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
National Postcodes to be introduced
Options
Comments
-
-
TheBustedFlush wrote: »Out of continuing curiosity, what bit of your address did you leave out, if any?
Number and street name?
You had your own name, a suburb name and Limerick and an Eircode. Would the Eircode address database have provided any further detail to the address you wrote on the envelope? And was it your address at which post is usually delivered with your name on it?
I left out the house number and the street name but I had my personal name, suburb name, and "Limerick "V94 ****"
The letter arrived yesterday with an An Post "Correct Address" label affixed to the right of where I had written. My (fully correct) postal address was written in manually on the label.
I presume that it went from Fermanagh to Belfast to Dublin to Cork to Limerick to me. It only takes three hours to drive between the post box and my house.0 -
Why didn't you send it by taxi?Trouwe Ier wrote: »I presume that it went from Fermanagh to Belfast to Dublin to Cork to Limerick to me. It only takes three hours to drive between the post box and my house.0 -
Why didn't you send it by taxi?
You'd certainly wonder if there was that much difference between the carbon footprint of me driving direct from Brookeborough to Limerick or the letter taking its circuitous route!
Lesson learned for the future: Only put stamps on those envelopes you know you'll need to use.0 -
We’ve already had two customers give us an Eircode and nothing else. In one of those cases it was a handwritten scrawl that only got handed to me when we were loading the van – pretty fecking useless when we don’t have a landline and even getting a mobile signal there can be a dice throw. And then we had the hilarity of a few customers (I will say they were well meaning so I don’t blame them) asking how to get an Eircode for their delivery locations, and I had to patiently explain to them that those sites don’t have (and can’t get) Eircodes (which was, quite rightly, greeted with incredulity).
Hard to ignore it given that it is actively making my job needlessly more difficult.
For myself I’m familiar enough with using it to know roughly where the area is. The key information is that I can tell where locations are with reference to each other, and with a map I can do a job. We’ve a couple of foldup maps that we’ve marked with GPS lines, Loc8 lines and OpenPostcode lines that seem to do the job pretty well (sadly, OpenPostcode never took off). Just blu-tac it to the portable holder and you’re good to go.
In practice, if the load is one of ours, we get the truck or van loaded first, and then when I’m back at the office I can send the detailed information needed to find the delivery point to the driver’s phone. If the load isn’t one of ours the information can be in, shall we say, a dire format. In that case where you have a handwritten scrawl listing deliveries you just need to be able to know where they are with reference to each other in order to plan the load – finding the specific locations isn’t our problem at that point.
There are times when I do need to break out the Excel. If we need to watch the tachograph hours, have multiple loads, have to watch weights on the pin, etc, it helps to have draft loading in a format that can do the calculations and be dragged’n’dropped for easy manipulating. I know my offsets with Loc8 are wrong, but as long as I can get Excel to tell their relative locations it can be made to work. The only reason any of this is possible is because, even if I don’t know its exact algorithm, it is fundamentally a geocode. Personally I prefer GPS since Cartesian coordinates are piss-easy to work with, but any geocode can be made work in this way.
Most of the above has been described before in this thread. People who work in the industry, who do planning or freight forwarding, know the basic mechanics involved – and I’ve yet to personally meet anyone who does this for a living who hasn’t shared my complaints of Eircode.
If only the ‘debate’ could even get to that.
Any postcode is, fundamentally, an information code. There are all sorts of things you can build into it (eg: error correction, non-similar codes when locations are in close proximity, compressibility, etc.), and each brings about its own unique set of constraints. Being a geocode is also a constraint. It then becomes a question of assigning weights to each constraint (i.e. determining how important each requirement is).
Finding the ‘best’ code in this context is a technical discussion involving group theory, computer science and information theory. Honest question – how many people in this thread would even be capable of comprehending the discussion let alone participating in it? And bonus kicker – anyone here have any confidence whatsoever that the decision-makers of Eircode even have an inkling about any of this????
Shame that the metric for ‘winning’ didn’t involve usability….
Since you seem so confident with your bold pronouncement of “Eircode gives us a modern functional postcode which will make postal and package/courier deliveries more accurate and efficient and can be used for navigational directions right to the individual address”, can you tell me how you can possible consider it the superior code when it cannot be used for addresses lacking Eircodes (something that, for example, Loc8 can do)?
