Talk:List of postcode districts in the United Kingdom

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Sanction[edit]

This page has been created after this deletion discussion. See also the talk page. -- RHaworth 19:38, 2005 Mar 10 (UTC)

Credit[edit]

This page is derived from http://3lib.ukonline.co.uk/pocketinfo/data/postuk.csv, which is a CSV file found at PocketInfo - Travel. It was developed by Mike Cockerill who has given permission for its use here. -- RHaworth 18:41, 2005 Mar 10 (UTC)

Comments[edit]

This list is terribly out of date - it contains lots of postcodes that have been changed since 1998. It contains an odd mix of postal and wrong traditional counties, and contains postcodes that have never existed such as B41 (Yes I know the original Birmingham 41 district covered Smethwick, but that was superceded by B66 et al.) Owain 19:40, 10 Mar 2005 (UTC)

  • Yes, I know. Note, in particular, that HS, Outer Hebrides is missing. Following the closure of the VfD debate, I wanted to get a list into Wikipedia asap. I found the CSV file on the web, liked its format, got the author's permission and used it. I think people will enjoy updating it and I can see some arguments developing over what county names to use. -- RHaworth 19:54, 2005 Mar 10 (UTC)

I have just noticed that the order of districts is left-aligned alpha, eg. ML10 comes before ML2. 'Twas thus in my source. I will try and work through and correct. -- RHaworth 20:52, 2005 Mar 10 (UTC)

Do we really need the n/a post town and postcode sections? Surely if postcode districts aren't listed under existing post towns then they are implictly n/a? As it is it seems a little messy. Owain 11:47, 17 Mar 2005 (UTC)

  • Keep the n/a's - they are doing no harm and they don't take up much space. -- RHaworth 00:39, 2005 Mar 18 (UTC)
OK, it's just that a lot of postcode areas don't have contiguous blocks of district numbers any more, so maybe all tha n/as could be grouped into one big n/a per postcode area... Owain 09:34, 18 Mar 2005 (UTC)

I have added some Y0.. postal groups, they were missing (I am not sertain for the cities of YO41, YO42, YO43, YO51, YO60, YO61, YO62, but the postal codes exist for certain!) AceT 17:39, 1 July 2006 (UTC)[reply]

BN23 is now Westham, and BN24 is now all Pevensey.

Counties[edit]

Writing as one brought up in Westmorland, I have no objection to reversion to the traditional counties. But to lump all of Yorkshire together is throwing away useful information. Can we please use the proper Ridings? And why has Doncaster been left in the fictional South Riding?

I have done the area containing Haworth as an example. Is there a Wikipedia Manual of Style approved way of representing ditto? Warning. If you are going to convert to using dittos, make sure that the postal districts are in the right order first, eg. move CO8 and CO9 to before CO10.

-- RHaworth 00:39, 2005 Mar 18 (UTC)

Suggestion[edit]

This table makes little sense at first glance: this will likely put off readers who may go elsewhere to find the answer they're looking for. Wouldn't it make more sense if the list is indexed by postcode prefix to have that information in the first column on the left hand side? -- Francs2000 | Talk [[]] 16:59, 13 Jun 2005 (UTC)

Yes. Do you have a round tuit handy? —Phil | Talk 09:42, July 14, 2005 (UTC)
Not just at this moment, no. -- Francs2000 | Talk 10:31, 14 July 2005 (UTC)[reply]
I'm sectionising the article more thoroughly which ought to make editing a bit easier. —Phil | Talk 16:30, July 14, 2005 (UTC)

Improvement or Deletion?[edit]

I've had a go at sorting out what this page should look like for just the Aberdeen area:

Area District Post Town
AB Aberdeen AB10 Aberdeen
AB11 Aberdeen
AB12 Aberdeen
AB13 Milltimber
AB14 Peterculter
AB15 Aberdeen
AB16 Aberdeen
AB21 Aberdeen
AB22 Aberdeen
AB23 Aberdeen
AB24 Aberdeen
AB25 Aberdeen
AB30 Laurencekirk
AB31 Banchory
AB32 Westhill
AB33 Alford
AB34 Aboyne
AB35 Ballater
AB36 Strathdon
AB37 Ballindalloch
AB38 Aberlour
AB39 Stonehaven
AB41 Ellon
AB42 Peterhead
AB43 Fraserburgh
AB44 Macduff
AB45 Banff
AB51 Inverurie
AB52 Insch
AB53 Turriff
AB54 Huntly
AB55 Keith
AB56 Buckie

Even with access to some databases of addresses and the PO files of changes, it was pretty time consuming. People still use the old postcode areas, wrong post towns or add in Aberdeen in addition to the correct one. I've avoided any questions of counties or countries on the basis that they're not needed for postal purposes.

Given the effort needed to get it something like right and then maintain it, I'm tending towards the view it should be deleted. Comments? --Cavrdg 13:38, 9 August 2005 (UTC)[reply]

It is time consuming, yes — but there should be enough of a geographical spread of UK wikipedians to keep their own bit up-to-date. I've just spent a while tidying all the tables up, so please don't delete it just yet! :) Owain 14:26, 9 August 2005 (UTC)[reply]
You won't get consensus to delete it just because it'll take an inordinately long time to clean up. I know a lot of users that will object to the proposal. I feel Owain's suggestion is a good one, I'll happily do the Buckinghamshire ones (MK, HP, NN, OX, SL, LU etc) and maybe a few more besides. -- Francs2000 | Talk 00:36, 10 August 2005 (UTC)[reply]

