Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

Eircode - its implemetation (merged)

1…5859616364…69

Comments

  • #1802
    plodder
    Registered Users, Registered Users 2 Posts: 7,750 ✭✭✭plodder


    sondagefaux wrote: »
    Who said it was useful? It's a bit of meaningless fluff to fill space in a newspaper, like those articles you see that say something like '80% of Irish men don't like their partner eating crisps in bed' which turns out to be based on some survey done by a PR company.
    or like statistics that say 90% of people know their eircode ..

    But, would it not be better if statistics were useful? The CSO reported that the mean sale price of houses in month 11 of 2017 in Galway as €243,000

    That seems on the low side, but vast swathes of rural Galway, and Clare are included there along with Galway city. You'd expect the figure to be higher for Galway city, but you're out of luck. They are all lumped in together. It turns out that the mean price for Galway city was €261,000, but again that is an average across the whole city, with I imagine quite some variation.

    At least in Dublin there are approx 35 different routing keys. So, the stats are a lot more granular and therefore useful.
    If you want a more finely grained breakdown of indebtedness or property prices or whatever than is available when figures for these are published on a county basis, you can either do the research yourself, or campaign for the CSO to publish them on a different basis. Given that the Census records the Eircodes of all Census addresses, and that the Property Price Register shows the details of all residential properties sold in the state, a simple change to make it compulsory for all properties sold or transferred to have their Eircodes included in the register, would make it much easier for the CSO to map property prices.
    What other basis could the CSO use? They already publish them by Eircode RK, county and local authority. Those are the only criteria that make sense to the public at large, and which can be easily reported in the media.

    It wouldn't look so good for the newspapers to be reporting "The small area with the highest growth in property prices in Nov 2017, was area number 783718929". It means nothing to anyone.


  • #1803
    sondagefaux
    Registered Users, Registered Users 2 Posts: 1,443 ✭✭✭sondagefaux


    plodder wrote: »
    sondagefaux wrote: »
    Who said it was useful? It's a bit of meaningless fluff to fill space in a newspaper, like those articles you see that say something like '80% of Irish men don't like their partner eating crisps in bed' which turns out to be based on some survey done by a PR company.
    or like statistics that say 90% of people know their eircode ..

    But, would it not be better if statistics were useful? The CSO reported that the mean sale price of houses in month 11 of 2017 in Galway as €243,000

    That seems on the low side, but vast swathes of rural Galway, and Clare are included there along with Galway city. You'd expect the figure to be higher for Galway city, but you're out of luck. They are all lumped in together. It turns out that the mean price for Galway city was €261,000, but again that is an average across the whole city, with I imagine quite some variation.

    At least in Dublin there are approx 35 different routing keys. So, the stats are a lot more granular and therefore useful.
    If you want a more finely grained breakdown of indebtedness or property prices or whatever than is available when figures for these are published on a county basis, you can either do the research yourself, or campaign for the CSO to publish them on a different basis. Given that the Census records the Eircodes of all Census addresses, and that the Property Price Register shows the details of all residential properties sold in the state, a simple change to make it compulsory for all properties sold or transferred to have their Eircodes included in the register, would make it much easier for the CSO to map property prices.
    What other basis could the CSO use? They already publish them by Eircode RK, county and local authority. Those are the only criteria that make sense to the public at large, and which can be easily reported in the media.

    It wouldn't look so good for the newspapers to be reporting "The small area with the highest growth in property prices in Nov 2017, was area number 783718929". It means nothing to anyone.

    On the one hand you want the CSO to publish more granular statistics, on the other hand you don't...

    The CSO could easily publish statistics on the basis of combining several Small Areas, with names given to the combined areas, and maps showing their boundaries. They already exist...

    The fact that it chooses not to says nothing about Eircodes one way or the other.

    Eircodes could be used to do this. Just because nobody is choosing to do this doesn't mean that Eircodes themselves are flawed.

    If this is an issue (and I don't really see why it is), campaign for the presentation of statistics in a way you want to see.