The bit that really bugs me is that Eircode was a decision made by a politician, and one thoroughly unjustified by the available evidence imo, and plenty of folks will queue up to defend it – which, as Bayberry correctly notes, is exactly the sort of attitude that will allow another dogs dinner in the future.
Not sure about that – under 500,000 locations per routing code is unlikely to be enough capacity, and even more so when you consider that Eircodes don’t actually define a specific area any coding system will be inherently arbitrary.
To do this you’d need to construct from the set of unused Eircode’s a mapping to areas of a fixed resolution. That part is certainly do-able, but it would require that no new address codes are generated by Eircode which is totally unrealistic.
The big flaw is that Eircode is tied to pre-indexed locations. Sure, Google could try to index every fixed area but I’m pretty sure the result will either have too poor a resolution or too large a fixed area to be useful.
And yet again you're whinging that a postcode, specifically designed for postal addresses, isn't a location code.
A bit like complaining that your car can't fly.0 -
Advertisement
-
And what exactly will An Post write on that label, if the eircode is for one of those 30% of addresses that don't have a unique address?
Eircodes are unique to each address.
If it's an address in a rural townland, the combination of the address and eircode will be enough to locate the individual building.0 -
Trouwe Ier wrote: »I left out the house number and the street name but I had my personal name, suburb name, and "Limerick "V94 ****"
The letter arrived yesterday with an An Post "Correct Address" label affixed to the right of where I had written. My (fully correct) postal address was written in manually on the label.
I presume that it went from Fermanagh to Belfast to Dublin to Cork to Limerick to me. It only takes three hours to drive between the post box and my house.
It would probably have taken that route anyway.
A recent order I placed on ebay was about 100 miles by the shortest road route from me.
Its actual journey involved pick up by a local courier from the sender in rural Cumbria, transfer of package from the sender's local courier's location to a depot in Carlisle, transfer from the Carlisle depot to a regional hub outside Leeds, transfer from the regional hub to a depot in Darlington, transfer from the Darlington depot to my local courier's location, delivery by my local courier to me.
If you look up just the Carlisle depot to Leeds regional hub to Darlington depot part of the package's full journey, you'll see that it's a lot more than 100 miles.
This is normal for all postal and courier companies. Items don't get delivered using the shortest possible route. They're delivered after a series of transfers from one of the delivery company's locations to its other locations until the final transfer to the delivery worker who is responsible for delivery to the recipient.0 -
marmurr1916 wrote: »Eircodes are unique to each address.
No they're not, Several addresses have multiple eircodes, one for each postal delivery point at the same address.0 -
Why would Google support them? They aren’t the national postcode and have no official standing. Had Loc8 been chosen as the national postcode, and their algorithm released, then absolutely Google would have had support because it would have been almost trivial to add it.Isn't loc8 a geocode? They've had 5 years to get it on google maps and they haven't.
The single strongest complaint I have about Loc8, namely that it isn’t open, applies even more strongly to Eircode.
It is good to know that your sniping appears to have nothing to do with the quality or lack thereof of either Loc8 or Eircode.Loc8 get sniped at because they are unprofessional and don't know how to conduct themselves as a respectable organisation and get taken seriously.
But the real hole in your argument is that it isn’t an apples to apples comparison. Had Loc8 (or, for that matter, any alternative geocode such as OpenPostcode) been chosen and they had government funding and support what would have been the result? That’s the real comparison you should be making, but won’t because…well…that part never gets addressed.
Of course I am complaining. In what fecking universe was it a good idea to spend €27 million on what should be a piece of national infrastructure but exclude most of the functionality???? In what fecking universe was it a good idea to solicit the expertise of a wide range of stakeholders, with almost all recommending functionality like this (and I personally haven’t seen a single submission that argued against including such), only to turn around later and completely disregard that common sense recommendation for no rational reason???? In what fecking universe was it sensible to deliberately design a piece of national infrastructure in a manner that crippled its interoperability (a point raised, and seemingly accepted, several times during the consultation process)????marmurr1916 wrote: »And yet again you're whinging that a postcode, specifically designed for postal addresses, isn't a location code.
When you read the documents in the run up to before Eircode went crazy with its design you see some glaring differences between what was expected and what was actually delivered. In the 2010 Joint Committee report it was stated “One of the main drivers for the introduction of a postcode system in Ireland is the forthcoming liberalisation of the postal market.” How does basing Eircode entirely on An Post’s delivery centres help do that exactly….???
