With the development of GBR are there opportunities to simplify the Routing Guide?

[deltic]

I seem to recall a 1960s? BR equivalent of the Routing Guide which was a very short document. We now have a very complex version which has grown like topsy to protect individual TOC revenues. As we move back to a single rail provider is there a case for simplifying the routing guide and if we did what benefits would it provide to passengers?

 

[andythebrave]

It would take a while to weed out the mickey takers but "any reasonable route" is a lot shorter.

 

[Bletchleyite]

It would take a while to weed out the mickey takers but "any reasonable route" is a lot shorter.

Trouble is you can't implement that on a journey planner. You have to have actual rules in terms of what a journey planner will sell.

 

[jfollows]

It should be scrapped, allowing the shortest route, any route within 3 miles and “any reasonable” route.
I’m saying that as one of the small number of passengers who understands most of the routeing guide as it exists today.
But it’s embedded in journey planners, as @Bletchleyite says, so it would be very hard to unravel.

 

[andythebrave]

Trouble is you can't implement that on a journey planner. You have to have actual rules in terms of what a journey planner will sell.

Add a sentence when using a planner "as the ticket you have selected permits travel by any reasonable route you are not restricted to the services shown" obviously not applicable to advances.

 

[deltic]

Would a first step be to always allow the quickest journey which may require a doubling back. At least in areas with ticket gates to prevent deliberate misuse.

 

[Bletchleyite]

Add a sentence when using a planner "as the ticket you have selected permits travel by any reasonable route you are not restricted to the services shown" obviously not applicable to advances.

You still have to have a set of rules to determine what itineraries it will issue, even if you then say "but you can use other ones if the guard doesn't mind".

== Doublepost prevention - post automatically merged: 17 Jun 2025 ==

Would a first step be to always allow the quickest journey which may require a doubling back. At least in areas with ticket gates to prevent deliberate misuse.

Because of the anomalies this would create I think you'd have to look at fares first, such as creating station groups over larger areas. I can see sense in a Milton Keynes fare group for example. Wolverton and MKC already take the same fares for all but very local journeys, and Bletchley for most longer ones.

 

[Haywain]

Would a first step be to always allow the quickest journey which may require a doubling back. At least in areas with ticket gates to prevent deliberate misuse.

I can't really see that happening as it would lead to overcrowding of long(er) distance services.

 

[Bletchleyite]

I can't really see that happening as it would lead to overcrowding of long(er) distance services.

Would it? I reckon you'd only get a small number of Bletchley passengers doubling back at MKC (and I'd expect some of those already do it using the WMT Crewe service because they don't attempt to pass the gateline and so have no idea it's not allowed, and tickets by and large are not checked on board south of Northampton). Most people prefer a direct service even if slower.

 

[thedbdiboy]

It should be scrapped, allowing the shortest route, any route within 3 miles and “any reasonable” route.
I’m saying that as one of the small number of passengers who understands most of the routeing guide as it exists today.
But it’s embedded in journey planners, as @Bletchleyite says, so it would be very hard to unravel.

If your baseline is 'Any Reasonable' you don't need the 3 mile rule. Rules iike that create loopholes and gaming the system. These days, a standardised Journey Planner algorithm coupled with AI could probably manage 'Any Reasonable' quite successfully, letting a wide range of routes and journey options apply to open tickets but weeding out the smartypants that want to justify the use of their London to Scotland ticket via Cornwall

 

[Benjwri]

If your baseline is 'Any Reasonable' you don't need the 3 mile rule. Rules iike that create loopholes and gaming the system. These days, a standardised Journey Planner algorithm coupled with AI could probably manage 'Any Reasonable' quite successfully, letting a wide range of routes and journey options apply to open tickets but weeding out the smartypants that want to justify the use of their London to Scotland ticket via Cornwall

I fail to see how bring AI into it would make it any more simple. It would do the opposite. It becomes ‘your route if valid if the magic black box says so’. Not to mentions AI is notorious for differing answers each time.

If you’re needing AI to interpret what the rules are, it’s too complicated.

 

[Bletchleyite]

I fail to see how bring AI into it would make it any more simple. It would do the opposite. It becomes ‘your route if valid if the magic black box says so’. Not to mentions AI is notorious for differing answers each time.

