Mobile Broadband is best for Leitrim

garrethmcdaid

I think this figure of €2.4bn needs to be removed from the debate.

It arose because I estimated that it would cost €3.8bn to buy the eircom network, and €1bn to upgrade it, which is a reasonable estimation given eircoms current market cap and corporate debt levels.

Someone else said it would cost half that (€2.4bn) to rollout an entirely new network to *every* house in the country.

Its hardly a scientific analysis.

bealtine

garrethmcdaid wrote: »

I think this figure of €2.4bn needs to be removed from the debate.

It arose because I estimated that it would cost €3.8bn to buy the eircom network, and €1bn to upgrade it, which is a reasonable estimation given eircoms current market cap and corporate debt levels.

Someone else said it would cost half that (€2.4bn) to rollout an entirely new network to *every* house in the country.

Its hardly a scientific analysis.

It's as scientific as your analysis, yours is a guesstimation as is the other figure.
The other figure is based on analysis done by IoffL.

clohamon

garrethmcdaid wrote: »

The document makes quite clear that the NBS constitutes State Aid and that the designation of EDs is a sufficiently mitigating factor to allow to to proceed.

The Blue areas were areas of proposed services which, if live by 30th June 2008, would become red areas; otherwise they were removed from the map. They are therefore not relevant to the final scope of the scheme. It is the green (unserved) areas that are relevant.

EDs are not mentioned in the Decision.

cgarvey

garrethmcdaid wrote: »

This report detailed the limitations of Mobile Broadband, which we are all aware off, but did not propose an alternative solution.

IrelandOffline, to my knowledge, have the rough workings and calculations done, but not published. There are engineers & industry experts involved in this, so it carries some weight. However, waiting for IrelandOffline to publish a solution is indicative of a complete and total government failure in this regard (as if NBS wasn't, already).

It's not nice to turn such a necessary utility as broadband in to a political football, but the fact is the government have failed to deliver on this in a number of ways.

garrethmcdaid wrote: »

Fixed line connectivity [..] but its completely impractical
[..]
if we wanted to use the eircom infrastructure, we'd have re-nationalise eircom

Agreed, although those points were not made (or made weakly) on your initial blog post. I don't recall IrelandOffline suggesting that a national broadband scheme be eircom-based or fixed-line-based.

garrethmcdaid wrote: »

However, IrelandOffline, to my knowledge, have never acknowledged that the entire NBS is subject to approval of the EU Commission vis a vis State Aid Rules, and the State is not allowed to fund new masts in areas where existing LOS services are available or have the immediate potential to become available.

Not sure why IO should have to acknowledge anything! Are you suggesting that because the EU approves the scheme, it must be OK? So the state can't fund EDs that have LOS operators, but can fund EDs that have 2 other 3G operators (1 of which the contractor piggy backs for service already)? To stop beating around the bush, why should Three be funded to provide 3G service in an area where they have no coverage (and rely on other networks for coverage), when, quite clearly, they gain by providing their own voice service. They get to extend their voice network (which is where the money is; data is just an inconvenience / loss-maker for operators) with a government subsidy, while other operators lose out on the voice revenue that they would have carried on Three's behalf.

garrethmcdaid wrote: »

Finally, the potential of 4G LOS technologies like HSPA+ is not being acknowledged by IrelandOffline.

HSPA+ being considered 4G is probably a stretch (given that it's still a 3GPP initiative and still CDMA-based, designed to fit in with current 3G technology. For the same reason, it will still be completely unsuitable for mass broadband rollout. Yes, 3G/3.5G/4G will have a place in any national broadband scheme (i.e. a scheme that is actually national, and is actually broadband, not the current NBS farce), but relying on it as the only solution is wrong in too many ways. HSPA will still have cell contention / cell size issues. In ideal circumstances, users will get higher speeds (and even latency is still a debatable topic), but it does little for the users at mid cell or cell edge. Next Gen 3G is touched on, but equally disregarded by IO in that briefing doc. The document served to prove why relying on 3G alone is a bad idea, and it has yet to be proven to be inaccurate (despite invitation).

garrethmcdaid wrote: »

HSPA+ will greatly reduce both latency and contention on mobile networks

Reduce? Yes, most likely. Greatly reduce? No

garrethmcdaid wrote: »

and provide theoretical speeds of up to 22mbps.

Yeah right. We have 7.2Mbps now, don't we?!

garrethmcdaid wrote: »

The choices both the current Government, or any Government, are faced with in terms of rural broadband are basically to wait for the market to deliver fixed line connectivity (which will probably never happen) or to fund (to the maximum extent possible under EU Law) the rollout of 3G/4G technology.

