LLU - Competition or not?

crawler

OK, this debate is one that is currently gaining ground in Europe and is one I (as Crawler, nothing else) personally would welcome the views of posters here.

The debate loosely translates into how much competition LLU really provides or indeed if it is real competition or not.

In the NOT camp the following is the arguement

1) LLU is not platform competition therefore no new network is actually built.
2) Copper is a medium with physical restrictions and the same copper is used by all ( ref point 1)
3) the incumbent still DOES control the line and can disconnect at will in the even of contract breach (as we have seen)
4) There is a distinction drawn between the physical line and the customer. the LLU operator "owns" the cusotmer but not the line
5) By virtue of the fact a monthly rental is paid per line by the LLU operator it can not be true competition
6) The service at best can only be described as a temporary service distinguisher.
7) Even as newer DSLAM technology is deployed, it will only be newer until the incumbent operator upgrades the network (as in the case of 21CN with BT and eircom's NGN build at that point the new services will be available to all resellers thus eliminating the advantage in "6"
8) TRUE competition can only be achieved by way of new platforms of delivery and total seperation from incumbent operators using multiple technologies

In the IS camp

1) Operaotrs can distinguish services in any number of ways, thus ensuring that their product is always better than the incumbent
2) Operators can overlay services on top of the unbundled line, which is something that the incumbent operator may find time consuming to do
3) The access path is only a small enabling part of the overall solution if multiple services are provided (such as voice, TV, security, home working. teleworking, hosting etc etc)
4) Platform competition may be best, but is very costly and may not lead to global availability, thus diminishing the value
5) Financing an asset is similar to leasing an access path - it's a sound business practice.
6) The rental fee paid could be offset agianst having to maintain the network therefore the cost elements are the same - it is best not to reinvent the wheel - the copper does not know who it is working for.

So there you have it.

what think ye?

Sponge Bob

A Very very good summary of all the issues.

overall its a good thing in the absence of any other universal path and any universal right of way to install an alternative. Were the wireless telegraphy act wayleaves amended for all carriers then maybe it would be different

probe

At the end of the day we are down to real world practical realities.

1)Wireless has scalability limitations in urban areas, reliability issues everywhere and is unsuitable or provides sub-standard service for some types of applications (eg VoIP).

2)Fibre is great. However the investment required to install it – particularly in a spread out, dysfunctionally “planned” country such as Ireland is huge and nobody seems prepared to make it. Except in new housing developments where the service charges are on the high side - and it becomes a monopoly in its own right!

3)Cable (a la cable coaxial cable TV) is for urban areas only and has scalability limitations when it comes to providing very high speed capacity for multi-media applications.

4)Satellite is expensive, has poor latency and is a non-runner.

5)This leaves you with LLU. The infrastructure is largely in place. Backhaul is still a problem in some areas and this could be addressed by a combination of unbundling eircom’s backhaul and using the ESB national fibre system and the metropolitan WANs.

Some specific points in relation to your posting:

a) A “new network” can be built using LLU. The existing telephone exchanges become redundant as users are moved to VoIP. Speeds available are increasing each year as DSL technologies improve, and if you start installing the MSAN nodes in street cabinets in the case where customers are further away from the exchange building – distance from the “exchange” becomes less and less relevant. The new network makes competition for cable TV operations, phone, internet access etc and several operators, including the incumbent can compete in each area.

b) Regulations can be put in place to prevent disconnection and force it to be done in an orderly manner with due notice to the end user allowing them time to make alternative arrangements.

c) Someone has to maintain the copper pair – so loop unbundling fees have to be paid. It doesn’t mean that there is no competition. When the incumbent starts losing large numbers of customers to LLU they suddenly wake up – as we see with France Telecom.

.probe

bk

Tom Young wrote:

Crawler, PM and I'll give you a view. I'm based in Brussels.

Most of what you have stated is good. Many arrangements on LLU will be based on regulatory forebearance arrangements.

I am reluctant to put too much effort into my posts as the audience here don't know the landscape or politics well enough.

