A bit of history
Location map (49 kb png) or Google Maps
Observed performance of a voice line - based on initial connect speeds - (Two png files, totalling 12 kb)
Partial map of Telstra's plan (65 kb png)

For further details, see the Link archives

Hi Phil, Col

At our meeting in August, installation of a SCAD was proposed as part of the solution to network problems in the Mt View/Mt Bright/Mt Baker area. The submission from the Communications, Electrical and Plumbing Union to the Senate Environment, Communications, Information Technology and the Arts References Committee Inquiry Into the Australian Telecommunications Network <http://www.aph.gov.au/senate/committee/ecita_ctte/tele_network/subm issions/sub96.doc> indicates that the dial-up data rate performance of a SCAD is limited to 26 kilobits per second (kb/s).

I am achieving 28.8 kb/s on my existing line, albeit with fairly expensive electronics compensating for some infrastructural peculiarities. 26 kb/s would constitute a degradation in service quality. Am I missing something here?

Please reply by e-mail. Your response - or not - will be promulgated this weekend.

Response - phone message 18 October (mp3: 38 kb)

Greetings Linkers,

I'm hoping this will lead to an exchange of interest to the list.

The response to the message below was a message on my answering machine: Phil Lammert asking me to phone him. I'm not going to do that - this discussion belongs in the public domain. After all, Australia's telecommunications infrastructure remains public property (despite rabid political manoeuvring).

As far as I can tell, a SCAD will bring optical fibre closer to the area in question. The number of lines might increase, but the potential bandwidth of each line will decrease. It looks to me like a solution for a past century, not the present. There are no plans to bring more realistic bandwidth into the area, as far as I know.

For a little background, see < http://www.aph.gov.au/senate/committee/ecita_ctte/tele_network/submissions/sub72.doc> and < http://www.aph.gov.au/senate/committee/ecita_ctte/tele_network/submissions/sub29.htm>.

Is it sensible for Telstra to spend money providing a solution that will not meet current needs, let alone those of the future? I'd welcome opinions.

At 09:56 AM 20/10/02 +1000, David Boxall wrote:
> Is it sensible for Telstra to spend money providing a solution that
> will not meet current needs, let alone those of the future? I'd
> welcome opinions.

David, I do hope you are going to get this proposal to the current inquiry some how. Of course it's not sensible! It's almost as if the company is trying to emulate the water drought - let's squeeze them out of existance and then we won't have to worry about no bush no more...... Another phrase: passive/aggressive.

Sorry Telstra, but some 'solutions' are *not* 'solutions'.

Jan

JLWhitaker Associates
Melbourne, Victoria, Australia
jwhit@primenet.com -- http://www.primenet.com/~jwhit/whitentr.htm

Hi Phil,

A week has passed and I have no answer to my enquiry.

Silence constituting assent, you have agreed that proposed work on the network in the Mt View/Mt Bright/Mt Baker area, while it might increase the quantity of services, will decrease aspects of the quality of service. You also implicitly acknowledge that the whole story was not presented at our meeting in August.

That said, the CEPU submission mentioned in my first message indicates that more modern, more capable solutions are available. Please tell me why the solution chosen is not one that will meet the current need. As both shareholder and customer, I'm deeply concerned at the proposal to waste Telstra resources on a project that will not do the job, when there is technology available that will.

Please reply by e-mail. As before, your response - or not - will be promulgated this weekend.

Response - phone message 24 October (mp3: 49 kb)

Hello again Linkers,

Forwarded below is my latest message to Telstra. Once again, there has been no e-mailed response, just another message on my answering machine from Phil Lammert. The message says that: - Telstra is "putting something in writing", which is "being approved at the moment". Sounds like any communication not oral is a big thing to Telstra. - "It's true we are putting a SCAD in there, but not for some time". If it's not going to do the job, does a delay make much difference? - "the SCAD that we put in there will be ISDN capable". Aren't they all? - "We're also doing a number of other things to ensure people get good Internet". What's good Internet? 26 kb/s isn't.