In the actual announcement from Eamon Ryan was this: “A locational code system, such as the one envisaged, will also unlock the potential across Government Departments for use of this spatial data for policy planning.” It certainly seemed that this was supposed to be locational from the outset.
From the 2010 report again were these aims of the postcode: “ to improve mechanisms to handle spatial data, facilitating health service planning ” (spatial data, almost like it was envisioned to be a geocode), “to facilitate emergency services” (which Eircode doesn’t for the reasons I explained in depth previously) and “to permit new types of “geographic” based services ” (directly contradicting any claim that this was never intended to be a location code).
From the principles of that 2010 report (repeating an earlier 2005 report): “It is a public postcode that is proposed, not a “hidden” or technical code” (Eircode utterly fails this) and “that the system of postcodes must be appropriate to the needs of a developing smart economy and that future proofing should be an essential yardstick” (Eircode threw interoperability out the window).
That Eircode did a complete about-turn at the 11th hour to ignore all of this was fecking crazy. I’m not buying the attempt at revising history that this was never intended to be a location code because given that a) I actually followed the process as it unfolded and b) anyone can easily look up the submissions and reports and see what was actually supported/recommended/proposed.0 -
Well, Loc8 ain't dead yet. If they can provide codes for non-Eircode locations, they have a use. And they certainly raised their public profile. :pac:
Maybe they could map each Eircode to a Loc8 code as well.0 -
Advertisement
-
dr.fuzzenstein wrote: »Well, Loc8 ain't dead yet. If they can provide codes for non-Eircode locations, they have a use. And they certainly raised their public profile. :pac:
Maybe they could map each Eircode to a Loc8 code as well.
You sir have just come up with the way to save Loc8.
Mapping the eircode is essential & then an additional layer of adding non eircode, such as car parks, tourist attractions, churches, etc.
If they did that (& changed their horrible interface), I'd use them.0 -
You sir have just come up with the way to save Loc8.
Mapping the eircode is essential & then an additional layer of adding non eircode, such as car parks, tourist attractions, churches, etc.
If they did that (& changed their horrible interface), I'd use them.
I think this would have been self-evident to Loc8, Go Code and OpenPostcode long before now.
4 code systems instead of one. Put them all together in an app and use what you like.0 -
TheBustedFlush wrote: »I think this would have been self-evident to Loc8,
I don't know. I've seen some of the posts that guy has made, both here & on Facebook & I'm not sure it works like that for him0 -
Why would Google support them? They aren’t the national postcode and have no official standing. Had Loc8 been chosen as the national postcode, and their algorithm released, then absolutely Google would have had support because it would have been almost trivial to add it.
The single strongest complaint I have about Loc8, namely that it isn’t open, applies even more strongly to Eircode.
It is good to know that your sniping appears to have nothing to do with the quality or lack thereof of either Loc8 or Eircode.
But the real hole in your argument is that it isn’t an apples to apples comparison. Had Loc8 (or, for that matter, any alternative geocode such as OpenPostcode) been chosen and they had government funding and support what would have been the result? That’s the real comparison you should be making, but won’t because…well…that part never gets addressed.
Of course I am complaining. In what fecking universe was it a good idea to spend €27 million on what should be a piece of national infrastructure but exclude most of the functionality???? In what fecking universe was it a good idea to solicit the expertise of a wide range of stakeholders, with almost all recommending functionality like this (and I personally haven’t seen a single submission that argued against including such), only to turn around later and completely disregard that common sense recommendation for no rational reason???? In what fecking universe was it sensible to deliberately design a piece of national infrastructure in a manner that crippled its interoperability (a point raised, and seemingly accepted, several times during the consultation process)????
When you read the documents in the run up to before Eircode went crazy with its design you see some glaring differences between what was expected and what was actually delivered. In the 2010 Joint Committee report it was stated “One of the main drivers for the introduction of a postcode system in Ireland is the forthcoming liberalisation of the postal market.” How does basing Eircode entirely on An Post’s delivery centres help do that exactly….???
In the actual announcement from Eamon Ryan was this: “A locational code system, such as the one envisaged, will also unlock the potential across Government Departments for use of this spatial data for policy planning.” It certainly seemed that this was supposed to be locational from the outset.
