Friaco pricing.

Fergus

The big step forward with FRIACO is that eircom lose the ability to time-charge use of their network.

Your ISP buys a fixed amount of circuit capacity, the ability to cope with X calls at once, for a fixed fee from eircom. And thats all eircom has to do with it. Peak or Off-Peak doesn't come into it, because the lines are assigned wholly to the ISP at all times.

At present, all calls, Internet and voice alike, share the one eircom network, and eircom can argue that they need to charge more in business hours to avoid over-use.

With FRIACO, too many calls to the ISP just causes the ISP's pre-purchased quantity of FRIACO lines to jam up.. it doesn't overflow into the rest of eircom's network, so eircom have no basis to impose their own usage-related pricing structure.

Basically, the ISP not eircom now has the freedom to decide the balance of price vs quality of service they want to sell at.

zynaps

Originally posted by Dangger Over use and abuse (such as just leaving a computer conected all day on dial up)[/B]

Err, how is that overuse or abuse?
That's the whole point of flatrate...

Would it also be overuse or abuse to leave your dsl/cable/wireless connection online all the time?
Because I believe that's how it tends to work.

If friaco comes out, and I have no feasible broadband options available, I'll probably get a second phone line/ISDN line and leave the PC on all the time on whatever flat rate package works well.

Also, someone mentioned earlier that the majority of people would tend to use the net in the evening anyway, as they're in work.
Perhaps this way they can have internet access in work, without having to pay multiples of hundreds per month for satellite or leased lines or something.

Why go sparingly on it? It's not gold or something...

zynaps

SkepticOne

Originally posted by zynaps
Err, how is that overuse or abuse?
That's the whole point of flatrate...

Would it also be overuse or abuse to leave your dsl/cable/wireless connection online all the time?
Because I believe that's how it tends to work.

Main difference between broadband (dsl/cable/wireless etc) and PSTN (56k and ISDN) is that PSTN uses up resources even when it is not transferring data. Flat-rate 56k and ISDN will emerge from FRIACO provided the wholesale price is low enough and the ISP market is competitive enough, but it will never be "always on". ISPs will always have ways of discouraging heavy use. They will still have to pay for the capacity that allows you to stay on.

zynaps

How does PSTN use more resources than other systems? Is it actually expensive for them to have someone online?
(I'm asking, not challenging what you said) :P

SkepticOne

Originally posted by zynaps
How does PSTN use more resources than other systems? Is it actually expensive for them to have someone online?
(I'm asking, not challenging what you said) :P

It is because a dedicated 64k channel from the ISP's modem to your modem needs to be kept open regardless of whether or not traffic is being transferred over it. The ISP, through FRIACO, will be buying a fixed block of such capacity from Eircom. In order to serve as many people as possible, they will want moderate use from as many people as possible. Extreme users will take up more of this fixed capacity and measures will be taken to curb such use if necessary.

Here's a simple analogy. You are sharing a house with friends. In order to save money, it is decided to by, say, breakfast cereal in bulk. Each person in the house pays a few quid a month and a big bag of cerial is bought and made available.

So long as the people in the house each eat roughly one bowl a day, everything is fine and there is no need to 'meter' the cereal consumption. (in a commercial environment there would be costs associated with such metering and billing).

Should someone arrive, however, and eat significantly more than others, then the whole thing breaks down, and the tennants then need to keep track of how much everyone is using.

Therefore flat-rate internet via FRIACO depends on everyone being considerate to others in the use of limited resources.

Another analogy is the all-you-can-eat buffet. To stay open, the buffet needs to attract as many non-gluttens as possible. If they only people going to such establishments are people who eat eight plates worth, the establishment would quickly shut down.

Fergus

Note that flat-rate ISPs in the UK and elsewhere actually have a ratio of several users to each FRIACO line they possess.

If you look at the wholesale price for FRIACO circuits you will probably find that is it higher than the monthly customer retail flat-rate service. AFAIK, in the UK a basic FRIACO port wholesale costs the ISP about STG27 per month (more for tandem), but of course the retail offerings are more like STG14/month.

This is because the ISP operates on the basis of only 10% or 20% of users being online at any one time, so for example if they have 1000 users they only actually rent 200 FRIACO lines.

This is why flat-rate dial-up never actually means always-on. The point is you can dial up anytime and stay on as long as resources allow, and for a fixed monthly fee. If the ISP did find they had a lot of users who were doing keep-alives etc to keep the circuit open all the time, they would have to do something about it.

De Rebel

Originally posted by Fergus
........ per month (more for tandem), but of .......

