The Northern Ireland Next Generation Broadband Project ( NGBP ) administered by the Department for Enterprise, Trade and Investment ( DETINI ) resulted in the award to BT Group of the contract to install FTTC equipment to provide local-loop data rates in excess of 2 mbps to 85% of Northern Ireland businesses. Unfortunately the basis on which this award was made was flawed and has resulted in the strengthening of an incumbent Communications Provider and the entrenching of life-limited technology.
Nestled in the paragraphs of Northern Ireland Executive's Investment Strategy 2008-2018 is an aspiration that the Province should remain abreast of technological developments in "speed and capacity"1. The Executive has had a reasonably progressive attitude to broadband deployment, with the award of a contract to BT Group in 2005 for the deployment of ADSL to all 191 exchanges taking fixed-line broadband penetration to 99% which is unprecedented in the UK.
It was with this foresight that the aspiration was given form with the publication of Tender 103380-2009 in April 2009 which solicited bids for the implementation of "next generation broadband speeds to 85% of businesses by 2011"2, an admirably short timescale compared to many network technology projects.
Th tender required the selected vendor to roll-out, operate and maintain the network on a retail basis and to provide wholesale access thereto, with an overall project duration of five years from the time of completion.
Of the 13 suppliers that were pre-qualified for biding only four were actually approved to submit bids and in the end only one, BT Group, felt that it could profitably meet the requirements of the tender and submitted a bid. On 3 December 2009 it was duly selected as the winner.
In order to establish the two Lots comprising the tender DETINI had conducted a study of business density on the basis of Output Areas, which were created for the purposes of the 2001 Census and which each contain approximateley 150 households4. Business premises were correlated to OAs by postcode and the resulting business density then determined Lot 1 or Lot 2 assgnment.
The technical requirements of the tender were as follows:
| Tender phase | Business Density / sq km | Minimum downstream data rate | Grant funding available |
|---|---|---|---|
| Lot 1 | < 10 | 2 Mbps | 8 to 12 million UKP |
| Lot 2 | >= 10 | 10 Mbps | 3 to 8 million UKP |
Grant funding for the tender was estimated at a maximum 18 million UKP over three financial years.
The tender specified only the data rates to be achieved and not the technology by which this was to be delivered. The use of the term "broadband" in the tender followed from the widely-quoted ( but technically incorrect ) OFCOM definition of a delivery technology exceeding so-called "narrowband" data rates of 128 kbps downstream5. Therefore it did not bind the bidders to a particular technical implementation such as multi-frequency delivery over phone lines.
The tender also required the vendor to supplement their retail services with wholesale access for third parties on the basis of equivalence of access.
The single criterion on which vendor selection was based was that of "most economically advantageous offer". In essence this meant the offer which required the lowest amount of public funding. The distortion that this introduced in the bidding process is reflected by the voluntary decision of three of the four qualified contenders not to submit a bid. Only BT Group, with a 26.4% share of the UK retail broadband market at the time of the tender issue6, felt able to subsidise the deployment sufficiently to meet the criterion. This directly contradicts DETINI's assertion that the state aid provided a direct and appropriate incentive to bidders.
The second failing was one of omission. By attempting to remain technologically neutral the tender failed to grasp the opportunity to specify requirements at the network protocol level. Given BT Group's indifference to implementing IPv6 routing this means that IPv4 will remain the incumbent protocol for businesses in Northern Ireland, stifling innovation and magnifying the risk to online activities beyond the projected IPv4 exhaustion date.
Thirdly the tender failed to specify any operational requirements beyond the local-loop link data rate. There was no specification of a minimum acceptable service uptime nor an SLA for fault resolution. So long as the winning bidder can demonstrate that their implementation achieves the specified data rate then their adherence to the tender is satisfied and any further service parameters are out of scope.
Finally, the concession that the winning bidder would be permitted to operate the network on a retail basis has arguably led to a distortion of the broadband retail market in Northern Ireland. Already BT Infinity billboard posters have been posted around Belfast and the Faster Broadband NI7 site redirects potential customers to the BT Wholesale site to check "their broadband speed availability"; a site which, of course, makes no reference to potential LLU providers for the premises.
The Northern Ireland Executive is to be commended for its forward-looking attitude towards Internet access for businesses and residents. Unfortunately, DETINI's realisation of the aspiration was flawed from the outset by a narrow focus on speed as the indicator of network maturity and the rejection of any selection criteria other than lowest public cost. The result was the selection of a conservative, life-limited technology and the exclusion of bidders that may have provided more innovative and stimulating solutions.