From the 2010 report again were these aims of the postcode: “ to improve mechanisms to handle spatial data, facilitating health service planning ” (spatial data, almost like it was envisioned to be a geocode), “to facilitate emergency services” (which Eircode doesn’t for the reasons I explained in depth previously) and “to permit new types of “geographic” based services ” (directly contradicting any claim that this was never intended to be a location code).
From the principles of that 2010 report (repeating an earlier 2005 report): “It is a public postcode that is proposed, not a “hidden” or technical code” (Eircode utterly fails this) and “that the system of postcodes must be appropriate to the needs of a developing smart economy and that future proofing should be an essential yardstick” (Eircode threw interoperability out the window).
That Eircode did a complete about-turn at the 11th hour to ignore all of this was fecking crazy. I’m not buying the attempt at revising history that this was never intended to be a location code because given that a) I actually followed the process as it unfolded and b) anyone can easily look up the submissions and reports and see what was actually supported/recommended/proposed.
Loc8 received government funding from Enterprise Ireland, they didn't tender for the contract to be the national postcode because they couldn't figure out how the tender process worked.0 -
Join Date:Posts: 19365
Loc8 received government funding from Enterprise Ireland, they didn't tender for the contract to be the national postcode because they couldn't figure out how the tender process worked.
The existence of Loc8, which appears strongly in many of your replies, has nothing to do with the faults and failings of Eircode. You assume that anyone who complains about Eircode and its failings is a supporter of Loc8 - that is not true.
To go with 139 An Post sorting offices as the basis for routing codes is a nonsense choice when An Post have over 2,000 post towns. If they had used these 2,000 or so as the basis of routing keys it would not be so bad. If these were evenly spread among the 2.2m addresses, it would mean 1,000 addresses per routing key - a much more usable granularity for non-An Post users.
That these 139 routing key areas are non-contiguous is absolute nonsense and they have not published a map of routing keys - why?
Why all the secrecy?0 -
Sam Russell wrote: »The existence of Loc8, which appears strongly in many of your replies, has nothing to do with the faults and failings of Eircode. You assume that anyone who complains about Eircode and its failings is a supporter of Loc8 - that is not true.
To go with 139 An Post sorting offices as the basis for routing codes is a nonsense choice when An Post have over 2,000 post towns. If they had used these 2,000 or so as the basis of routing keys it would not be so bad. If these were evenly spread among the 2.2m addresses, it would mean 1,000 addresses per routing key - a much more usable granularity for non-An Post users.
That these 139 routing key areas are non-contiguous is absolute nonsense and they have not published a map of routing keys - why?
Why all the secrecy?
I replied to a comment using loc8 as an example of a geocode that's not on Google, loc8 are the most high profile (for all the wrong reasons) of the geocodes, so I tend I think of them when I think of geocodes.
I'm well aware you can dislike both ericode and loc8 or have a dislike of eircode and no opinion on loc80 -
-
I replied to a comment using loc8 as an example of a geocode that's not on Google, loc8 are the most high profile (for all the wrong reasons) of the geocodes, so I tend I think of them when I think of geocodes.
I'm well aware you can dislike both ericode and loc8 or have a dislike of eircode and no opinion on loc8
Can you please, please, please learn to type eircode correctly? For someone who advocates on their behalf so much, it is depressing how often you call them ericode.
In the meantime, Eircode was designed from the ground up. Even if you never mention Loc8, it can still be argued that in designing something from the ground up, we should have gotten the best we possibly could for the money.
NOthing you have said has persuaded me that we did.0 -
Can you please, please, please learn to type eircode correctly? For someone who advocates on their behalf so much, it is depressing how often you call them ericode.
In the meantime, Eircode was designed from the ground up. Even if you never mention Loc8, it can still be argued that in designing something from the ground up, we should have gotten the best we possibly could for the money.
NOthing you have said has persuaded me that we did.
I'm not here to persuade anyone, I voice my opinion, take it or leave it.
(I typed "ericode" wrong on my phone once and it saved, now it autocorrects to that each time I type eircode, it's not that I don't know the correct name, it's just a typo)0 -
Join Date:Posts: 19365
I replied to a comment using loc8 as an example of a geocode that's not on Google, loc8 are the most high profile (for all the wrong reasons) of the geocodes, so I tend I think of them when I think of geocodes.
