NRE & WebTIS are failing to validate some of the 'shortest' routes

[319321]

09/09 1330 TheTrainLine.com is now validating all the test cases. WebTIS and NRE are not.
07/09 1020 All journey planners are now not validating some routes that should be valid
06/09 1001 All Journey Planners are now validating all the test cases specified. It's been a strange week at JourneyPlanner central.
04/09 1434 The Prittlewell > Southend East via Upminster is not validating using NRE, but is validating with WebTIS and thetrainline.com
04/09 1339 All journey planners are now validating the shortest route for each of the test cases defined in post #12
04/09 1133 All journey planners are now failing to validate the shortest routes.

Original Post Below ---------------------
It would appear that the National Rail Enquires Journey Planner is implementing a fares-check rule for shortest-distance journeys making many previously available routes invalid.

It also means that the only route that validates involves a fixed interchange. Since the 'no longer than by three miles' sentence was deleted from the Routing Guide in November last year, this has meant that many 'shortest' journeys which involved a fixed interchange failed to validate under that rule, but the journey that was entirely by rail did validate. Now it would seem that since the routing guide was updated this month, the opposite is now true.

Is this yet another change that, rather than defining the term "shortest", simply changes the terms meaning to whatever the rail industry thinks it should be day-by-day?
--- old post above --- --- new post below ---
I am assuming that it applies to 'direct trains' - I am unable to think of an example to test it against though. I would be grateful if someone who knows of a route that only validates under the 'direct trains' rule can find out if it still validates.

 

[yorkie]

It would appear that the National Rail Enquires Journey Planner is implementing a fares-check rule for shortest-distance journeys making many previously available routes invalid.

Thanks for letting us know.

I know you didn't mean this but I thought I'd best point out, to avoid confusion, it doesn't actually make them invalid. It's just that National Rail Enquiries (NRE) is falsely reporting the routes as invalid.

People can still use tickets on the shortest route in accordance with condition NRCoC 13 (a)(ii) and neither NRE nor the Routeing Guide cannot invalidate those routes.

NRE needs to ensure that routes valid under NRCoC 13(a)(ii) are always offered as permitted routes without any reference to the Routeing Guide whatsoever. Booking engines should do likewise.

It also means that the only route that validates involves a fixed interchange. Since the 'no longer than by three miles' sentence was deleted from the Routing Guide in November last year, this has meant that many 'shortest' journeys which involved a fixed interchange failed to validate under that rule, but the journey that was entirely by rail did validate. Now it would seem that since the routing guide was updated this month, the opposite is now true.

The '3 mile rule' is still mentioned but it only gets a brief mention and it's relevance (other than the limited circumstances described on page F1 of THE NATIONALROUTEING GUIDE IN DETAIL) isn't made clear.

Perhaps someone can ask the Department for Transport (DfT) - who are subject to FOI - if the apparent withdrawal of the '3 mile rule' was a 'material change'.

If it was a materiel change, they need to have approved it and I believe they should supply us with their reasoning.

If it wasn't a material change, then it remains in place in my opinion, and the DfT should then insist that ATOC re-instates it as well as taking appropriate action.

The shortest route is covered in NRCoC 13(a)(ii) and the Routeing Guide is separately listed under 13(a)(iii), so the 3 mile rule needs to be integrated into 13(a)(ii), to avoid confusion, in my opinion. Perhaps we can ask the DfT to do this as a matter of urgency to avoid this sort of problem reoccurring in future.

Also if there is no "3 mile rule" then how on earth do they deal with the situation whereby one route is shorter than the other by the National Rail Timetable (NRT) while another route is shorter by the electronic data used by booking engines?! Under the "3 mile rule" both would remain valid.

The loss of that rule also causes problems when you have avoiding lines and can invalidate trains which call at stations just off the main line because it's a few chains longer (e.g. Lewisham). This clearly isn't what they intend, so we need to get the 3 mile rule enshrined into the NRCoC to avoid such 'errors' in my opinion.

