Last month, we said that there would be more frequent updates than before. The only downside to this is that making a blog post every time we release an update bombards you with trivial updates. We’ve decided to make posts every couple of weeks, wrapping up the changes since the last post.

Here’s what’s changed over the past few weeks:

• We’ve changed the way we handle cookies, which means there’s a new pop-up when you first visit the site on a device. Some people have reported problems with this – if you’re seeing problems, such as the popup never disappearing, please email us
• Our new East Kent map covers Faversham to Kearsney, Ramsgate, Martin Mill and Minster.
• The Brighton area map has been redrawn, and now covers the Coastway route as far as Lancing, with route indications at Brighton
• The West Coastway map now covers Shoreham-by-Sea to Emsworth – including Littlehampton and Bognor Regis. It links up with the Horsham map at Arundel, and the Guildford map at Emsworth/Warblington.
• The Catford Loop has been added to the West Dulwich to Otford, Teynham and Sheppey map
• The resignalled Shepperton branch now shows routes, signals and track circuit indications
• Train operator codes are now shown on schedule pages so you can easily see who operates a train. The FRGT code shown where we can’t show you the train running number (or headcode) has now been replaced with a simple padlock symbol – you can still click on it to show the schedule though
• We’ve made some efficiency savings to the site so pages load faster and feel more responsive

We will continue making updates live as soon as they’ve gone through testing, and we’re making a real effort to try to get through the backlog of support tickets – please bear with us!

Now that we have a vastly improved server infrastructure, we’re able to do releases on a much more frequent basis. This week, there’s been a minor release every couple of days, and we’ve done the following:

• At Ferryhill South Junction, the berth for signal T474 populates when Y474 at Northallerton is occupied. We think this is a cosmetic error within the signalling system itself, with no impact on safety. We’ve put a note on the map to mention this
• Between Horsforth and Headingley, berth 3888 never showed trains. We’ve fixed this
• The Rugby map now has route indications for about a third of the map, and we’ve fixed an error with the positioning of some signals in the yard
• Many of the route indications at Sittingbourne showed incorrectly. This was down to a problem with data mapping on our side – we were one bit out! We’ve fixed this, and in the coming hours, the incorrect routes should clear out. We’ve also added route and signal indications to the Sheppey branch
• Signals TL2047 and TL2049 near North Kent East Junction to the right of the London Bridge map now show their aspect
• We’ve removed the berth for signal 1001 at Meadowhall (Sheffield), as this is a distant signal
• At Clapham Junction, we’ve added some more route indications on platforms 1 to 11
• Trains through the tunnel near High Brooms disappeared, so we’ve added an extra berth so don’t disappear
• Platform 1 at Willesden Junction can have a train terminate and run back north to Harlesden. We’ve added the missing berth, which is useful when London Overground terminate trains there
• At Nunhead, we had signal 452 listed twice, and we’ve corrected this
• Trains to and from Selby at Temple Hirst Junction between Doncaster and York now show in all berths on the line, and route indications for trains to and from the branch will now appear
• Signal G426 at Cheltenham Spa is now shown as a ground position light (GPL), not a main aspect
• Missing signal B66 near Nailsea & Yatton is now on the map, so trains will no longer disappear there
• Some errors crept in around the junction to Yate – two signals were missed, and some route indications were incorrectly drawn
• We now have route indications around Newport

We’ve also started planning for the migration to our new back-end server infrastructure, which will start with our train describer processor. Once this is live, the hit-and-miss linking between trains on the map and schedule data will be much, much more robust. More on that in later weeks.