I'm well aware you can dislike both ericode and loc8 or have a dislike of eircode and no opinion on loc8
You have not explained why Eircode is so angled towards An Post (who have said they have no need of a post code and will not be using it) nor why only 139 post towns were used instead of a more representative number of their over 2,000 post towns. Nor have you given a rationale as to why there is no map of the routing keys. Nor as to the reason for the atmosphere of secrecy surrounding Eircode and its implementation. The do not even have a 'Contact us' on their website.
Proposers of this Eircode system have both championed it as being all that is required to address a property and then ridiculed people who expected An Post to deliver letters with Eircodes on them as the address. You cannot have it both ways. Eircodes are not sufficient for apartments without further information, nor are they sufficient for rural addresses without a pocket computer (or satnav when/if they implement them).
For the record I do not have any opinion about Loc8 nor about any other coding system. I just feel that a dog's dinner of a solution (similar to the dreadful eVoting disaster) has been foisted on us by an incompetent civil service. We deserve better.
However, it is relatively easy (but unlikely) for the Minister to pull the plug (a la Seamus Brennan when a very poor design of direction signs were being foisted on Dublin). Most of the design work would carry over*.
* Just increase the number of routing keys to 2,000 for a start. This would allow the random bit to remain the same. If they were logically arranged, it could make the system quite good. People could them omit the random bit for when privacy concerns arise and I for one would then support them - although with reservations.0 -
Advertisement
-
Sam Russell wrote: »You have not explained why Eircode is so angled towards An Post (who have said they have no need of a post code and will not be using it) nor why only 139 post towns were used instead of a more representative number of their over 2,000 post towns. Nor have you given a rationale as to why there is no map of the routing keys. Nor as to the reason for the atmosphere of secrecy surrounding Eircode and its implementation. The do not even have a 'Contact us' on their website.
Proposers of this Eircode system have both championed it as being all that is required to address a property and then ridiculed people who expected An Post to deliver letters with Eircodes on them as the address. You cannot have it both ways. Eircodes are not sufficient for apartments without further information, nor are they sufficient for rural addresses without a pocket computer (or satnav when/if they implement them).
For the record I do not have any opinion about Loc8 nor about any other coding system. I just feel that a dog's dinner of a solution (similar to the dreadful eVoting disaster) has been foisted on us by an incompetent civil service. We deserve better.
However, it is relatively easy (but unlikely) for the Minister to pull the plug (a la Seamus Brennan when a very poor design of direction signs were being foisted on Dublin). Most of the design work would carry over*.
* Just increase the number of routing keys to 2,000 for a start. This would allow the random bit to remain the same. If they were logically arranged, it could make the system quite good. People could them omit the random bit for when privacy concerns arise and I for one would then support them - although with reservations.
I'm not sure why he in particular "has to explain" anything.
Presumably post towns were used because this is what was originally recommended as the starting point for any code design - it had to define these.
The number of post-towns that needed to be defined was best advised by the USP since they have a requirement to deliver mail to every part of the country - no other company is required to do this - and to provide that mail delivery service at a national uniform cost in terms of postage stamp prices. And obviously mail does not just mean letters, it means parcels too.
I suspect - but do not know - that if 2,000 post-towns were used as the first part of the code then you would have to use a different set of characters to denote these. And those might be too granular for An Post purposes, and potentially, at the same time be seen to be too specific to them.
I agree with you on the map of the routing keys, but it would appear that they can be crowd-sourced relatively easily once the database emerges into public. No reason it couldn't have been issued by Eircode.
Not sure what you mean by the atmosphere of secrecy surrounding its implementation - I would have thought it was anything but secret given the almost daily conversation about it on boards such as this.
As for not being contactable, it took me all of 5 secs to find this in their FAQs:
How do I contact Eircode?
For queries please email us at hello@eircode.ie or submit a query. You can also write to us at:
Block C
Maynooth Business Campus
Maynooth
Co. Kildare
W23 F854.
0818 300 005
"Proposers of this Eircode system have both championed it as being all that is required to address a property and then ridiculed people who expected An Post to deliver letters with Eircodes on them as the address. You cannot have it both ways. Eircodes are not sufficient for apartments without further information, nor are they sufficient for rural addresses without a pocket computer (or satnav when/if they implement them)."
I haven't read any claims that Eircode was all that was required to address a property. It was promoted as a unique location code for Irish addresses - in other words, a code - combined with an address - that uniquely identifies it and its location. That's been my reading of it.
