Eircode - its implemetation (merged)

Sam Russell wrote: »

The only way to do it is to give unique addresses to all areas, and to have a statuary authority responsible for the maintenance of those addresses. A fixed format for addresses should be implemented from this.

However, the opportunity for this has been missed. Only Revenue could enforce this now through the LPT.

I would agree, but it's an entirely separate issue and doesn't relate to eircode or any post code.

Creating unique addresss outside of a postcode implementation still wouldn't solve the problem you raise "I don't know their exact address or their house location on a map"

If they gave every house a unique address, then you'd still have to actually ask that person for their new unique address. Which is ironic because giving every household a unique address is exactly what they have done with eircode. But as you've protested before, you don't feel comfortable asking people for it, as I remember you saying "you don't know them that well and might not have their phone number, or might only see them at funerals or Christmas"

You're "problem" is extremely spurious, I don't see how ANYTHING could solve it for you, except you know, actually asking the person "what's your address and postcode" which you seem to have an aversion to and expect a government department to solve it for you

ukoda wrote: »

I'm struggling to think of a postcode design that lets you find an address on a map when you've no idea which house it is?

Geo only code - can't do it
Heirarchical code - can't do it

The way to solve this is for people to share their eircode (or ANY postcode)

There's also no where to find a persons mobile number without asking them but we do ok with this.

The only way to do it is to give unique addresses to all areas, and to have a statuary authority responsible for the maintenance of those addresses. A fixed format for addresses should be implemented from this.

However, the opportunity for this has been missed. Only Revenue could enforce this now through the LPT.

Sam Russell wrote: »

As I have said many times, if you know someone's address in the non-unique area, there is no way to find their Eircode unless you can find it on an OS map or you ask them yourself. This will limit its adoption in the very area it is needed most. Knowing your own phone number, or your own Eircode is one thing. For those that wish to communicate with you - well they need to know it as well. I can find out landline numbers if it is in the phonebook or it is available on directory inquires but otherwise I am out of luck. With Eircode, it is as I stated above.

I'm struggling to think of a postcode design that lets you find an address on a map when you've no idea which house it is?

Geo only code - can't do it
Heirarchical code - can't do it

The way to solve this is for people to share their eircode (or ANY postcode)

There's also no where to find a persons mobile number without asking them but we do ok with this.

oscarBravo wrote: »

Fair enough - it was probably overstating the case to say that it's one of the biggest hurdles.

As for your second hurdle, they came up with a way around that: they posted out Eircodes. Granted, many people managed to lose their code, but if I give you something and you lose it, it's a little unfair to complain to me that you don't have access to it.

As I have said many times, if you know someone's address in the non-unique area, there is no way to find their Eircode unless you can find it on an OS map or you ask them yourself. This will limit its adoption in the very area it is needed most. Knowing your own phone number, or your own Eircode is one thing. For those that wish to communicate with you - well they need to know it as well. I can find out landline numbers if it is in the phonebook or it is available on directory inquires but otherwise I am out of luck. With Eircode, it is as I stated above I am out of luck.

oscarBravo

hans aus dtschl wrote: »

I don't think that's accurate. You're saying that really high up on the list of issues Eircode faces is that people believe that it's possible to satisfy all conflicting requirements? That's a strange thing to say. I'd say that's pretty low on the list of hurdles they face. Way below things like end user adoption, adoption by mapping system providers, adoption by the national postal company, etc.

I work in providing various automated systems to customers myself, it's absolutely normal to have issues with rollout. It's never yet been acceptable for me to say "one of the biggest problems we'll have is that because people think it can be perfect, it won't make everyone happy".
If anyone at Eircode is thinking like that (I'm not suggesting they are), they've already lost.

They'll be able to address a lot of the issues they face. Some of the bigger issues that they face are
End user adoption - end users aren't compelled to use it when dealing with government at the moment, if they were then adoption would be faster than currently. It's their primary issue. It's literally the metric by which their success will be measured, long term.
Rural access - many rural people who the system will benefit most don't have great access to internet, so the lookup website is not as useful to them as it is to someone in a city (where ironically the system is of least benefit). They could alleviate this by having other ways for people to find postcodes. Print a basic map, put eircodes in the phone book, etc, there are ways of addressing this.
Map providers not using Eircode - they're working on this I think - it's a critical issue
GPS/SatNav providers not using Eircode - they're working on this I think - same as above
Couriers/Post not using Eircode - this will improve to a degree if the above are improved
Emergency services uptake - I'm not sure if this is critical in determining the success of Eircode
.....
....
somewhere way down here is "people think it can satisfy all requirements"

Fair enough - it was probably overstating the case to say that it's one of the biggest hurdles.