Tom please let us all know, unlike the Broadband forum, you will find that most people here on the IOFFL forum have a very in depth knowledge of the landscape, politics and technology involved here.

Most of the people here on this forum are either current or past members of IOFFL and have been heavily involved with all the operators, Comreg and at EU level or work in the industry.

So please leave it open to widespread debate.

bk

Crawler, I think your post is an excellent overview of both the pros and cons of LLU.

I think we all recognise that LLU is certainly not a perfect solution.

However on a practical level it is quiet clear from the examples of both the UK and French market, that it does have a very strong impact on competition and a wider availability of services in the market.

I can guarantee that without stronger competition from cable, fibre and wireless, Eircom would simply not rollout faster speeds and services without the presence of LLU competition in the market.

It would be a concern for the LLU operators that eventually the incumbent will eventually rollout similar technologies to compete, however looking at the example of the UK and France market, it seems even with the incumbent (BT) rolling out new tech and faster speeds, that LLU continues to go from strength to strength, with even more operators deciding to take the LLU route.

uncle_sam_ie

probe wrote:

a) A “new network” can be built using LLU. The existing telephone exchanges become redundant as users are moved to VoIP. Speeds available are increasing each year as DSL technologies improve, and if you start installing the MSAN nodes in street cabinets in the case where customers are further away from the exchange building – distance from the “exchange” becomes less and less relevant. The new network makes competition for cable TV operations, phone, internet access etc and several operators, including the incumbent can compete in each area.
.probe

How far will MSAN nodes stretch DSL from the exchange. And,what speeds would you get at four miles?

bk

probe wrote:

2)Fibre is great. However the investment required to install it – particularly in a spread out, dysfunctionally “planned” country such as Ireland is huge and nobody seems prepared to make it. Except in new housing developments where the service charges are on the high side - and it becomes a monopoly in its own right!

Yes, that is exactly the problem I'll be facing in two months. Moving into a new apartment, Smart are the ones delivering all the TV, phone and BB services over fibre.

Problem is Smart might not exist next week.
Have no choice on service levels and pricing, BB is only 2m.
It is quiet expensive.
Can't go with anyone else as no phone line or cable and no dishes, antennae allowed on apartment

probe wrote:

3)Cable (a la cable coaxial cable TV) is for urban areas only and has scalability limitations when it comes to providing very high speed capacity for multi-media applications.

Cable can be superb, as long as the network is designed and run correctly, not like UPC.

However with cable you face the same potential problem as fibre, of a monopoly, I forsee this becoming a problem when UPC introduces a phone service and new developments end up only with UPC.

crawler

All - thanks for the views so far.

It is a hard one - and one that does not have a right answer therefore any considered viewpoint is right - ultimately the market will decide what is right based on the regulatory and commercial environment but they are all interlinked.

In relation to Ireland my only concern is that LLU could stifle investment in platform competition - in an ideal world we would have perfect LLU and platform competition but how does one impact on the other and the investment required?

Interesting that people noted France and the UK - this is where the debate originates (see it was not just me going all "Mission Impossible" for Digiweb!) and if you look at what Bulldog/C&W have done in the UK in relation to unbundling it really does question the long term benefits of unbundling. On the flip side if the incumbent is losing RGUs hand over fist to LLU operators, they are then forced to react and upgrade and/or innovate. To counter that though the incumbent will always know what the competition is at as they are using the metallic paths owned by the incumbent, this allows the incumbent to react with full knowledge. If the platofrm is completely independant of the incumbent they are toally in the dark and will be forced to react anyway due to fear of the unknown and the "overnight change" effect.

The question is though - is this REAL competition....sigh....?

Couple of comments :-

If you were building new Coax plant you would be on drugs if you didnt build it using Siamese (Coax and twisted pair cable) or Tiamese (Coax, TP and Fibre) cables, little diffence in cost, even if you didnt use it but a much more valuable asset.