If you’re needing AI to interpret what the rules are, it’s too complicated.

Indeed.

Add to that that "any reasonable" only really works if the sanction for not being reasonable in the eye of several different people along the way is just to be told not to do it again or at worst charged an excess. With strict enforcement involving penalty fares and prosecutions (even if it is only supposed to be an excess) it has to be possible to find out in a completely black and white way if a route is valid or not.

 

[miklcct]

It's said that for a while, and only plays if the code entered has the first stop outside the London zones, which is correct.

Would a first step be to always allow the quickest journey which may require a doubling back. At least in areas with ticket gates to prevent deliberate misuse.

I think a double back should be disallowed under any circumstances, as it will create break of journey loopholes, unless you want to ban break of journey completely.

Indeed.

Add to that that "any reasonable" only really works if the sanction for not being reasonable in the eye of several different people along the way is just to be told not to do it again or at worst charged an excess. With strict enforcement involving penalty fares and prosecutions (even if it is only supposed to be an excess) it has to be possible to find out in a completely black and white way if a route is valid or not.

Reasonable is a subjective word which can't be a basis for rules to be enforced. For example, if I travel from Brighton to London via Barnham, is it reasonable given that there are 6 direct trains per hour on the main line? What if the main line is closed instead? If I double back via a further station even the train journey takes longer, is it reasonable because the next direct train is a long wait? If I deliberately choose a slower route because I don't want to change trains at an unfamiliar London Terminal, preferring a smaller station with fewer platforms to navigate, am I reasonable?
Or if I deliberately travel via Clapham Junction and Willesden Junction to London Euston because I don't want to use the deep tube underground, is my route reasonable? No one can give a definitely answer.

 

[Benjwri]

I think a double back should be disallowed under any circumstances, as it will create break of journey loopholes, unless you want to ban break of journey completely.

Under any circumstances? How would one travel from Bristol to Pilning?

 

[Bletchleyite]

Under any circumstances? How would one travel from Bristol to Pilning?

I don't think it's reasonable to be telling passengers they can't go by any route they like, there just needs to be a means of pricing it. To avoid anomalies, the fare system can reflect the distance by rail to some extent. For instance rather than saying people from Ormskirk can't go to Preston at all on a Sunday (well, not without a split), which is presently how the system works, you simply have a Route Liverpool ticket priced the same as one from Liverpool to Preston, and route the other one "via Rufford", then the passenger has an option to make the journey they want and the fare system doesn't gain an anomaly.

In BR days they did generally accept Ormskirk-Preston via Liverpool on Sundays as long as you didn't break your journey (though with a fully ungated system they couldn't really stop you) but that hasn't been the case for a long time. Presently Trainline just tells you you can't, which is just silly. (It doesn't seem, unlike Trainsplit, to sell splits over routes where there isn't at least one through ticket).

 

[Adam Williams]

These days, a standardised Journey Planner algorithm coupled with AI

That's gone super well for Trainline, hasn't it..

 

[thedbdiboy]

I'm actually trying to look forward here. I realise that the current crop of AI assistants are pretty ropey; What I was looking ahead to was a specifically configured AI application that would be more permissive on open tickets than simply refusing any specific route that wasn't programmed in. That was exactly how the human brain mk1 dealt with Any Reasonable before the advent of journey planners. If someone was taking an unusual route the presumption factors such as time of day and engineering work would be taken into account, and unless the route was clearly being used to avoid a fare or for nefarious purposes a reasonably benign approach could be used.

 

[Benjwri]

I don't think it's reasonable to be telling passengers they can't go by any route they like, there just needs to be a means of pricing it. To avoid anomalies, the fare system can reflect the distance by rail to some extent. For instance rather than saying people from Ormskirk can't go to Preston at all on a Sunday (well, not without a split), which is presently how the system works, you simply have a Route Liverpool ticket priced the same as one from Liverpool to Preston, and route the other one "via Rufford", then the passenger has an option to make the journey they want and the fare system doesn't gain an anomaly.

In BR days they did generally accept Ormskirk-Preston via Liverpool on Sundays as long as you didn't break your journey (though with a fully ungated system they couldn't really stop you) but that hasn't been the case for a long time. Presently Trainline just tells you you can't, which is just silly. (It doesn't seem, unlike Trainsplit, to sell splits over routes where there isn't at least one through ticket).