I'll reframe: one of the biggest problems in any discussion about Eircode is the perception in some quarters that it could have been designed in a way that would have exempted it from criticism. As such, it's constantly being compared to either a putative perfect system (where it will inevitably fall short), or to a system that would have better addressed the requirements of whomever is doing the criticism (where it will inevitably fall short).

As for your second hurdle, they came up with a way around that: they posted out Eircodes. Granted, many people managed to lose their code, but if I give you something and you lose it, it's a little unfair to complain to me that you don't have access to it.

hans aus dtschl

oscarBravo wrote: »

One of the biggest hurdles Eircode faces is the belief, counter to all logic, that it's possible to design something to perfectly satisfy all conflicting requirements.

I don't think that's accurate. You're saying that really high up on the list of issues Eircode faces is that people believe that it's possible to satisfy all conflicting requirements? That's a strange thing to say. I'd say that's pretty low on the list of hurdles they face. Way below things like end user adoption, adoption by mapping system providers, adoption by the national postal company, etc.

I work in providing various automated systems to customers myself, it's absolutely normal to have issues with rollout. It's never yet been acceptable for me to say "one of the biggest problems we'll have is that because people think it can be perfect, it won't make everyone happy".
If anyone at Eircode is thinking like that (I'm not suggesting they are), they've already lost.

They'll be able to address a lot of the issues they face. Some of the bigger issues that they face are
End user adoption - end users aren't compelled to use it when dealing with government at the moment, if they were then adoption would be faster than currently. It's their primary issue. It's literally the metric by which their success will be measured, long term.
Rural access - many rural people who the system will benefit most don't have great access to internet, so the lookup website is not as useful to them as it is to someone in a city (where ironically the system is of least benefit). They could alleviate this by having other ways for people to find postcodes. Print a basic map, put eircodes in the phone book, etc, there are ways of addressing this.
Map providers not using Eircode - they're working on this I think - it's a critical issue
GPS/SatNav providers not using Eircode - they're working on this I think - same as above
Couriers/Post not using Eircode - this will improve to a degree if the above are improved
Emergency services uptake - I'm not sure if this is critical in determining the success of Eircode
.....
....
somewhere way down here is "people think it can satisfy all requirements"

oscarBravo

Irish Steve wrote: »

Some are just disappointed that an opportunity to put an all encompassing system into place has been missed...

There are no all-encompassing systems. No matter how our postcodes were designed, there would have been flaws from one perspective or another.

One of the biggest hurdles Eircode faces is the belief, counter to all logic, that it's possible to design something to perfectly satisfy all conflicting requirements.

Irish Steve

ukoda wrote: »

I think it's more a case of how loud the people who dislike the design are, as opposed to how many of them there are.

brutal honesty, the vast majority neither know nor care about how good or bad the design and implementation is, as it doesn't significantly affect their day to day life.

Some are just disappointed that an opportunity to put an all encompassing system into place has been missed, some are incandescent about the whole process because they wanted to be involved and were excluded for strange reasons, and some are hacked off that the eventual solution seems to have been very much tailored towards the requirements of a nanny state rather than facilitating the state AND business that has a requirement for an aid to planning quicker, faster and cheaper deliveries without spending a fortune on complex IT solutions in order to get that information.

The underlying issue was that An Post neither wanted or needed any form of post code system, as they'd already built their own bespoke solution that covered the requirement, and I have to admit to some concern about the decision to incorporate a very strange An Post structure at the core of the system, given the lack of interest from An Post in the first place.

What we've ended up with is a system that can facilitate the tracking of people to an address with very specific and defining accuracy, which for state services is a big plus, as the implementation very specifically excludes any form of manipulation to use an invalid code, but without access to the specific geocode information or a computer, right now, it's not a lot of help to the majority of people in the country.

If Navteq and the like do eventually manage to negotiate sensible access rights, we might then see Eircodes implemented in GPS systems, and if that happens, it will make finding obscure locations anywhere in the country a non event, if it doesn't happen, then Eircode will become as irrelevant to day to day living as Loc8 will become, if a system is not taken on board by the major players of Sat Nav systems, regardless of what platform it actually runs on, the use of that system will be crippled, and I have major concerns that the Eircode suppliers will be too greedy and as a result, the product will not be adopted.