  • #1804
    antoinolachtnai
    Registered Users, Registered Users 2 Posts: 9,817 ✭✭✭antoinolachtnai


    There is no agreed shared frame of geographical reference for collecting information and providing statistics.

    Postcodes could have given us this, by making county, ED and SA divisions more universal and useful.

    Instead we have a new, rather odd geographical entity, the routing code.

    It is a pity introduction of the postcode wasn't taken as an opportunity to make things simpler. Instead it has made everything more complicated.


  • #1805
    BonnieSituation
    Registered Users, Registered Users 2 Posts: 19,088 ✭✭✭✭BonnieSituation


    antoinolachtnai wrote: »
    There is no agreed shared frame of geographical reference for collecting information and providing statistics.

    Postcodes could have given us this, by making county, ED and SA divisions more universal and useful.

    Instead we have a new, rather odd geographical entity, the routing code.

    It is a pity introduction of the postcode wasn't taken as an opportunity to make things simpler. Instead it has made everything more complicated.

    It hasn't at all made things more complicated.

    There are simpler and more effective ways of disseminating statistical data than the RK. No sane person who deals in spatial data (and I am one) would use the RK for serious analysis (unless the RK was the focus, but I digress), but the RK is a tangible way to disseminate data to the lay person just as the county, LA or province is as well.

    The house prices yoke on the previous page is merely a bit of fluff. If someone wants to drill down to ED or SA level they are more than welcome to do it. But most won't and will just read it and go "Sure look, house prices are up in X, Y and Z" and the they turn the page.

    The Routing Key is an effective part of a successful postal code system. If it ends up being used for something else that it wasn't intended for then it's hardly its or An Post's fault.

    Just look at the UK experience, they implemented postcodes in the 60s and 70s and the you see them being used for car insurance, school access etc... it's always the way with spatial data; people will find a use if they need to.

    ---

    That being said, I was looking for a postcode yday for a house in Manchester and just sniggered that every house on that street had the same one.

    I never realised that that phenomena existed in large cities but there you go. Only ever saw it on rural outposts.

    Eircode ftw.


  • #1806
    antoinolachtnai
    Registered Users, Registered Users 2 Posts: 9,817 ✭✭✭antoinolachtnai


    Loads of 'sane' people will use the for analysis, just as loads of 'sane' people use the Dublin postcodes and the counties outside Dublin.

    They use it for the simple reason that it is readily accessible to them.

    It has made things more complicated in that there used to be just one readily accessible system for collecting and analysing address data - this was counties combined with dublin postal zones. Now there are two - (a) postal counties combined with dublin postal zones and (b) routing codes. (b) does not map straightforwardly to (a).

    Neither (a) nor (b) map straightforwardly to legal counties, EDs and SAs. It would make it a lot easier (maybe not for you, but for regular people) if the new system mapped straightforwardly to the 'sophisticated' system of EDs and SAs. But it doesn't.


  • Advertisement
  • #1807
    BonnieSituation
    Registered Users, Registered Users 2 Posts: 19,088 ✭✭✭✭BonnieSituation


    antoinolachtnai wrote: »
    Loads of 'sane' people will use the for analysis, just as loads of 'sane' people use the Dublin postcodes and the counties outside Dublin.

    They use it for the simple reason that it is readily accessible to them.

    So are the SAs and EDs... I can link you to them if you want. There are even shapefiles!!!
    It has made things more complicated in that there used to be just one readily accessible system for collecting and analysing address data - this was counties combined with dublin postal zones. Now there are two - (a) postal counties combined with dublin postal zones and (b) routing codes. (b) does not map straightforwardly to (a).

    How? All we have now is more ways to analyse (albeit clumsily).

    I agree that using the Dublin codes was a bad one, but alas, it hardly warrants any bellyaching. I mean, if you have a problem with routing key sizes then surely you can take any statistical analysis that uses "Dublin 8" as its basis with a grain of salt.

    But your issue is resolved by the more granular and useful EDs and SAs.
    Neither (a) nor (b) map straightforwardly to legal counties, EDs and SAs. It would make it a lot easier (maybe not for you, but for regular people) if the new system mapped straightforwardly to the 'sophisticated' system of EDs and SAs. But it doesn't.