OK. Let's see if we can get it done. AL was fairly easy, so that's the As done and updated on the article page. --Cavrdg 13:56, 10 August 2005 (UTC)[reply]

Hang on, are we totally changing the structure of the table? Are we dropping the county column? What happened to my pretty table colours? Owain 18:26, 10 August 2005 (UTC)[reply]
  • Speedy revert to Owain's last versions. Cavrdg's mods are terrible: he has thrown out the county and country columns. This list is long enough as it is, making a separate row for every district is going to make it even longer.
I'm not convinced about the need for county and country in this list. They're not needed for postal purposes and, if people want to know more about the places, they can follow the links to the articles.
A vertical list with the key field first is much easier to look down to find the one you want. I think the ones in the As are not unreasonable but I do baulk a bit at having getting on for 50 rows for Birmingham. Perhaps blocks of ten where they all point to the same post town would be better. B1 - B9 Birmingham, B10 - B19 Birmingham, etc. --Cavrdg 08:03, 11 August 2005 (UTC)[reply]
MRSC has added to the destruction by throwing out the London Postal District names - OK they are not "Post Towns" but that should be explained by note at the top of the article rather than throwing out the data. The only other source on Wiki is London postal district which is not in a nice columnar format such as this list.
My only complaint about Owain's format is that it is a little difficult to see where one area ends and the next begins but that is easily solved by the line I have put in between Birmingham and Bath. -- RHaworth 06:48:53, 2005-08-11 (UTC)
In fact I am very puzzled as to why Owain has been putting   cells at the end of each line - they don't have any effect if the internal borders are suppressed. (I specifically left them out when I created the article in order to show the break between each area without any page lengthening overhead.) -- RHaworth 07:10:12, 2005-08-11 (UTC)
It made the table borders look terrible with the border-collapse: collapse style. Not to mention the fact that it probably isn't valid HTML :) Owain 08:31, 11 August 2005 (UTC)[reply]

YO95[edit]

Does YO95 mean anything to anyone? http://www.streetmap.co.uk shows a place near Pocklington and it appears in a Department for Work and Pensions pdf list but, otherwise, it seems a bit obscure. --Cavrdg 14:55, 31 August 2005 (UTC)[reply]

It's a non-geographic code isn't it? Looks like streetmap is playing up. Pocklington is YO42. Owain 12:50, 1 September 2005 (UTC)[reply]

I suppose so. There are postcodes for PO Boxes but for ones like YO91 (Nestle / Rowntree) and NE99, Streetmap points to the relevant city centre with the warning The location for this post code is not accurate. You should confirm your destination before travelling so I wondered if the point near Pocklington was the site of a new development of some sort. --Cavrdg 18:49, 1 September 2005 (UTC)[reply]

Data quality[edit]

This page is getting more filled with errors. A quick glance at the postal county shows several postcodes from the London postal county in the Essex or Middlesex postal counties. Bishops Stortford is now moved to Essex?? and since when was "West Riding of Yorkshire" a postal county?

This whole list needs to be populated again with some reliable data. MRSC 16:01, 19 September 2005 (UTC)[reply]

They are not errors, they are the correct traditional counties. For example, if you enter E1 1AA into http://www.capscan.co.uk/mcd5demo.htm it correctly shows Middlesex as the traditional county; Similarly, E4 8AA correctly shows Essex. There is no such thing as a postal county anymore, they are referred to as 'former postal counties' in all the PAF documentation. Owain 12:48, 20 September 2005 (UTC)[reply]

BTx codes <> Post towns[edit]

These are missing on the list for some reason. Does anyone know which post towns tie in with which codes?

Post towns

Antrim, Armagh, Augher, Aughnacloy, Ballycastle, Ballyclare, Ballymena, Ballymoney, Ballynahinch, Banbridge, Bangor, Bushmills, Caledon, Carrickfergus, Castlederg, Castlewellan, Clogher, Coleraine, Cookstown, Craigavon, Crumlin, Donaghadee, Downpatrick, Dromore, Dungannon, Enniskillen, Fivemiletown, Hillsborough, Holywood, Larne, Limavady, Lisburn, Londonderry, Maghera, Magherafelt, Newcastle, Newry, Newtownabbey, Newtownards, Omagh, Portrush, Portstewart, Strabane

Postcode districts

BT18, BT19, BT20, BT21, BT22, BT23, BT24, BT25, BT26, BT27, BT28, BT29, BT30, BT31, BT32, BT33, BT34, BT35, BT36, BT37, BT38, BT39, BT40, BT41, BT42, BT43, BT44, BT45, BT46, BT47, BT48, BT49, BT50, BT51, BT52, BT53, BT54, BT55, BT56, BT57, BT58, BT59, BT60, BT61, BT62, BT63, BT64, BT65, BT66, BT67, BT68, BT69, BT70, BT71, BT72, BT73, BT74, BT75, BT76, BT77, BT78, BT79, BT80

MRSC 10:35, 20 September 2005 (UTC)[reply]

I can look them up. I was going to before the re-organisation anyway! :) Owain 12:41, 20 September 2005 (UTC)[reply]
Excellent! MRSC 13:35, 20 September 2005 (UTC)[reply]
I've had a go. There are a few oddities. BT58 is described here as contains business addresses only
and BT29 seems to be used both for Crumlin and, for places around the International Airport, Belfast. --Cavrdg 19:49, 20 September 2005 (UTC)[reply]
Brilliant. BT58 is probably similar to postcodes in Wapping in the E98 group (e.g.) E98 1NW used for News of the World. MRSC 20:46, 20 September 2005 (UTC)[reply]

Main post town[edit]

Now the field denoting what the mnemonic letters stand for has been removed, we need a different way to indicate what they mean. How about emboldening the post town that the area is named after? This still won't help in the case of TD, TS, and others that aren't named after post towns though. Perhaps the last column could be used to indicate what the area is, rather than the pretty useless England, Scotland, Wales, Northern Ireland that's there now? Owain 10:47, 21 September 2005 (UTC)[reply]

Resource for missing districts and other thoughts[edit]

We now have around 2742 districts listed. According to ONS Postal Geography there were 3064 in May 2005 so we're still missing quite a few. I've made about 50 changes on the As to Ds and I think we now have all of the geographic ones for those, at least.

