Is anyone actually going to take responsibility for the bus Working Timetables?

My prediction last week that the number of errors on the TfL WTT pages would exceed 500 by this week was comfortably exceeded. That’s out of about 3200. .

@jamesevans - is anyone actually going to take hold of this? Not necessarily today, given Covid-19 but … ever?


That would appear to be a “no” then.

I have placed a list of the 537 files involved at

That is a 17 per cent error rate. Quite astounding, in the wrong sort of way.

@mjcarchive @jamesevans

Is there something we developers can do to help?

What concerns me most is that the data that is being passed out this way is separate from the (legal) interface between TfL and the bus contractors. In an ideal world, we developers should be seeing the same datasets, at the same time, between the two responsible parties.

I’m very keen personally for the bus data being provided to the public to be as accurate as possible, otherwise people will just use their cars or Uber and make this whole exercise futile.

@briantist @jamesevans

If TfL gave us the slightest clue about what might be going wrong, then maybe developers or even the less sophisticated like me could help.

My guess - and that is all it is - is that the problem is a system which would work perfectly if the human input to it worked perfectly colliding with a human input which is, well, human. That is not always the human’s fault, if they have been given insufficient guidance on what does and does not matter (such as having current and expired files correctly and unambiguously identified).

Years ago I can remember watching some reply “yes” to all the prompts that were designed to ensure stuff was checked. When asked, they said they always replied “yes” because it got them past the prompt!

To update from last week, there are now 560 dodgy links. listed at

There are also some which link to files which are no longer current but which haven’t actually stomped over a newer version. I have not included these in the 560.

1 Like

An improvement this week, with more than half the errors being corrected via either loading even more recent schedules or by restoring the correct versions. 272 dodgy links remain, compared with 560 last week.

Interestingly, quite a few of the new schedules loaded carry dates from June or July, which is in line with the suspicion that the distinction between current and expired files is not always being made correctly, but as nobody from TfL is saying anything about the nature of the problem we are left speculating.

The history of all this suggests that there is a fair chance that the corrections will get overwritten by the same old errors next week … but we’ll have to wait and see. Also to see whether teh remaining errors get dealt with.

The 272 dodgy links are listed at