Fergus,

There is a lot of coverage of "Single Tandem FRAICO" in the ComREG Directive, especially the fact that it will not be provided at this point.

Could you explain in 2 or 3 paragraphs or point me to an explenation of the advantages of "Single Tandem FRAICO", the implications for users given its non availibility here and the potential cost considerations.

Sorry if I am the only one on the palnet that does not understand the issue, but ComREG appears to make a big deal about it, without actually explaning the issues.

Thanks

apologies if this is OT, but it seemed appropriate to follow on from the previous post

Fergus

Well, I'm no specialist and I apologise to telecom engineers if I'm making a blithering idiot of myself, but anyway..

In a simple sense, everyone's phone line is hooked into a local telephone exchange. These are primary exchanges. I think there are around 47 of them in Ireland. There may be smaller remote exchanges and so on, but everyone essentially has their own primary.

Then there are tandem exchanges. These are exchanges that connect the primary exchanges together. I think there are around 13 tandems, and they are heavily interconnected. They are sort of the 'core' of the voice phone system.

Every local primary exchange is connected to the tandem network.
If you make a call to someone who is not connected to the same primary exchange as you, the call goes via the tandem exchanges out to the other primary exchange you are trying to reach.

Now, this is important in regard to FRIACO. The current FRIACO product being proposed only allows ISPs to connect to their user's calls via the primary exchanges. The calls are not allowed to use the tandem network. This means that if you make your FRIACO call to your favorite ISP, they must have their office hooked in directly at your primary exchange, otherwise you can't reach em. With this type of FRIACO, an ISP that wants national coverage needs to have their own links directly into each of the 47 primary exchanges. As they are picking up their calls from the eircom network at the exchange where they originate, so eircom don't have to worry about the call needing to be routing through their precious core tandem network.

If the ISP can't reach every exchange, they can pay eircom extra for carrier services.. sort of like hiring leased lines or dedicated capacity through eircoms network, which would allow them to bring connections from the primary exchanges they can't reach directly right back to their offices. However, this obviously adds a lot to their running costs.

This was the original type of FRIACO available in the UK. However, pressure from ISPs led OfTel to get BT to introduce Single Tandem FRIACO. This means the ISP only has to hook in at one of the Tandem exchanges in the core network, and from there they can pick up calls from any primary exchange. This is obviously a lot simpler for the ISP. The cost of these type of FRIACO circuits though is higher, because obviously more resources within the telephone network are being allocated for each call.

The low number of primary exchanges in Ireland compared to UK (they have something like 800+ I think) may make things simpler here. I don't know if UTV/Esat have interconnect capability already in place at all the primaries. So it's hard to know what impact lack of Tandem FRIACO will have.

De Rebel

Originally posted by Fergus
Well, I'm no specialist and I apologise to telecom engineers if I'm making a blithering idiot of myself, but anyway..

Fergus

That explains it very well thanks, and the wording of the directive now makes a lot more sense. It looks like this must have been the key sticking point which caused the breakdown of the pre Christmas discussions.

There is an appendix of exchanges in the Directive, it looks like there are 16 Tandem and 48 primary. Therefore by forcing the OLOs to connect to the primary exchanges they need three times the number of connections, or else pay eircom to carry the transit traffic on the core. No wonder Eircom banned them from the Tandem Exchanges to "protect the integrity of the core network"

Do I smell a rat or what..........

pork99

The current FRIACO product being proposed only allows ISPs to connect to their user's calls via the primary exchanges.

Are they going to be allowed to get away with this sh1t?

Fergus

The same arguments were made by BT in the UK, but OfTel got an independent panel of experts to look into the situation. IIRC, they essentially concluded that while there were issues with tandem overload, it was quite possible for BT and the ISPs to plan around it in advance and invest what was required.. on the basis that FRIACO demand would never get too out of control as users (and the voice service itself) would be switching to real IP / broadband networks in the short to medium term anyway.

That was over two years ago and I don't recall the British telephone system since coming crashing down due to epidemics of linux distro downloading

The report from OfTel on tandem capacity etc is
here.

"BT Internet customers have enjoyed unmetered surfing since February [2000], and we have no intention of taking that away. In each case, people are paying a fair price for a good unmetered deal." "unmetered Internet access is entirely possible if you have a business model based in the real world."

British Telecom August 2000

Fergus

Originally posted by pork99
Are they going to be allowed to get away with this sh1t?

It's not that bad. The critical point is to get any kind of FRIACO introduced as this will mean for the first time, eircom have to sell wholesale fixed circuit *capacity* as opposed to charging for *time*. This is a huge step forward.