Subloop unbundling not an option in Ireland yet so VDSL etc is difficult/impossible - ADSL2+ with rate adaptation is as good as it get and SDSL/VDSL for short loops. Same in relation to a new network (but you still can only use the existing copper runs - i.e. no new reach network platform.

watty

Yes we need platform competition as LLU is not true competion, as done in ireland it is not competition at all.

LLU is useful only if it is cheap enough and fast (less than week) to implement inculding number so that people can take the risk of delivering innovative services. Really though with migration to IP for all services and perhaps limits being reached on speed of twisted copper, the LLU concept generally may be getting past its "sellby" date as soon as we get a working LLU. It was 6 to 10 years ago we needed it.

Crawler wrote:

If you were building new Coax plant you would be on drugs if you didnt build it using Siamese (Coax and twisted pair cable) or Tiamese (Coax, TP and Fibre) cables, little diffence in cost, even if you didnt use it but a much more valuable asset.

Definately. The expensive bit is the trench. Even in Kenya where you can get fiber to your door because digging trenches is cheap, they put fiber.

probe wrote:

1)Wireless has scalability limitations in urban areas, reliability issues everywhere and is unsuitable or provides sub-standard service for some types of applications (eg VoIP).

Disagree. Wireless systems can be scalable, reliable and low latency suitable for all applications. There are many different Wireless Technologies. I'm not saying Wimax (which version anyway?) is the answer.

To scale wireless you need to reduce cell size. Frequency is an issue. Low UHF gives uncontrollable cell size but good building penetration.

Microwave (2GHz up) is poor building penetration, but very scaleable cell size.

The GSM frequencies were well chosen.

Higher microwave gives stacks of Bandwidth, so plenty speed, but only LOS.

Wireless does have problems. But it has huge advantages, the problem it is has a bad reputation due to some operators not having sufficinet backhaul, too big a cell or doing unsuitable installs or using the wrong technology for the band.

probe

bk wrote:

Yes, that is exactly the problem I'll be facing in two months. Moving into a new apartment, Smart are the ones delivering all the TV, phone and BB services over fibre.

Problem is Smart might not exist next week.
Have no choice on service levels and pricing, BB is only 2m.
It is quiet expensive.
Can't go with anyone else as no phone line or cable and no dishes, antennae allowed on apartment

Have you done any research on the old "Telegraph Acts" (if necessary following them into more current legislation). I would be surprised if they didn't give legal powers to the Minister (for Posts & Telegraphs) to install a phone line, and these powers would probably have been delegated in the botched eircom privatization process to that company. If you have signed a purchase or rental contract for property, any clauses purporting to prohibit you from using eircom may well be illegal and therefore unenforceable. It would be appalling if you can't have eircom services - they are such a lovely cuddly company whose broadband services NEVER BREAK DOWN!

.probe

probe

watty wrote:

Disagree. Wireless systems can be scalable, reliable and low latency suitable for all applications. There are many different Wireless Technologies. I'm not saying Wimax (which version anyway?) is the answer.

To scale wireless you need to reduce cell size. Frequency is an issue. Low UHF gives uncontrollable cell size but good building penetration.

Microwave (2GHz up) is poor building penetration, but very scaleable cell size.

The GSM frequencies were well chosen.

Higher microwave gives stacks of Bandwidth, so plenty speed, but only LOS.

Wireless does have problems. But it has huge advantages, the problem it is has a bad reputation due to some operators not having sufficinet backhaul, too big a cell or doing unsuitable installs or using the wrong technology for the band.

No EU15 country is using wireless for broadband to any extent except IRL. If you take a VDSL speed of at least 50 Mbits/sec as a "multi-media efficient target" for domestic subscribers I think it will be difficult/impossible to deliver this via wireless particularly in an urban environment.

If you did achieve anything like this (with tiny cell site sizes) you'd have cellsites all over the place and you would quickly find a political limit to scalability due to objections from various quarters.

And there are health issues with unknown risks from large scale heavy use over long periods within close proximity to the source.