I agree, but I disagree that any reasonable route is a good idea. You yourself talk about how ‘generally’ things were accepted. There shouldn’t be ambiguouity because if there is people will unknowingly get caught out and that just isn’t fair

That's gone super well for Trainline, hasn't it..

Any press is good press right?

I'm actually trying to look forward here. I realise that the current crop of AI assistants are pretty ropey; What I was looking ahead to was a specifically configured AI application that would be more permissive on open tickets than simply refusing any specific route that wasn't programmed in. That was exactly how the human brain mk1 dealt with Any Reasonable before the advent of journey planners. If someone was taking an unusual route the presumption factors such as time of day and engineering work would be taken into account, and unless the route was clearly being used to avoid a fare or for nefarious purposes a reasonably benign approach could be used.

Not everything needs to be AI. Regardless of how ropey they are now you’re suggesting there should be AI because you want restrictions that need interpreting based on a huge amount of factors. Routing shouldn’t be up for interpretation, certainly not when the penalty is prosecution. People need to be certain of what will be accepted, and shouldn’t have to consult a magic AI which no one understands the exact criteria for.

As I said before the AI answer will always differ. It might accept a route one time but not the other. Do I have to bring a screenshot of it allowing my route?

 

[Farigiraf]

For instance rather than saying people from Ormskirk can't go to Preston at all on a Sunday (well, not without a split), which is presently how the system works, you simply have a Route Liverpool ticket priced the same as one from Liverpool to Preston, and route the other one "via Rufford", then the passenger has an option to make the journey they want and the fare system doesn't gain an anomaly.

While having two separate tickets (a Mon-Fri via Rufford and a Sunday via Liverpool) would be a good idea, I don't think it would be very popular with OPSTA (Ormskirk, Preston and Southport Travellers' Association) etc. if you increase prices on Sundays to match the Liverpool-Preston fare simply because GBR won't run a Sunday service. If you keep the price the same however, then you would have to either restrict break of journey (most likely unpopular) or adjust fares to every station south of Ormskirk to meet the Ormskirk-Preston via Liverpool fare, or else you would get people starting short. IMO the best thing to do would be to reroute the Stagecoach 2A bus via Croston, ensure it connects with Merseyrail in Ormskirk, and accept rail tickets on this service on Sundays.

 

[Starmill]

Apparently the new version of the NRCoT will just do away with the routing guide and concept of permitted routes entirely.

== Doublepost prevention - post automatically merged: 17 Jun 2025 ==

I think a double back should be disallowed under any circumstances

Is this a serious suggestion? How would you expect anyone to travel from St James Park to Axminster? Or Hartford to Warrington Bank Quay? Or a hundred other examples?

== Doublepost prevention - post automatically merged: 17 Jun 2025 ==

For instance rather than saying people from Ormskirk can't go to Preston at all on a Sunday (well, not without a split), which is presently how the system works, you simply have a Route Liverpool ticket priced the same as one from Liverpool to Preston, and route the other one "via Rufford", then the passenger has an option to make the journey they want and the fare system doesn't gain an anomaly.

You just replace one anomaly with another, on a Sunday the journey time would be triple and the price quadruple. Why bother doing all that work for something so useless?

 

[crablab]

Apparently the new version of the NRCoT will just do away with the routing guide and concept of permitted routes entirely.

I know this thread is a bit speculative, but could you expand on this a little as you've stated it in a fairly authorative manner?

 

[Starmill]

I know this thread is a bit speculative, but could you expand on this a little as you've stated it in a fairly authorative manner?

Unfortunately not as it's just the standard railway rumour mill. Nothing is agreed until everything is agreed and all that.

 

[crablab]

Unfortunately not as it's just the standard railway rumour mill. Nothing is agreed until everything is agreed and all that.

Fair enough
What would be the intention if the Routeing Guide was removed? Advances everywhere?
Even with Oval, there is still a definition of routes and stations that contactless ticket is valid at.

 

[Bletchleyite]

You just replace one anomaly with another, on a Sunday the journey time would be triple and the price quadruple. Why bother doing all that work for something so useless?

How's it useless? Not everyone finds the bus an acceptable mode of transport. If that ticket was offered I bet sales would be nonzero; just telling people they can't do something is silly.