People looking to test it on its own to see whether mail would be delivered were never told this would work, nor was it promised by Eircode or An Post that it would from what I have read. In fact, Eircode/An Post deliberately said it should be included as the last line of the address , not "ah sure, don't bother writing in the address at all, we'll find ya anyway".
What's a pocket computer? If you mean a smartphone, say so. a lot of the population has them and that number will continue to grow.
Satnav adoption takes time - if the legislation to use eircodes only passed in June, then it follows that the relevant licences to issue the databases could only have happened after that. And I'd reckon that the software updates for satnav software probably only happens 2-3 times a year, and for Ireland possibly less. However, I'd take a bet that some of the satnav companies have one eye on the Christmas market and will look to have it on their devices, if they can, for that time - otherwise, their sales may go through the floor, if they're not available in time. Just a guess on my part, I could be wrong.
Google test and sample data for integration with their own systems, no matter who is issuing it. So again, if they are only receiving a large enough sample or full database now, then it's going to take a bit of time for it to be on their systems, but I can't see them ignoring it with their European headquarters based here for this kind of location data.
And yes, I think it could have been better.0 -
Sam Russell wrote: »You have not explained why Eircode is so angled towards An Post (who have said they have no need of a post code and will not be using it) nor why only 139 post towns were used instead of a more representative number of their over 2,000 post towns. Nor have you given a rationale as to why there is no map of the routing keys. Nor as to the reason for the atmosphere of secrecy surrounding Eircode and its implementation. The do not even have a 'Contact us' on their website.
Sorry you seem to under the impression I have some obligation to explain this to you. I have no such obligation. I will post my opinion the same way everyone else does here.0 -
Join Date:Posts: 19365
TheBustedFlush wrote: »I'm not sure why he in particular "has to explain" anything.
Presumably post towns were used because this is what was originally recommended as the starting point for any code design - it had to define these.
The number of post-towns that needed to be defined was best advised by the USP since they have a requirement to deliver mail to every part of the country - no other company is required to do this - and to provide that mail delivery service at a national uniform cost in terms of postage stamp prices. And obviously mail does not just mean letters, it means parcels too.
I suspect - but do not know - that if 2,000 post-towns were used as the first part of the code then you would have to use a different set of characters to denote these. And those might be too granular for An Post purposes, and potentially, at the same time be seen to be too specific to them.
I will just answer this point.
There are 25 characters in the Eircode alphabet, so a three character code could give 25 x 25 x 25 combinations = 15,625 so plenty of room. However, just using 15 characters at the start, and using a selection of characters in position three (they currently allow 'W') it would be simple to get to 2,500 combinations. So you have 15*10*20 would give 3,000. Omitting some letters might reduce the chance of rude or unpleasant combinations. The result would be Letter;number;character. I would think that would be OK as a routing code if a three character code was essential, but expansion to four characters would not be the end of the world.
More routing keys would not impact An Post in the slightest since their sorting system is fully mechanised, and it already deals with those post towns. In fact it would be an advantage to them.
On another issue, now why would the Universal Service Provider try to set up a fault ridden system when they have no need of it? They would not try to scupper it so their competitors would be disadvantaged by it - that would be bizarre, so it must be some other reason but I cannot think of one myself. They appear (by the fact they hand write addresses when they only have the Eircode) not to have coded it into their systems, so they have not used their advanced sight of Eircode to ready their sorting system which would suggest they will not be implementing it any time soon.
And yes, I think it could have been better.
So we agree on one thing.0 -
As one who appears to have been in the "soft" pro-Eircode camp, I would have to agree with the comments about granularity.
If the routing key identifiers had two letters and a digit then, potentially 15x15x10 i.e. 2250 could have been created with three characters.
By assigning just three characters to the random identifiers, then 24 x 24 x 24 i.e 13,824....but round down to 10,000....to take account of any obscenities etc....unique identifiers could have been created.
So, there would have been enough capacity, potentially for 22,500,000 addresses: a "comfort factor" of 10.
For sensitivity's sake, applying this logic across the whole island and assuming something like 3,100,000 addresses, then each of 2,250 routing key areas would have, on average, 1,377 addresses, so most towns with a population of greater than 4,000 people could have at least routing key code. The average size of the routing key area would be about 38 km2 but there would be a huge standard deviation.
Post codes would look like this: VE7 X5H (symmetrically 3+3 and one digit less to remember).
However, it wouldn't work in Dublin as you would need more than 10,000 unique identifiers in the likes of D15 or D24...then you would have to filter out routing key combinations like "FU" or "VD" etc.