In my view wireless should be kept for mobile applications and used sparingly.

Wireless may however be part of the solution in low population density rural areas.

.probe

bk

probe wrote:

Have you done any research on the old "Telegraph Acts" (if necessary following them into more current legislation). I would be surprised if they didn't give legal powers to the Minister (for Posts & Telegraphs) to install a phone line, and these powers would probably have been delegated in the botched eircom privatization process to that company.

I haven't moved in yet, but I'll certainly be investigating that.

However to be honest I'm more interested in NTL, I currently have NTL TV and BB in the same area, so I know it is available.

SeaSide

The one thing that you have done in your analysis is to leave out the financial element. An unbundled line does allow additional services to be offered but the voice, Internet or TV that is received down that line is the same albeit at a different level of quality in terms of contention etc. What LLU allows an operator to do is to control this quality and not be dependent on the incumbants 48:1

probe

watty wrote:

Definately. The expensive bit is the trench. Even in Kenya where you can get fiber to your door because digging trenches is cheap, they put fiber.

I don't live in Kenya, but this would not be a problem in the town I live in, where 20 years ago the powers that be had the foresight to create cut and cover type tunnels under the footpaths. The utility workers can open a “manhole” in the footpath and go down a ladder and walk around to install/inspect/repair their electricity cables, phonelines, cable TV (nearly 300 channels of TV, movies and radio with 12 languages on offer), water, electricity and anything else that might be down there.

The road surfaces and footpaths are always perfect because they never get dug up for utility repair/installation jobs.

The same tunnels also have vacuum pipes the size of sewer pipes that carry rubbish from apartment buildings to a central waste processing facility in real time (no waiting for the weekly bin man!). Waste is separated at this facility – glass sent for recycling etc. Most of the rubbish is thermally treated (ie that which does not contain chlorine because only materials containing chlorine create dioxins eg clean wood is safe to burn but newspapers made from wood are not because they often use bleach to make the paper). In addition to providing more electricity than we consume in the “urban village” (we export the surplus), the tunnels carry water at 95C which goes into heat exchangers in each dwelling to heat the place in winter and provide hot water all year around. Because the town is on the coast, the ecological energy system takes advantage of the water temperature differences between sea and air temperatures in the energy transfer system. The system also produces water at 5C in summertime which goes into the same heat exchangers to provide air conditioning. So the electricity bills are tiny despite the warm climate and even then electricity costs about 9c/kw (compared with 15c for ESB).

If they wanted to put fibre in the commune it would be very cheap to install – but as everybody is drowning in internet bandwidth and TV channels as it stands, there is no rush.

.probe

probe

crawler wrote:

Subloop unbundling not an option in Ireland yet so VDSL etc is difficult/impossible - ADSL2+ with rate adaptation is as good as it get and SDSL/VDSL for short loops. Same in relation to a new network (but you still can only use the existing copper runs - i.e. no new reach network platform.

That is just a bureaucratic matter for comreg to wake up and deal with. Deutsche Telekom are pushing it through in France and Germany where HDTV (one or more HDTV sets in a household), BlueRay type VoD and internet surfing at the same time in many households will lead to a traffic jam on ADSL2+ loops. Might as well sort it now (as they did 20 years ago with the utility tunnels in the story above).

.probe

probe

uncle_sam_ie wrote:

How far will MSAN nodes stretch DSL from the exchange. And,what speeds would you get at four miles?

The MSAN is open technology. It can use fibre, ADSL2+, VDSL, virtually anything and the distance depends on the technology being used. Over very long distances (over 10km) where you have multiple DSL repeaters you are probably stuck with 2 Mbits/sec which while better than dial-up and UMTS and is OK for browsing subject to an acceptable contention rate from your ISP, is unsuitable for TV and video on demand. That's the price one pays for the luxury of living in the countryside. One also has to travel further to do shopping etc. Just one of life's tradeoffs. There is nothing to stop the DSL modem from providing TV services however if it has a DTT receiver chip in the box.