There is a much clearer relationship to true running cost when you are talking about providing a pre-agreed fixed capacity within a single exchange (FRIACO), as opposed to the crazy situation to date where eircom come up with this vague per-minute charge based on all sorts of possible costs throughout their network. I hope this will mean it's not so easy for eircom to try on their usual crazy wholesale over-pricing.

BoneCollector

as opposed to the crazy situation to date where eircom come up with this vague per-minute charge based on all sorts of possible costs throughout their network

I think this is whats known as the errcom Random Number Generator
Which is soon to lose it Mystical Powers over the internet user

Richard Barry

It seems to me that eircom have managed to pull the wool over ComReg’s eyes.

To quote from ComReg0302:

“In reviewing the draft product description containing the OLO requirements, some serious technical difficulties surrounding the implementation of Single Tandem FRIACO were identified. These difficulties could potentially adversely affect the integrity of the eircom network to the detriment of other users. eircom proposed an alternative, but this proposed solution was not economically efficient”.

Good old FUD (fear, uncertainty and doubt). It served IBM for decades and kept their profits sky high.

If 1,000 ISP customers move from 1891 access to 1893 flat rate, the same tandem capacity that hitherto served the 1891 traffic will surely be automatically redeployed to handle the 1893 traffic. The only problem that has to be managed is people staying online longer under 1893 than when they were on a pay as you go basis. This will generally arise during peak hours (because there should be spare transit capacity in the network outside the 09-17h Mon-Fri timeframe).

A problem of this nature can surely be managed as it arises by putting a temporary local ceiling on the number of tandem channels used by 1893 dial-up in exceptional circumstances where insufficient capacity exists in the network fabric. At the same time the ISPs in question and eircom need to bang heads together to sort the problem out (in terms of adding capacity or whatever).

In any event, it still escapes me why the parties are bothering with the FRIACO route at all. The telecommunications industry worldwide is going IP – which means DSL. It provides a better quality solution for the end user and does not tie up switching equipment. The way FRIACO is progressing in Ireland, it will be just as geographically onerous for an ISP to provide coverage on an unmeasured dial-up basis as it is to roll out DSL.

ComReg would surely be better off mandating more favourable tariffs for wholesale bitstream – based on a range of DSL speed options and flavours (say from 128k upwards). The current wholesale tariff in Ireland (which only covers the segment between the local eircom POP and the customers’ premises) is more expensive than the end to end retail tariff (which includes the ISP feed) in most other countries. Something radically wrong somewhere with the costings?!

Finally, ComReg 0302 doesn’t mention the word “ISP” once. Presumably a reflection of a vision of keeping the Irish ISP business a nice closed shop for OLOs (ie licensed telcos).

R

PS
BT charge GBP 507 per annum per single tandem FRIACO port – which is effectively an ISDN channel in terms of capacity. An ISP with 100 ports (channels) could serve between 500 and 800 customers. Based on 800 sharing 100 ch., the cost to the ISP per customer per month in real money would be about EUR 8 (for the single tandem FRIACO connection).

Fergus

I agree that the Eircom argument against Tandem capacity is FUD. Remember, they've managed it in the UK with a network many times more complex.

A direct move to IP would be better, but I think it assumes eircom would act outside of self-interest. Their bitstream price is forced into place in part by a lot of legal pressure on the regulator, not reality.

FRIACO critically is required to break the costing structure that sustains their per-minute / per-megabyte monopolistic profits on the existing PSTN. One of the main reasons I think the i-Stream product is so expensive is because they are attempting to maintain an equivalent cost per downloaded MB to that of dial-up, so as not to risk any impact on their excessive usage revenue.

The other thing about FRIACO (particularly ST Friaco) is that it creates a chance for competitors to enter the market, because it means far less network build investment for them to get started.

I guess essentially when new telecoms technology delivers 10 times the performance for the same or less money than before, eircom always try to charge 10 times as much anyway. It's just one long-running con trick and its time is up.

SkepticOne

Originally posted by Richard Barry
In any event, it still escapes me why the parties are bothering with the FRIACO route at all. The telecommunications industry worldwide is going IP – which means DSL. It provides a better quality solution for the end user and does not tie up switching equipment. The way FRIACO is progressing in Ireland, it will be just as geographically onerous for an ISP to provide coverage on an unmeasured dial-up basis as it is to roll out DSL.