Nevertheless, using the current syntax, up to 1500 routing key areas could still have been created.
The system has benefitted me but they have messed up with many of the routing key areas. By my reckoning, there are about 23 in Cork City and County but only 2.4 in Limerick City and County.....and V94 is the most irrational of the lot.0 -
Sam Russell wrote: »The existence of Loc8, which appears strongly in many of your replies, has nothing to do with the faults and failings of Eircode. You assume that anyone who complains about Eircode and its failings is a supporter of Loc8 - that is not true.
To go with 139 An Post sorting offices as the basis for routing codes is a nonsense choice when An Post have over 2,000 post towns. If they had used these 2,000 or so as the basis of routing keys it would not be so bad. If these were evenly spread among the 2.2m addresses, it would mean 1,000 addresses per routing key - a much more usable granularity for non-An Post users.
That these 139 routing key areas are non-contiguous is absolute nonsense and they have not published a map of routing keys - why?
Why all the secrecy?
As someone who generally supports Eircode, with a few reservations, I would totally agree with these two criticisms.
I think an 'official' map has not been published yet for no reason other than prioritising getting the system running over satisfying the nerds. It's not like it's possible to keep this information secret, and it would have been good PR to put out a high-res PDF.
Though I agree with you on the smaller areas, it will still be possible to make areas of any size you like, it's not hard to write an algorithm to sort lat/long into separate areas. It may not be as useful as you think, though, since different uses require different granularity. Nevertheless, I think many people would have defaulted to those areas as a rule of thumb and they would have been fine, but that's in a parallel universe, to borrow a phrase.
But some of the other criticisms are a mixture of nonsense, default anti-everything whining, and desperate post-rationalisations of why differences with other systems are required features.
On balance, there are huge advantages to this system. There are no advantages to fighting lost battles.0 -
On balance, there are huge advantages to this system. There are no advantages to fighting lost battles.
One of the key issues with Ireland is a tendency to go "yerrah it'll do" and that is the primary argument in favour of Eircode as designed. IT could have been so much more.
Personal view is that while it may be a losing battle in terms of fixing the post code system to something better and more in line with best international practice given the opportunity it was and wasted, it may have beneficial effects on decisions in the future; it may encourage more people to voice opinions before it gets to this stage with future projects. It may make our decisions makers more inclined to have a longer term view rather than, as seems to be the case, look at short term monetary gains which come with a long term opportunity cost.
As such, I see advantages to pointing out where things could be done better and that there is an increasing demand towards doing things better in this country.
That, in itself, may be tremendously valuable in the future.0 -
Carawaystick wrote: »No they're not, Several addresses have multiple eircodes, one for each postal delivery point at the same address.
Individual buildings may have multiple different addresses within them. Each address within the building has its own eircode which is unique to that unique address within the building.
An example I gave earlier in the thread was the Wilton Shopping Centre in Cork.
Each unit has its own address and its own eircode within the same building.
However, a building is not necessarily an individual address in its own right, especially if it's subdivided into different units.0 -
Join Date:Posts: 19365
As someone who generally supports Eircode, with a few reservations, I would totally agree with these two criticisms.
I think an 'official' map has not been published yet for no reason other than prioritising getting the system running over satisfying the nerds. It's not like it's possible to keep this information secret, and it would have been good PR to put out a high-res PDF.
Though I agree with you on the smaller areas, it will still be possible to make areas of any size you like, it's not hard to write an algorithm to sort lat/long into separate areas. It may not be as useful as you think, though, since different uses require different granularity. Nevertheless, I think many people would have defaulted to those areas as a rule of thumb and they would have been fine, but that's in a parallel universe, to borrow a phrase.
Smaller areas would allow the user to decide how to incorporate them. For example whether to serve areas (now much smaller) from Depot 1 or Depot 7 - and not having to decode the full Eircode, convert to Lat/Lon and then check which is the appropriate depot - not so easy when the routing keys are non-contiguous, and not easy to do from the top of your head.But some of the other criticisms are a mixture of nonsense, default anti-everything whining, and desperate post-rationalisations of why differences with other systems are required features.
On balance, there are huge advantages to this system. There are no advantages to fighting lost battles.
I am not anti-everything at all. I am not championing any system but I do feel a much better system could have been achieved.
I do not like the random codes, but hey ....