Indicative of the failed thinking. There are many technologies that can deliver broadband (real broadband, not what every regulatory authority, bar Ireland, call midband) reliably. IO have no problem in acknowledging (as seen as you want that so much!) that fixed-line is not suited to the all of the last 10% coverage. However FWA and a better delivery of fixed line (such as FTC, for example) would negate the need for midband in many EDs and urban areas completely ignored by NBS.

garrethmcdaid wrote: »

The development of greater mast capacity will also allow for the fluid rollout of 4G technologies like HSPA+, and whatever else comes down the line

Terminology disputes aside, to plan for what might come down the line, when there are already suitable technologies (like FWA) available, seem a bit off, to me.

4G (not 3.5G) does look promising to fill the last X%, but there is no non-line-of-sight technology suited to mass broadband roll out over a large geographic area right now (and none in the immediate pipeline). There probably will be, in time, but NBS is about now.

garrethmcdaid wrote: »

mobile broadband in future, which is likely to compare very well to what becomes available over fixed lines.

Source? I've not read any opinion from any research or industry expert to suggest that. You need only look at how fixed line is advancing to quickly realise mobile isn't maintaining pace, nevermind catching up.

garrethmcdaid wrote: »

As such, I do think the NBS is the best solution for Leitrim, in that the alternative appears to be for the State to do nothing and allow people figure things out for themselves.

So the alternative couldn't involve a scheme that looked at delivering cost-effective broadband using multiple technologies? None of the latter NBS tenders were interested in offering anything other than their core offering, but that can be pretty much put down to tender design, surely?

garrethmcdaid wrote: »

I currently operate off a 512kbps wireless link provided by a local supplier. A good 3G connection would vastly improve my connectivity.

Eeeek!! Let's hear you say that *after* you've tried to operate off both. Unless, of course, your current FWA provider is absolute rubbish. If they are, perhaps 10k of NBS money could sort out a lot of that and help it become profitable/self-sufficient (rather than going to 1 operator to further their profit in their core business adding a token data service with little to no guarantees for a handsome sum).

garrethmcdaid wrote: »

BTW, if someone could refer me to Ireland Offline's proposals re. rural broadband, I'd be very appreciative.

No such thing exists, in published form. There might be something published, but IO doesn't have the resources of the Dept of fish, so it may take time. Bear in mind, IO is a voluntary organisation still in the reformation stages dealing with a lot of queries. The 3G briefing doc took some effort, and time, and was not corporately or state funded. You'd expect if IO could come up with independent (yet-to-be-disproved) research that the department could have organised some consultants to do similar for small money.

thebman

garrethmcdaid wrote: »

Any business that requires >2mbps and is located in a non-serviced rural area has strategic planning issues.

I'll challenge you to give me an example of any country in the EU where rural areas are served by fixed line or fixed wireless services offering speeds greater than 2mbps.

Short sighted IMO. There are many reasons a business can't locate in a town centre. It would be beneficial to many rural businesses to be able to take orders online and advertise their service online. Their business may not be centered around being online but it would be massively beneficial to them.

As they get more business, they pay more taxes and more vat comes in from purchases and the investment is paid off.

The NBS was never about supplying 20mbps to rural areas. It was about supplying a scalable service.

And for what its worth, I run a business in a rural area on a 512kbps connection. If I need to VC I can do so at my local County Enterprise Board. I don't use Skype because all my Ireland and UK calls are free under my BT Business Package.

Good for you. 3 won't deliver that level of service IMO. People will be lucky to send/receive emails at 3 am on the service judging from this thread:
http://boards.ie/vbulletin/showthread.php?t=2055115306

And I have used the service when it was first launched and a year later as I have 3 family members that got the service as it was the only service available in the area.

3 have no problem flooding an area with modems when they know the service will degrade. Why was a provider with such a poor track record of service even considered for the NBS?

I know the NBS has contention commitments/minimum service commitments and they are appalling IMO but I doubt they'll even be enforced when 3 are found not to be fulfilling their obligations.

Demand for 20mbps in rural Ireland is minimal in the extreme. It would be complete folly to spend multiple billions on wired broadband to every rural home when 3G can satisfy 95% of market demand for €170m.

People seem to have very short memories.

The Community Broadband Scheme was by no means perfect, but it was offering fixed wired broadband to rural homes for around €40 per month 4 years ago, but demand was no where near what was originally anticipated.

Nobody is suggesting everyone in rural areas needs 20Mbps. A usable level of service and a reliable service is required though. 3 are utterly incapable of providing this using their track record on this site, other sites about Ireland and other sites about 3's service in other countries.

There was even an Irish website dedicated to the problems with the service until the owner was taking them to court and his solicitor recommended he shut it down as it might damage his case.

It is a disgrace that this company was awarded the NBS IMO.

watty

The Minister basically told us that the penalities for 3 Not meeting the NBS specs are secret.