ComReg would surely be better off mandating more favourable tariffs for wholesale bitstream – based on a range of DSL speed options and flavours (say from 128k upwards). The current wholesale tariff in Ireland (which only covers the segment between the local eircom POP and the customers’ premises) is more expensive than the end to end retail tariff (which includes the ISP feed) in most other countries. Something radically wrong somewhere with the costings?!

FRIACO (and the flat-rate products based on FRIACO) and DSL compliment each other, in my view. One of the characteristics of the Irish market is a generally very low internet use as pointed out by Eircomtribunal. Whilst some telcos may say that this is due to "lack of compelling content" or some other rubbish, it is clear to most of us that it is the high cost of staying online for even reasonable amounts of time. This is despite the quite high ownership of computers with internet connections in Ireland. We must therefore grow the number of active internet users in Ireland and this is best done initially with narrowband flat-rate services as most people own computers with built-in modems.

One of the things that telcos bang on about (one telco in particular) is the lack of demand for broadband and how the tax-payer needs to stimulate it with subsidies. I don't agree with this, but assuming for the sake of argument that it is the case, then, flat-rate narrowband will stimulate this demand in the long run simply by educating people to the possibilities of the internet. The internet itself is the best teacher of the uses of the internet. Demand will therefore be increased for DSL and the lack of demand argument will be removed from the telco's repetoire of excuses.

To compliment FRIACO, there needs to be an alternative for the more experienced users to move over to. DSL will fill this need once the price comes down to realistic levels. There is no reason that work can't begin on this now.

Richard Barry

Originally posted by Fergus
I agree that the Eircom argument against Tandem capacity is FUD. Remember, they've managed it in the UK with a network many times more complex.

A direct move to IP would be better, but I think it assumes eircom would act outside of self-interest. Their bitstream price is forced into place in part by a lot of legal pressure on the regulator, not reality.

FRIACO critically is required to break the costing structure that sustains their per-minute / per-megabyte monopolistic profits on the existing PSTN. One of the main reasons I think the i-Stream product is so expensive is because they are attempting to maintain an equivalent cost per downloaded MB to that of dial-up, so as not to risk any impact on their excessive usage revenue.

The other thing about FRIACO (particularly ST Friaco) is that it creates a chance for competitors to enter the market, because it means far less network build investment for them to get started.

I guess essentially when new telecoms technology delivers 10 times the performance for the same or less money than before, eircom always try to charge 10 times as much anyway. It's just one long-running con trick and its time is up.

>>>A direct move to IP would be better, but I think it assumes eircom would act outside of self-interest. Their bitstream price is forced into place in part by a lot of legal pressure on the regulator, not reality.

Quite. But two issues still remain:

1) The new authority (ComReg) needs to revisit wholesale bitstream pricing and justify to the public why Irish wholesale prices for 35% of the broadband product exceed typical European retail pricing for 100% of the product.

2) Scale down the wholesale bitstream speed to FRIACO type levels (eg 128k – 2 channel isdn) and scale down the cost pro rata.

It should deliver an IP platform that provides far better value to the consumer, has a longer shelf life for the ISP and has no network congestion implications for eircom – FUD or otherwise.

>>>The other thing about FRIACO (particularly ST Friaco) is that it creates a chance for competitors to enter the market, because it means far less network build investment for them to get started.

It would be nice to think that was the case. However if you or I decided to go into the internet business tonight, under the current FRIACO proposals, where would the cost savings lie? (compared with promoting a DSL platform as the minimum entry point) and ignoring the fact that we haven’t a clue about costs because despite years of timewasting “negotiations” and regulatory whatevers eircom have managed to keep the crunch issue of cost out of the picture even at this late stage?:

1) We would have to buy or create IP transportation capacity into every eircom switching entity in the country to offer nationwide FRIACO. We’d have to do much the same to offer a DSL based service. A non trivial, competition-eliminating proposition whichever way you look at it. You might say lets concentrate on area X for starters – but this makes marketing difficult. You run an advert on virtually any medium and you will get lots of punters who are out of your coverage area who will cost you money to say no to and who themselves might feel pissed off at your inability to deliver.

2) Where is FRAICO cheaper than DSL under the current proposals? The cost of the modem basically. Not that a DSL modem is more expensive than a 56k dial-up unit but the customer has to upgrade his modem to a DSL box and a dedicated “modem” (ie the card in the DSLAM) is required at the service provider’s end. But surely these one off costs are trivial compared with what people are over paying month in month out (EUR 40 – 500+) for an appallingly slow, second rate, analog dial-up product that delivers about 20% of the potential value of the internet? And if you have only one PSTN line, everyone else in the house is offline (voice or internet) while you are surfing in non-DSL mode.