David,

Correct me if I'm wrong ... from my knowledge, admittedly incomplete, of the area we're talking about, Mt View is somewhere between Wollombi and Cessnock?

My point is to ask a question: going along with the contention that a SCAD would be inappropriate, what solutions would be suitable in what's a moderately remote location?

Richard Chirgwin

On 28 October 2002, Richard Chirgwin wrote:

> Mt View is somewhere between Wollombi and Cessnock?
The most distant subscriber is about 15 km from Cessnock exchange. Though rural in character, Telstra classifies the area as urban.

> what solutions would be suitable
I'm not qualified to say. That's one very good reason for posting to Link. Perusing the CEPU submission and allowing for the fact that broadband is not an issue, perhaps ANT-1, I-RIM or C-MUX? According to the CEPU submission, the dial-up potential of a SCAD is lower than the service level usually achieved (albeit with some effort and expense on my part) at present. What would you suggest?

Broadband is not an issue purely on cost grounds. Once the price drops to something realistic, it will become an issue. That said, wouldn't adept management (given the costs of getting personnel, equipment and materiel into the area) make allowance for it?

David,

The 15km is the starting point. That's way beyond ADSL, and in fact beyond the spec for a 56k modem (around 8km from memory) - which explains why you had to put in effort to get 28k dialup reliable.

OK; you're able to get 28k with effort; what would a non-technical user get in this circumstance? If I bought a decent modem but didn't try and tweak the connection, what would I expect? - because even from an engineer's point of view, something that boosted 99% of users from 19.2 to 26k is good even if the other 1% lose 2kb/second.

I missed a really dumb question, of course. So I'll ask it now: how close does the fibre from Cessnock come to the area we're talking about? Is Internet connection sharing feasible where you are - and would Hunterlink play along?

Richard

>If I bought a decent modem but didn't try and tweak the connection, what would I expect?
If you bought the modem I recommend for this area (Maestro Woomera), left it at the default Expert mode and allowed time for it to adjust itself to the line, I expect you would get 26.4 to 28.8 kb/s most of the time. That doesn't require any technical knowledge, just a willingness to spend ~$300 on a modem.

There is a problem with distinctly non-technical people buying a computer "with the Internet". What they get is the cheapest internal modem (WinModem?) that passes muster in the shop. Those things frequently fail to even connect out here. Try convincing a retired farmer that, after spending ~$3000 on a new system which the vendor told him would connect to the Internet, he needs to spend another $300. This after a series of bad experiences have turned him off the idea of the Internet altogether. Shouldn't the network be up to the task without such expense & aggravation for subscribers?

> how close does the fibre from Cessnock come to the area
There is fibre along Wollombi Road, which comes within 5 km. At least, I believe that to be the case. I don't think Telstra has ever given me a straight answer about the infrastructure. Our phone services use copper from the exchange. The fibre parallels the copper, I believe.

Our line comes into the area via Bellbird, over Mount View at Bimbadeen lookout. The fibre runs through Bellbird, but probably comes closest to the area at Millfield (closer to Wollombi). There's a (rather too large) map at <http://users.hunterlink.net.au/~dddab/images/Map3.png> which might help.

> Is Internet connection sharing feasible
Distances are too great for cable. The area is too mountainous for wireless. Difficult.

> would Hunterlink play along?
They've been very accommodating in the past, but I believe the company recently consolidated with a larger one which has a reputation for less flexibility.

(Ouch!!! $3k for a PC? Surely somebody could drive to Sydney to do it for a third the price?)

OK, that was blurted out and can be ignored...now, I'm going to keep dissecting this thread not to irritate you, David, but because what you've written illustrates how complex the world can become. What was one problem is now three: 1) Long-run copper infrastructure; 2) Thin population (making it difficult to pay for a better service even if technically feasible); and 3) Industry deception (not even blaming the retailer here, I'd be the retailer believes what the vendors tell him).