 

[319321]

Although not worded as you wanted, I made a request for information about this change from the DfT and they denied having information about it.
https://www.whatdotheyknow.com/request/removal_of_the_provision_that_ro#incoming-674425

 

[yorkie]

In that case I believe that the original wording still applies, so the 3 mile rule is still in effect, in my opinion.

I believe ATOC must, therefore, have mistakenly removed it and therefore I trust they will re-instate it very soon after reading this thread (hopefully tomorrow )

 

[319321]

In that case I believe that the original wording still applies, so the 3 mile rule is still in effect, in my opinion.

I believe ATOC must, therefore, have mistakenly removed it and therefore I trust they will re-instate it very soon after reading this thread (hopefully tomorrow )

The three mile rule disappeared in November last year though, why would they suddenly reinstate it?

I asked this question regarding where the evidence for the three mile rule was since it disappeared from section F earlier this year, but didn't get an answer.

The reason I am interested in such matters is I am in dispute over a ticket I bought in connection with a journey plan. I will post the whole saga here and it will give some of you a right good laugh, but only when the matter has come to a conclusion.
--- old post above --- --- new post below ---
Hang on, I may be being mad!
--- old post above --- --- new post below ---
OK, Page F1 Section B states:

If a journey does not have an origin routeing point and a destination
routeing point, its permitted route is (1) the shortest route served by a regular
passenger service or (2) a route no more then 3 miles longer or (3) any other route
permitted by an easement

Prior to November 2014, this was supplemented with a rule that even if a journey has an origin routing point and a destination routing point, routes up to three miles longer than the shortest route would be valid:

Journeys on direct trains or taking the route of shortest distance or a distance longer
by no more than 3 miles are always following a permitted route. Journeys following a
route given by the Routeing Guide are following a permitted route.

This provision was removed in November 2014.

Journeys between the stations shown on the ticket by direct trains; or by the
shortest route which can be used by scheduled passenger services (including
change of trains); are always following a permitted route. Journeys following a
route given by the Routeing Guide are following a permitted route

Are you saying that this shouldn't have been removed, and if it is a material change the DfT should have been aware of it?

 

[maniacmartin]

I'm not convinced that requesting an internal review of the handling of the FoI request was the right thing to do. Assuming that their response is correct and they were not consulted by ATOC, then as far as freedom of information is concerned, the DfT appear to have obeyed all the rules in replying with the information they hold as requested.

Your request for them to force ATOC to go through the proper process falls outside of the FoI provisions and I believe should be made through separate correspondence with the DfT (and not to the FoI department). I wouldn't hold your breath tho.

 

[yorkie]

The three mile rule disappeared in November last year though, why would they suddenly reinstate it?

It's not disappeared completely; it's in the Routeing Guide in Detail.

According to the result of your FOI request, a material change has not been requested. So it's either an illegal attempt at a change or, perhaps more probably I'd like to think, a mere oversight and mistake by ATOC/RSP.

I asked this question regarding where the evidence for the three mile rule was since it disappeared from section F earlier this year, but didn't get an answer.

The reason I am interested in such matters is I am in dispute over a ticket I bought in connection with a journey plan. I will post the whole saga here and it will give some of you a right good laugh, but only when the matter has come to a conclusion.

Feel free to send me a PM if I can be of assistance.

Also don't forget to refer the matter to Passenger Focus.

Do you think a breach of consumer law (see Code of practice on retail information) has taken place? If so, additionally you could require the ORR to investigate such a potential breach.

OK, Page F1 Section B states:

Prior to November 2014, this was supplemented with a rule that even if a journey has an origin routing point and a destination routing point, routes up to three miles longer than the shortest route would be valid:


This provision was removed in November 2014.


Are you saying that this shouldn't have been removed, and if it is a material change the DfT should have been aware of it?

If it was a deliberate act to abolish the rule, then they would require DfT approval.

If it was a mistake, then I would expect both the DfT to be following this up as well as ATOC/RSP hastily reinstating it.

However I believe that, to avoid confusion, the '3 mile rule' should be integrated into NRCoC Condition 13(a)(ii) rather than stated in the Routeing Guide. This is because routes permitted in the Routeing Guide are completely separate, and covered under 13(a)(iii). It would also ensure this sort of nonsense does not occur again.