HDPA isn't scaleable, so even if there is 1.2Mbps indoor on 95% of the ED on launch (unlikely, but then they get the grant for that ED). What if 36:1 contention is exceeded 6months later or speed is only 120k 6 months or a year later? We don't know.

The increases to speed of HSPA from 7.2, 14.4, 21MBs etc affect progressively tinyier parts of a cell. That's basic physics. There will never be substantially more capacity or speed unless you triple the number of masts.

LTE is 4G, not any form of HSPA. Even LTE only manages close to very basic entry level DSL. The much hyped 100Mbps is 4 times the spectrum of 3G sectors and shared among all users and only close to mast. Lightly loaded it's 4 to 5 times faster and heavily loaded 8 times faster than HSPA. That is for most users still less than 1Mbps if there are 20 simultaneous downloads in a sector.

Mobile mast densities are based on economical coverage for voice calls. A Mobile network designed for universal Nomadic / portable indoor Data coverage needs nearly 6 times as many masts. Fixed wireless because it uses outdoor directional aerials can have x16 the capacity/Speed product for less masts in the same size spectrum.

Indoor coverage in a room with window facing mast is about -6dB, that's equivalent to twice the distance. In a room with windows not facing mast the signal can be down -20dB! that is over 8 times distance.

Compare vodafone 3G coverage and 3's coverage maps for same masts. Compare 3's indoor and outdoor coverage with internationally accepted models.

512k FWA can be up to 10x better than HSPA and unlike 3G has reasonable latency which doesn't vary much (HSDPA is 90ms to 2000ms, with 120ms to 170ms as typical off peak/light load).

oscarBravo

garrethmcdaid wrote: »

The Community Broadband Scheme was by no means perfect, but it was offering fixed wired broadband to rural homes for around €40 per month 4 years ago, but demand was no where near what was originally anticipated.

If you're talking about the Group Broadband Scheme, there was no shortage of demand: the alleged lack of demand has been repeatedly cited by various ministers who claimed that there were no applications for the scheme - which might be explained by the fact that no applications were permitted after mid-2005.

It's a bit rich to claim that there's no demand for something that people are not allowed to ask for.

watty

The GBS was badly designed and people told to wait and then it was cancelled.

They used that as an excuse to structure the NBS such that the only people that it was worth while to finally tender where people doing a 3G Phone roll out.

Where was the excellent regional WISPs, Digiweb, UPC, Smart, Magnet, IBB/Imagine in the tendering?

BT pulled out. IFA/Motorola pulled out. 3 Hutchinson and eircom/meteor the two shorlisted. The process should have been killed then and gone back to drawing board with different advisor when it was obvious no viable solution was forthcoming. Instead they believe what NO ONE ELSE in the WORLD believes, that a 3G/W-CDMA based service can deliver controlled contention and a minimum 1.2Mbps and an always on service suitable for VOIP. The people that sell the base stations don't believe it. I have their research and papers.

eircom was priced out of it even though they were going to mostly piggy back on Meteor because it was subsidized by voice calls and had a real rather than potential Satellite supplier. It wouldn't surprise me if 3 /Avanti was 1/2 the price for Satellite. Which at the end of the day is the ONLY thing the NBS is supplying that would not be happening anyway. All going well I predict 5x better Non-NBS satellite will be available as cheap as NBS next year, possibly before Avanti's bird is launched. 3's Sat solution/ supplier was supposed to be available nearly a month ago (June 2009) and is still in a lab with no completion date no launch date last time I looked.

The GBS was badly designed and they blocked it.

The NBS was so badly designed only one answer was possible.

Kenya and South Africa etc can do this stuff better.

trekkypj

Re the whole demand question...

I don't think it's particularly helpful to claim there's no demand for NGN in rural areas. That's the same argument that was used to delay again and again rollout of DSL in the first place.

There can only be demand if people are aware of the potential for its use. If people don't know what a 20Mbps connection can do for their needs, they're hardly going to wave flags and scream "Give me fibre!". You can't measure demand unless there's a prospect of availability for a reasonable cost. It's a supply-driven service. If you build it, you'll get customers. The numbers subscribing to 3G internet services in areas in rural Ireland where there's nothing else is evidence of this. These people didn't know what 3G was about except that it was available and it could supposedly do things their dial up connection couldn't do. That's the reality in rural areas.

Many of the most successful businesses in Ireland both in terms of longevity and likelihood to survive are located outside the cities. They may not have flashy headquarters or big name directors but they are successful and they would benefit hugely from NGN broadband in terms of promoting their business, and being able to videoconference with BTB clients, potential investors and the wider world requires proper infrastructure.

Saying things such as 'if they can get DSL then they don't need NGN' and 'if they need NGN they should move their business to the towns where they can get it' is the wrong attitude to take.