1) We already know about the cable runs from town. But look at the location of the fibre - 5km away. So if the fibre were interrupted to terminate (say) a DSLAM, the users may still be out of reach. And (in this scenario) the DSLAM would be more expensive than in suburbia: it needs power (and power backup), and it needs to be outdoor hardened, etc.

Also - it would be a massive coincidence if the copper runs from the nearest exchange were convenient to the fibre location ... so if a DSLAM went in as near as possible to the users, connection would demand new copper to be laid from the DSLAM to somewhere.

In short, physics and logistics are against you for broadband - and even for a reliable modem connection (except with the expensive Maestro).

2) probably doesn't need further discussion.

3) The industry has a fair bit to answer for here. You relate the $3000 PC with a crappy WinModem - and you're right. But I've also remarked on Link that the limitations of 56k modems have been known for years. So what happened? The industry has lied to users; people learned to believe they would get something, when they were never going to get it; and Web page design has tailed along.

I'm as happy as the next guy to criticize Telstra. But I think it's a mistake to make Telstra the lightning rod for the B.S. marketing of others. Imagine that the govt decided to massively expand a taxpayer program; for the PC industry, it means an expanded market, courtesy of public funds, as a reward for misleading its customers for a decade. That's hard to swallow...

Richard Chirgwin

<...>

I think you have to accept that getting city standard services into remote regional areas is hard and is not likely to happen in the short term.

The problem is that most CBD orientated businesses design their services to suit the city standard, and having walked away from the remote/rural areas they now think that they can impose their city standard based services on those areas that they have abandoned. The banks are one obvious area that I am having a shot at here, but I also think all levels of government are just as guilty of walking away.

I think in the short to medium term the best that governments can do, and I think it is a federal government responsibility, is mandate and legislate that those business/services/authorities who wish to continue to provide a service to remote/rural areas shall design the services that they provide to accommodate the lowest common denominator, ie. what suits the remote/rural communities. If that means that the service does not have the flash, and eye candy, and single browser functionality that they would like - then tough - them's the rules.

In order to prevent businesses et al. from just walking away totally from r/r and hence reducing the services provided, the government would have to include in the legislation an incentive for them to continue to deliver such services to r/r.

OK, pie in the sky, pigs might fly...

-- Howard.

At 9:55 AM +1100 1/11/02, Howard Lowndes wrote:
> I think you have to accept that getting city standard services into
> remote regional areas is hard and is not likely to happen in the
> short term.

Given the broad acceptance of this idea, you're right, it's inevitable that regional & remote areas won't get city-standard service. However, all it would take to overcome this inevitability would be a shift in sentiment to the idea that it's not inevitable after all.

The current "reality" has clearly not always prevailed. Getting POTS into the bush in the first place was a much greater logistical challenge than getting it into the cities, for the same reasons that it's easier to get fibre to the cities than the bush. Nonetheless, it was done; the postmaster general considered it essential and worthwhile, and the Universal Service Obligation was put in place to ensure that logistical problems wouldn't impact on equity of access to this essential service.

I wasn't around when it happened, I don't know the history; can a more senior institute member elucidate? Maybe it was a tough battle to get POTS outside the cities; perhaps we can learn lessons from how this fight was fought last time around.

Regards,

V.

At 09:55 AM 1/11/02 +1100, Howard Lowndes wrote:
> In order to prevent businesses et al. from just walking away totally
> from r/r and hence reducing the services provided, the government
> would have to include in the legislation an incentive for them to
> continue to deliver such services to r/r.

I'm wondering if it is worthwhile distinguishing the level of remoteness better. I'm in an outer suburb area of Melbourne, not far from a town called Cranbourne that has to pay STD to ring to Melbourne in most cases [Telstra does provide a regional calling plan that allows 90km capped calls at $1.99, but that's still 8 times the cost of a local call]. It does come up as an election issue, but obviously nothing is done about a minute aspect like this, no matter how hard the voters scream about it.