.probe

watty

Microcells use milliwatts. The more cell density increases the aerials can be smaller and the power lower. It becomes safer (if there was a risk, which has never been proven) and less obtrusive.

With very small cells, then 24GHz, 47GHz etc can be used with TINY aerials and easily 200Mbps. Fibre between streetlights and each streetlight the ultimate microcell, tests are already running on these kind of systems.

I agree that the priority for Wireless is Mobile services. When TV goes 100% digital 760MHz to 864MHz should be removed from Fixed Broadcast TV and licenced to Mobile IP services. Also when GSM is closed, this too should be relicenced as technology neutral IP services.

750MHz to 1.2GHz is simply the best band for mobile services in terms of balance between cell size flexibility and building penetration. Aerials for the mobile "modem" are well established minature technology too.

probe

watty wrote:

Microcells use milliwatts. The more cell density increases the aerials can be smaller and the power lower. It becomes safer (if there was a risk, which has never been proven) and less obtrusive.

We don't know that low power is safer - particularly with digital radio (though intuitively I suspect it may be). We do know that long distance is safer.

With very small cells, then 24GHz, 47GHz etc can be used with TINY aerials and easily 200Mbps. Fibre between streetlights and each streetlight the ultimate microcell, tests are already running on these kind of systems.

Urban only solution. Requires laying of backhaul to each streetlamp. (Hanging fibre from lamppost to lamppost is 1950s, ESBish, oirish, sloppy, polluting - unacceptable today. Back to street dig-up. Or else some sort of linear "mesh" backhaul which increases latency, greatly lowers end to end bandwidth available to the user, introduces component failure reliability problems (eg the lamppost 6 node is out, therefore houses served by 6 to 99 are offline too, type of issue).

I agree that the priority for Wireless is Mobile services. When TV goes 100% digital 760MHz to 864MHz should be removed from Fixed Broadcast TV and licenced to Mobile IP services.

Ireland will be like Cuba on the analog TV front - it will go on forever due to RTE and government incompetence and procrastination and cluelessness. Much of the automobile stock on the streets of Havana are of 1950s US manufacture. Ditto for the analog TV stock in Ireland - it will be the only country using analog TV in another few years. The government will have to start up a factory to make analog standard definition TV sets to meet replacement demand as people's TVs break down!

.probe

watty

Long distance means HIGHER power and less safe for the user beside the modem.
We DO know that whatever the risk might be less power is safer. 800W of microwave at 4Ghz cooks you from the inside. 800W at 40MHz does not, nor does 800mW at 4GHz.

Street lights need cables (underground) and are easier to rewire than houses. Half as much digging at least. Did I suggest anywhere overhead wires?

The lack of a joined up thinking plan for DTT can be discussed here:
http://www.boards.ie/vbulletin/forumdisplay.php?f=56

Slutmonkey57b

probe wrote:

I don't live in Kenya, but this would not be a problem in the town I live in, where 20 years ago the powers that be had the foresight to create cut and cover type tunnels under the footpaths. [......]

.probe

May I be the first to say, in true Irish fashion... You Cun.t.

probe

watty wrote:

Long distance means HIGHER power and less safe for the user beside the modem.
We DO know that whatever the risk might be less power is safer. 800W of microwave at 4Ghz cooks you from the inside. 800W at 40MHz does not, nor does 800mW at 4GHz.

On the distance front I am taking things to their logical conclusion! E.g. someone on the Aran Islands is not going to be too worried (and I am sure that there is no health risk to them) from a GSM cellsite located in Dublin 4 running at twice the approved safe power output, situated next to a school, because the operator has difficulty getting other sites in area and is taking a chance on not being caught by ComReg. In other words when you are out of range, you are out of harm. Nothing more than that should be inferred.

.probe

probe

Slutmonkey57b wrote:

May I be the first to say, in true Irish fashion... You Cun.t.

May I say in true French fashion - je regrette de dire que je n'ai pas une notion de ce que vous parlez !

.probe