Businesses should not have to move from their base of operations, firstly. Most businesses in rural areas are tied to the location due to supply chains, logistics and planning requirements, so moving for net connectivity is simply unrealistic and is an absurd attitude to take. Secondly, urban Ireland is OVER developed and unbalanced in favor of places like Dublin and Cork. We need further development outside existing urban areas to facilitate a more even growth and development.

If we are to get out of this recession we have to create lots of jobs for people. To do this we must create opportunities for businesses to locate outside major urban areas while still having top class infrastructure - this is vital. That's why we need fibre to the cabinet everywhere possible, and equivalent alternatives (NOT just slow DSL) available everywhere else. And lest we forget, the emerging generation are a generation of network users. They do everything through the net and through their iPhones, through Twitter, LinkedIn, Facebook and other means. NGN will be essential in their world. It's not just for us but for generations to come.

The cost to build now is high, but it must be done sooner or later. The later we leave it, the higher the cost of not doing it sooner and the greater the likelihood of our careless attitude relegating us for good to the status of a second string economy.

garrethmcdaid

cgarvey wrote: »

IrelandOffline, to my knowledge, have the rough workings and calculations done, but not published. There are engineers & industry experts involved in this, so it carries some weight. However, waiting for IrelandOffline to publish a solution is indicative of a complete and total government failure in this regard (as if NBS wasn't, already).

Nobody is 'waiting for IrelandOffline to publish a solution'.

The primary reason why the NBS will use 3G is that no practical or economically viable alternative exists. The absence of an alternative proposal from IO supports that view, which is why I alluded to it.

cgarvey wrote: »

It's not nice to turn such a necessary utility as broadband in to a political football, but the fact is the government have failed to deliver on this in a number of ways.

I certainly haven't turned it into a political football. I was responding to claims made by local FG and SF reps in my area that the NBS was deficient because the Government should be digging cables into every home in rural Ireland.

cgarvey wrote: »

Agreed, although those points were not made (or made weakly) on your initial blog post. I don't recall IrelandOffline suggesting that a national broadband scheme be eircom-based or fixed-line-based.

I don't recall IO suggesting anything re. the NBS, other than drawing attention to the obvious shortcomings of 3G, but clearly anything other than 3G will involve huge cost and a massive logistical effort.

cgarvey wrote: »

HSPA+ being considered 4G is probably a stretch (given that it's still a 3GPP initiative and still CDMA-based, designed to fit in with current 3G technology.

This whole terminology thing is excessively pendantic. Fast Ethernet is CDMA-based, and I don't imagine too many people would have a problem with having a 100mbps connection in their homes.

cgarvey wrote: »

For the same reason, it will still be completely unsuitable for mass broadband rollout.

The NBS is not about 'mass broadband rollout'. Its about providing a solution to areas that will never be served by the private sector. My reference to HSPA+ is indicative of the fact that mobile data communications are advancing at an exponential rate vis-a-vis wired data communications, which adds to the legitimacy of using mobile in the NBS.

cgarvey wrote: »

IO have no problem in acknowledging (as seen as you want that so much!) that fixed-line is not suited to the all of the last 10% coverage. However FWA and a better delivery of fixed line (such as FTC, for example) would negate the need for midband in many EDs and urban areas completely ignored by NBS.

Suggesting that FWA can service *all* of the last 10%, at a reasonable cost, doesn't stack up. FWA could never service the area I live in, which is home to large plots of commercial forestry and where the land is neither flat nor overlooked by high vantage points. You would literally have to erect a transmitter or repeater for every house.

More to the point, private sector operators are already supplying a service in most areas where the topology and character of the landscape can facilitate FWA.

The reason why we need 3G in the last 10% if because it isn't economically viable to provide FWA in these areas.

cgarvey wrote: »

No such thing exists, in published form. There might be something published, but IO doesn't have the resources of the Dept of fish, so it may take time. Bear in mind, IO is a voluntary organisation still in the reformation stages dealing with a lot of queries. The 3G briefing doc took some effort, and time, and was not corporately or state funded. You'd expect if IO could come up with independent (yet-to-be-disproved) research that the department could have organised some consultants to do similar for small money.

Nobody expects IO to produce a detailed proposal, but some sort of consensus re. outline cost, technology mix and mode of delivery would be helpful.

3G is the path of least resistance to bringing the last 10% into the loop, and its future-proofed. The alternatives, if they exist, involve massive expensive and massive logistical effort, and in my view, do not represent either equitable or efficient use of public funds.

The Greens have been beating the drum about proper planning for decades, and for years our opponents have been telling us that there's no problem with thousands of people being dispersed across the countryside in single dwellings that are at a remove from main service arteries.

As much as we'd like to, we can't provide a ultra-fast broadband connection into every home, any more than we put a hospital and train station on every street corner.