For England and Wales, the Proviser house price site has a list that includes almost all the geographic districts. You need to click on the initial letter of the postcode in the box top left and then scroll down to see the list.

I'm not sure that using &hellip; (…) in the lists is helpful. M32 … M34 could mean M32, M33, M34 or M32, M34. I also think we should list the districts in numeric order even if it means returning to an earlier posttown.

For next steps, I think it would be good to create stub articles for any post towns that don't have them and, where a post town has many districts, include some information on the area covered as has been done with London. Telling someone that M16 is in Manchester doesn't give much information. Saying it includes Old Trafford would be more interesting. The danger is that this list could become very long with mini-articles about the more interesting districts. We'd need to limit the expansion to links to places in the district. --Cavrdg 07:38, 25 September 2005 (UTC)[reply]

I have found and verified 2923 districts, when interrested I can post them here, mostly I only have the Postal code, not the rest. AceT 19:38, 11 July 2006 (UTC)[reply]

Askham/Askam[edit]

Is Askham in Furness a typo or a variant for Askam in Furness? Morwen - Talk 17:35, 13 December 2005 (UTC)[reply]

No, it's a typo! Kijog 19:03, 16 December 2006 (UTC)[reply]

KY postal area[edit]

This appears to be showing coverage instead of post town. MRSC 10:38, 6 August 2006 (UTC)[reply]

I've corrected this. A quick glance reveals more places that have been added that are not post towns. MRSC 10:53, 6 August 2006 (UTC)[reply]

Grantham[edit]

Grantham is twice, under LN and NG. was it recoded? Morwen - Talk 08:31, 7 October 2006 (UTC)[reply]

In 2004 it was NG31-33 (with no other Granthams listed). There are no changes listed on major changes. Is LN more prestigious than NG? :) MRSC 08:49, 7 October 2006 (UTC)[reply]
Quite. Morwen - Talk 09:37, 7 October 2006 (UTC)[reply]

London postcode prefix E[edit]

E18 is South Woodford, though the area does not exists as such it is more the LU station that lends its name, however, Woodford is in Essex and takes the postcode IG8 from Ilford. E19 was proposed for Woodford Green, where to keep with alphabetical listing, E18 would be Woodford and E19 would be Woodford Green. IG8 0HN would become E19 0HN. E20 is the postcode for the fictitional area of Walford, used on the BBC soap [ Eastenders ] 82.69.38.93 14:11, 12 November 2006 (UTC)Matthew Davis[reply]

Jersey and Guernsey?[edit]

Aren't JE and GY missing from the list? —The preceding unsigned comment was added by 217.152.120.139 (talk) 10:24, 24 April 2007 (UTC).[reply]

Elipses indicating ranges of postcodes?[edit]

The character … has been used three times in the article. It seems to be used to indicate a range of directly consecutive integers as the numeric part of the district code for CH and SK, but this use makes little sense for ST. Also, there are many consecutive ranges where it's not used. Does anyone here recall inserting it, and what they meant when they did? Postcodes (talk) 18:25, 22 November 2007 (UTC)[reply]

Simple importer python code[edit]

Known issues

  1. Needs to deal with cases where post towns contain both linked and unlinked text.
  2. Needs to determine the field operated on explicitly, not rely on whether it contains commas to determine it's the postal districts field.
  3. Should provide extra information for postal districts as a field, INSERTing NULL if not supplied.
  4. A separate record should be INSERTed for each post town if there is a list.
  5. Some re-factoring is required. Postcodes (talk) 23:56, 26 November 2007 (UTC)[reply]
# This code is released under the GPL version 2

import BeautifulSoup
import httplib2
import sre
import codecs
import MySQLdb

h=httplib2.Http()
#r=h.request("http://en.wikipedia.org/w/index.php?title=List_of_postcode_districts_in_the_United_Kingdom&printable=yes")
#NOTE: Please copy the file locally before experimenting with this, e.g. use
#wget "http://en.wikipedia.org/w/index.php?title=List_of_postcode_districts_in_the_United_Kingdom&printable=yes" -O pcdistlist.html
#and
#r=h.request("http://localhost/~username/pcdistlist.html")
 
pcs=BeautifulSoup.RobustHTMLParser(r[1])
elipregex=sre.compile("([0-9A-Z]*)\s*"+unichr(0x2026)+"\s*([0-9A-Z]*)")

fout=codecs.open("wppcds.sql", mode="wb", encoding="utf8")
db=MySQLdb.connect(user="username", host="host.example.net", db="postcodes", passwd="password")
tables=pcs.body.div.div.div.div.nextSibling.nextSibling.nextSibling.nextSibling.div
table=tables.findNextSibling("table")
fieldnames={}
fields={}
lastfields={}

db.query("DELETE from postcodes.postcode_districts")

