I have argued before that ticket barriers are inappropriate for an airport railway: they are an obstacle to passengers, particularly those with luggage and small children. Foreign visitors are unlikely to know how to fiddle the ticketing system or want to tangle with the local police: mistakes in ticket purchase tend to be genuine.

Until recently, barrier gates have been seen as essential for train stations in Great Britain. The Department for Transport has put a requirement to gate stations in its recent franchise agreements - and hit trouble, particularly in places like Sheffield (where a key footbridge links two parts of the city) and York (where the station is a historic structure).

In the recent Greater Anglia franchise consultation document, the requirement to gate stations is diluted. Prospective franchisees are invited to demonstrate the case for gating "if this is considered an appropriate solution". There is an implication that increased staff visibility and a ticketless travel monitoring system might also be appropriate.

This looks like progress.

Ottawa has long been seen as one of the leaders in Bus Rapid Transit (BRT) technology. It has built a network of segregated rights-of-way served by fleets of buses.

Now there are plans to replace this with light rail.

The reason is that the segregated routes - Transitways - run to, but not through, the city centre. There, buses have to share roads with traffic - and there are too many of them! Daily ridership is nearly a quarter of a million: this needs 2600 buses on the streets each day. Within 15 years the system will have run out of capacity.

Solution? Light rail - and light rail in tunnel through the city centre, because a segregated right of way is an absolute necessity.

In 2001, a demonstration project, the O Train, opened in the city's suburbs.

This was called light rail, although in any European city it would have been regarded as a perfectly normal train service operated by ordinary diesel multiple units (which are virtually unknown in North America).

The main defect of the service was that it went from almost no-where to nearly no-where - although, on the way, it did serve a local college and a shopping mall.

It terminated - frustratingly - a couple of kilometres short of the airport. Even more frustratingly, the little-used freight tracks on which the train ran continued on virtually to the terminal: it was difficult to see exactly where they did go, because they were so covered with weeds!

Let us hope that, in the creation of the new light rail system, the city's airport is included in the network.

According to page 133 of the 2009 issue of the Department for Transport's publication, Transport Trends, aviation (international and domestic) now accounts for 23% of all petroleum consumed by transport in the UK.

Rail accounts for 1.3%, and road most of the rest.

Road consumption was growing through the 1980s, but has now levelled off.

See the DfT Transport Trends website.

The headline above is one you are unlikely to see in a newspaper. It's not news. However, the 685 fatalities in 18 accidents were news - because they are so infrequent.

The tragic thing is that giving air and rail accidents undue prominence in the press gives people a distorted idea of how safe those modes actually are, certainly by comparison with road.

I do wonder how many road accidents and fatalities the media are responsible for by their practice of endlessly reporting on rail accidents and thereby telling people railways are "unsafe", for example.

I also deplore the practice of reporting level crossing crashes as "Train wrecks" or "Rail accidents". In 99.9% of cases - at least - the train was the innocent victim: it was the car driver who was responsible, by misuse of the crossing.

The figures above - 685 fatalities, 90 accidents to all aircraft types, are for 2009: for 2008, the figures were 502 and 109.

In the UK, rail accidents seem to be getting more destructive - usually because of the high closing speed of trains in collision. (And, please note, railway privatisation had no impact on the steadily downward trend in UK rail accidents: Andrew Evans has comprehensively researched this). Presumably as aircraft sizes increase, we'll see the same kind of effect - fatalities/accident are likely to increase.

But rail and air are exceptionally safe ways to travel!

The German Aerospace Centre (DLR) has developed a system to prevent trains colliding with each other. Each train would carry a radio system: as soon as two trains were within radio range of each other, they would exchange information about their position, speed, loading gauge and planned route. If the system detected a potential collision, drivers would be warned.

The system is seen as a backup to existing signalling systems.

