Defence of the Dark Fibre Arts: Ofcom delays plans to force BT to open its network Ofcom has delayed plans to force incumbent telco BT's Openreach to open up its dark fibre network. For now. The UK comms regulator confirmed it will not attempt to reintroduce a temporary remedy for Dark Fibre Access, which would have required BT to provide a restricted form of DFA for Ethernet until March 2019. Ofcom has …
This post has been deleted by its author
Street furniture? Probably not, or not in quantity. So I think it's unlikely it'll be mass-market/retail due to cost and hassle. Same with poles as that's a different set of PITA regarding access, plus boring bits like checking poles aren't rotted, safe splicing at height regs. Plus the fibre's more expensive than duct-friendly 'blow me' fibres. Plus the benefit of dark fibre is being able to run it at your own speed, so I'd want to be using it for PoP or backbone connections, ideally via in-span interconnects. Which means digging up roads to install chambers. Or smaller digs because for popular destinations, there's a high probability of BT chamber being near a competitors.
Which of course is why BT doesn't want this. Councils might, because it could mean less roadworks for civils to build parallel intfrastructure.
Focusing on lucrative areas? Of course. That's where the money is. Although depending on how the product ends up, it could mean OLOs can justify a business case to PoP smaller towns. Or if there's some high-bandwidth anchor customers that could justify the cost. For retail, it'd be a harder business case, ie trying to service the £9.95/month or less customer with a decent 1Gbps+ product.
Does it really surprise anyone that a watered down DFA product that is limited to 1Gbps has few takers when the product itself and its price is based on the 1Gbps EAD variants?
The whole reason a service provider wants dark fibre is so that they are NOT constrained in terms of bandwidth and can determine the speed of the circuit themselves through the choice of active equipment.
Openreach's argument in their appeal of the original (and much better) DFA determination that Ofcom had defined the market wrongly is perhaps technically right, but fundamentally flawed and should have been thrown out at the time.
Granted - the market for DFA in rural parts of the UK is commercially different from city centres, but so is the market for lit fibre services. And Ofcom, Openreach and the Service Provider community have all accepted that - despite those differences - EAD services at up to 1Gbps cost the same regardless of where they are delivered.
DFA was based on EAD 1G, i.e. the processes were to be broadly the same (without the active equipment), the pricing was based on it ('EAD 1G minus'), DFA was to be available wherever EAD 1G was available. The only difference is the potential to run greater transmission speeds at very little additional costs.
The courts - and the regulator - have missed an opportunity here - to accept for DFA the same market inequality as exists for EAD, using that acceptance as a precedent and force Openreach to provide the product that the whole service provider community has been waiting for for so long.
It looks like we will just need to wait a few years more...
I can't go into detail, but I agree with Flak fully.
I do have a tiny bit* of sympathy for Openreach - they have probably sized their fibre network on the assumption that they won't be selling bits of it as dark fibre, which is a very low margin product. Once you have used all the fibres in a bundle, laying a new fibre bundle can be almost as expensive as the first one. In theory, you just blow a new fibre bundle down the conduit - in practice, conduits collapse and get blocked, often due to HGVs, and in popular areas get full anyway, which is why (D)WDM makes sense.
The way IEEE and ITU standardization is moving, I expect in the long term all commercial premises will be fibre-connected and be able to buy services from a choice of providers over the same Customer Equipment - relevant standards are IEEE 802.1ag and ITU-T Y.1731, but operators and providers will take a very long time to implement such customer friendly services, and it may well take some regulatory prodding. The service provider and the network operator will be able to be different - much like you can buy power or gas from many companies, but it is delivered by a single distribution network operator to your premises. (If you are a really big power customer, in the right geographical spot, you can get power from two different distribution networks, but it isn't common. It is easier to get service from two different telecomms network operators.)
You can almost already do this: I have seen services provided by one provider where they provide their own (ADVA) CE, which is connected via a short length of patch fibre to BT Openreach's ADVA which uses BT Openreach's service to connect to the provider's PoP. The provider's ADVA was there so they could manage and monitor their service, and of course, BT Openreach's ADVA was there so they could manage and monitor their service. With the right standards and regulatory practices in place, you would not need to stack hardware in this manner.
*Only a very, very, tiny bit.
Ubiquitous fibre is the nirvana.
The further down the OSI stack you move, the harder it is to virtualise or disaggregate.
At Layer 3 this already exists (the Internet), with 'over the top' services completely independent of the underlying infrastructure (net neutrality is a discussion for another day).
I suspect the ADVA solution you are referring to being Layer 2 Ethernet.
When you get to Layer 1, infrastructure competition is really the only thing that provides true choice.
>"Openreach's argument in their appeal of the original (and much better) DFA determination that Ofcom had defined the market wrongly is perhaps technically right, but fundamentally flawed..."
Don't believe Ofcom's partisan propaganda. This wasn't BT v Ofcom, it was BT, Virgin Media and CityFibre v Ofcom: i.e. "engineers who actually invest in infrastructure" v "quango lawyers who pontificate about infrastructure". The Competition Appeal Tribunal ("CAT") aren't stupid: the panel comprised two experienced economists, and a high court judge. Ofcom didn't lose because of a technicality, they lost because they didn't do their job: they decided that they wanted to impose a dark fibre remedy on the market, in the face of evidence that it would severely damage infrastructure investments, and they 'made up the evidence' to try to bolster that position. Hence they lost.
"BT appealed on the grounds that...the dark fibre remedy was disproportionate, including because it would undermine infrastructure based competition for VHB services from providers such as Virgin Media and CityFibre. A hearing took place in April-May 2017 over sixteen hearing days, in which the Tribunal heard BT’s arguments in relation to market definition and the competitive core... The imposition of the DFA remedy was... scheduled for September 2017, to allow those issues to be considered, if appropriate, before the implementation date of the DFA remedy. Given that Ofcom’s market definition will now need to be reconsidered, the hearing has been vacated."
Source: https://www.monckton.com/cat-overturns-ofcoms-dark-fibre-remedy (accessed 13 April 2018)
Full details are on the CAT site: http://www.catribunal.org.uk/237-9285/1260-3-3-16-British-Telecommunications-.html
The full, 199-page, judgment is here: http://www.catribunal.org.uk/files/1260_BT_Judgment_CAT_25B_101117.pdf - a right riveting read! Perhaps one of the weekend ;)
IOW it's not OFCOM f**king up as usual its the other broadband suppliers turning their noses up at this.
And they're right. It still leave OpenReach in the driving seat.
The real f**kup was OFCOM's failure to construct a legal argument that would stand up in court.
Given how much control they were asking BT to give up a court case was inevitable.
Let's see if OFCOM can do better next time (and how far down the road will that be).
It was their first time actually in court and I am sure it was all new and scary with that bloke in the wig being very unhelpful by not telling what they should to do, well that is what they think but they will have to ask BT if they are correct once they start answering the phone again
This post has been deleted by its author
'Given how much control they were asking BT to give up?'
What we need to do is set up an organisation with the power to tell BT and other providers what to do. We could call it the Office for Communications or OFCOM for short.
Do BT vet OFCOMs HR choices FFFFS!
Ofcom's job isn't to tell companies what to do, it's to regulate the market. If the government wants to tell industry players what to do, step one on the list is "don't sell the industry you already own".
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