The choices are therefore to do something or sit around thinking about what we could do if we had lots of money and the moon was made of cheese.

bealtine

garrethmcdaid wrote: »

This whole terminology thing is excessively pendantic. Fast Ethernet is CDMA-based, and I don't imagine too many people would have a problem with having a 100mbps connection in their homes.

I lost interest in your polemic at this point, it's obvious you haven't a clue what you are talking about.

Ethernet cdma and 3G cdma are two utterly different things.

Just to be clear ethernet cdma is : "carrier detect multiple access" a networking technology. (To be precise it's now known as CSMA/CD and is a layer 2 protocol)

and
3G cdma is Code division multiple access
http://en.wikipedia.org/wiki/Code_division_multiple_access

Go read.

garrethmcdaid

bealtine wrote: »

I lost interest in your polemic at this point, it's obvious you haven't a clue what you are talking about.

Not when it comes to Layer 2 differences in Ethernet and 3g, no, but that's hardly germane to the discussion.

oscarBravo

garrethmcdaid wrote: »

Not when it comes to Layer 2 differences in Ethernet and 3g, no, but that's hardly germane to the discussion.

Layer 1, I'm afraid - and it goes to the very heart of the discussion.

garrethmcdaid

bealtine wrote: »

It's as scientific as your analysis, yours is a guesstimation as is the other figure.

My figure is based on the current market valuation of eircom and its payload of corporate debt, both of which are verifiable.

bealtine wrote: »

The other figure is based on analysis done by IoffL.

Where is this available?

garrethmcdaid

thebman wrote: »

Short sighted IMO. There are many reasons a business can't locate in a town centre. It would be beneficial to many rural businesses to be able to take orders online and advertise their service online.

Why would a business need 20mbps (or even 4mpbs) to take orders online or advertise their service online?

Are you suggesting that rural businesses are hosting their own web servers?

I can think of very few applications that are pervasive in small rural businesses where alternatives do not exist and that require NGN.

To me mind, the NBS was never about rural businesses. Its about access to knowledge and learning for the next generation of Irish workers.

garrethmcdaid

clohamon wrote: »

The Blue areas were areas of proposed services which, if live by 30th June 2008, would become red areas; otherwise they were removed from the map. They are therefore not relevant to the final scope of the scheme. It is the green (unserved) areas that are relevant.

EDs are not mentioned in the Decision.

All academic.

My original point was that the NBS was never going to be able to pepper the countryside with sufficient masts to reduce contention and boost latency, given the effect that this would have on private incumbent providers.

PogMoThoin

garrethmcdaid wrote: »

My original point was that the NBS was never going to be able to pepper the countryside with sufficient masts to reduce contention and boost latency, given the effect that this would have on private incumbent providers.

No, but many of the FWA providers already offer Real broadband in the areas put forward for NBS. I know for a fact that alot of the the areas in Mayo already have FWA coverage, others have Edge coverage with O2. The NBS is bringing nothing new, just wasting money on a mobile network (that 3 would have built anyway) thats todays tech and not planning for the future.

Imagine what a FWA provider could have done in their area with just €1 million to improve their coverage?

garrethmcdaid

trekkypj wrote: »

Re the whole demand question...

I don't think it's particularly helpful to claim there's no demand for NGN in rural areas. That's the same argument that was used to delay again and again rollout of DSL in the first place.

I can only speak from my own experience. I tried to get a GBS going in my area. I went door to door to at least 40 houses and could only get expression of interest from 7. It didn't happen, even though Last Mile were doing the school up the road and were prepared to offer a residential FWA solution.

I also stood as a candidate in the recent Local Elections, and canvassed at least 700 rural houses. Granted, people had plenty to be concerned about, but only 2 voters mentioned broadband access, and they were located in a village less than a mile form an exchange. Their issue was the delay in getting eircom to light the exchange, not rural broadband.

trekkypj wrote: »

There can only be demand if people are aware of the potential for its use. If people don't know what a 20Mbps connection can do for their needs, they're hardly going to wave flags and scream "Give me fibre!". You can't measure demand unless there's a prospect of availability for a reasonable cost. It's a supply-driven service. If you build it, you'll get customers. The numbers subscribing to 3G internet services in areas in rural Ireland where there's nothing else is evidence of this. These people didn't know what 3G was about except that it was available and it could supposedly do things their dial up connection couldn't do. That's the reality in rural areas.

I'd agree with that, but I'd also question the equity of the State spending vast amounts of money on delivering the sort of access that is being talked about here. If it were clear that such an investment would produce a dividend for the economy, then the decision would be straightforward, but that's questionable in the extreme.

trekkypj wrote: »

Many of the most successful businesses in Ireland both in terms of longevity and likelihood to survive are located outside the cities. They may not have flashy headquarters or big name directors but they are successful and they would benefit hugely from NGN broadband in terms of promoting their business, and being able to videoconference with BTB clients, potential investors and the wider world requires proper infrastructure.

