Other than back hoe fade telephone cables are not affected to the same extent by the other causes in the same way as water pipes are not affected by electrolysis whereas telephone cables can be.
Overhead fibre/coax is extremely common around the world. US/Japan especially. One of the reasons they have many more last-mile networks than us is that they're not so precious about this.
1. They may not be, but your customers are. Aerial is cheaper to install but costs more to maintain in the long term. I do first in maintenance on a major TV repeater, if the TV signal goes off or is degraded there can be as many as 300 calls evey 30 minutes from a repeater that servers 50000 people. In fact when the olympics was on I was paid to sit and baby sit a transmitter for 5 hours just in case something happened.
2. Customers seem to be expecting quicker and quicker repair times as people rely on these systems more and more they will require better and better reliability and repair times. I am already seeing it, with residential customers asking for compo if their service is down for more than a day.
3. If you do not live in a city then it is highly unlikely that there will be a maintenance presence. It is not uncommon for telstra to not have a splicing van within 4 hours of a lot of places. That is because it is rare for something to get cut. I expect that will be the same for NBN mk2. As a result of there not being a need on a regular basis there will not be to many of these vehicles around period
4. This brings me to the mess that will happen when cyclones. bush fires etc etc happen. A set of tools for a copper jointer cost sub $1000, it doesn't matter if they get wet, dusty etc etc. An OTDR and a fusion splicer cost $50k and they need to be kept dry clean etc. This requires that the network be built in such a manner as to be protected from the elements. Fibre is a very different animal to copper and coax
Quite frankly most people have little understanding or appreciation for external plant and street furniture.
That is true, but my experience is extensive in that area, especially in copper and fibre. The one thing about fibre is that it is not well suited to frequent re entry. The fibre on power poles around australia is all on very well maintained main transmission routes. With all the greenies and the move to aerial bundled cables in street distribution the same level of maintenance is unlikely /is not going to continue. I was in Brisbane last week and the weather was wet and windy. Energex had 6000 customers without power all day, as they fixed some others were reported. This was predominantly due to trees falling on power lines. This is quite common now. If that had been fibre it would have taken a week to fix it, not same day. Then there is the cost issue, fibre cable is cheap but splicing and enclosing it is a different story. It would not be out of the question for a fibre break to cost $3 to 5k to fix a simple break where a copper break could be fixed for $300. If you are paying off a new network you do not want high maintenance costs. Lets put it this way if I had a choice of providers and one was under ground and the other on power poles, I would be with the underground preferably ducted network.
They finally notice things that have been there for years and get all precious about the risk, ignoring the fact that nothing has happened.
Weather events do happen regularly
Copper cable is easy to locate and make temporary repairs quickly, not to mention copper is a lot tougher than fibre.
I think that's debatable - fibre is quite tough. You can make it as tough as you want - depending what you order. (Ever seen the armoured submarine cable for shallow waters?)
On power poles faults are easy to find/fix - you just look up!
I hope you don't want me to take that seriously. Get someone to show you a bit of fibre. It doesn't have to come down to break. It deosn't stretch like copper. You need to use an OTDR to locate faults and to do this you need to have access points without splitters etc.
Often fibre repairs get a bad rap because the cable is quite strong
I am not sure about that
- so by the time it's snapped it's really messed up (eg. ever seen it fibre really messed up because of a big earth drill pulling and snapping it?
That is underground cable. I haven;t seen to many direction drills that go through thin air. From my point of view those cables should have been located properly
Cu cables tend to be so heavy they break in different ways.
Actually quite often the bearer breaks and the cable falls on the ground and the cable is still connected through resulting in many customers still having a phone
Doing Cu repairs is time consuming and hard on large cables.
And should never be required. There exist cable locators and vacumn excavation equipment. I have no sympathy for careless operators who get big bills. The thing about underground is that if a ducted cable is cut it is generally in good weather and a quick repair can be undertaken.
Fibre splicing these days can be a lot quicker - you can much more easily run temporary fibre cables than Cu ones.
perhaps on cables > 100 pr, but such damage is preventable if good practice is followed. I am yet to be able to order a storm or wish a bush fire away. That is the distinction. Using your reasoning we should probably not go ahead with the NBN because Telstra's underground cables are going to be cut by all the cowboy operators constructing the NBN.
I'm not sure if you've ever watched someone repair or punch down a 1200 pair street cable, but it's a lot harder than doing fibre.
Believe it or not I have watched, in fact I have done it. What really makes it interesting is when it is a randon jointed cable which means both ends need to be identified as well. But I do think such cable cuts are a lot rarer than cables coming down in storms. In fact I would recon there would be less than 2 nation wide annually all over Oz. Stom damage from wind or trees would be probably 20000 to 50000 anually I think that is a material difference. In fact the numbers could be 10 to a 100 times higher than that.
I fellow can locate and make temp repairs to copper cable, fibre splices need to be prepaired and protected making temporary repairs that would take 15 minutes on copper take 4 or 5 hours and more than 1 person on fibre. If the $43Bn cost estimate is based on 70% aerial deployment then it may well blow out to $100BN + if it were to all be put underground.
If you've got real cable damage then fibre ain't going to be hard to fix. Maybe you need better splicing guys? I can give you some references ...
No its not hard, it just takes longer than copper and cost a lot more. By the way the last time I looked there weren't to many 1200 pr cables in the distribution network, most distribution cables are between 2 and 100 prs, with most 30 or below. I can joint a 50 pr in 45 minutes. I can do a 10 pr in 10 minutes. Try getting your gun fibre spicer to do that. By the way FTTH design is differnet to IEN design and will involve a lot more fibre in residential streets
So, why underground it unless necessary? The rest of the world has moved on from this curiously Australian dislike of overhead.
It provides for better reliability in storms and fire events, end of story. Also there is a report that all carriers that have external plant are required to fill in regarding getting all the communications cables under ground which is a requirement of the telecommunications act
My analysis would be that if Rudd is as keen as he is to build FTTH then he needs to buy/ re nationalise Telstra and build out the Telstra network and separate it etc. That would be a cheaper way to do it.
It'd cost $40b to buy Telstra, then another $43b to do FTTH. So instead of $43b you're out $83b. What's the point of that?
It would cost $41BN to $50bn to buy Telstra which would give you the ability to fund $6BN per annum out of free cash flow to extend the existing fibre network not build it from scratch, so my point is that it would cost $41BN to do what they want, change the rules and refloat it for probably $41BN =$0 cost to the taxpayer. There are many other reasons to do it this way as well
Might I also add that not all government assistance needs to be in cash form. Governments could also use their business as a catalyst to encourage investment etc. This would result in a better outcome for taxpayers who ultimately pay for all of this.
I don't think people have thought some of the investment part through.
I am absolutely sure of that. The whole process has been flawed from the start
Nor why the government is doing it.
The government are doing it to save face and get even with Telstra for not submitting a valid bid. It is as simple as that. It also looks like good politics, but when interest rates start to climb again and go through the roof the mood may change. They dont even have a business plan yet. How dumb is that announce a $43 Bn project before they have a business plan. That is assuming this bunch can do it for $43Bn which is unlikely. There are less risky ways to achive this end which will cost a lot less. If the governement has a lazy $43 BN to invest on a nation building project then they should be looking at the Bradfield scheme to solve the nations fresh water problems. As I said to someone the other day, we humans need food and water to live but we can survive with out internet, but better still we can have both for the same money. FTTH will happen without this grand standing, once the business case exists.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the AusNOG