 

[thedbdiboy]

The three mile rule was never part of the original routeing guide permission, it was added later to provide some flexibility with regard to mileages on loops etc. However it is (as we have seen) a very imperfect device.

However, it does mean that this talk of the DfT having to expressly approve it's removal all alteration is moot - as it was never in fact a regulatory obligation on day 1. In the present climate there would undoubtedly need to be some discussion but that isn't quite the same thing.

 

[button_boxer]

It's not disappeared completely; it's in the Routeing Guide in Detail.

But now only for the case where the origin and destination share a routeing point.

 

[Andrew1395]

I think the 3 mile allowance for local journey calculations was the original reason it was introduced.

 

[Paul Kelly]

It would appear that the National Rail Enquires Journey Planner is implementing a fares-check rule for shortest-distance journeys making many previously available routes invalid.

Are you able to share some examples you've been looking at? There's definitely something strange going on here. I tried a journey that has been mentioned in another thread recently, Blackwater to Wembley Stadium. London Group fails the fares check as a destination routeing point, but the journey is still permitted via the tedious route of Guildford, Effingham Junction, Surbiton, Clapham Junction, London Victoria and Marylebone as that is the shortest route.

Continuing from Guildford via Woking and Waterloo (the obvious route) is less than 3 miles longer than via Effingham Junction and Victoria, but NRE isn't showing any itineraries this way. It forces you onto the stopping train from Guildford to Clapham Junction via Effingham Junction and then with the added connection time at CLJ and further detour via Victoria you miss two trains from Marylebone to Wembley that you could have got, meaning you arrive ONE HOUR later than if you'd taken the direct train from Guildford to Waterloo. That really is ridiculous.

 

[319321]

I have re-checked the journeys I used to test the system, and they still force the passenger to take a fixed interchange (all the fares checked have a route ANY PERMITTED). Some tests are
- Bearsted to New Hythe via Sevenoaks
- Bekesbourne to Sturry via Dover Priory
- Holmwood to Betchworth via Horsham
- Prittlewell to Southend East via Upminster

I am assuming its because of the introduction of the new NFM64 fares check rule, but I could be wrong .....
--- old post above --- --- new post below ---
ATOS / WEBTIS NOW ALSO IMPLEMENT THE FARE CHECK RULE

When I checked Sunday the routes I specifed above were still validating via WebTIS (I used the c2c site). However, the routes are no longer validating using this planner either.

They do still validate using thetrainline.com though.

 

[kieron]

It would appear that the National Rail Enquires Journey Planner is implementing a fares-check rule for shortest-distance journeys making many previously available routes invalid.

I don't think this involves a fare check rule of any sort. Northampton-Lockerbie, for one, is still shown as being valid on the shortest route on NRE, even though it fails the fare check.

It does appear that someone is still working on this. When I had a look at this last night, Lower Sydenham-Anerley via New Beckenham* wasn't shown as being valid except with a Travelcard. When I looked again this afternoon, it was shown as also being valid with single and return tickets. This is a route which involves two stations with no common routeing point, but doesn't involve travel through any routeing points, so I was wondering how they were going to handle it if they did decide to carry out a fare check here.

* You may need to delete any nationalrail.co.uk cookies you have before following this link for it to load correctly.

 

[319321]

OK, a further update. I tried checking the Bearsted > New Hythe via Sevenoaks using my mobile phone on the mobile version of the NRE site and it shows up as a valid route. However the route still does not validate using main web site.

Even weirder, using the mobile site I tried searching for Southend East > Prittlewell via Romford. Rather than offer me the correct fare of £2.90 for an anytime ANY PERMITTED single, it quotes a fare of £25.30! When clicking the 'Buy Now' button, I get a 'Something went wrong' error though. Going from Prittlewell > Southend East via Romford, I get quoted a fare of £24.70!
--- old post above --- --- new post below ---

I don't think this involves a fare check rule of any sort. Northampton-Lockerbie, for one, is still shown as being valid on the shortest route on NRE, even though it fails the fare check.