The other issue is population density in relation to distance from a non-r/r, whatever that is. I'm thinking here about the spread of a Queensland or WA or even western NSW compared to a state like Victoria. I'm not sure how to draw a line about what fits the definition, but you gotta think there would be extreme rural, as in stations of huge hectare size far from any town, and then there is moderate rural, say the NSW situation or western Victoria, and hobby farm rural or regional towns like Ballarat, and it's surrounding areas.

Another one that is almost the reverse is people who live in apartments/units. There are limitations on what cabling is allowed in some of those situations. We sometimes had a problem in older installed areas of running out of pairs of cable as density increased. I wonder if the same thing will happen in the multi-unit developments where the density of population is actually increasing.

One would think the services would vary according to these factors.

Bottom line is that the description that is used, if too large grained [city v r/r] sometimes doesn't really describe the problem. It's like Telstra saying they reach over 90% of the population. Not that difficult when those 90% live on the coast in the cap cities.

Jan [rambling on a Saturday morning]

Hi Linkers,

Still no response of any substance from Telstra.

On Fri, 1 Nov 2002, Richard Chirgwin wrote:
> (Ouch!!! $3k for a PC?
Not as bad as it sounds. The system in question included two printers (laser & ink jet) and a scanner.

> if the fibre were interrupted to terminate (say) a DSLAM,
> the users may still be out of reach. And (in this scenario)
> the DSLAM would be more expensive than in suburbia: it
> needs power (and power backup), and it needs to be outdoor
> hardened, etc.
If it makes any difference, there's already a RAM-8 in the line. I gather that needs a 300 volt power supply, so that much of the infrastructure is already in place.

Two weeks since Phil Lammert promised me information:
<http://users.hunterlink.net.au/~dddab/audio/20021024.mp3>, a letter arrives. Signed by Phil - looks promising: "Did you know Telstra is working hard for you in your area to give you an Internet service that suits your needs?

Telstra BigPond(TM) Home Dial-up Internet gives you dial-up access from a fixed phone" ... - a sales speil.

There was also a CD-ROM. Has Telstra taken over from AOL as supplier of shiny trinkets?

Looks like I'm not going to get any information out of Telstra, after all. That doesn't surprise me. Getting information from Telstra has always been like pulling teeth. In Telstra's case, the teeth of a living Allosaurus.

Last time, I was told that the information I requested wasn't available. The CEPU submission to the senate ecita inquiry confirms that Telstra's records leave much to be desired, so I guess that's believable.

As one of the owners of the network, I feel entitled to information about it. Shouldn't the information that is available be freely so, preferably online?

I've been told that the most cost effective solution for my area would be something called C-MUX. The only reference I found (apart from the CEPU submission) was a page from Korea that was far too heavy for my bandwidth. Comments/suggestions on the idea would be most welcome.

<...>

Phil,

Three weeks have passed since your phone message: < http://users.hunterlink.net.au/~dddab/audio/20021024.mp3>. I've yet to receive the promised information.

I understand that all phone services are multiplexed in one way or another. The problem with a SCAD is that it limits dial-up data to a maximum of 26 kb/s. In addition SCAD technology supports, not ADSL, but only the outmoded & expensive ISDN. Telstra has been known to employ technologies that do not impose those limits. Why are we not worthy of the best?

All services being multiplexed, bandwidth to each service should be proportional to:- total bandwidth available to the multiplexer, divided by the number of simultaneous users (ignoring limits imposed by technology & customer-side conditions). SCAD technology requires an optical fibre feed, so:
-   what fibre capacity do you propose installing?
  -   how many fibres; how many lit, how many dark?
Where do you propose terminating the fibre? Use the map at < http://users.hunterlink.net.au/~dddab/images/Map3.png>, if you don't have a suitable one.
What are the subscriber numbers (total & estimated simultaneous users) to be served by the proposed SCAD?