while(table != None):
	row=table.tr
	while(row != None):
		hcell=row.th
		cell=row.td
		if hcell!=None:
			hcellno=0
			while(hcell != None):
				if len(hcell.contents) > 0 and len(hcell.contents[0].string.strip()) > 0: # check the cell has something in it
					fieldnames[hcellno]=hcell.contents[0].string.strip()
				else:
					fieldnames[hcellno]="Region" # TODO: fix this in the WP page
				hcell=hcell.findNextSibling("th")
				hcellno=hcellno+1
		else:
			cellno=0
			while(cell != None):
				if len(cell.contents) > 0 and len(cell.contents[0].string.strip()) > 0: # check the cell has something in it
					if cell.a != None:
						t=cell.a.contents[0].string.split(",")
						if len(t) == 1:
							fields[fieldnames[cellno]]=t[0].strip()
						else:
							t=[x.strip() for x in t]
							fields[fieldnames[cellno]]=t
					else:
						t=cell.contents[0].string.split(",")
						if len(t) == 1:
							fields[fieldnames[cellno]]=t[0].strip()
						else:
							t=[x.strip() for x in t]
							fields[fieldnames[cellno]]=t
				cell=cell.findNextSibling("td")
				cellno=cellno+1		
			same=True
			if len(lastfields) == 0 and len(fields) != 0: same=False
			for f in lastfields:
				if not fields.has_key(f) or lastfields[f] != fields[f]: 
					same=False
			if not same: # skip blank table rows

				# Construct and output SQL queries.
				pdf = None
				for f in fields.keys():
					if type(fields[f]) == type([]):
						pdf=f
					else:
						fields[f]=fields[f].replace("'", "\\'")
				if pdf != None and len(fields[pdf]) > 1:	
					temp=fields.copy()
					for pd in fields[pdf]:
						temp[pdf]=pd.replace("'", "\\'")
				# Expand elipses if present TODO: if a single entry has an elipsis
						s=elipregex.search(pd)
						if s != None:
							sttpd=s.groups()[0]
							endpd=s.groups()[1]
							pdprefix=sre.sub("[0-9]*", "", sttpd)
							pdrange=[pdprefix+str(elpd) for elpd in range(int(sre.sub("[A-Z]*", "", sttpd)), int(sre.sub("[A-Z]*", "", endpd))+1)]
							temp2=temp.copy()
							for rpd in pdrange:
								temp2[pdf]=rpd
								query="INSERT INTO postcodes.postcode_districts (`"+("`,`".join(temp.keys()))+"`) VALUES ('" + \
				("','".join(temp2.values()))+"');"
								query=query.replace(" districts", " district")
								try:
									db.query(query)
								except:
									print("Error running query: "+query)
									print db.error()
								fout.write(query+"\n")
						else:
							query="INSERT INTO postcodes.postcode_districts (`"+("`,`".join(temp.keys()))+"`) VALUES ('" + \
				("','".join(temp.values()))+"');"
				# Correct "Postal districts" to "Postal district"
							query=query.replace(" districts", " district")
							try:
								db.query(query)
							except:
								print("Error running query: "+query)
								print db.error()
							fout.write(query+"\n")
				else:
					query="INSERT INTO postcodes.postcode_districts (`"+("`,`".join(fields.keys()))+"`) VALUES ('" + \
				("','".join(fields.values()))+"');"
				# Correct "Postal districts" to "Postal district"
					query=query.replace(" districts", " district")
					try:
						db.query(query)
					except: 
						print("Error running query: "+query)
						print db.error()
					fout.write(query+"\n")
			lastfields=fields.copy()
		row=row.findNextSibling("tr")
	table=table.findNextSibling("table")
	fieldnames={}
	fields={}
fout.close()

Districts for Glasgow?[edit]

The article on Ballieston describes it as Ward 47 of Glasgow, though it's postal district is G69. Would it make sense to have districts for Glasgow as per London? Do the postal districts G1, G2, G3, G4, G5, G11, G12, G13, G14, G15, G20, G21, G22, G23, G30, G31, G32, G33, G34, G40, G41, G42, G43, G44, G45, G51, G52 and G53 (and maybe some others?) have names in the same way? Postcodes (talk) 16:46, 26 November 2007 (UTC)[reply]

Three extra fields: obsolete, partial and comments[edit]

Obsolete would store whether the postcode is one currently officially allocated by the Royal Mail. It would have values of Y, N, ? or the year it became obsolete. Partial would store whether the postcode covers more than one post town. It would have values of Y, N or a range, eg 9xx for Pevensey B24 Comments would hold any other information, such as "one delivery point" for Pevensey BN24. Any information repeated a significant number of times should probably have its own field.

Any thoughts on this?

Postcodes 23:54, 26 November 2007 (UTC)[reply]

Since there have been no objections, I will go ahead and implement this soon. If this is a problem for you, speak up now! If you are working on changes to this article and do not give notification, even if they are substantial it's quite reasonable to expect you to integrate your changes into the new layout. Postcodes 13:48, 30 November 2007 (UTC)[reply]
OK, Done! Postcodes 16:58, 30 November 2007 (UTC)[reply]

Non-geographic postcode districts[edit]

Should these be included in this list? If so, they should have a field to indicate their status. There's no obvious pattern to for them. Further, these are often allocated to a particular business. Rather than the generic entry "Businesses", these should be listed. Postcodes (talk) 00:02, 27 November 2007 (UTC)[reply]

I've started putting in the details for NGPDs, but I've only done A-H so far. The PDF linked to in the refs gives the real postcode district for some NGPDs, so I've used these instead of Y in the field where available. Postcodes 16:57, 30 November 2007 (UTC)[reply]

Subdivision of postal districts of London?[edit]

