Great Western Coffee Shop

From Bristol Post (https://www.bristolpost.co.uk/news/bristol-news/bristol-train-disruption-after-cable-7975824)


Looks like it was the same yesterday and the day before - in fact last departures from Severn Beach appear to have been at 13:00 and 15:00 on 24th December!

How long does it take to fix a signalling fault?  Can't things like pilot(wo)men be used to keep the service running?

Industrial action....

Taxis running, leaving Avonmouth when the train arrives and Severn Beach ten minutes BEFORE the train would normally leave.

Service on "The Beach" seem to be running hourly between Temple Meads and Avonmouth, passing each other at Clifton Down.   Flagged as "VVR" on Real Time Trains - not a code I have seen before.

It's not that new, but has got far more common. Officially:

But with the added note "VSTP is a mess.".

I think this is all part of the current practice of recasting much if not all of the timetable the night before. That's both a change (from the WTT), and also the basis from which changes are defined (so the TOCs can redefine cancellation).

Have you a link to that page, or a full list of all the codes you can copy here ... please ...? Thanks!

Apparently, buried cable troughs and a high ground water level are hampering finding out where the fault is, let along repairs.  Plus of course it's very much a secondary line.

This from Tom on RTT website 8th June 2021 https://www.railforums.co.uk/threads/realtimetrains-website.189152/page-28
To tidy this off:

VST - STP schedule under VSTP arrangements
VVR - VAR schedule under VSTP arrangements, note that this does not discriminate between a VST that has received a variation, a WTT that has received a VAR. You can also get stacks of VARs on top of each other in VSTP world. Or indeed a totally new schedule that has come through as a VAR.
VCN - schedule CANcelled under VSTP arrangements, sometimes they come with the nightly service plan agreement for the next day's services which appear as a PG cancel too, but sometimes not.

tl;dr VSTP is a mess

(Way over my head!)

Mine too ... I have tried to come up with a complete list in "English":

As it is planned to be:
WTT - Working Timetable

Variations:
VAR - Varied from the working timetable well ahead of time
STP - Short term varied
VST(P) - Very short term varied
VVR - varied and then varied again

Cancellations:
CAN - Cancelled well ahead
VC(A)N - Varied and then cancelled

I do apologise if my rather rushed post this morning misled. I was passing on RTT's "official" explanations, assuming their wording obviously havd nothing to do with "The Railway".

There is accessible support information for the data feeds in the Open Rail Data Wiki (https://wiki.openraildata.com/index.php?title=SCHEDULE), but it isn't up to date.

The Schedule feed contains the WTT entries, which are there from one timetable change date to the next. There is an entry per service, indexed by UID number. Services added during the timetable's currency have STP entries. These are "targeted"  to be in place twelve weeks in advance, though obviously some happen much later than that.

Changes are made by added entries with the same UID, and these apply on stated days only. Adding a CAN entry means a service does not run on that day. Adding a VAR entry means the service runs, but with differences.

Very Short Term Planned entries come in a different data feed (called VSTP). They use the same format as in the Schedule feed; they are interpreted as overriding those. No more detail is offered by the Wiki. I think they were originally expected to all be new entries, i.e. equivalent to STP but too late for one. Now it seems they have discovered that the other entry types are also possible and are using them. The "V" in these means "Very short term", i.e. in the last 48 hours.

VST  new services added in the VSTP
Anything in the Schedule - WTT, WTT+VAR, STP, and I think STP+VAR and even VST, can be altered by VVR or cancelled by VCN.

My take on this is that the 48 hour cut-off was the last time a train schedule could be sensibly altered, as at that point the schedule was turned into train plans (allocate stock and crew, plan station movements, etc..). Any later change or cancellation was dealt with operationally, as a deviation from the current timetable. But a new train was still worth adding, so that everyone at least had a UID to work to. Since it wasn't in the train plans yet that wasn't a constraint; and it would have to be improvised into existence in any case.

If that was how it worked in the olden days - say five years ago - now the TOCs have a need to change the schedule very late so as to make their cancellations vanish. Thus they are now using these new entry types, for which the new codes are appearing in RTT (where they were invented, I think). This evolution is not reflected in the documentation available (of course much of that was not intended as public).

Another evolution: The VSTP feed is updated in real time; the Schedule feed is only updated overnight. Logically, there should be nothing in the Schedule feed that is new and applies within the next two days. However, the Wiki implies that may happen, so I guess realtimeness has been spreading throughout the system.

I think it's more like this:

WTT - in the timetable as published (may be planned to vary with date)
STP, VST - added to the timetables since the WTT
VAR, VVR - changed since introduction
CAN, VCN - planned cancellation (so won't appear in the current timetable)

VST, VVR, and VCN are late changes and may not be reflected throughout the system

Good to see services running from Severn Beach again today - first train left at 06:01 and scheduled hourly, through there's at least one 2 hour gap - late cancellation due to crew issues.