The limitations of SCAD technology will lead to demands for its replacement, soon after installation in this area. What advantage does it offer that justifies such waste?

Yesterday, a letter arrived by Express Post (fast snail mail?) from Telstra. Below is a hurried retype of the text.

I haven't had much chance to think about what it says. What is not said seems to carry more meaning.

I did notice that they couldn't get my name right twice running. They seem to be questioning the CEPU's assessment of SCADs, as well.

15 November 2002

Dear Mr Boxhall,

Proposed installation of Small Capacity area Distributed System SCaDs

Thank you for your emails dated 17 and 20 October 2002 and your comments regarding Telstra's proposed installation of SCaDS technology in your local area.

At a meeting in the Bellbird Community Hall on 10 August 2002, Telstra Country Wide Area General Manager for the Hunter Valley and The Lakes Region, Phil Lammert, outlined Telstra's network development options to improve service and meet future demand for telecommunications services in the area. The network development options that were discussed at that meeting were cable augmentation, SCaDS and Optical Fibre cable installation.

Cable augmentation

The immediate plan to meet growth and increase the options to deliver faster Internet service in your area is to augment the existing copper cable network with the extension of spare capacity . The planned project includes the auxiliary installation of Optical Fibre cable for the future . This project will provide for identified growth in basic telephone services and ISDN products offering 64kbps and above. The project is expected to be delivered in February - March 2003.

SCaDS and Optical Fibre cable installation

SCaDS technology is a current multiplexer used to support services on the Optical Fibre cable network. SCaDS currently provide basic telephone service with a narrowband connect speed normally up to 28kbps channels. They may also be equipped to support Telstra's ISDN product that may provide data up to 128kbps.

In relation to your comments regarding Telstra's proposed upgrading of its network and the affects that this may have on your current Internet speed of 28.8 kps, you will be eligible for either the general digital data service (GDDS) or the special digital data service (SDDS) with a 64 kps capability. Telstra currently supplies GDDS through a basic rate integrated services digital network (ISDN). This service is delivered over the ordinary telephone network, but is different from the standard telephone service because it can transmit data at a higher rate.

For more information about ISDN Home and how you can get connected, call Telstra on Frecall 1800 356 285 or visit http://telstra.com.au/onramp/res.htm.

For customers outside the GDDS area or who cannot access an ISDN service, SDDS is provided by way of a BigPond satellite downlink service. Telstra can assist with the acquisition and installation of satellite equipment associated with the SDDS, such as the satellite dish and mount.

To find out more about Telstra's full range of BigPond Internet services visit Telstra BigPond at www.bigpond.com/broadband or call the Telstra BigPond Broadband Information Hotline on 131 282.

In addition, Telstra is also available to assist you with queries through the Internet Assistance Program. The program is a joint initiative between Telstra and the Commonwealth Government to achieve an effective data rate of at least 19.2 kps to assist internet users across Australia. The Program is available to customers of Internet and telephone companies using Telstra's fixed telephone network. For information or assistance, visit: http://www.iapselfhelp.com/customer/index.cfm

We trust this information has been of assistance in addressing your emails. Please do not hesitate to contact me if you have any further queries.

Signed by somebody for Phil Lammert

<...>

Response - letter dated 15 November 2002 (two png's totalling 28 kb)

Phil,

Thank you for your letter of 15 November <http://users.hunterlink.net.au/~dddab/Telstra021115.html>. It is evident that the proposed solution will degrade the dial-up data potential of services in this area, while failing to provide equitable broadband access. The economics of the solution are thus questionable. What alternatives were considered before this one was chosen?

More than five weeks have passed since my last message to you. What are your answers to the questions posed?

Greetings Linkers,

[. . .]

Subsequent requests for information from Telstra have met with deafening silence. Work has started, however. There are some quite impressive looking pits going in. It looks like conduits are being installed in parts, where presently cable is just ploughed into the ground.