I've asked about this in the London postal district talk page, but should the list cover sub-divisions of postal districts in London (if this is the right way to put it)? For example, should SW1E (Vauxhall) be included? If not, why not? Postcodes (talk) 10:13, 30 November 2007 (UTC)[reply]

SW1E is not the postcode for Vauxhall. The source being used isn't a definitive list of postcode districts, rather a tool for statistical use, giving a "default" district where the actual postcode is not known. MRSCTalk 18:16, 7 December 2007 (UTC)[reply]

Reformating of data[edit]

We now have this data repeated twice in the encyclopedia. Once on each article such as AB postcode area and again here. This is bad in terms of keeping it up to date. I propose to templatise the tables in each of the postcode area articles and then transclude them in this article in place of the current list. The tables can then be updated once and it will appear in both places. MRSCTalk 18:26, 7 December 2007 (UTC)[reply]

That seems logical. So preserving the current fields would give a template something like:
{{Infobox postcode_district |
region = |
area = |
postal_district = |
post_town = |
coverage = |
local_authority =|
former_postal_county =|
obsolete =|
partial = |
non-geographic = |
comments =}}
If I import the area code articles, I can generate these infoboxes and insert them in existing articles. These would then be aggregated into the pages you describe, either with a python wikipediabot or by some other mechanism.
Would that be useful, do you think? Or would it be better to output a table of PCDs per Area with all the fields listed above?
A few questions and thoughts on merging the data
  1. Are the tables in the XX_postcode_area articles consistently structured?
  2. What is the difference between Post Town and coverage?
  3. Does coverage mean that there are postcodes of the given PCD in the named area or place? If so,
    1. Couldn't the list be hugely long, i.e. if a PCD covers a large number of small places?
    2. Should we restrict the list to places of a population larger than a certain number? If so, what should that number be? What about places where the population is not known?
    3. Should we include places that are only partly covered by a PCD? If so, how should we indicate the partial coverage?
  4. It will almost certainly require changes to ensure all the records can be JOINed (USING `Post Town` and `Postal District`, I'd think). Do you think all the "Post Towns" in List of postcode districts in the United Kingdom are actually official Post Towns? Is there a list somewhere that you know of?
One other thing: I'm not sure what the point of the Area field is. Postcodes (talk) 10:19, 10 December 2007 (UTC)[reply]

I'm thinking more of something like this:

{{AB postcode area}} {{AL postcode area}} {{BA postcode area}}

with the same data transcluded on the individual article page (i.e. only the template to update).

This is fine, provided the data from this article is not lost. So it would need to be copied from here to the other articles first, and any inconsistencies resolved. Once it was all across the contents could be replaced with a series of include statements. Postcodes (talk) 13:28, 17 December 2007 (UTC)[reply]

I don't like the non-standard width. Not sure how to fix that. MRSCTalk 15:37, 14 December 2007 (UTC)[reply]

The inconsistent width can probably be fixed by setting CSS styles, e.g. width:100%; on the tables in the included articles. Postcodes (talk) 13:28, 17 December 2007 (UTC)[reply]

Drop "Region" column[edit]

What does anyone think of dropping the "Region" column from the table? If it was accurate, it would contain the country the former postal county is in. It isn't accurate (Chepstow is in England/Wales and Chester solely in England, for instance). Its sole redeeming feature is that its less controversial than the county column was.

If someone does think it should stay, would they prefer to use a version of the postal country or the country/countries each post town overlaps with? —Preceding unsigned comment added by Aoeuidhtns (talkcontribs) 14:56, 1 May 2009 (UTC)[reply]

Duplicate lists of postcode districts[edit]

The article Outward postcode list was replaced by a redirect to List of postcode districts in the United Kingdom by User:MRSC on 2009-09-07T11:26:25 but was reverted back to a full list by 90.240.21.145 on 2009-10-26T12:16:57j. Some minor amendments have subsequently been made.

The two articles are essentially identical in remit, though they currently present the data with different attributes and ordering.

Both articles also overlap substantially with the table of postcode districts that form the core of the articles for each of the 123 postcode areas (AB postcode area#Coverage through to ZE postcode area#Coverage, and the parallel articles for Guernsey, the Isle of Man and Jersey).

This means that postcode districts for each area are listed without synchronisation in at least three articles.

In Outward postcode list, the co-ordinates are unsourced and so it is unclear whether the data are suitable for Wikipedia, are copyright or constitute original research.

In List of postcode districts in the United Kingdom (itself transcluded from subpages /A–H, /I–P and /R–Z), the Obsolete, Comments and Region columns are currently of limited value. Information is sorted by post town within each postcode area and the article accurately but confusingly introduces itself: "This is a list of the post towns of the United Kingdom."

Three suggestions to improve this situation.

1. It seems clear that Outward postcode list should be reverted to a redirect.

2. As a bigger step, would it also be reasonable to convert List of postcode districts in the United Kingdom to a transclusion of the contents of each the postcode areas? (The three subpages could then be deleted.) That would require 123 transclusions (120 if Crown dependencies are excluded, but I suggest that it would be unhelpful not to make the list comprehensive). There are obvious disadvantages to scattering the content across so many pages. But it would remove all of the above duplication and make each individual page quicker to view and edit. If (and only if) mass transclusion works well at a technical level, UK-wide data could still readily be checked on the combined page as now. Alternatively, a simple Coverage page could be created for each area which could be transcluded into AB postcode area#Coverage etc. Would this be helpful and, less obviously, would it be technically practicable?

