[LINK] more myki pain

[Tom Worthington]

Jan Whitaker wrote:
> At 09:28 AM 11/04/2010, Tom Worthington you wrote:
>>>
>> http://www.theage.com.au/victoria/new-tram-plan-could-get-myki-back-on-track-20100409-rysb.html

Stilgherrian interviewed me on "why transport smart card projects go 
bad": 
<http://www.zdnet.com.au/why-transport-smart-card-projects-go-bad-339302352.htm>.

The point of the Age article was that if you have a lot of people trying 
to get off the tram at once, it will not matter how fast the reader is, 
it will still be a bottleneck. The suggested solution was to remove the
need for scanning on exit. Or perhaps adopt the car e-tag system: 
passenger would put the ticket in their hat, to be read by a scanner 
placed above the door, . ;-)

> Yes. They can't make it work TECHNICALLY because the comms break too 
> often on board and between the tram and the central data hub. ...

The news reports suggest that there are difficulties with the readers 
near the doors of the trams communicating with the central controller on 
the tram and from there to the central fare system. The obvious cause 
would be the high levels of electrical and radio frequency interference 
on a tram caused by the power for the electric motors. However, this is 
a little too obvious and any competent engineer will have checked that 
and tried installing filters, shielding and the like.

As you say, the ticket system should not depend on a continuous 
communications link back to the central control. That might work for the 
trains, where the readers are stationary, or on trains which have a high 
reliability communications system engineered for safety, but it would 
not work so well for the average tram or bus. I doubt the Miki system 
has been designed in this way. If it has then there is a simple 
solution: sack whoever designed it and start again.

> The system is NOT distance based. It's zone based. ...

Sorry, I assumed that the zones were geographic and so based on 
distance. Are they instead congestion based: that is are you charged 
more to travel in the inner city because it is congested, than for an 
equivalent length trip in an outer area?

In any case my point was that if there was just one flat fee the system
would be much simpler: you just debit the passenger one standard unit
when they start their journey. That may not achieve all that is wanted, 
but the Victorian Government needs to aim for a system within its level 
of ability.

The NSW government seems to have learnt the lesson that if you have a
complex ticket system fail, then you should introduce a simpler system,
not an even more complex one. The new NSW "MyZone" system is a
rationalisation of fares using the existing ticket machines:
<http://www.myzone.nsw.gov.au/>.
This will not provide much benefit to the public, but is
likely to work. Also it will provide an easier base for the Perl 
Consortium to build a smart card system from: 
<http://www.tomw.net.au/blog/2010/04/sydney-electronic-ticketing-system.html>.

ps: Another technique might be to use the microchips in pets for 
electronic tickets. You could then take your dog for a ride. ;-)


-- 
Tom Worthington FACS HLM, TomW Communications Pty Ltd. t: 0419496150
PO Box 13, Belconnen ACT 2617, Australia  http://www.tomw.net.au
Adjunct Lecturer, The Australian National University t: 02 61255694
Computer Science http://cs.anu.edu.au/user/3890