It does appear that someone is still working on this. When I had a look at this last night, Lower Sydenham-Anerley via New Beckenham* wasn't shown as being valid except with a Travelcard. When I looked again this afternoon, it was shown as also being valid with single and return tickets. This is a route which involves two stations with no common routeing point, but doesn't involve travel through any routeing points, so I was wondering how they were going to handle it if they did decide to carry out a fare check here.

* You may need to delete any nationalrail.co.uk cookies you have before following this link for it to load correctly.

Have you got any other examples of esoteric routes that only validate using the shortest route rule? Or the direct trains rule?

I only said it was because of a fares check because thats the only recent major change that Rail Settlement Plan have made to the routing guide. I'm not sure what else would cause such a sudden change in route validity.

 

[Andrew1395]

The journey planners have always used NFM64 Fares in fares checking. The only change is that it is now explicit on the Routeing Guide web pages. See wording from the forums guide to Routeing:-

Although no mention is made in the Routeing Guide that a particular version of the fares database should be used, booking engines carry out this fares check based on the data in a version of the fares database released in September 1996 (NFM64), and therefore any route permitted at that time is considered to have 'protected route status', as confirmed by ATOC in correspondence regarding a journey from Southport to Manchester, for which Liverpool remains an appropriate RP for Southport due to the fares check being satisfied at that time.

 

[319321]

OK, so it might not be that the NRE Planner is applying a fares check rule to these routes. What is it doing then?

 

[crispy1978]

I made a journey on Monday from Scarborough to Accrington. I changed at Leeds and Todmorden as per NRE. This is a quicker journey than changing at York/Leeds and getting the Blackpool North train.

The conductor on the LDS-MCV train told me to change at Hebden Bridge, when I said that I'd been advised by NRE to change at TOD he asked my connection time then said that was fine.

Technically this is a marginally longer route than going direct through Hebden Bridge and Burnley Manchester Road.

Is this a valid route, or the sort of thing the 3 mile routeing guide is trying to cut out? Just don't want to get caught out by a technicality!

 

[319321]

I've checked the routes I listed in my previous post and they now all validate using the National Rail Enquiries planner.
TheTrainLine.com is also validating the routes.

However, WebTIS is still not validating the routes.
--- old post above --- --- new post below ---
Update @ 1127 - All the shortest routes have disappeared from NRE again. WebTIS is still not validating the shortest route.
--- old post above --- --- new post below ---
Update @ 1133 thetrainline.com is failing to validate the test cases.

 

[strowger]

Deleted

 

[319321]

At 1339 I tried a couple of the test cases in each of the journey planners, and the routes now seem to be validating. I will check again early this evening.
--- old post above --- --- new post below ---

Who is to say that a £2 single between two stations in London is not, in fact, valid via Wick and Thurso when a chancer asserts it to be? Printouts are easy to fake.

Interesting you say that....when testing some routes around the London area, I was being quoted anytime fares between stations that normally cost around £13 as being around £3.70. It wouldn't let me buy them though.

Something very strange has been going on.

 

[AngusH]

Printouts are easy to fake.

Absolutely. There's no way to tell once the page is on paper whether it was printed from a valid routing web page or not. (It would be a perfectly accurate copy of the page, but the page might not correct)

Relying on printed webpages isn't a verifiable approach.


One idea for better assurance is a scannable 2d QR code on the printout that is readable by a device held by the ticket inspector or anybody else, that includes all of the routing data in compressed form (embedded in the barcode) and is digitally signed by the railway/issuer. It should be possible to compress most journey routings to fit within 100-200 characters and still have some space for a digital signature.


You could then know that you had a valid route approved by a suitable authority and you could present it at any time for validity testing. Or test it yourself using your mobile phone.



I believe that would be difficult to fake if proper security is in place.

 

[Hadders]

How's that going to work in an area without a decent phone signal or train without reliable wi-fi.

 

[AngusH]

How's that going to work in an area without a decent phone signal or train without reliable wi-fi.

The trick is that you're not embedding an internet link, you're actually storing the whole routing data in compressed machine readable form in the actual barcode. It wouldn't need any external access at all.

I've got a working prototype of this type of thing, I'll see if I can post more details in a separate thread.