If you don’t find it economic to pay the DSL modem costs in your usage circumstances you are almost certainly better off on balance staying with a pay as you go dial-up option. I’m not trying to force DSL down everybody’s throat. I do feel that few people have thought through the FRAICO logic except for eircom. And we know the bias that their thought patterns have followed. DELAY being #1. A welcome sideshow to take the spotlight off DSL. And as I stated in my previous posting, NO ISP NEED APPLY. ComReg appear to have locked them out.

>>>I guess essentially when new telecoms technology delivers 10 times the performance for the same or less money than before, eircom always try to charge 10 times as much anyway. It's just one long-running con trick and its time is up.
How much would the 2,6 Ghz Intel Pentium processor cost today if its cost was based on the speed, power and price of the original 8086/88?

Neither ComReg nor eircom would appear to have heard of Moore’s law!

R.

De Rebel

Fergus and Richard thanks for a most enlightening sharing of information.

This sounds like trouble. Re-reading COMREG0302, the second page of Section 1(Background) makes worrisome reading….. There are four paragraphs on the second page: I am inserting my conclusions – have I got it correct?

In reviewing the draft product description containing the OLO requirements, some serious technical difficulties surrounding the implementation of Single Tandem FRIACO were identified. These difficulties could potentially adversely affect the integrity of the eircom network to the detriment of other users. eircom proposed an alternative, but this proposed solution was not economically efficient.

Sounds like ComREG are accepting an argument which OFTEL have dismissed. Round 1 to Eircom

The restrictions required to ensure network integrity in the proposed product description for the implementation of Single Tandem FRIACO would have led to a product offering which would have prohibited new entrants from availing of the service. The necessary commercial constraints which the product description would have imposed would have made the product non-commercially viable.

ComREG recognises the implications of loosing Round 1, but is unable to do tackle the issue head on (Does that sound familiar) Round 2 to Eircom

In light of the of the above technical and commercial constraints surrounding the Single Tandem proposed product description ComReg believes that it is more practical and expedient and in the interests of users, to intervene in the negotiations and introduce FRIACO on a phased basis, by freezing the product description for phase 1(Primary level FRIACO) in the form set out in Annex 1 of this Decision Notice. ComReg considers that the product description in Annex 1 will allow operators connected at Tandem level to offer an equivalent service to those connected at primary level by utilising Interconnect Extension Circuits. ComReg will also continue to work with the industry in seeking to introduce a commercial and technically viable product offering for Single Tandem FRIACO.

The parts in italics (my emphasis) is the Irish/ComREG solution to the Irish problem. A competitor, be it OLO or new entrant ISP, will have three options (1) offer a limited service restricted to a few of the 48 primary exchanges loosing economies of scale and making marketing and product identification a nightmare (2) Connect to every one of the 48 Primary Exchange – more money for Eircom, and a major obstacle for new entrants (3) Connect to a Tandem Exchange and use ICE to transit the core network as set out in Appendix 1 of COMREG0302; presumably paying Eircom mucho zillions for the privilege. In other words a convoluted mess, enabling the incumbent to disincentivise new entrants and monopolise the core network to strip the OLOs of the advantages which Competition and Market Liberalisation were to confer on them. Round 3 to Eircom

ComReg, recognising the benefits that will arise from the delivering of a FRIACO product to the marketplace in a timely fashion and for the clarity of all parties, sets out the complete process, with dates, for the development of inter-operator processes, terms and conditions and cost orientated pricing associated with the product description in Annex 1 of this Decision Notice.

Here comes another half assed, half baked, over priced product; it’s the best we an do for ye, jeeze we’d love to do better, but sure wouldn't half a loaf better than no bread,
I won’t declare Round 4 until I hear what the wholesale price is set at…...

Originally posted by Richard Barry
BT charge GBP 507 per annum per single tandem FRIACO port – which is effectively an ISDN channel in terms of capacity. An ISP with 100 ports (channels) could serve between 500 and 800 customers. Based on 800 sharing 100 ch., the cost to the ISP per customer per month in real money would be about EUR 8 (for the single tandem FRIACO connection).

That’s superb information Richard. Are those ratios (100 ports:500/800 customers) the norm in the UK? Have you any more of this price/ratio information in relation to other countries?

Regarding the point that we would all be better employed focusing on DSL rather than analogue based solutions: there really is no argument. Its pathetic that in this country we are getting our little hearts into a flap over FRAICO in 2003. But the wholesale Bitstream pricing has eliminated any likelihood of competition/realistic pricing for DSL. What’s ominous, if my reading of the ComREG document is anywhere near the truth, is that wholesale FRAICO could be going exactly the same way.

