Response to Ofcom's plan of work 2021/2022
Ofcom are consulting on their plan of work for the next 2 years.
Ofcom's proposed plan of work is published here https://www.ofcom.org.uk/__data/assets/pdf_file/0022/209128/plan-of-work-202122.pdf
The Centre for Public Data are encouraging responses to review addressing data in the UK. The 2 recommendations that they're pushing for are:
In its new work plan, Ofcom should review the following:
Costs of PAF
Since 2013, the Royal Mail’s costs for updating PAF have remained relatively static. However, cloud technologies and collaborative editing techniques have seen the costs of delivery drop elsewhere.
We recommend Ofcom review whether the costs of PAF are still fair and proportionate.
Governance of PAF
PAF has an influence on a broader data market, outside of postal services. Its governance creates legal uncertainty and financial cost for many UK data users.
We recommend Ofcom review whether PAF’s current governance is appropriate, and how it could be improved while still ensuring that the postal services market operates satisfactorily.
Their longer response is here, which I recommend reading: https://docs.google.com/document/d/17iTZOaqNbeyBCxfqBD7O_WonJ46lAXvcgRDwQFJgBQo/
Are you happy with these points to be raised in a response from OSMUK? Are there other recommendations or elaborations that we should raise as part of our response?
Tony Shield Mon 25 Jan 2021 3:38PM
I would like OSMUK to formally respond to the consultation supporting the CPD's response. I think it would help if were were able to respond with data from a poll of members about whether an open licensed PAF would be supportive of OSM's aims. Quoting the numbers of addresses with postcode and without postcode in the OSM database might be helpful. Requesting thee PAF to be extended to contain UPRN's may also be beneficial.
Brian Prangle Wed 27 Jan 2021 4:16PM
We should broadly agree with the CPD repsonse but be more explicit than their call for a governance review and ask for a full comparative cost benefit analysis of making PAF available under an open licence using Gareth's points ( if there's evidence form other OSM chapters we should ask for this via the OSMF Advisory Board
Gareth L Wed 3 Feb 2021 10:51PM
I've drafted this so far. Consultation responses need to be sent by Friday, but would prefer to complete any changes on Thursday before sending.
We suggest that Ofcom include a review of address data, notably the PAF, as part of your 2021/22 work plan. Ofcom should review the following:
1. Costs of PAF
Since 2013, the Royal Mail’s costs for updating PAF have remained relatively static whilst it continues to produce IP for Royal Mail.
We recommend Ofcom review whether the costs of PAF are still fair and proportionate, including a comparison of costs to equivalent services in other countries, and whether a collaborative maintenance approach, such as that demonstrated by OpenStreetMap or the Geoplace could further reduce costs and improve quality.
2. Governance of PAF
PAF has an influence on a broader data market, outside of postal services.
The Geospatial Commission has observed that postcodes are “the most commonly recognised form of an address by the public”.
The governance of PAF creates legal uncertainty and financial cost for many UK data users, including parties taking part in Ofcom’s planned Open Communications initiative, due to the PAF featuring in address data or used in the derivation of other data sets.
We recommend Ofcom review whether PAF’s current governance is appropriate, and how it could be improved, specifically with a wider cost benefit analysis of making PAF available under an open license while still ensuring that the postal services market operates satisfactorily.
Particular feedback points from you please:
IP generation for Royal Mail, this is true. a report from 2012 showed that it earned approx £27M. I don't know whether that number has gone up or down since, and if that value potential is factored into the amount of money paid to RM to manage the PAF, which is why I haven't specifically cited it, but do push for a review of costs.
The second point is highlighting the fact that postcodes are more than just post system references and that realisation must be factored in when deciding governance/licensing. The encouragement to review the costs and supressed benefits as consequence of PAF data 'polluting' other datasets.
I was tempted to put some numbers in on postcode inclusion on OSM, but it didn't lend itself succinctly to either point, and the data is full of caveats like 'fewer than 20% of all postcodes exist on at least 1 building or address point within OSM"
Any thing that absolutely should not be in there?
Gareth L · Sun 24 Jan 2021 6:05PM
The longer response already includes a statement about being unsure how Ofcom and the Geospatial Commission's responsibilities interrelate, but for information, OSMUK's response to the geospatial commission strategy call for evidence in 2018 included the following regarding addressing:
Q5: Do you anticipate that any changes will be needed to the both address data and the wider address ecosystem, to support emerging technologies? Please provide evidence of value to support any proposed changes.
The country's postcodes are commonly used as a pseudo-identifier when matching positional data from different sources. We agree with the commissions observation that it is “the most commonly recognised form of an address by the public”. For emerging technologies to succeed it is likely that they will need to gain the support of the public. As such we believe that initial focus should at first be on providing greater and more frequent access to postcode and PAF data under an open data licence. It is our view that the PAF is a specific geospatial database and is included in the commission's view of geospatial data types (Question 1). Unlocking access to PAF will also ease the route to the open data release of the National Address Gazetteer; something which we are keen to see as soon as possible.
An additional and valuable first step would be the immediate release of Ordnance Survey’s “Code-Point with Polygons” data under an open data licence. Our members have identified a number of errors with the existing “Code-Point Open” dataset, often related to the introduction of new postcodes. A monthly release (or more frequently / as live) would help to ensure that corrections are published with less delay.
In Northern Ireland the Central Postcode Directory (CPD) maintained by NISRA, and Pointer (the address database) maintained by Land & Property Services (LPS) should be released as open data.
Furthermore, the recent trend in public sector (gov.uk) registers to provide address data only as a UPRN, must be stopped until the relevant UPRN data is also made available under the OGL. Since the public don't have access to the database that decodes these into full addresses, providing address data only as a UPRN removes much of the usefulness from address-containing datasets that are converted to registers. We believe it should be a requirement to have an openly licenced address register containing those UPRN records that appear in other gov.uk registers. A true open national address database is a must.
Thinking longer term, the rise of ‘floating transport’ (e.g. dockless bicycle hire schemes) and alternative delivery locations (e.g. parcel collection boxes, delivery to car boot) is likely to result in a change in the way we consider addresses. Ensuring that our systems are flexible enough to adapt to this change is important and will act as an enabler to new service offerings.