 

[kieron]

Have you got any other examples of esoteric routes that only validate using the shortest route rule? Or the direct trains rule?

Not really. That's just a route I noticed a while ago, and thought of when I saw the talk about fare check (as it happens, there are no fares for the route in the "NFM64 fares" data). I was surprised to see it recommend different tickets for exactly the same journey when I searched for it a few hours apart, hence why I mentioned it.

I only said it was because of a fares check because thats the only recent major change that Rail Settlement Plan have made to the routing guide. I'm not sure what else would cause such a sudden change in route validity.

Sorry; I just think of "fare check" as referring to the concept of that name in the routeing guide. There have been some recent changes with this, as you say, and these have had effects such as removing various valid routes to Lockerbie.

Most of the things mentioned in this thread don't relate to changes in the routeing guide fare check. In particular, the three examples in post 12 each have a common routeing point, so you wouldn't carry a fare check out when choosing a route.

At least it's been reverted (on NRE, anyway) now.

I made a journey on Monday from Scarborough to Accrington. I changed at Leeds and Todmorden as per NRE. This is a quicker journey than changing at York/Leeds and getting the Blackpool North train.

The conductor on the LDS-MCV train told me to change at Hebden Bridge, when I said that I'd been advised by NRE to change at TOD he asked my connection time then said that was fine.

Technically this is a marginally longer route than going direct through Hebden Bridge and Burnley Manchester Road.

Is this a valid route, or the sort of thing the 3 mile routeing guide is trying to cut out? Just don't want to get caught out by a technicality!

[t depends. If the Leeds-Todmorden train you caught went through Bradford, it would be valid based on map NP.

If it went through Mirfield, on the other hand, then it wouldn't be a mapped route as the only mapped routes via Mirfield are ones which go via Manchester and Blackburn.

The shortest route (on tables 39 and 41) is via Bradford and the area of Todmorden which doesn't go through the station, and is 112.75 miles long.

The national timetable still doesn't list a distance between Todmorden and Burnley, even though it refers to trains going that way. railmiles.me gives the distance as 8.9 miles. Adding this to the total for Scarborough-Mirfield-Todmorden and Burnley-Accrington from the timetable gives a total of 114.15 miles, 1.4 miles more than the shortest route.

I think it may be worth getting in touch with Northern about this route. I don't know if it was like this before the Todmorden Curve opened, but there are a few opportunities to catch a train via Mirfield and Burnley, and these aren't shown as being valid according to the routeing guide. And I suspect the reason they're not shown is more due to inertia than to any hoped for commercial benefit.

I think it would be best to come from someone familiar with the journey. I don't really know how much of a problem these restrictions are.

 

[34D]

It depends. If the Leeds-Todmorden train you caught went through Bradford, it would be valid based on map NP.

If it went through Mirfield, on the other hand, then it wouldn't be a mapped route as the only mapped routes via Mirfield are ones which go via Manchester and Blackburn.

The shortest route (on tables 39 and 41) is via Bradford and the area of Todmorden which doesn't go through the station, and is 112.75 miles long.

The national timetable still doesn't list a distance between Todmorden and Burnley, even though it refers to trains going that way. railmiles.me gives the distance as 8.9 miles. Adding this to the total for Scarborough-Mirfield-Todmorden and Burnley-Accrington from the timetable gives a total of 114.15 miles, 1.4 miles more than the shortest route.

I think it may be worth getting in touch with Northern about this route. I don't know if it was like this before the Todmorden Curve opened, but there are a few opportunities to catch a train via Mirfield and Burnley, and these aren't shown as being valid according to the routeing guide. And I suspect the reason they're not shown is more due to inertia than to any hoped for commercial benefit.

I think it would be best to come from someone familiar with the journey. I don't really know how much of a problem these restrictions are.

Might I suggest that someone asks about Brighouse to Burnley. This journey is quicker to change at Todmorden (5 min connection) than waiting 30 minutes at Hebden Bridge with a later arrival time.