    But regular people just don;t care about this stuff. We might think they do. But they don't. As I said, the lay man will look at such articles or analyses and go "Huh, interesting" and move on.

    The rest of us will use SAs and EDs.


  • #1808
    antoinolachtnai
    Registered Users, Registered Users 2 Posts: 9,817 ✭✭✭antoinolachtnai


    Having more ways to analyse in itself makes the situation more complicated. Some data will be published by county/zone and some will be published by RK.

    How am I supposed to map from a postal address to a small area without needing to incur charges or otherwise get access to proprietary databases?


  • #1809
    sondagefaux
    Registered Users, Registered Users 2 Posts: 1,443 ✭✭✭sondagefaux


    BonnieSituation wrote: »
    antoinolachtnai wrote: »
    There is no agreed shared frame of geographical reference for collecting information and providing statistics.

    Postcodes could have given us this, by making county, ED and SA divisions more universal and useful.

    Instead we have a new, rather odd geographical entity, the routing code.

    It is a pity introduction of the postcode wasn't taken as an opportunity to make things simpler. Instead it has made everything more complicated.

    It hasn't at all made things more complicated.

    There are simpler and more effective ways of disseminating statistical data than the RK. No sane person who deals in spatial data (and I am one) would use the RK for serious analysis (unless the RK was the focus, but I digress), but the RK is a tangible way to disseminate data to the lay person just as the county, LA or province is as well.

    The house prices yoke on the previous page is merely a bit of fluff. If someone wants to drill down to ED or SA level they are more than welcome to do it. But most won't and will just read it and go "Sure look, house prices are up in X, Y and Z" and the they turn the page.

    The Routing Key is an effective part of a successful postal code system. If it ends up being used for something else that it wasn't intended for then it's hardly its or An Post's fault.

    Just look at the UK experience, they implemented postcodes in the 60s and 70s and the you see them being used for car insurance, school access etc... it's always the way with spatial data; people will find a use if they need to.

    ---

    That being said, I was looking for a postcode yday for a house in Manchester and just sniggered that every house on that street had the same one.

    I never realised that that phenomena existed in large cities but there you go. Only ever saw it on rural outposts.

    Eircode ftw.

    That's always been the case. Almost every UK postcode covers multiple addresses. It's one thing when a postcode for an urban area covers multiple houses in the same street, it's another thing when those houses get subdivided into flats. Inner London is really bad for this. A postcode that once covered 20 houses might now cover 60+ flats.


  • #1810
    sondagefaux
    Registered Users, Registered Users 2 Posts: 1,443 ✭✭✭sondagefaux


    antoinolachtnai wrote: »
    Having more ways to analyse in itself makes the situation more complicated. Some data will be published by county/zone and some will be published by RK.

    How am I supposed to map from a postal address to a small area without needing to incur charges or otherwise get access to proprietary databases?

    And some could be published by townland or barony or riding...


  • #1811
    Alun
    Registered Users, Registered Users 2 Posts: 21,548 ✭✭✭✭Alun


    sondagefaux wrote: »
    That's always been the case. Almost every UK postcode covers multiple addresses. It's one thing when a postcode for an urban area covers multiple houses in the same street, it's another thing when those houses get subdivided into flats. Inner London is really bad for this. A postcode that once covered 20 houses might now cover 60+ flats.
    ... and conversely in the countryside a small hamlet of one or two houses or even an individual house can have it's own postcode.


  • Advertisement
  • #1812
    BonnieSituation
    Registered Users, Registered Users 2 Posts: 19,088 ✭✭✭✭BonnieSituation


    sondagefaux wrote: »
    That's always been the case. Almost every UK postcode covers multiple addresses. It's one thing when a postcode for an urban area covers multiple houses in the same street, it's another thing when those houses get subdivided into flats. Inner London is really bad for this. A postcode that once covered 20 houses might now cover 60+ flats.

    Aye. I didn't even cop it was that bad though. I foolishly assumed it was more granular in England.

    TBH most of my usage has been in the north and it's grand.

    As I said, eircode ftw. :)
    sondagefaux wrote: »
    And some could be published by townland or barony or riding...