3. More simply, would it also be better to sort postcode districts in numerical order and abandon the attempt to group districts by post town in List of postcode districts in the United Kingdom? Some postcode districts are split across multiple towns and post towns are already listed more succinctly at List of post towns in the United Kingdom so this order may be confusing or complicating in a list of districts. If the proposed transclusions were applied using the existing area lists, this reordering would occur as an automatic consequence.

Richardguk (talk) 04:51, 27 October 2009 (UTC)[reply]

Agreed. I think there is a lot of overlap on this subject. Points 1. and 2. are good ideas, although I think it would be better to leave the Crown dependancies with their own List of postcodes in ... articles as they are not part of the UK, therefore users might not think to search the UK titled article for these (yet the crown dependancies should still be transcluded in the UK article as you say). Point 3: In terms of coverage and possibly post towns, I think that should be just left to the individual AB postcode area#Coverage/PL postcode area#Coverage etc articles, as no doubt local wikipedians can tabulate this information and it doesn't mean an entire coverage artcile needs to be maintained should anything change. It would also be ideal that any subsequent tables created can be auto-sorted by the reader as in Outward postcode list, but initlialised for postcode districts in alphanumeric order. Zangar (talk) 13:00, 30 October 2009 (UTC)[reply]
Thanks for the feedback.
Re Crown dependencies, we are in agreement. The existing Guernsey/Man/Jersey postcode articles correspond closely to the UK XX postcode area articles and having their areas in the UK list does not prevent people from seeing the individual articles. In fact, I was wrong to think that they are missing from the current lists of districts. The page where they are currently missing is List of post towns in the United Kingdom, having been deleted by User:MRSC on 2009-07-28T21:54:01. I therefore propose to revert that deletion for the reasons above.
Re sort order: if <onlyinclude> tags were carefully placed within the coverage table on each area page, so as to include the content but not the headers of each area table, then I think it would be technically possible to produce one (very large) sortable table by transclusion.
  • We could also add a hidden sortkey for proper numerical sorting so that, for example, AB2 appears after AB1 and not AB19.
  • For post towns to be properly sortable, districts with multiple post towns would need to be split into multiple rows.
On the other hand, one advantage of listing districts by town instead of numerically is that we can greatly abbreviate consecutive district ranges. For example, EDINBURGH could be listed on one row as "EH1–EH13, EH14 (part), EH15–EH17, EH91*, EH95*, EH99*" (asterisks indicating non-geographic districts). This makes the table much shorter than listing the districts across 20 rows and would be a fair reason for not transcluding the table from the detailed ones in the individual area articles.
Judging whether it is better for districts to be listed separately or in post town ranges is complicated by the wild variation between remote areas and conurbations:
  • most areas have fewer than 10 post towns, but the maximum is 62 post towns (LL postcode area);
  • most areas have fewer than 23 districts (counting separately part districts split by post town), but the maximum is 82 (BT postcode area);
  • two-thirds of post towns cover only one district (or part of one district) but the maximum is 51 districts (GLASGOW post town).
Apologies for dithering, but I wonder now with regard to points 2 and 3 whether it is better merely to refine and make the most of the status quo. (Even so, that would not justify retaining Outward postcode list since the essential difference is in format not content.)
Richardguk (talk) 16:02, 30 October 2009 (UTC)[reply]
No problems! Now, that is a big question about how to order it, and the arguments for both are quite valid. As you say, based on the generated list size, it would seem better to list districts by postal town - although I think then the resulting list would be a more detailed version of List of post towns in the United Kingdom and would render that article obsolete (which I don't think would be too bad). I personally would be in favour of this to keep things a little cleaner and avoid duplication of information. Although if there is a general concensus that there is a need for these seperate articles, then I think by default this acticle would need to be listed by district.
  • Good spot with the Crown dependancies, I'd say go ahead and make the revert.
  • With regards to the coding for table sortability, if you're happy to make the necessary changes when the general transclusion takes place, that'd be great
  • I don't think there's much justify in keeping the Outward postcode list, unless there are any other objections
I've also noticed that this article doesn't wikilink to any of the specific postcode article pages, I'll try to do that soon
Zangar (talk) 16:54, 2 November 2009 (UTC)[reply]
Cheers! In line with the EDINBURGH example above, have illustrated how a simple condensed untranscluded sortable table might look at User:Richardguk/List of postcode districts. A hidden sortkey ensures that, for example, BA10 sorts directly after BA9 not BA1. For a live article, there would of course also be an introductory section, navboxes and categories added at the top and tail, along the lines of the existing article. — Richardguk (talk) 13:25, 3 November 2009 (UTC)[reply]

I much prefer the list created by Richardguk. In any event, Wikipedia:Articles for deletion/Outward postcode list is a fork of what we have now, so I've nominated it for deletion. MRSC (talk) 18:21, 4 November 2009 (UTC)[reply]

Thanks. I've contributed to the Outward postcode list AfD discussion.
Regarding the proposed new table, it occurs to me that abbreviating ranges (as with "AB10–AB12, AB15, AB16, AB21–AB25, AB99*" instead of "AB10, AB11, AB12, AB15, AB16, AB21, AB22, AB23, AB24, AB25, AB99*") might cause problems for people who use "Find" or search engines to look for a particular district (such as AB11). The table gets uglier and the first two columns wrap if the Districts column is too wide. Perhaps we could list each district in a comma-separated list (still in one table row for each post town, to keep the list short) but separating long lists of districts with <br/> where they are in different multiples of 10 (thus: "AB10, AB11, AB12, AB15, AB16,<br/>AB21, AB22, AB23, AB24, AB25,<br/>AB99*"). The existing prototype already has line breaks for the London alphabetical districts so this extends the practice.
Richardguk (talk) 09:01, 6 November 2009 (UTC)[reply]

Outward postcode list debate[edit]

Post-merger feedback[edit]

In view of the ("tentative") outcome of the deletion/merger debate, and the consensus of support for the revised layout demonstrated at User:Richardguk/List of postcode districts, I have replaced Outward postcode list with a redirect (categorised as unprintworthy) to List of postcode districts in the United Kingdom. I have also replaced the entire list at the latter page with a sortable table along the lines discussed. (Incidentally, Outward postcode list was previously by far the worst offending article on Wikipedia for ambiguous wikilinks, having links to 512 disambiguation pages prior to being replaced with a redirect.[2]) Presumably the three sub-pages (List of postcode districts in the United Kingdom/A-H, .../I-P, .../R-Z) can now be converted into redirects to the main article page, but I have not done this yet.