 

[styles]

In a general sense? Yes.

In a practical sense? Yes, but it's far from simple.

GBR should be an opportunity to simplify our complex ticketing system, but without setting explicit fares for every point to point (and the trouble this would cause for open access operators), the routeing guide exists for a reason.

No doubt it can be simplified, but there's a ton of 'edge cases' which have built up over time, which are linked to the physical route options on the ground, and need to be accounted for, so we'll need something in place still.

It also needs to be consistent, so that guards and revenue protection can give consistent experiences to passengers.

We could of course go down the compulsory reservation route which many countries default (but not limit) to, but I don't think this is desirable personally.

 

[Starmill]

How's it useless? Not everyone finds the bus an acceptable mode of transport. If that ticket was offered I bet sales would be nonzero; just telling people they can't do something is silly.

I agree that it is indeed deeply silly to just go "nuh nuh nu nuh nuh you can't do that" which is essentially what the industry does nowadays. But if you took the two return fares it'd cost £27.10 Ormskirk to Preston day return, and £26.10 single. Just under quadruple the price is kind of an embarrassment don't you think?

 

[miklcct]

Is this a serious suggestion? How would you expect anyone to travel from St James Park to Axminster? Or Hartford to Warrington Bank Quay? Or a hundred other examples?

Yes. I am serious. In this case two tickets should be offered, because by doubling back, you are travelling further than the destination of your ticket which isn't something normally allowed. If you sometimes allow it, it will just cause confusions.

Practically, the ticket machine should give a warning why such a ticket is not available, as someone trying to find a route in the route map will not expect to travel further than the destination.

For example, if I walk into Luton station and want to travel to Kentish Town, I will look at the route map first, which shows a direct route. If there is a train at the departure board which calls at Kentish Town, great. If not, I will cross-reference the calling point with the route map to see if I can change at an intermediate station, for example, St Albans or West Hampstead Thameslink.

The process is exactly the same as travelling on the London Underground using a tube map, where double-backing is not a normal circumstances and is only used when engineering works or disruptions render certain platforms unavailable.

 

[Bletchleyite]

I agree that it is indeed deeply silly to just go "nuh nuh nu nuh nuh you can't do that" which is essentially what the industry does nowadays. But if you took the two return fares it'd cost £27.10 Ormskirk to Preston day return, and £26.10 single. Just under quadruple the price is kind of an embarrassment don't you think?

You wouldn't sum the fares, you'd just avoid the anomalies by setting it the same as the highest fare for any component of the journey, which is probably Liverpool to Preston. Yes, it's expensive, but it's better than telling people they can't make the journey.

From a very quick look it appears that £26.40 Anytime Short Return, £22.90 Anytime Day Return, £18.30 Off Peak Day Return and £17.50 Anytime Day Single, which are the same as the fares from Lime St except for the lack of complexities like TPE Only fares, are what is used for Kirkdale and Sandhills where one would be assumed to go via Liverpool. These should be the "route Liverpool" fares for the entire of the Northern Line from Ormskirk to Sandhills inclusive, as well as there being "route Rufford" fares for all those stations. Presently Walton to Preston doesn't even have a route Liverpool fare which really is ludicrous. "Yer'll af'ter go ter Laime Stsreets fer dat mate!"

 

[Starmill]

Fair enough
What would be the intention if the Routeing Guide was removed? Advances everywhere?
Even with Oval, there is still a definition of routes and stations that contactless ticket is valid at.

Something like the Dutch system where whatever the website returns can be considered permitted. My guess is it'll be something very, very much open to the interpretation.

 

[Bletchleyite]

Something like the Dutch system where whatever the website returns can be considered permitted. My guess is it'll be something very, very much open to the interpretation.

But presumably the Routeing Guide's mechanism would remain present in the background, as there has to be *something* that tells the journey planner what to return. I can see sense in simplifying the public terms and conditions to "if the journey planner returns it then it's valid" rather than the Guide being the determiner (similarly I could see sense in moving away from having separate textual peak restrictions and making the "unpublished" ones the actual ones that apply, so instead of reading a restriction code to see if you're valid on a walk up you'd type it into some sort of planner) but there still needs to be the thing in the background to make the planners work.