It was the case for a long time that the digital mapping of Ireland wasn't used by the GPS suppliers, so vast tranches of the country had NO coverage on GPS, and one of the reasons for that was the price being sought by Ordnance Survey for their mapping data, there was no vision that enabling people to find their way around the island would be good for so many other services.

Time will tell, right now, I'm not at all confident that the long term use of Eircode is going to achieve what it should have been capable of, partly because of poor design and implementation strategies, and partly because of the lack of vision for the future that some of the suppliers seem to be showing with their requirements for access to the database information.

hans aus dtschl wrote: »

Yep, but isn't it interesting how many people seem to dislike the design?
For all the technical advantages it has, irrespective of how few flaws it has, many people dislike the design.

Well, I find that interesting, anyway.

I think it's more a case of how loud the people who dislike the design are, as opposed to how many of them there are.

hans aus dtschl

BowWow wrote: »

That didn't take long. Thread is now going down that usual road of how the wheel Eircode could have been designed.........

Yep, but isn't it interesting how many people seem to dislike the design?
For all the technical advantages it has, irrespective of how few flaws it has, many people dislike the design.

Well, I find that interesting, anyway.

stoneill

By some fortune or misfortune, the town of Delta, British Columbia has the postal code V4G1N4.

just noticed an Eircode endorsement on another boards forum.

http://www.boards.ie/vbulletin/showpost.php?p=98301724&postcount=2716

Some couriers starting to use it obviously.

xband

It'll be interesting to see if it does get implemented in the various commercial mapping services.

Given how long it's taken so far, I wouldn't hold my breath though. This isn't a very big market and I can't really see some of those services getting tied into paying for database access or implementing technically complex solutions for a market that's about the population of the English East Midlands.

gctest50

ukoda wrote: »

I just said "in theory" I'm not saying eircode will or should do this, I'm just simply pointing it out as a feature of this type of code that isn't available in other codes.

And as you know, eircode could offer non geographic options too, like routing key "ESB"

Isn't "a geo-location code could offer non geographical codes" a bit of a contradiction

Wouldn't that be so useful for the Travelling Community ?

If they are leaving for the UK / Europe, it could be forwarded to Pavee Point or something, bit like Parcel Motel

clewbays wrote: »

Interesting analogy plodder. If I understand you correctly you are saying that traditional postcodes and landline numbers are both hierarchical but that mobile numbers and eircodes are not! Your argument reminds me of the relatively recent restructuring of landline phone numbers for some area codes to accommodate expansion - an extra digit was added. That is the danger of a hierarchical eircode - you either make it longer to accommodate An Post sorting offices, small areas, and uniqueness or you risk future revisions to address new builds.

I can't envisage any situation where a code would need to be made longer. You need to create new small areas from time to time as new buildings are built. So, you allocate new small area codes to accommodate this. That wouldn't normally affect existing postcodes at all.

marmurr1916

clewbays wrote: »

Interesting analogy plodder. If I understand you correctly you are saying that traditional postcodes and landline numbers are both hierarchical but that mobile numbers and eircodes are not!

Landlines and hierarchical postcodes are so 20th century...

clewbays wrote: »

That is the danger of a hierarchical eircode - you either make it longer to accommodate An Post sorting offices, small areas, and uniqueness or you risk future revisions to address new builds.

Indeed. Imagine a rural townland with 20 houses. Using a UK style system, you could use one postcode for the entire townland; or a variation of a UK style system with 20 individual postcodes arranged hierarchically. Then someone comes along a builds a new housing estate containing 100 houses in the townland...

OT:

http://www.irishexaminer.com/ireland/handy-new-an-post-delivery-system-ticks-all-the-boxes-374563.html

clewbays

plodder wrote: »

True, but mobile phones numbers are attached to people who aren't tied to a particular location. I'm not sure how you would code them hierarchically or what benefit they would be. Postcodes are tied to fixed locations though.

Interesting analogy plodder. If I understand you correctly you are saying that traditional postcodes and landline numbers are both hierarchical but that mobile numbers and eircodes are not! Your argument reminds me of the relatively recent restructuring of landline phone numbers for some area codes to accommodate expansion - an extra digit was added. That is the danger of a hierarchical eircode - you either make it longer to accommodate An Post sorting offices, small areas, and uniqueness or you risk future revisions to address new builds.

