Monday, 12 September 2011

Can You Do Any Better?

A challenge thrown down ...
One or two blog commentators have been a little sniffy with fbb's crusade against bad publicity from Travel South Yorkshire. One suggested that fbb "had a grudge against TSY" and a subequent e-mail asked "Could you do any better?"

Three guesses as to fbb's answer! No, on second thoughts, you are only allowed one guess.

It is not easy to reproduce a whole timetable leaflet in blog format, but the ever adventurous chubby one will have a go. Recent TSY leaflets all share the same structure; a front cover, general "useful" information, a list of stops, a map and a timetable.

So readers are invited to compare fbb's offering with TSY's "official" publication. The September 5th route 123 revision is typical.
TSY shows Broomhill and Beighton as major destinations, although the route serves neither. See "Betjeman's Beautiful Broopmhill [2]" (read again) for the Broomhill saga and here is Beighton High Street with absolutely no sign of a 123. The First bus is a service 30, which, together with the recently added X52, are the only buses from Sheffield to Beighton (thanks, anonymous, for your correction, see comments below).
fbb corrects this blunder ...
... and adds important destinations Endcliffe Student Village, University Western Bank and Crystal Peaks to the main list, two of which are completely ignored on the important front cover. These are significant places that many people will want to travel to.

Waterthorpe is removed because that is the name of an estate, also NOT served by the 123, although the more usefully listed Crystal Peaks shopping centre is on the northern edge of that area. Intake is replaced by Normanton Spring because, once again, the bus does NOT serve the traditional former Intake tram and bus terminus, but IS the only bus service via Normanton Spring, so worth a mention, mayhap?

Most important of all, Fulwood is replaced with Crimicar Lane, long standing (since 1955) Sheffield bus destination. This removes a deep-seated source of confusion with TSY's data.
The so-called "list of stopping points" repeats all the errors of the front cover and adds a few more. The 123 does not run via Hunters Bar ...
... which is a roundabout with an old tollgate on it; Western Bank appears twice; Sheffield Centre appears to have moved to outside the University (which it isn't) and Woodhouse is in the wrong place. Finally TSY's list is just SOME stops, about a quarter of them. So fbb offers ...
... a more accurate alternative (click to see the whole panel).

TSY has invested (heavily) in a very clever map creation computer program from a Swedish firm called Kartor.
This provides the maps for the bus leaflets, as here for the 123.
The tiny little rings are the bus stops, unnamed and useless because there are no road names, but with a very odd selection of place names which have been added by TSY's skilled (?) staff. It is not at all clear why we need two labels for Ranmoor or two for Endcliffe Student Village when the stops are on opposite sides of the road but clearly in the same location. Kartor's map shows ALL roads used, however occasionally, and, especially in the city centre, is just plain incomprehensible. fbb's offer (here showing only the route west of the centre)...
... split into two for bloggability ....
... attempts to resolve some of these inadequacies. AND there is room for important road names if these are desired. fbb has also created a city centre enlargement ...
... to explain the rather convoluted routes, widely different according to direction. In almost all cases, the  indicated stops match those on the stops list. Incidentally, fbb's full map takes up the same amount of space as TSY's.

Finally we come to the timetables.
The most glaring error is that TSY's IT systems are totally incapable of dealing with "loops" and the 123 has a substantial "one way" section at the Crimicar Lane end. fbb has corrected this error ... 
... although the operator would need to specify realistic times at Hallam Grange Road (see map) to replace fbb's educated guess. Other changes to remove misplaced and confusing locality names follow the pattern of the earlier adjustments. At the Woodhouse end of the route there is another loop in the evenings when journeys, paid for by TSY, serve the Skelton Road area of the village. But TSY unhelpfully shows the journeys as running to Woodhouse only. fbb clarifies this without adding clutter to the lists of times.
Clicking on these last two panels will show all the times.

So, sorry for the rather lengthy bloggathon, but fbb is satisfied that there is a real and valid alternative to TSY's rather unhelpful efforts. Of course, there is a huge difference between the two processes. TSY's is expensive, professional, totally IT produced and, according to TSY, absolutely hunky-dory and far better than sliced bread.

fbb's is amateur and uses cheap and cheerful software costing a total of £70! There is one other difference, of course, fbb knows where the buses go.

There is a suspicion, unproven as yet, that TSY's super system takes much longer to implement that fbb's. Maybe someone at TSY Towers will let us know?
Wow, nice offices! 

Next blog : Tuesday September 13th


  1. I wonder what you would consider to be the data necessary for the names of stops? The problem seems to be inconsistency, but I am not sure if consistency is possible. You could have three elements:-

    a) Town/city
    b) District/suburb
    c) Precise location - defined by road name or landmark (e.g. a church, pub or shop)

    The system probably needs to make all three fields mandatory - I don't think that would be a great problem, although perhaps in some smaller villages the first two would be the same.

    The programming of the system then needs to be setup to display that data intelligently. For example, if fields a) and b) contain the same data, then obviously only one needs to be displayed. If the route only runs within one city, so that for the display of the timetable, the contents of field a) will be the same for all stops, then it would not need to be displayed.

    The summary route description (on the leaflet cover) would probably best use field b), omitting the repeats.

    But then I have had experience of databases, and know quite well that such a set of rules would not adequately cover every instance. You can probably think of plenty of examples. Perhaps an extra field or two would be needed, or perhaps a manual override should be allowed for situations where the rules are inadequate. The TSY leaflets seem to allow complete freedom in this respect, and that clearly doesn't work.

    The other important point is to provide adequate space for the full names of the stops. In my local area, the two main journey planners (Deutsche Bahn and EFA-BW) refer to a stop as 'Seniorenzentrum am K', yet the full name is 'Seniorenzentrum am Kurpark'. The full name does not appear at all in the journey planners, although typing the full name in the DB site does bring up the correct one - as well as several others!

    Your point about knowing where the buses go is also crucial - anyone inputting data into a computer system for presentation to other people needs to understand that data themselves, otherwise the well known 'garbage in, garbage out' situation will inevitably prevail.

  2. Despite your suggestions, most of Sheffield's problems stem from the "mandatory" use of a district name. These are applied unthinkingly and unnecessarily. More significantly, some names have been invented by TSY to make the system work. This creates errors.

    The best stop descriptor system is that once used by Glasgow Corporation. Stops at other than obvious suburban "centres" were described as "Main Road" AT "Side Road"; e.g. Milngavie Road at Burnbrae Avenue. Almost fool- and computer- proof!

    The xephos naming convention only uses the district name for the main stop(s) thereat. Other roads have a district (ONLY where approprate) demoted to second place, e.g. in Sheffield:- "Hemsworth" for the terminus and shops, but Blackbrook Road Hemsworth for the approaches.

    This also works but is less specific.

  3. Inordinate amounts of money and effort is being expended to develop ever more innovative ways to deliver public transport information; noone wants to spend money or manpower on the grinding job which would be to ensure the data is 100% correct in the first place. Such a waste.

  4. Just a quick note - Beighton to Sheffield and vice versa is also served by the X52 route these days.