I do think an attempt to standardised addressing in Ireland, and I include the requirement to number houses in named streets that have names.
I do think a better attempt at dealing with non-unique addresses should have been attempted even if it took time.
That and smaller routing keys and we would be fine.
I do like the fixed length code, but D6W was a big mistake.0 -
Why would Google support them? They aren’t the national postcode and have no official standing. Had Loc8 been chosen as the national postcode, and their algorithm released, then absolutely Google would have had support because it would have been almost trivial to add it.
The single strongest complaint I have about Loc8, namely that it isn’t open, applies even more strongly to Eircode.
It is good to know that your sniping appears to have nothing to do with the quality or lack thereof of either Loc8 or Eircode.
But the real hole in your argument is that it isn’t an apples to apples comparison. Had Loc8 (or, for that matter, any alternative geocode such as OpenPostcode) been chosen and they had government funding and support what would have been the result? That’s the real comparison you should be making, but won’t because…well…that part never gets addressed.
Of course I am complaining. In what fecking universe was it a good idea to spend €27 million on what should be a piece of national infrastructure but exclude most of the functionality???? In what fecking universe was it a good idea to solicit the expertise of a wide range of stakeholders, with almost all recommending functionality like this (and I personally haven’t seen a single submission that argued against including such), only to turn around later and completely disregard that common sense recommendation for no rational reason???? In what fecking universe was it sensible to deliberately design a piece of national infrastructure in a manner that crippled its interoperability (a point raised, and seemingly accepted, several times during the consultation process)????
When you read the documents in the run up to before Eircode went crazy with its design you see some glaring differences between what was expected and what was actually delivered. In the 2010 Joint Committee report it was stated “One of the main drivers for the introduction of a postcode system in Ireland is the forthcoming liberalisation of the postal market.” How does basing Eircode entirely on An Post’s delivery centres help do that exactly….???
In the actual announcement from Eamon Ryan was this: “A locational code system, such as the one envisaged, will also unlock the potential across Government Departments for use of this spatial data for policy planning.” It certainly seemed that this was supposed to be locational from the outset.
From the 2010 report again were these aims of the postcode: “ to improve mechanisms to handle spatial data, facilitating health service planning ” (spatial data, almost like it was envisioned to be a geocode), “to facilitate emergency services” (which Eircode doesn’t for the reasons I explained in depth previously) and “to permit new types of “geographic” based services ” (directly contradicting any claim that this was never intended to be a location code).
From the principles of that 2010 report (repeating an earlier 2005 report): “It is a public postcode that is proposed, not a “hidden” or technical code” (Eircode utterly fails this) and “that the system of postcodes must be appropriate to the needs of a developing smart economy and that future proofing should be an essential yardstick” (Eircode threw interoperability out the window).
That Eircode did a complete about-turn at the 11th hour to ignore all of this was fecking crazy. I’m not buying the attempt at revising history that this was never intended to be a location code because given that a) I actually followed the process as it unfolded and b) anyone can easily look up the submissions and reports and see what was actually supported/recommended/proposed.
In its most recent iteration it was designed to be a postcode. That's what it is, that's what it does. If you want a code that allows you to find an electricity pole in the middle of a field, don't use eircode which wasn't designed for this task.
PS: would it kill you to apply a bit of brevity and levity to your posts?0 -
Advertisement
-
Sam Russell wrote: »Smaller areas would allow the user to decide how to incorporate them. For example whether to serve areas (now much smaller) from Depot 1 or Depot 7 - and not having to decode the full Eircode, convert to Lat/Lon and then check which is the appropriate depot - not so easy when the routing keys are non-contiguous, and not easy to do from the top of your head.
I am not anti-everything at all. I am not championing any system but I do feel a much better system could have been achieved.
I do not like the random codes, but hey ....
I do think an attempt to standardised addressing in Ireland, and I include the requirement to number houses in named streets that have names.
I do think a better attempt at dealing with non-unique addresses should have been attempted even if it took time.
That and smaller routing keys and we would be fine.
I do like the fixed length code, but D6W was a big mistake.
So basically street address in rural areas, with currently nameless roads being assigned names and houses being assigned numbers.
That would take years to achieve, meet huge public resistance and require an enormous amount of money to implement.
Ireland has a fairly unique addressing system for most rural areas - any postcode needs to acknowledge that and work with it rather than trying to completely change the addressing system.0
This discussion has been closed.
Advertisement