I really doubt that. VC services are available at County Enterprise Boards. I doubt very much if there are businesses in rural blackspots who require VC services on a daily or even weekly basis.

trekkypj wrote: »

Saying things such as 'if they can get DSL then they don't need NGN' and 'if they need NGN they should move their business to the towns where they can get it' is the wrong attitude to take.

Why? How is that different from saying if they need access to the N4 they should move near the N4, or should we be building roads on a case by case basis?

trekkypj wrote: »

We need further development outside existing urban areas to facilitate a more even growth and development.

Yes, we need to promote commercial activity outside of Dublin and Cork, but the last thing we need is industrialisation of the countryside.

trekkypj wrote: »

They do everything through the net and through their iPhones, through Twitter, LinkedIn, Facebook and other means. NGN will be essential in their world. It's not just for us but for generations to come.

You don't need NGN to use any of those services. NGN for Twitter?

bealtine

garrethmcdaid wrote: »

My figure is based on the current market valuation of eircom and its payload of corporate debt, both of which are verifiable.

eircom is only worth what is likely to be paid for it

clohamon

garrethmcdaid wrote: »

All academic.

My original point was that the NBS was never going to be able to pepper the countryside with sufficient masts to reduce contention and boost latency, given the effect that this would have on private incumbent providers.

The design of the scheme, the geography of it - the map, the EDs. and ultimately the technologies employed were all matters for the department and the Minister.

They were given as much latitude as they wanted, and given an extra €30M. The reason this scheme is not reaching all unserved areas has nothing to do with the EU state aid rules.

Masts aren't mentioned in the decision.

watty

bealtine wrote: »

I lost interest in your polemic at this point, it's obvious you haven't a clue what you are talking about.

Ethernet cdma and 3G cdma are two utterly different things.

Just to be clear ethernet cdma is : "carrier detect multiple access" a networking technology. (To be precise it's now known as CSMA/CD and is a layer 2 protocol)

and
3G cdma is Code division multiple access
http://en.wikipedia.org/wiki/Code_division_multiple_access

Go read.

Modern ethernet uses Switches. Only hubs and coax ethernet still use CDMA (carrier detect multiple access)

3G uses Code division multiple access is a Spread spectrum technology invented for 2 user person to person Secure Communications. I had one of the military units here. It's in the frequency domain. Ethernet's carrier detect multiple access is in the time domain and is a completely unrelated protocol/technology.

It's very simple and cheap to implement. But each additional code looks like noise to the other user(s). The maxium number of codes is 15 (strong signal). Then as signal is weaker they have to use fewer codes overall for everyone. A single user close to mast can have all 15 codes.

If there are 5 users and one is at limit of range then only 5 codes are available for EVERYONE, so the speed for the single original user drops to a bit less than 1/15th, not 1/5th. Thus it scales poorly loosing a huge amount of capacity and range as users added.


Real cells usually have 3 overlapping petals or pizza slices on 3 different 5MHz channels. For clarity the diagram shows a single omnidirectional sector.

If there are 20 users then the available number of codes in use (according to total signal strength of weakest connected user) have to be time slices. Thus with two people sharing a code the speed halves on that code and the latency can double.
Latency is typically 150ms but thus can vary from 80ms to 2000ms very rapidly. This is very bad jitter and causes huge difficulties for many applications especially VOIP.

cgarvey

Ignoring your "pedantic" point (which many, a lot more technically capable than you and I combined have addressed)....

garrethmcdaid wrote: »

The primary reason why the NBS will use 3G is that no practical or economically viable alternative exists. The absence of an alternative proposal from IO supports that view, which is why I alluded to it.

My point, again, is that 3G is a technically inferior product to, say, FWA. For the last 10%, we can obviously discount cable access and most phoneline/DSL broadband (although the latter could definitely be brought to many many homes, including in NBS areas with some proper USO enforcement and/or grant funding). So we're left with 3G, satellite and FWA. FWA could be delivered to many areas of the NBS, at less cost and a vastly superior user experience.

Your opinion seems hinged on 3G being future-proofed. There's nothing to support that 3G will imporve beyond what we have now in cable/FWA technologies. In fact, there's little to suggest that 3G will catch up with those technologies (who, themselves, continue to advance).

Mobile "broadband" will advance, for sure. But, during the lifespan of the NBS it will not reach anywhere near the user experience of the basic DSL access we have now, or FWA.

So, my point is that 3G will have a "better-than-nothing" place in a proper/real NBS (that was actually broadband and was actually nationwide). To rely on it as the sole technology is short-sighted and, IMO, a waste of public funds. The gain Three Ireland get (being able to extend their voice network in to areas that were previously uneconomical for them, and in areas where they relied on roaming arrangements) does not outweigh the benefit they'll deliver.

garrethmcdaid wrote: »

I certainly haven't turned it into a political football.