Sorry if I sound critical and/or cynical and while I do acknowledge and commend the efforts of the Minister and ComREG, this does not sound too promising.

What chances of wholesale FRAICO being charged at Eur 8 per customer here…………

zynaps

A bit OT, but if it is inefficient to have FRIACO going through the whole network of exchanges if necessary, would it be possible for an OLO with dslams in the exchanges to somehow stick multiple dialup connections into a dsl, and thus remove the pstn traffic?
Since their dsl ports may have enough bandwidth to carry many dialup lines, there would only be traffic from the user to their own exchange if they could have some conversion thing where 20 dialup lines are served by one 1024k port, or whatever...

Yes, I don't know what I'm talking about, but it seemed like a good idea at the time....

zynaps

Fergus

When FRIACO was originally introduced in the UK it was on the same primary exchange only basis as being proposed here. The regulator recognised this was a problem and tandem interconnect was going to be needed. At the start of 2001, tandem FRIACO was introduced. 'Stage 1' was to allow tandem FRIACO, with a view to moving traffic off it onto an IP network in 'Stage 2'. Here are some points made at the time (DMSU = a type of tandem):

28. During the Director’s consideration of Worldcom’s request, BT submitted that the expected substantial increase in traffic, which would result from unmetered interconnection for Internet purposes, would exceed the capacity of the Public Switched Telephone Network ("PSTN") as currently configured. In particular, the DMSU switches would not be able to handle the expected increase in traffic;

29. In view of BT’s arguments the Director sought independent technical expert advice on the options available for investment to enable unmetered interconnection at the tandem layer, that is, the layer of the PSTN which carries trunk traffic which is switched by a Tandem Switch;

32. Having considered the independent technical experts’ and consultants’ advice, as well as the further submissions from BT and other operators, the Director is now in a position to amend his Direction of 26 May 2000 to provide for interconnection at the Tandem Switch;

44. Stage 1 lasts until 31 January 2003 by which date BT could reasonably be expected to have expanded its network to deal with the expected increase in Internet Traffic. During Stage 1 BT shall be obliged to provide FRIACO switched through any of BT’s Tandem Switches, to a Point of Connection situated at that Tandem Switch, that is single tandem FRIACO;

45. The Director considers that in Stage 1 if Internet Traffic continues to grow as estimated there could be a shortage of capacity at one or more Tandem Switches and that management of such capacity is therefore required;

46. Such management can be achieved by reasonably rearranging the traffic of those operators requesting or being supplied with single tandem FRIACO, pursuant to this Direction. This will create additional capacity at the Tandem Switch. Such rearrangements are also necessary to ensure that network integrity is maintained;

47. In order to encourage operators to complete the required rearrangements set out in this Direction it is reasonable and appropriate to provide for the ability of BT to cease its provision of single tandem FRIACO to operators which fail to complete the required rearrangements;

53. However, due to the limitation of capacity at BT’s Tandem Switches, the Director considers it reasonable to place a maximum limit on the number of ports per operator per Tandem Switch, in order to reduce the likelihood that any one operator is able to obtain all the available single tandem FRIACO ports at the more popular Tandem Switches. This limit may be revised by the Director;

The full directive is available here.

Richard Barry

>>>But the wholesale Bitstream pricing has eliminated any likelihood of competition/realistic pricing for DSL.

At the expense of repeating part of an earlier posting (but it is an important issue IMHO) I believe that the wholesale bitstream pricing should be opened up again, by the new authority, for the reason explained earlier.

AND the wholesale offering should be expanded to cover slower speed offerings with a pro-rata reduction in the cost – eg wholesale 256k should cost about EUR22 per month and wholesale 128k should cost about EUR 11 per month and wholesale 64k should cost about EUR 6 per month – based on the existing offer for 512k.

The pricing of ESAT BT’s 256k proposed offering compared with the pricing of their 512k offering is a precedent for this. In France, Detusche Telekom is offering 128k DSL to residential subscribers for EUR 15 per month (TTC U/T) on special offer at the moment.

France Telecom’s own Wanadoo ISP offers 64k service to cable customers for EUR 25 per month (TTC U/T) (which is expensive in my view, given the alternatives). Wanadoo 128k is EUR 30 on DSL (TTC U/T). Best speed/value in France is free.fr EUR 29,99 per month (TTC U/T) for 512k service.