    Or garda district


  • #1813
    BonnieSituation
    Registered Users, Registered Users 2 Posts: 19,088 ✭✭✭✭BonnieSituation


    Alun wrote: »
    ... and conversely in the countryside a small hamlet of one or two houses or even an individual house can have it's own postcode.

    And converse to that two houses miles from each other on the same stretch of road... having the same code.

    It's truly a mess.


  • #1814
    sondagefaux
    Registered Users, Registered Users 2 Posts: 1,443 ✭✭✭sondagefaux


    Alun wrote: »
    sondagefaux wrote: »
    That's always been the case. Almost every UK postcode covers multiple addresses. It's one thing when a postcode for an urban area covers multiple houses in the same street, it's another thing when those houses get subdivided into flats. Inner London is really bad for this. A postcode that once covered 20 houses might now cover 60+ flats.
    ... and conversely in the countryside a small hamlet of one or two houses or even an individual house can have it's own postcode.

    I don't know if there are any rural postcodes that only cover one house, although there are many special postcodes that cover individual organisations or even individual departments within organisations (eg: DVLA postcodes).

    But they are generally reserved for organisations that deal with large volumes of mail.

    The vast majority of postcodes cover multiple addresses. My own postcode covers 38 different addresses. I have an even numbered street number. The odd numbered house directly opposite mine has a different postcode, which it shares with all the odd numbered houses on my street.


  • #1815
    Alun
    Registered Users, Registered Users 2 Posts: 21,548 ✭✭✭✭Alun


    sondagefaux wrote: »
    I don't know if there are any rural postcodes that only cover one house, ...
    There most certainly are, my niece lives in one, although (lucky her!) the house and grounds cover an area bigger than most housing estates :D


  • #1816
    sondagefaux
    Registered Users, Registered Users 2 Posts: 1,443 ✭✭✭sondagefaux


    Alun wrote: »
    sondagefaux wrote: »
    I don't know if there are any rural postcodes that only cover one house, ...
    There most certainly are, my niece lives in one, although (lucky her!) the house and grounds cover an area bigger than most housing estates :D

    Not exactly typical. As I said, the vast majority of postcodes cover multiple addresses.


  • #1817
    Alun
    Registered Users, Registered Users 2 Posts: 21,548 ✭✭✭✭Alun


    BonnieSituation wrote: »
    And converse to that two houses miles from each other on the same stretch of road... having the same code.
    I can't say I've ever seen that. I too am familiar with the north of England, and an area I'm very familiar with in the Yorkshire Dales has distinct postcodes for very small villages / hamlets just over a mile apart.


  • #1818
    Alun
    Registered Users, Registered Users 2 Posts: 21,548 ✭✭✭✭Alun


    sondagefaux wrote: »
    Not exactly typical. As I said, the vast majority of postcodes cover multiple addresses.
    I never said they were typical, but they're a lot more common than you may think especially in sparsely populated rural areas.


  • #1819
    PDVerse
    Registered Users, Registered Users 2 Posts: 160 ✭✭PDVerse


    All opinions are not equal. The opinions of experts outweigh the opinions of non-experts if all we have are competing opinions. Let's take an example of using Small Areas in the Eircode design. Experts explained that there would be considerable changes to Eircodes required after Small Areas were changed for each Census. Non-experts made pronouncements like the following

    https://www.boards.ie/vbulletin/showpost.php?p=96345955&postcount=7385
    https://www.boards.ie/vbulletin/showpost.php?p=96350729&postcount=7404
    https://www.boards.ie/vbulletin/showpost.php?p=96356110&postcount=7416

    Less than a year after Eircode launch Census 2016 was undertaken which required changes to Small Area boundaries to reflect the change in household formation and to protect privacy. This results in 43,362 Eircodes that have a different Small Area designation for the 2016 Census than they had for the 2011 Census. Either a large scale project would be required to re-allocate new Eircodes for existing addresses and ensure they were updated in everyone's databases, or the accuracy and usability of this feature would have eroded each Census.

    Now that we have data we no longer need to rely on opinion.