Please note:

A. Postcode districts:

1. Each cell in the postcode district column has a hidden sort key to ensure that AA10 sorts after AA9 and not before AA2. AA0 is sorted after A99 and before AA1 (even though Royal Mail sometimes sorts district 0 after district 9, treating it as an alias for district 10).

2. Districts within the same post town are grouped into a comma-separated list in one row to reduce the size of the table.

3. Rows are, by default, ordered primarily by postcode area and secondarily by the number of the lowest district in the comma-separated list for each post town (and, in the event of a split district causing a tie, tertiarily by post town name). This order is chosen because the page name would cause most readers to expect to see a list in postcode district order, even though we are grouping by post town to save space.

4. Instead of using dashes or ellipses for consecutive ranges, each district is listed, so that searches can find any valid district.

5. Districts split across post towns are listed in each place as "AANN (part)".

6. No attempt is made to indicate which sectors (or part-sectors) correspond to each post town, on the basis that this detail is likely to be harder to maintain accurately over time. Also, sectors and part-sectors are easily misinterpreted by those unfamiliar with the structure and terminology of postcodes (particularly with the space separator that sectors require), so a simpler but clearer list is advantageous. If there is consensus on this, it may be helpful to add a note so that future editors understand the rationale.

7. An asterisk is used to denote wholly non-geographic districts. This needs a little more explanation in the introductory section to the article.

8. For ease of maintenance, no attempt is made in this list to describe the organisations using any non-geographic districts, there are no references or notes within the table, and no reference is made to localities within post towns or to non-postal placenames.

9. "GIR*" is included under BOOTLE (area L) because GIR 0AA is assigned to the BOOTLE post town and is effectively an alias for L30 4GB.

B. Counties:

10. Counties were mentioned in the debate (by 84.68.42.171). Unfortunately, the boundaries of most administrative and ceremonial counties (and even of England, Scotland and Wales) do not correspond closely to most postcode boundaries. For sortable columns to be useful, we need to avoid having rows that cover multiple placenames. There is therefore a new column showing the former postal county for each post town, which is a simple hierarchy but one which is likely to be broadly meaningful to most readers interested in a particular part of the country.

11. Because the former postal county has officially been changed from Leicestershire to Rutland for much of OAKHAM (LE15) and MARKET HARBOROUGH (LE16), the county for these is given as "Leicestershire / Rutland" and so they will sort at the end of Leicestershire if sorted by the county column. Rationale: It seems more likely that people seeking Rutland will try looking alphabetically under Leicestershire than vice versa.

12. There is no special hidden sort key for counties, so counties prefixed "County" (Durham and the Northern Ireland counties), "East" (Lothian, Sussex), "Isle(s) of" (Man, Scilly, Wight and the Scottish islands), "Mid" (Glamorgan, ...lothian [sic]), "North" (Humberside, Yorkshire), "South" (Glamorgan, Humberside, Yorkshire) or "West" (Glamorgan, Lothian, Midlands, Sussex, Yorkshire) will sort according to their full (prefixed) name and not under the substantive part. Rationale: Most of subdivisions and alternative names are sufficiently obvious to justify not adding the complication of a special sort key in the sortable county column.

C. Piping:

13. Since a sortable table may cause rows to become separated, every row is fully linked. This, unfortunately, necessitates many duplicate links for postcode areas and counties, but linking only some rows would be unhelpful to readers who use the sort functionality.

14. Post towns are hardcoded in CAPITALS and (except for LONDON) piped to the corresponding placename article.

15. As with postal links on other pages, piped wikilinks are used so that LONDON links to London postal district. For the equivalent postal county in the last column, the link is directly to London itself so that there is a choice of postal and geographical link, at the expense perhaps of causing mild confusion to people finding LONDON and London taking them to different articles.

16. Pipes are used to avoid linking to disambiguation and redirect pages to reduce the likelihood of future double-redirects. (The only exception is DULAS which currently has a disambiguation page but no article page.)

17. To ensure that every post town has a corresponding geography article, a stub article needs creating for Dulas, Anglesey as well as for Llanbedrgoch, Llwyngwril, Marianglas and Tyn-y-Gongl (LL area); Roy Bridge (PH area); and Clarbeston Road and Glogue (SA area).

D. Wrapping:

18. CSS styles are used to prevent the post town and county columns from wrapping, at the expense of increasing the size of the wikicode and HTML.

19. To prevent the district column becoming excessively wide, comma-separated lists of districts are split with a <br/> where a post town has districts that span multiples of 10 or (within central London) where there are alphanumeric districts in traditional district 1 (this is as proposed during the course of the merger debate). Even so, there remain some wide lines (which may wrap), such as with SW10–SW19 and with W1A*–W1W. Are these manual line breaks beneficial?

