nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Child Item

Eircode design

PDVerse

Hello folks. If anyone has a question about any aspect of Eircode design I'm happy to answer.

Find more posts tagged with

Quick Links

Comments

plodder

oscarBravo wrote: »

Intuitive? You think postcodes should be intuitive?

Let's see: I have an acquaintance with the postcode "8700". What does that tell you, intuitively? How about "SY16 1AD"? Or "80112"? "3343810"?

SY = Shrewsbury area (quite intuitive)

SY16 = "SY16 area covering the areas of Newtown, Powys.The area contains approximately 6,935 households with a population of about 16,021 (2011 census)"

SY16 1AD (see it here)

It's as intuitive as it can be. You could take a list of all the top level postcode areas in the UK, and hazard a guess at where many of them are, or you could learn them off easily (if you wanted).

But, more important, all of the above is free information. It can be presented/used any way you like. You are not restricted by the particular capabilities of an app (that could be withdrawn at any time). You only pay for information down at the household level in the UK.

OldMrBrennan83

This post has been deleted.

TheChizler

And that point brings us back to people's problems with the business model (which IMO currently includes an adequate number of free lookups) and not a problem with the design.

plodder

Patww79 wrote: »

This post has been deleted.

Hardly. It couldn't be Belfast (BT), Dundee (DD), or Oxford (OX) could it?
or Liverpool (L) or Manchester (M).

Do you think you could guess where the following Eircode routing keys are:

A41, T21, H91, F31, E21, P25?

ukoda

plodder wrote: »

Hardly. It couldn't be Belfast (BT), Dundee (DD), or Oxford (OX) could it?
or Liverpool (L) or Manchester (M).

Do you think you could guess where the following Eircode routing keys are:

A41, T21, H91, F31, E21, P25?

Why would anyone need to guess the first part of any postcode in either ireland or the UK.. Another example of a fictitious "problem"

plodder

TheChizler wrote: »

And that point brings us back to people's problems with the business model (which IMO currently includes an adequate number of free lookups) and not a problem with the design.

It is a problem with the design because:

a) the number of free lookups is only for use by people, not software and its not for commercial use.

b) the app which has been provided to cater for some use cases (which it does admirably) doesn't cater for all use cases.

The web site I linked to above wouldn't be possible with Eircode unless it was hosted by someone who already has an Eircode subscription.

TheChizler

plodder wrote: »

TheChizler wrote: »

And that point brings us back to people's problems with the business model (which IMO currently includes an adequate number of free lookups) and not a problem with the design.

It is a problem with the design because:

a) the number of free lookups is only for use by people, not software and its not for commercial use.

b) the app which has been provided to cater for some use cases (which it does admirably) doesn't cater for all use cases.

The web site I linked to above wouldn't be possible with Eircode unless it was hosted by someone who already has an Eircode subscription.

What have they got to do with the design of the code? Any provider of any code is either going to have to release the algorithm or database for looking up the location. How they do that is a business decision.

ukoda

plodder wrote: »

It is a problem with the design because:

a) the number of free lookups is only for use by people, not software and its not for commercial use.

b) the app which has been provided to cater for some use cases (which it does admirably) doesn't cater for all use cases.

The web site I linked to above wouldn't be possible with Eircode unless it was hosted by someone who already has an Eircode subscription.

It's not at all a design problem. You know perfectly well that eircode data can be released for free if they so choose. There was a decision made to monetise it. That decision could be reversed and the design would not have to change. They would simply publish the ECAD somewhere and presto it's now a 'free' postcode for all purposes. No design changes needed.

plodder

ukoda wrote: »

It's not at all a design problem. You know perfectly well that eircode data can be released for free if they so choose. There was a decision made to monetise it. That decision could be reversed and the design would not have to change. They would simply publish the ECAD somewhere and presto it's now a 'free' postcode for all purposes. No design changes needed.

Releasing the data (or some of it) for free would certainly mitigate many of the problems, but as you well know also, the decision to use a random code and no hierarchy was partly to prevent people from making use of the system without having the database. PDVerse admitted this. So, you can call it a design feature. I call it a design problem.

ukoda

plodder wrote: »

Releasing the data (or some of it) for free would certainly mitigate many of the problems, but as you well know also, the decision to use a random code and no hierarchy was partly to prevent people from making use of the system without having the database. PDVerse admitted this. So, you can call it a design feature. I call it a design problem.

Call it what you want, but it is wrong to claim or insinuate that eircode cannot be 'free' because of the way it's designed. That is simply not true. The way it's designed allows for it to either be free or monetised. It was decided to monetise it. And it can be decided to release it for free and there is nothing in the design to stop the ECAD being published on some government website.

plodder

ukoda wrote: »

Call it what you want, but it is wrong to claim or insinuate that eircode cannot be 'free' because of the way it's designed. That is simply not true. The way it's designed allows for it to either be free or monetised. It was decided to monetise it. And it can be decided to release it for free and there is nothing in the design to stop the ECAD being published on some government website.