  • #1820
    [Deleted User]
    Posts: 31,118 ✭✭✭✭ [Deleted User]


    There's no reason why eircodes can't be grouped together geographically for statistical purposes, all you need is an application that places them on a map and a "lasso" tool to select them and assign a "small area code".

    It isn't a big deal, with current processing power it would be very easy to cut-n-paste clusters of eircodes from one small area to another should the boundaries move.


  • #1821
    antoinolachtnai
    Registered Users, Registered Users 2 Posts: 9,817 ✭✭✭antoinolachtnai


    PDVerse wrote: »
    All opinions are not equal. The opinions of experts outweigh the opinions of non-experts if all we have are competing opinions. Let's take an example of using Small Areas in the Eircode design. Experts explained that there would be considerable changes to Eircodes required after Small Areas were changed for each Census. Non-experts made pronouncements like the following

    https://www.boards.ie/vbulletin/showpost.php?p=96345955&postcount=7385
    https://www.boards.ie/vbulletin/showpost.php?p=96350729&postcount=7404
    https://www.boards.ie/vbulletin/showpost.php?p=96356110&postcount=7416

    Less than a year after Eircode launch Census 2016 was undertaken which required changes to Small Area boundaries to reflect the change in household formation and to protect privacy. This results in 43,362 Eircodes that have a different Small Area designation for the 2016 Census than they had for the 2011 Census. Either a large scale project would be required to re-allocate new Eircodes for existing addresses and ensure they were updated in everyone's databases, or the accuracy and usability of this feature would have eroded each Census.

    Now that we have data we no longer need to rely on opinion.

    It really wouldn't be much of a project. 40000 changes just isn't that many changes. You just have to manage the splits over time. It would require some work and planning to keep the thing converged and consistent, certainly. This is really an issue over 20-50 years, not 5-10 years.

    How are the RKs, which form 40 percent of the code going to be kept current as An Post is reconfigured and the flat mails business is wound down?


  • Advertisement
  • #1822
    PDVerse
    Registered Users, Registered Users 2 Posts: 160 ✭✭PDVerse


    Deleted User wrote: »
    There's no reason why eircodes can't be grouped together geographically for statistical purposes, all you need is an application that places them on a map and a "lasso" tool to select them and assign a "small area code".

    It isn't a big deal, with current processing power it would be very easy to cut-n-paste clusters of eircodes from one small area to another should the boundaries move.

    Lasso tools aren't required to perform this task. Either the point-in-polygon routine is run in a desktop GIS or in a spatially enabled database. It's a one line statement that executes in seconds.


  • #1823
    PDVerse
    Registered Users, Registered Users 2 Posts: 160 ✭✭PDVerse


    antoinolachtnai wrote: »
    It really wouldn't be much of a project. 40000 changes just isn't that many changes. You just have to manage the splits over time. It would require some work and planning to keep the thing converged and consistent, certainly. This is really an issue over 20-50 years, not 5-10 years.

    How are the RKs, which form 40 percent of the code going to be kept current as An Post is reconfigured and the flat mails business is wound down?

    Routing Keys were designed to be future proof. The number of Eircodes that have a different Routing key now compared to launch? Zero. I expect that to be the case for the rest of my lifetime.

    The kindest thing I can say about your assessment of the impact of the 40k changes is its a nice example of the back-fire effect.


  • #1824
    [Deleted User]
    Posts: 31,118 ✭✭✭✭ [Deleted User]


    PDVerse wrote: »
    Lasso tools aren't required to perform this task. Either the point-in-polygon routine is run in a desktop GIS or in a spatially enabled database. It's a one line statement that executes in seconds.
    Exactly, no biggie, anything that can relate to their relative co-ordinates will do.


  • #1825
    PDVerse
    Registered Users, Registered Users 2 Posts: 160 ✭✭PDVerse


    Deleted User wrote: »
    Exactly, no biggie, anything that can relate to their relative co-ordinates will do.

    Just so we're clear, the large project I'm referring to is the dissemination of the 40k changed Eircodes to each household, the update of all existing databases that have the old code to change it to the new code, writing of routines to accept the old code and change it to the new one should it continue to be provided. A major headache.