May be better if the question is realistic than asking about enthusiast type things such as 4 different trains scarborough to blackburn (which I suggest normal people wouldn't do)

 

[319321]

Update 06/09 1020
For the test cases, I pretended I wanted to make a journey on a Mon 7 Sep @ 1015

- Bearsted to New Hythe via Sevenoaks
RJIS - Does not validate
ATOS - Does not validate
Trainline - DOES validate

- Bekesbourne to Sturry via Dover Priory
RJIS - Does not validate (but does from 2142)
ATOS - Does not validate at all
Trainline - Does not validate (but does from 2142)

- Holmwood to Betchworth via Horsham
RJIS - DOES validate
ATOS - Does not validate
Trainline - DOES validate

- Prittlewell to Southend East via Upminster
RJIS - Does not validate
ATOS - Does not validate
Trainline - Does not validate

 

[strowger]

Deleted

 

[kieron]

Might I suggest that someone asks about Brighouse to Burnley. This journey is quicker to change at Todmorden (5 min connection) than waiting 30 minutes at Hebden Bridge with a later arrival time.

You can, of course, suggest anything you like. It may help if you explain what you did, what happened, and what you expected to happen.

Brighouse and all of the Burnley stations are associated with the Halifax/Sowerby Bridge routeing point, so the situation is rather different to Scarborough-Accrington. Or Mirfield-Burnley, for that matter.

 

[319321]

Update 07/09 2351
Latest status on whether or not the shortest routes are validating

- Bearsted to New Hythe via Sevenoaks
RJIS - DOES validate
ATOS - Does not validate
Trainline - DOES validate

- Bekesbourne to Sturry via Dover Priory
RJIS - DOES validate
ATOS - Does not validate at all
Trainline - Does not validate (but does from 2142)

- Holmwood to Betchworth via Horsham
RJIS - DOES validate
ATOS - Does not validate
Trainline - DOES validate

- Prittlewell to Southend East via Upminster
RJIS - Does not validate
ATOS - DOES validate
Trainline - DOES validate

So it's getting better.
--- old post above --- --- new post below ---
Update 09/09 1330:

- Bearsted to New Hythe via Sevenoaks
RJIS - Does not validate
ATOS - Does not validate
Trainline - DOES validate

- Bekesbourne to Sturry via Dover Priory
RJIS - Does not validate
ATOS - Does not validate
Trainline - DOES validate

- Holmwood to Betchworth via Horsham
RJIS - DOES validate
ATOS - Does not validate
Trainline - DOES validate

- Prittlewell to Southend East via Upminster
RJIS - Does not validate
ATOS - DOES validate
Trainline - DOES validate

TheTrainLine.com now validates all the test cases. RJIS and ATOS are still not compliant, but the routes that validate and don't validate seem to be constantly changing.
--- old post above --- --- new post below ---
Update 10/09 1200:

Although thetrainline.com is the only planner validating the routes, the other two are at least now stable, each only returning the valid route in one out of the four test cases.

- Bearsted to New Hythe via Sevenoaks
RJIS - Does not validate
ATOS - Does not validate
Trainline - DOES validate

- Bekesbourne to Sturry via Dover Priory
RJIS - Does not validate until 2036
ATOS - Does not validate
Trainline - DOES validate

- Holmwood to Betchworth via Horsham
RJIS - DOES validate
ATOS - Does not validate
Trainline - DOES validate

- Prittlewell to Southend East via Upminster
RJIS - Does not validate
ATOS - DOES validate
Trainline - DOES validate

 

[yorkie]

Who is to say that a £2 single between two stations in London is not, in fact, valid via Wick and Thurso when a chancer asserts it to be? Printouts are easy to fake..

That would be a very serious fraud. When an East Coast Guard found half the occupants of a 1st class coach to be holding Baldock-North Sheen 1st class singles, with itineraries and reservations via Aberdeen & Inverness, the tickets were withdrawn because of suspected fraud but the tickets were still being offered so it was very easy for a BTP officer to ascertain that the tickets were correct and all was above board.

If anyone wishes to have evidence of an itinerary, my strong advice is to book online with the itinerary clearly shown and obtain reservations where available. It's not clear whether or not an NRE printout would suffice so I wouldn't rely on that.