I have never insinuated or claimed it. I've spent the last couple of months calling for some data to be released for free...

ukoda

plodder wrote: »

I have never insinuated or claimed it. I've spent the last couple of months calling for some data to be released for free...

Your post above clearly says it's a design problem that software can't look up eircode for free. It's not. It's a political decision that stops 'free' use of eircode. The ECAD will work perfectly well with software and doesn't give a crap if it's paid for or not.

plodder

ukoda wrote: »

Your post above clearly says it's a design problem that software can't look up eircode for free. It's not. It's a political decision that stops 'free' use of eircode.

Let me explain it this way. Eircode was designed in such a way that all practical usage of it revolves around a database and there is very little useful information to be derived from collections of Eircodes without having access to the database.

That is not the case with the UK postcode. So, it is not simply a political question as to whether the database should or should not be monetised.

It is tied directly to the design of the code.

Therefore, that is a design problem in my opinion.

The ECAD will work perfectly well with software and doesn't give a crap if it's paid for or not.

Of course, but the fact that there is an ECAD was a design decision. You don't need access to anything like the ECAD to make use of the UK postcode for sorting/grouping locations. That is a design issue.

oscarBravo

D.L.R. wrote: »

Not sure why you're so angry that I don't think eircode is a well designed system. But whatever, its my opinion. Design by committee, too many cooks, take your pick.

I'm not angry; I just don't understand your argument. You haven't bothered to defend it, so it's clear that you don't either.

Your criticism of Eircode is, sadly, completely typical of the initial Irish reaction to almost anything: "it's crap". Not "it's pretty good, with some caveats", or "hmm, I wouldn't have done it that way, but I can see the arguments for why it would be". Just "it's crap", with no effort to analyse or critique in any meaningful way.

I'm still at a loss as to what "intuitive" even means, where a postcode is concerned. plodder would have us believe that the UK postcode is "intuitive", because an "SY" postcode obviously refers to the Shrewsbury area. Quite why it doesn't intuitively refer to Salisbury, Sandy, Selby, Shaftesbury, Spilsby, Stanley, Stavely, Sudbury or Swanley is unclear, but anyhoo. He then goes on to "intuit" population statistics from the same code, while studiously ignoring the rest of the codes I provided.

Given that plodder has now admitted that he wasn't, in fact, addressing my question about intuitiveness but instead continuing to grind an axe about implementation in a design thread, I'd still like to know what, exactly, "intuitive" or "accessible" even means in the context of a postcode?

ukoda

plodder wrote: »

Let me explain it this way. Eircode was designed in such a way that all practical usage of it revolves around a database and there is very little useful information to be derived from collections of Eircodes without having access to the database.

That is not the case with the UK postcode. So, it is not simply a political question as to whether the database should or should not be monetised.

It is tied directly to the design of the code.

Therefore, that is a design problem in my opinion.

Of course, but the fact that there is an ECAD was a design decision. You don't need access to anything like the ECAD to make use of the UK postcode for sorting/grouping locations. That is a design issue.

I'm not disputing any of the facts. I'm simply saying it's wrong to say the design of eircode prohibits it use for free. It does not.

It was designed to have a database yes. That was a good decision and I don't believe it had anything to do with monetising it. It was the right thing to do to use the Geodirectory as the database.

It doesn't matter if it has a database, that has no bearing at all on wether or not you can or can't charge for it. Take loc8code, a non database code yet you are required to licence it and pay a fee if you want to use it for more than 15 look ups a day (same as eircode) this code is also useless for delivery planning as it forces everything into a square ignoring roads and natural boundaries

plodder

Well you do come across as a bit angry. But, that's by the way.

dictionary.com wrote:

Intuitive: easy to understand or operate without explicit instruction

In this context, I'd say easy to recognise without being told in advance.

Some UK postcode areas:
Belfast (BT), Dundee (DD), or Oxford (OX)
or Liverpool (L) or Manchester (M).

Eircode routing keys:
A41 (Ballyboughal Dublin), P17 (Kinsale), H91 (Galway) , R42 (Birr), F42 (Roscommon), Y21 (Enniscorthy)

Do you not think the UK ones are more intuitive? Granted to be fair the D codes are, in the same vein as the UK ones (as in D stands for Dublin, rather than we are already familiar with them), but it's the rest of the system ...

ukoda

plodder wrote: »

Well you do come across as a bit angry. But, that's by the way.

In this context, I'd say easy to recognise without being told in advance.

Do you not think the UK ones are more intuitive? Granted to be fair the D codes are, in the same vein as the UK ones (as in D stands for Dublin, rather than we are already familiar with them), but it's the rest of the system ...