    The 40k affected households could spend a couple of years frustrated that one website only accepts the old code while another only accepts the new one.


  • #1826
    antoinolachtnai
    Registered Users, Registered Users 2 Posts: 9,817 ✭✭✭antoinolachtnai


    PDVerse wrote: »
    Just so we're clear, the large project I'm referring to is the dissemination of the 40k changed Eircodes to each household, the update of all existing databases that have the old code to change it to the new code, writing of routines to accept the old code and change it to the new one should it continue to be provided. A major headache.

    The 40k affected households could spend a couple of years frustrated that one website only accepts the old code while another only accepts the new one.

    This would be a big problem, because the operator does not seem to have any coherent procedure in place for dealing with a deprecated eircode. (See https://www.boards.ie/vbulletin/showpost.php?p=105318507&postcount=1759)


  • #1827
    sondagefaux
    Registered Users, Registered Users 2 Posts: 1,443 ✭✭✭sondagefaux


    Alun wrote: »
    sondagefaux wrote: »
    Not exactly typical. As I said, the vast majority of postcodes cover multiple addresses.
    I never said they were typical, but they're a lot more common than you may think especially in sparsely populated rural areas.
    What percentage of all UK postcodes do not cover multiple addresses?


  • #1828
    antoinolachtnai
    Registered Users, Registered Users 2 Posts: 9,817 ✭✭✭antoinolachtnai


    PDVerse wrote: »
    Routing Keys were designed to be future proof. The number of Eircodes that have a different Routing key now compared to launch? Zero. I expect that to be the case for the rest of my lifetime.

    What were the steps taken to make sure the routing keys were future proof?

    (How will it cope now that the customer for whom they were designed is, as expected, going through the biggest rationalisation and transformation program in its history and is beginning to wind down its traditional mainstay product?)

    This is all just self-contradictory. On the one hand, we are supposed to believe that An Post's operations will continue to be run with the same boundaries forever, no matter how the population shifts and no matter how the product mix changes.

    On the other hand, we are supposed to believe that just because an SA gets split or combined somewhere, it would mean that an SA based code would immediately require wide-scale revisions. Sure, over decades you have to cater for shifts in population but that is all very feasible. It makes an impact when you issue new codes, sure, but it doesn't necessarily mean you ever have to change an old code.


  • #1829
    PDVerse
    Registered Users, Registered Users 2 Posts: 160 ✭✭PDVerse


    antoinolachtnai wrote: »
    This would be a big problem, because the operator does not seem to have any coherent procedure in place for dealing with a deprecated eircode. (See https://www.boards.ie/vbulletin/showpost.php?p=105318507&postcount=1759)
    The process for handling deprecated Eircodes has been in place since the first quarterly update in October 2015. The issue raised in the quoted post proves my point. Updated data and clear procedures exist, but it is up to each organisation to implement the updates. Every retired Eircode is stored in the database, with relevant information to allow it to continue to be used or updated to the new Eircode if that is the scenario. It takes us less than an hour to update our database with each quarterly update, and we generally release to live the next working day. Some may only update their data once a year.


  • #1830
    [Deleted User]
    Posts: 31,118 ✭✭✭✭ [Deleted User]


    PDVerse wrote: »
    Just so we're clear, the large project I'm referring to is the dissemination of the 40k changed Eircodes to each household, the update of all existing databases that have the old code to change it to the new code, writing of routines to accept the old code and change it to the new one should it continue to be provided. A major headache.

    The 40k affected households could spend a couple of years frustrated that one website only accepts the old code while another only accepts the new one.
    I wasn't aware that householders were getting their eircodes replaced!
    Not a good move so close after the launch of the system.


  • Advertisement
  • #1831
    PDVerse
    Registered Users, Registered Users 2 Posts: 160 ✭✭PDVerse


    Deleted User wrote: »
    I wasn't aware that householders were getting their eircodes replaced!
    Not a good move so close after the launch of the system.

    They aren't. They would be if I had designed a hierarchical postcode using Small Areas.


Advertisement