Sam Russell wrote: »

But Eircodes do not allow a translation from one address to a different one this you have stated many times - (I am not bothered to track down examples before you ask). You use an example of moving from one side of a site to another but what if you move from one side of town to the other? For example, if revenue moved income tax from Limerick to Sligo, then they would need a new Eircode. However, a geo-location code could allow non-geographic codes which do not have locations. [I know Eircode could too but they said they will not] Government, Eircom (eir) and ESB, etc would love these for handling regular post.

I just said "in theory" I'm not saying eircode will or should do this, I'm just simply pointing it out as a feature of this type of code that isn't available in other codes.

And as you know, eircode could offer non geographic options too, like routing key "ESB"

Isn't "a geo-location code could offer non geographical codes" a bit of a contradiction

ukoda wrote: »

Actually one of the benefits they claim in the eircode design is that the geo coordinates can change but the post code remains the same, so essentially you could "move" a postcodes location a few feet up the road or across the road or where ever and the postcode itself wouldn't have to change. Something that can't be done with a pure geo code as a geo code is completely tied to one exact location.

Could be a useful feature in say a campus scenario where they move a department across the campus, it could in theory keep its postcode but the location would simply get updated on the database to new coordinates.

Similarly if a company had a huge site and they moved their main entrance to say another road, the coordinates could be changed but again they wouldn't have the hassle of getting a new postcode like they would if we used Geo based ones.

But Eircodes do not allow a translation from one address to a different one this you have stated many times - (I am not bothered to track down examples before you ask). You use an example of moving from one side of a site to another but what if you move from one side of town to the other? For example, if revenue moved income tax from Limerick to Sligo, then they would need a new Eircode. However, a geo-location code could allow non-geographic codes which do not have locations. [I know Eircode could too but they said they will not] Government, Eircom (eir) and ESB, etc would love these for handling regular post.

Sam Russell wrote: »

I asked them the other day. They pointed me to a website that gave that response - not very active stance by them.

Ah ok I see. I think until they are actually ready to launch the standard response will always be "we are working on it" companies don't tend to say anymore until they have a launch date

ukoda wrote: »

You asked them in Aug? It would be interesting to go back to them now and ask for an update? If you're so inclined, I'd be interested to see what they say

I asked them the other day. They pointed me to a website that gave that response - not very active stance by them.

plodder wrote: »

True, but mobile phones numbers are attached to people who aren't tied to a particular location. I'm not sure how you would code them hierarchically or what benefit they would be. Postcodes are tied to fixed locations though.

Actually one of the benefits they claim in the eircode design is that the geo coordinates can change but the post code remains the same, so essentially you could "move" a postcodes location a few feet up the road or across the road or where ever and the postcode itself wouldn't have to change. Something that can't be done with a pure geo code as a geo code is completely tied to one exact location.

Could be a useful feature in say a campus scenario where they move a department across the campus, it could in theory keep its postcode but the location would simply get updated on the database to new coordinates.

Similarly if a company had a huge site and they moved their main entrance to say another road, the coordinates could be changed but again they wouldn't have the hassle of getting a new postcode like they would if we used Geo based ones.

gctest50 wrote: »

There isn't that much order to mobile phone numbers and we all get along fine with them, you could have 0871315466 and your wife could have 0838274222

& everyone that needs to know your numbers has them

True, but mobile phones numbers are attached to people who aren't tied to a particular location. I'm not sure how you would code them hierarchically or what benefit they would be. Postcodes are tied to fixed locations though.

gctest50 wrote: »

It'll be great for areas where you have 2 or more with the same name

eg John Russell A23 XH56 and John Russell A23 DF45 and John Russell A23 U76

kind of thing would have prevented this :

so .......... GoSafe need to start using it asap

.

To actually solve this problem, they'll have to make Eircodes mandatory when applying for driving licenses, vehicle tax and registration etc. I wouldn't have a problem with that fwiw.

They should also make it mandatory for entries on the property price register. if it's not mandatory for applications like these, then Eircodes won't be of much use for them.

gctest50

plodder wrote: »

Oh, like Eircode?

Barcodes themselves are designed so they can be read easily by laser scanners. That's a different issue really. But, the product/article numbers encoded in barcodes are hierarchical though. It would be fairly chaotic if they were random.

So, I don't think hierarchical data is going anywhere really.

There isn't that much order to mobile phone numbers and we all get along fine with them, you could have 0871315466 and your wife could have 0838274222

& everyone that needs to know your numbers has them