It does not meet the criteria of the definition of intuitive as it would have to be deciferable without instruction and there's no way you can tell what DD stands for unless someone tells you or you look it up. You can guess. But that's different.

plodder

ukoda wrote: »

It does not meet the criteria of the definition of intuitive as it would have to be deciferable without instruction and there's no way you can tell what DD stands for unless someone tells you or you look it up. You can guess. But that's different.

Okay. Let's be even more specific. If we define intuitive in this context as meaning that if you understand that UK postcode areas (a bit like Irish car reg plates) are formed from the initial or final letters of syllables or words in the names of the relevant place

could you then infer or in some cases guess that:

L stands for Liverpool
M for Manchester
OX for Oxford.

I think you could

If you didn't know that SY was actually Shrewsbury, that's fine. We are talking in general terms and having learned this, we can see the pattern and we are comparing with a system that is considerably less intuitive, because it is largely random.

Why is Kinsale P17 and the surrounding areas P72, T12, and P43? If there is a pattern there, I don't see it.

ukoda

plodder wrote: »

Okay. Let's be even more specific. If we define intuitive in this context as meaning that if you understand that UK postcode areas (a bit like Irish car reg plates) are formed from the initial or final letters of syllables or words in the names of the relevant place

could you then infer or in some cases guess that:

L stands for Liverpool
M for Manchester
OX for Oxford.

I think you could

If you didn't know that SY was actually Shrewsbury, that's fine. We are talking in general terms and having learned this, we can see the pattern and we are comparing with a system that is considerably less intuitive, because it is largely random.

Why is Kinsale P17 and the surrounding areas P72, T12, and P43? If there is a pattern there, I don't see it.

I already told you I can try guess. What benefit is that to me? I'll have to look it up to sure. What is the benefit? What scenario is this helpful in? Another fictional scenario that has no real life application. Clutching at straws.

And by the way, if I was unfamiliar with the UK postcode and asked to guess where L was, I'd guess London.

oscarBravo

plodder wrote: »

Well you do come across as a bit angry. But, that's by the way.

I'm immensely frustrated at the Irish habit of criticising things without bothering to actually think about them first, but that's not the same as being angry.

Do you not think the UK ones are more intuitive?

OK, scouts' honour: don't look these up.

Intuitively, what areas do these British postcodes refer to -

ME? SD? SM? TD? ZE?

Now, leaving that aside for a second, I can concede that the UK postcode system is mostly more intuitive than the Irish one. Yes, you can often guess roughly where a postcode might be. The idea that the Irish postcode system is - and I quote - "shockingly bad" for no other reason than that you can't match the first few characters to a location using guesswork is, with the greatest of respect, a very stupid criticism.

I'm still waiting for someone to tell me what's intuitive about "8700", "80112" or "3343810". Or is every postcode system outside of the UK also "shockingly bad"?

clewbays

oscarBravo wrote: »

How? And, from their perspective, why?

Remember: An Post didn't want a postcode.

Well, that's the simple truth of it, we have adopted a routing key system that only has meaning for An Post who we are saying has absolutely no need of it and that Eircodes would weaken An Post's competitiveness if they encouraged people to use them.

They must be really pissed off that people use Dublin 1, Dublin 2 etc, when looking for ManyPlaces street would have made it much harder for couriers.

plodder

oscarBravo wrote: »

OK, scouts' honour: don't look these up.

Intuitively, what areas do these British postcodes refer to -

ME? SD? SM? TD? ZE?

I'd heard of one them before ZE = Zetland. An old name for the Shetland islands. Of course, none of them are intuitive. And I never said that all of them are anyway.

Now, leaving that aside for a second, I can concede that the UK postcode system is mostly more intuitive than the Irish one. Yes, you can often guess roughly where a postcode might be. The idea that the Irish postcode system is - and I quote - "shockingly bad" for no other reason than that you can't match the first few characters to a location using guesswork is, with the greatest of respect, a very stupid criticism.

I'm still waiting for someone to tell me what's intuitive about "8700", "80112" or "3343810". Or is every postcode system outside of the UK also "shockingly bad"?

Fair enough, but I'm somewhat familiar with the UK system. I've no idea where those other ones are from.

ukoda

plodder wrote: »

I'd heard of one them before ZE = Zetland. An old name for the Shetland islands. Of course, none of them are intuitive. And I never said that all of them are anyway.

Fair enough, but I'm somewhat familiar with the UK system. I've no idea where those other ones are from.

You've still failed to answer how knowing ZE is Zetland (now Lerwick) is any benefit to you?

If you're going there you will look up the actual postcode of where you need to go, it will take you to probably a street with 20 houses. In Ireland, you do the same thing, but it takes you to the exact building you want to go to

sondagefaux

plodder wrote: »

oscarBravo wrote: »

Intuitive? You think postcodes should be intuitive?