Richardguk (talk) 05:11, 23 November 2009 (UTC)[reply]

17. checkY Remaining missing post town articles now created: Llanbedrgoch, Llwyngwril, Marianglas (redirect to Marian-glas), Tyn-y-Gongl, Glogue. Redirects previously created for Dulas, Anglesey (DULAS), Roybridge (ROY BRIDGE), Clarbeston Road (CLARBESTON ROAD). — Richardguk (talk) 13:21, 23 December 2011 (UTC)[reply]

Reliable sources / Wikileaks mirror[edit]

This article has one major problem in that it has no sources cited for the data. I am a little concerned that it may in fact be a mirror of the Postzon geo data "leaked" onto Wikileaks. see here MRSC (talk) 22:06, 9 January 2010 (UTC)[reply]

I see now what has happened here. After Richardguk's changes the anon user 95.146.185.191 (talk) replaced the entire list with a copy of what was originally posted to Outward postcode list, removing all links and with the misleading edit summary "added coverage and latitude/longitude etc". I have restored Richardguk's version (and made some changes to the key/notes).
Back to the issue of sources, I have the Address Management Guide and subsequent postcode updates which gives a source for everything except former postal county. Do we have a citation for this information? MRSC (talk) 12:08, 10 January 2010 (UTC)[reply]

Page size[edit]

{{fn}} has been nominated for deletion and I came here to replace it with {{ref}}. When I opened the page for editing it caused Firefox to freeze. After I got it open I noticed that it contained a lot of unnecessary formatting, "style="white-space:nowrap;"|" was used 3000. I removed all of them and used {{nowrap}} where necessary. It's used in the four headers. It's not required in the following circumstances, where the town or county is a single word, where the link is [[Alford, Aberdeenshire|ALFORD]], [[Cleveland, England|Cleveland]] or [[Avon (county)|Avon]]. It is required, where the link is {{nowrap|[[Stockton-on-Tees|STOCKTON-ON-TEES]]}}, {{nowrap|[[St Albans|ST. ALBANS]]}}, {{nowrap|[[Inverness-shire]]}} or {{nowrap|[[Barra|Isle of Barra]]}}. Given that the fourth column title "Former postal county" seems to be longer that any of the counties it might not be required there either but... In all {{nowrap}} is used 554 times and reduces the page size quite a bit. something lame from CBW 12:18, 20 March 2010 (UTC)[reply]

Merger proposal[edit]

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
The result of this discussion was no consensus to merge MRSC (talk) 21:11, 9 December 2013 (UTC)[reply]

91.85.32.166 has proposed that List of post towns in the United Kingdom be merged into List of postcode districts in the United Kingdom. The reason stated is: "The above [list of postcode districts] page is just this [list of post towns] page but better and more readable." — Richardguk (talk) 23:17, 12 December 2012 (UTC)[reply]

  • Oppose: Though the list of districts is grouped by post town (and therefore includes all of the information in the list of post towns), the list of districts is so long that it is much less convenient for checking post towns at a glance. Also, post towns change less often than postcode districts, so it is helpful to have the more stable post town list separate from the more volatile postcode districts list; conversely, having both lists allows for cross-checking doubtful entries, where vandalism or good-faith mistakes may have introduced errors into one of the lists. — Richardguk (talk) 23:18, 12 December 2012 (UTC)[reply]
  • Support - have just taken a look and Richard might have a point if the post town list was actually ordered in alphabetical order of the post towns. But it's in order of the post codes, so both lists are basically duplicated, except more information is given in the postcode district page. Postcodes barely change, the last change was over 2 years ago, and there were 2 changes in the last 6 years. The point about "having both lists allows for cross-checking" basically confirms that it's unnecessary duplication and even sounds like the pages are using each other as a primary source, and not adhering to Wikipedia guidelines. Rapido (talk) 14:10, 17 December 2012 (UTC)[reply]
  • No merge - but it might be a good idea to reformat in an sortable table as an alphabetical list of post towns. MRSC (talk) 11:54, 31 December 2012 (UTC)[reply]
    The list of postcode districts is already sortable by post town (by clicking the relevant column heading, assuming the browser has JavaScript enabled). But the current list of post towns has the advantage of being very much more compact than a table with 1,500 rows (and hence more digestible to the casual reader, less demanding on older webbrowsers and useful for checking redlinks etc at a glance). Compactness is my main reason for opposing a merger. Exploding the post town list to a single row per town would therefore have the double disadvantage of losing the compactness and creating much more duplication between the two lists (unless the district list were in turn expanded to list only one postcode district per row, which would make that page exceptionally long by Wikipedia standards; though a district-per-row list would be helpful to indicate each district's valid sectors (0–9), to reflect the granularity of non-geographic status; it is currently hard to verify which districts are wholly non-geo, as distinct from districts which contain a mix of geo and non-geo sectors). — Richardguk (talk) 17:12, 31 December 2012 (UTC)[reply]
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

External links modified[edit]

Hello fellow Wikipedians,

I have just modified 5 external links on List of postcode districts in the United Kingdom. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 23:07, 22 May 2017 (UTC)[reply]

External links modified[edit]

Hello fellow Wikipedians,

I have just modified one external link on List of postcode districts in the United Kingdom. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 04:50, 10 September 2017 (UTC)[reply]

Post Towns (?)[edit]

In the list in the article, many places in the post towwns column are in CAPITALS. The article doesn't specify the significance of that. It seems unlikely to be "Special Post Towns" as there are more than 110 entries. Could someone who knows, update the article to explain. -- SGBailey (talk) 23:52, 19 November 2022 (UTC)[reply]