>>>>What chances of wholesale FRAICO being charged at Eur 8 per customer here…………

One has to look at FRIACO pricing differently to DSL. Under FRIACO the ISP rents an ISDN channel for a fixed monthly price from the incumbent. The only difference between this ISDN channel and a regular retail one is that it has a free phone tariff for incoming calls originating within a specified geographic area. The geographic area applying under the current ComReg proposals will be relatively small. Therefore an ISP is going to have to roll out lots more points of presence. They went through the same scenario in GB under v1 of FRIACO and it didn’t work because few ISPs had the resources to put modems “in every exchange area”.

OFTEL subsequently mandated single tandem FRIACO offer under which BT offered free incoming calls to FRIACO ports from a much wider geographic area for a slightly higher fee per port (ie about EUR 800 per annum equivalent per ISDN channel). The ISP needed fewer points of presence to deliver a national service as a result.

The ratio of ISDN channels to subscribers signed up is a matter for the ISP. It is no different to the existing situation. If you were to knock on ten neighbours doors (who you knew were internet users) – chances are you will find one or at most two online at any point. So statistically speaking with your ISP’s hat on, if you were offering them a FRIACO service you would only need 1 or 2 channels to service your ten neighbours. Same logic applies to an office phone system. If you have say 100 extensions you might need 10 “outside lines” – perhaps 15 or 20 depending on your type of business. Obviously a lot more if you were running a call centre.

The FRIACO wholesale cost per customer therefore can’t be precisely determined until you know what modem to subscriber ratio will work for your client base.

If you compare the costs of BT FRIACO which provides a 64k internet connection with DSL it is truly expensive and bad value for money. Which brings me back to my original point. DSL is the only way to go!

R.




TTC = including VAT
U/T = unlimited traffic

sceptre

Originally posted by Richard Barry
At the expense of repeating part of an earlier posting (but it is an important issue IMHO) I believe that the wholesale bitstream pricing should be opened up again, by the new authority, for the reason explained earlier.

Agreed - broadband technology is probably the future of the entire network rather than just for the few who really want it. More efficient from the telco end, more possibilities from the customer end.

bk

I completely agree with Richard Barry that broadband and IP based platforms are the future.

However it takes years for anything to happen in Ireland and for this reason there is a need for FRIACO.

1) Many people, particularly in rural areas are too far from an exchange to use ADSL (maybe we will get other xDSL tech in the future that will make it possible), so for these people FRIACO will be the only option (well also Satellite and Wireless perhaps).

2) IMO the most important reason for FRIACO is a stick to beat Eircom, as we are all aware Eircom makes a vast amount of money from PSTN and ISDN services due to the per second billing. This is the main reason Eircom have been slow to roll out ADSL, it will destroy the time based revenue model.

However with FRIACO rolled out, it will remove Eircoms time based model anyway and therefore I believe we will see them push ADSL more heavily then.

Brian

SkepticOne

It is obvious from an engineering point of view that DSL is more efficient. That is not the point here. The point is that even in countries with widespread affordable broadband (not Ireland obviously), most people stick with dial-up because it suits their needs as they see it. This will change with time, but right now people are happy.

In Ireland, because of the expensive metered service people have had to put up with, most people don't bother with the internet at all - even those with internet connections. This cuts them off from a vast amount of information that has been developed outside of Ireland but is charged for excessively by monopolistic telcos who did not contribute to its creation.

Whilst work appears to be underway on flat-rate, we must make sure it is done properly and whatever products emerge will be affordable and attractive.

This does not mean that pressure should not be brought to bear on excessive DSL prices and associated bitstream prices, but it is not a simple case of bitstream vs. FRIACO or DSL vs. flat rate. These two services cater to vastly different markets.

neverhappen

“In reviewing the draft product description containing the OLO requirements, some serious technical difficulties surrounding the implementation of Single Tandem FRIACO were identified. These difficulties could potentially adversely affect the integrity of the eircom network to the detriment of other users. eircom proposed an alternative, but this proposed solution was not economically efficient”.

..

If 1,000 ISP customers move from 1891 access to 1893 flat rate, the same tandem capacity that hitherto served the 1891 traffic will surely be automatically redeployed to handle the 1893 traffic. The only problem that has to be managed is people staying online longer under 1893 than when they were on a pay as you go basis. This will generally arise during peak hours (because there should be spare transit capacity in the network outside the 09-17h Mon-Fri timeframe).

How close is the network to exceeding its limits at the moment ?

Is there really going to be that much of an increase in traffic during the working day because of flat-rate? I would have thought that, for daytime users in general, they will use the connection more or less the same (as dictated by their work requirements) but with "certainty of cost".