Let's see: I have an acquaintance with the postcode "8700". What does that tell you, intuitively? How about "SY16 1AD"? Or "80112"? "3343810"?

SY = Shrewsbury area (quite intuitive)

SY16 = "SY16 area covering the areas of Newtown, Powys.The area contains approximately 6,935 households with a population of about 16,021 (2011 census)"

SY16 1AD (see it here)

It's as intuitive as it can be. You could take a list of all the top level postcode areas in the UK, and hazard a guess at where many of them are, or you could learn them off easily (if you wanted).

But, more important, all of the above is free information. It can be presented/used any way you like. You are not restricted by the particular capabilities of an app (that could be withdrawn at any time). You only pay for information down at the household level in the UK.

As your example shows, the SY postcode covers addresses in Wales (Powys). Shrewsbury is in England. I'm not convinced that a postcode derived from the name of a town in England is 'intuitive' when it's used for addresses in Wales...

Hazard a guess where many of them are?

NW = certain parts of north-west London.

NE = certain parts of north-east London? No - Newcastle-upon-Tyne

WC = Winchester? No - West Central London

BT = Belfast, but also every address in Northern Ireland...

Can't really see why an address in Co. Armagh with a BT postcode is 'intuitive'...

Sam Russell

Mod: Can we leave the UK post code system now - it's been done to death.

beauf

Its all irrelevant now anyway. It done and dusted. They just make it mandatory in a range of public services. Then its a success regardless.

plodder

sondagefaux wrote: »

As your example shows, the SY postcode covers addresses in Wales (Powys). Shrewsbury is in England. I'm not convinced that a postcode derived from the name of a town in England is 'intuitive' when it's used for addresses in Wales...

Hazard a guess where many of them are?

NW = certain parts of north-west London.

NE = certain parts of north-east London? No - Newcastle-upon-Tyne

WC = Winchester? No - West Central London

BT = Belfast, but also every address in Northern Ireland...

Can't really see why an address in Co. Armagh with a BT postcode is 'intuitive'...

Once you understand their system (like ours) is based around 'post towns' rather than administrative areas, I think it makes sense. I don't recall you complaining about that aspect of Eircode.

And of course there are always anomalies - London's areas being the biggest. But, it doesn't change the overall point, especially since Dublin's preexisting codes didn't have the same problem.

It's a bit like looking at the car reg plate system and saying they aren't intuitive at all because one or two of them might be ambiguous. (WD could have been Wexford or Waterford for example).

Alan_P

plodder wrote: »

Once you understand their system (like ours) is based around 'post towns' rather than administrative areas, I think it makes sense. I don't recall you complaining about that aspect of Eircode.

And of course there are always anomalies - London's areas being the biggest. But, it doesn't change the overall point, especially since Dublin's preexisting codes didn't have the same problem.

It's a bit like looking at the car reg plate system and saying they aren't intuitive at all because one or two of them might be ambiguous. (WD could have been Wexford or Waterford for example).

I don't really understand your points, they seem to be going around in circles and not really based on anything about Eircode.

Can I clarify, if the Eircode database was available to freely download off a government web site, would all your objections go away ?

If that's true, then all your complaints are about the commercial model.

beauf

You don't understand the usefulness of having a number plate with abbreviations for county? Instead of codes?

Would they be more useful as random number that you need to use a database to read?

It's Human engineering 101.

sondagefaux

plodder wrote: »

sondagefaux wrote: »

As your example shows, the SY postcode covers addresses in Wales (Powys). Shrewsbury is in England. I'm not convinced that a postcode derived from the name of a town in England is 'intuitive' when it's used for addresses in Wales...

Hazard a guess where many of them are?

NW = certain parts of north-west London.

NE = certain parts of north-east London? No - Newcastle-upon-Tyne

WC = Winchester? No - West Central London

BT = Belfast, but also every address in Northern Ireland...

Can't really see why an address in Co. Armagh with a BT postcode is 'intuitive'...

Once you understand their system (like ours) is based around 'post towns' rather than administrative areas, I think it makes sense. I don't recall you complaining about that aspect of Eircode.

And of course there are always anomalies - London's areas being the biggest. But, it doesn't change the overall point, especially since Dublin's preexisting codes didn't have the same problem.

It's a bit like looking at the car reg plate system and saying they aren't intuitive at all because one or two of them might be ambiguous. (WD could have been Wexford or Waterford for example).

It does change the overall point - relying on pre-existing knowledge to understand a system ("Once you understand their system (like ours) is based around 'post towns' rather than administrative areas, I think it makes sense.") is the very opposite of 'intuitive' knowledge.

Many, many UK postcodes are no more 'intuitive' than Eircodes. ZE for Shetland is a very good example. BT for Co. Fermanagh is another very good example. NE for Newcastle rather than North-East London (when NW already means North-West London) is another very good example.