Neighbours have received maps of work proposed on their land. A scan of one map is at: <http://users.hunterlink.net.au/~dddab/Telstra_map.html> (65kB png). The terminology, abbreviations and numbers hold little meaning for me. Can anyone interpret?

[. . .]

Apparently, work is scheduled for completion next month. The little sheet-metal posts, that I gather marked junctions in the lines, have been replaced with concrete-lidded black plastic pits. The only conduits are, I'm told, in places where a plough cannot be used such as under roads which have been sealed since the original lines were ploughed in.

My median line speed has dropped to 26.4 kb/s. That is consistent with what the CEPU said to the Senate about SCADS: <http://www.aph.gov.au/senate/committee/ecita_ctte/tele_network/submissions/sub96.doc>. I'm also getting dropouts, which is a worry.

On a brighter note: About 500 metres to the south of my property begins the area served by the rural Paxton exchange. A friend in that area reports that her line speed has improved from 14.4 kb/s to over 40. No SCADS or Pair Gain there, obviously.

Col & Phil: would you care to give details?

[. . .]

We've been through periods of no dial tone and tangled lines. My neighbours have become quite adept at finding out which phone number goes to which house at any given moment. But for the efforts of the community, I doubt that the mess would be untangled even now. The contractors certainly seemed to be at a loss, I'm told. Work appears to be largely completed, however.

My modem reports line level stable at -16dB. Initial connect is usually above 42kb/s, but signal quality at that speed is typically worse than 60. Usable signal is generally achieved at about 36kb/s. Windows continues to report the initial connect speed throughout, of course.

According to neighbours who were in a position to see, the cables were buried remarkably deep (~1.5 metres). They report that, most of the time, two cables were laid together. A blue cable and a black one, both about 25mm in diameter.

Telstra still has not provided details about what has been done. The only response I've had is that they don't normally give subscribers the "nuts & bolts". Any ideas about what the two cables might be?

[. . .]

Hi David, Link,

Hazarding a guess based on some experience with Telstra and cabling:

The blue cable, if about 25mm would be a fibre-optic cable with alot of fibre in it - a fibre cable of about 8mm carries 16 fibres in my experience.
The black cable would be a waterproof cable of ordinary copper lines, either 50 or 100 pair.

If the cables were direct buried:
-it's unlikely you'll see any more upgrades
-assume less capacity in those cables than I have guestimated, AFAIK when they direct bury, they use a heavier casing on the cable.

If the cable are in a conduit:
-I am alot more confident in my guesses.
-Future capacity upgrades are very possible and relatively cheap.

Regards
Karl
=8)

[. . .]

In my experience, buried cables (either optical or copper) are typically double-jacketed with an outer nylon or termite barrier jacket. Depending on the installation method, ploughing-in or burying in a trench, different internal construction is used. When cable is to be ploughed-in, a steel strength member is included along with an outer corrugated steel armour. Other layers of protection may be included to resist rodents.

You will see that it is just about impossible to guess what cable is being layed. For the fibre (the blue cable), it can be anywhere from four cores to one hundred and forty four cores (don't even ask why I spelled that out! I started and then couldn't stop myself!). As for the copper cable (the black one), a 25mm cable can (depending on construction) contain anything from 25 pairs to 200 pairs. My best guess going on your description and my gut feel that the conductor size would be 0.64mm and not 0.4mm is 50 pair.

Regardless of cable size, the major cost when installing cable plant is not the cable itself, but the labour costs of the installation. The maintenance cost of properly installed and undisturbed cable plant is typically nil, so one would hope that the maximum cable size that could be justified based on a fifty-year projection would be used. Sadly, past experience with PMG, Telecom, and now Telstra leads me to believe that the minimum possible cable size will be layed in rural areas. After all, just why do you think rural subscribers are suffering today? It's simply because proper forethought was not applied when the initial cables were laid.

[. . .]

Russell