What about all the daytime voice traffic thats moved to the two mobile phone networks ?

Or does eircom suggest that all that "compelling content" will suddenly appear on the internet, coincidentally on the same day as flat-rate leaks onto the market ??

For off-peak (home) users, on-line time will increase but there should be plenty of capacity at those times.

Do COMREG currently, or have any plans to, use independent technical experts to examine any of these claims ?

Richard Barry

Originally posted by bk
I completely agree with Richard Barry that broadband and IP based platforms are the future.

However it takes years for anything to happen in Ireland and for this reason there is a need for FRIACO.

1) Many people, particularly in rural areas are too far from an exchange to use ADSL (maybe we will get other xDSL tech in the future that will make it possible), so for these people FRIACO will be the only option (well also Satellite and Wireless perhaps).

2) IMO the most important reason for FRIACO is a stick to beat Eircom, as we are all aware Eircom makes a vast amount of money from PSTN and ISDN services due to the per second billing. This is the main reason Eircom have been slow to roll out ADSL, it will destroy the time based revenue model.

However with FRIACO rolled out, it will remove Eircoms time based model anyway and therefore I believe we will see them push ADSL more heavily then.

Brian

SHDSL is available now. Range 12kms (up to 20kms with range extenders). Street cabinet based DSLAMs will also greatly enhance the range and speed of this (and any other xDSL) technology.

http://www.rad.com/Article/0,6583,10812,00.html

SHDSL would cover perhaps 98% of flat rate broadband requirements in Ireland.

The other 1-2% of long loop basket cases could be served by a flat rate analog dial-up FRIACO for “basket cases” - (ISDN wouldn’t work for these out of the way subscribers). Or wireless. These top-up FRIACO customers would be very dispersed and could not cause any overloading problems for eircom’s network.

While eircom wholesale bitstream charges remain artificially high, ISPs will be forced to charge high prices at the retail end.

If/when cheap FRIACO arrives, eircom will compete by offering cheap FRIACO itself to get more of the margin and stop customer migration to “OLOs”. I can’t see them cutting DSL prices because their mindset is more scared of DSL than flat rate dial-up. Don’t forget that rural customers will be the last to receive FRIACO because under the current proposed arrangements the roll out is just as “tedious” as a DSL rollout.

We are arguing about crossing various hurdles on the path to cheap ubiquitous DSL.

The only way in my view that we will get cheaper DSL in the final analysis is by regulatory action to slice and dice the wholesale bitstream offer. The logical time for this is now following the setting-up of ComReg.

If it is not done now, it will look as if the new authority is accepting the inherited arrangements by default.

R.

Richard Barry

Originally posted by neverhappen
How close is the network to exceeding its limits at the moment ?

I would doubt if there is much if any congestion at present. There may be a problem at one or two exchanges. As I indicated earlier a mechanism could be put in place for participating ISPs to financially or otherwise solve the problem with eircom.

It is surely a simple matter for ComReg to ask eircom to produce network saturation reports for switching entities that are in eircom's view likely to pose a problem?

If they refuse to/can't won't they are surely pulling the wool!

R.

SkepticOne

Originally posted by Richard Barry
SHDSL is available now. Range 12kms (up to 20kms with range extenders). Street cabinet based DSLAMs will also greatly enhance the range and speed of this (and any other xDSL) technology.

Should ComReg be making Eircom provide this service?

highlight

Eircom do provide a wholesale offer for it

bk

I agree with everything you say Richard, I just believe there is a place and need for FRIACO.

I'm sure that as soon as FRIACO is in the bag, IOFFL members and committee are going to go after the broadband issue 100%.

Up until now IOFFL has been campaigning for two things, unmetered internet access and affordable broadband. And as such has been dividing its resources between these two issues.
However if we get a well priced FRIACO product then IOFFL will be able to turn its considerable resources to the bb problem exclusively.

One thing that people haven't realised about the introduction of FRIACO is the MASSIVE benefit it will have for IOFFL. IOFFL have done what many people thought impossible, all the parties involved and the press now see IOFFL as a very professional and powerful organisation. Also IOFFL has gotten a massive amount of press from this and you can be sure that these new contacts in the press will be used to our full advantage in the campaign for BB.

We are nearing the end of round 1, we have Eircom on there knees, we just need to make sure that the knockout punch is delivered (A fair priced FRIACO product actually available to people).

We will be entering round 2 with a great deal more confidence, experience, respect, press and complete focus on bb

Richard we will need people like you and your ideas for this new battle.

Brian