Respectfully, I disagree. You've obviously carried the party line, but have done so in a way that just simply does not stand up to the critics (technical analysis, and all sorts of reports from all sorts of agencies). I firmly believe you would not be so positive about 3G were your party not in government. If I'm wrong, so be it, but nothing you'll say will change that opinion of mine.

garrethmcdaid wrote: »

clearly anything other than 3G will involve huge cost and a massive logistical effort.

If lessons had been learned from the GBS, I think there could easily have been an NBS that was meaningful and well delivered by a number of companies each with a proven record in broadband delivery (versus Three Ireland's abysmal history in delivering any midband connectivity). Instead, the government put an excessive value of having the logistics outsourced. I think, given the nature of the thing (last 10% being in difficult rural areas), one company and one solution was always going to be a disaster.

garrethmcdaid wrote: »

The NBS is not about 'mass broadband rollout'.

At risk of being pedantic again, I consider this to be a "mass" rollout in 3G terms. Given the intended population coverage of the proposed number of new masts, in the context of 3G contention and cell size.

garrethmcdaid wrote: »

My reference to HSPA+ is indicative of the fact that mobile data communications are advancing at an exponential rate vis-a-vis wired data communications, which adds to the legitimacy of using mobile in the NBS.

Nonsense! DSL has gone from 1Mb to ~20Mb in the same time 3G has gone from 384Kbps to 3-4Mb. I'm talking real world speeds achieved, as reported by users, rather than theoretical maximum speeds (which can't possibly be achieved unless you're the only person connected to a given sector, and 50 metres from cell centre, etc.)

garrethmcdaid wrote: »

Suggesting that FWA can service *all* of the last 10%

Don't know where you got that, and I've re-read my post.

garrethmcdaid wrote: »

More to the point, private sector operators are already supplying a service in most areas where the topology and character of the landscape can facilitate FWA.

Really? Any source for that?

garrethmcdaid wrote: »

The reason why we need 3G in the last 10% if because it isn't economically viable to provide FWA in these areas.

If the landscape allows, FWA can be delivered much cheaper than 3G. However, again, I don't know your basis for that statement (source?).

garrethmcdaid wrote: »

Nobody expects IO to produce a detailed proposal, but some sort of consensus re. outline cost, technology mix and mode of delivery would be helpful.

Agreed it would be helpful.

garrethmcdaid wrote: »

The Greens have been beating the drum about proper planning for decades

Would you care to outline what they've actually done about this since coming in to government? Will there be carrier-neutral ducting compulsory on all new developments from 2010? Will ducting be mandated on all road developments (so that we don't have to rip up new roads, within 3 months of them going down, to lay cable)? It's one thing to talk the talk / "beat the drum", it's an other thing to act on that once in a position to be able to do so.

garrethmcdaid wrote: »

As much as we'd like to, we can't provide a ultra-fast broadband connection into every home

Of course not, but we need to start considering it a utility and get the ball rolling on universal provision.

garrethmcdaid wrote: »

The choices are therefore to do something or sit around thinking about what we could do if we had lots of money and the moon was made of cheese.

"To do something" being hand a company with a apaulling history in the industry to provide coverage using an inferior technical solution in areas drawn up amidst plenty of controversy, and all in return for an enhanced voice network worth more to Three than any NBS money received. I can't help but think a little more sitting around and thinking, and a little more cheese would have helped.

trekkypj

garrethmcdaid: I'm not suggesting we lay down fibre to every home within the next year! But we have to face facts. The rural telecoms network is not being as well maintained as it should be. When it falls apart (as it inevitably will), something will have to replace it and 3G/LTE is not the answer.

I am saying that fibre should be laid to the exchange, and where feasible, the repeater/cabinet technology can be used to ensure universal coverage for high speed DSL (or FWA if necessary). And by high speed I mean being able to upload large files quickly, watch HD videos, download software, and make IP calls using Skype, Truphone or whatever. You can't do any of those things reliably on 3G (or indeed on sattelite) even if the speeds were there because of latency issues. That means a guaranteed 8Mbps service using existing DSL. (allowing for contention ratios).

Over the longer term you fiber connect homes in urban areas over two or three years. As there is existing ducting it should not be as disruptive or expensive to achieve.

Over a much longer term, say ten to fifteen years a, gradually replace poles with ducting. Prioritise getting the fibre to the village centres, to schools and businesses first. You could make it a condition that new homes and developments being built in rural areas must install ducting and fibre to the standard required as far as the road. Then just hook them up as the ducting and fibre gets installed on their road! That's one way to reduce the cost a bit.

Funding it is the tricky part. Government could do it but don't fully understand the need to think ahead for the future. Eircom needs huge investment and need to get DSL rolled out first for their own commercal reasons. The other operators on their own can't do it. I think myself what is needed is EU funding, matched by Government, contributions from Telecoms Companies, and even community fundraising. You could create a local broadband 'tax' like the council water rates, though I don't think that would be very popular.

We certainly don't want a repeat of the shambles of the Department of Posts and Telegraphs in the 1970s and early 80s which had people waiting months or years to get a phone line, and who were so pressed to get everyone connected all at the same time that they made use of split lines and pair gains and god knows what else - the very things that are preventing people from passing line tests for broadband today.

I'm not asking the Government or anyone else to find €5 billion to spend all in one year. I'm asking them, for once, just once to actually make a proper plan involving all interested parties. They should take the whole thing away from government departments and form an independent Broadband Commission:

• It should be run by representatives from all stakeholders.
  
• Give them a budget, plenty of engineers and surveyors and telecoms workers, and get the plan done right.
  
• Be realistic in how long it will take and cost it so it is possible to say with authority how much it can be done for.
  
• Make people install ducting to the road to save on costs if they want fibre broadband.
  
• Find and implement ways of raising funds from whatever sources are available. Create taxes, or seek EU assistance.... sell bonds to the public...
  
• Get the money in and get the job done within a reasonable period based on the money they are able to raise.

I don't think that's a bad plan.

bk

trekkypj I completely agree with your post, but why the following:

trekkypj wrote: »

Over a much longer term, say ten to fifteen years a, gradually replace poles with ducting.

Verizon, the largest telco in the US has proven that you can cheaply (€650 per home) hang fibre along existing poles, much cheaper then digging new ducting.

To put this into perspective, at this cost to run fibre to all of Irelands 1.6 million homes would only cost 1 Billion. Now I don't expect we could achieve the economy of scale of Verizon, so lets double the price to €1200 per home (a very reasonable figure as many similar sized European ISP's indicate it costs about €1000 per home). Now you are looking at a reasonable cost of just 2 Billion for every home in Ireland.

Then you do what they are doing in Australia, create a company 51% owned by the government and 49% owned by private ISP's (like Eircom, Vodafone, BT, o2, UPC, etc.) who transfer their existing fibre assets into the company and half finance it.

Realistically plan this and roll it out over a 5 year term.

So for just the cost of 1 billion * to the tax payer, you get a true next generation fibre network that allows us to finally match our rubbish claims of being a knowledge economy.

* The 1 billion wouldn't be needed in one year, it could be spread over 5 years or even over 30 years, like a PPP. Spread over 30 years, the revenue from line rental would likely easily pay itself back and likely more over.

Line rental of just €10 per month would bring in about €200 million per year from 1.6 million homes. Obviously some of that revenue would need to go to maintaining the lines and other costs, but it should be obvious that this would pay for itself over 30 years.

watty

Despite eircom's recent claims, a pole can last 75 years.

For the "one off" housing you can do a real 8M down/1Mbps up 20ms latency (up to 45km with directional aerial/dish) cheaper per user than Mobile Phone Mast rollout that can only manage sustained 500Kbpsdown 50kbps up @ 200ms (no mobile network has enough masts for even that).

trekkypj

bk wrote: »

trekkypj I completely agree with your post, but why the following:



Verizon, the largest telco in the US has proven that you can cheaply (€650 per home) hang fibre along existing poles, much cheaper then digging new ducting.

To put this into perspective, at this cost to run fibre to all of Irelands 1.6 million homes would only cost 1 Billion. Now I don't expect we could achieve the economy of scale of Verizon, so lets double the price to €1200 per home (a very reasonable figure as many similar sized European ISP's indicate it costs about €1000 per home). Now you are looking at a reasonable cost of just 2 Billion for every home in Ireland.

Then you do what they are doing in Australia, create a company 51% owned by the government and 49% owned by private ISP's (like Eircom, Vodafone, BT, o2, UPC, etc.) who transfer their existing fibre assets into the company and half finance it.

Realistically plan this and roll it out over a 5 year term.

So for just the cost of 1 billion * to the tax payer, you get a true next generation fibre network that allows us to finally match our rubbish claims of being a knowledge economy.

* The 1 billion wouldn't be needed in one year, it could be spread over 5 years or even over 30 years, like a PPP. Spread over 30 years, the revenue from line rental would likely easily pay itself back and likely more over.

Line rental of just €10 per month would bring in about €200 million per year from 1.6 million homes. Obviously some of that revenue would need to go to maintaining the lines and other costs, but it should be obvious that this would pay for itself over 30 years.

I wasn't aware of this, but that would make sense alright. Fibre on poles, eh? Certainly would reduce the cost some.

bk

trekkypj wrote: »

I wasn't aware of this, but that would make sense alright. Fibre on poles, eh? Certainly would reduce the cost some.

Great article here with pics on how Verizon do it, very cool:

http://www.bricklin.com/fiosinstall.htm