For details, see the DLR website

Collisions between trains do happen - notably, just outside Brussels earlier this year. But because of a fail-safe signalling system, disciplined and well-trained drivers and 175 years of learning from accidents, they don't happen often.

Sadly, far and away the most common collision is between a train and a vehicle on a level crossing. Something which could detect THAT and warn train drivers in advance really would be valuable.

Either that or an injection of that rare commodity, common sense, into the heads of all road vehicle drivers!

I flew from Montreal to LaGuardia on Tuesday on Air Canada's 8:50 flight.

A nice touch was that we were given what turned out to be a pretty reliable estimate of both the flight time and the gate-to-gate time. For the flight, the forecast was 58 minutes (in practice, 63): for gate to gate, it was an hour and 32 minutes (actually an hour and 22 minutes). Useful information, especially since it was pretty close to correct!

The bad bit was that, as usual when I fly Air Canada, I could not get the moving map part of the in-flight entertainment (IFE) to work. I don't think I've had a single flight with them when it's worked all the time: on most of the flights it hasn't worked at all. And it's the only bit of the IFE I ever use!

I had occasion today to try the two rail routes between Montreal and its airport - the commuter rail line operated by AMT, and the intercity line run by VIA Rail Canada.

Both were interesting and different: at some point I'll put a note of how to use both of them on our website, www.airportrailwaysoftheworld.com.

However, there were three other interesting things which happened on the rides.

Getting in to the commuter train to Dorval isn't easy - you have to climb four steep stairs, then slide a rather stiff vestibule door back to get into the passenger compartment.

At one of the intermediate stations, a lady who I guess was in her fifth decade managed to do this with a bicycle. This took some doing - especially the 90-degree turn at the top of the stairs while sliding open the stiff door!

Unfortunately, having done this, she was then told that for security reasons she couldn't take her bike on the train, and after some debate had to get off. Ouch! I wasn't quite sure why - the timetable says you can.

On the VIA Rail train back, I happened to choose a seat quite near the end of the coach. I was given very comprehensive information about how to break the window to get out in case of emergency. This involved pulling a tab, getting a hammer, breaking two skins of window, clearing the debris out with a seat cushion and putting my jacket over any remaining sharp edges. Wow!

Finally, the bilingual welcome to Montreal included a very stern warning about remaining seated until the train had come to a complete stand . Given the slow speed of approach of the train, this seemed to be a little excessive.

These last two events were rather reminiscent of air travel - the briefing if you sit in an emergency exit row, and the warnings as you taxi in to the terminal having landed at an airport.

Yes, both industries need to learn from each other, but were these two events over the top?

It was an interesting day!

There are some nice features about the National Rail web-site: I mainly use the journey planner, especially to check train times at the weekend.

You can, as with most journey planners, enter an origin and a destination as well as a preferred date and time of travel. You can specify whether the time is time of arrival at destination or time of departure from origin. Then up it comes with a selection of trains to meet your criteria.

When you get these, you can request a .pdf version of the timetable - what it calls the Pocket Timetable - with the times for your journey.

However, there's a bug in this bit!

If you specify that you want to depart at a specific time, fine, the printed timetable will take cognisance of this and times will start start at more or less the departure time you specify.

If however you specify that you want to arrive at a specific time, the timetable will assume that you actually want to depart at that time - so you'll arrive late!

Nice try!

This a question sometimes posed by those planning new high speed rail networks.

An airport is clearly a major traffic generator - and, probably, the larger the airport the larger its traffic potential.

At the same time, so is a city centre, a downtown, a Central Business District.

While it is true that these two points - airport and city centre - are not everyone's destination, it is likely that a city centre is (almost by definition) more central to a city region than an airport. Equally, it is often important for regeneration reasons to include a city centre stop.

What's the impact on journey time (and therefore attractiveness) of doing both - of stopping at both airport and city centre?

One fundamental assumption is that this can be done without significantly lengthening the route.

I did three calculations, based on an average speed of 250 km/h and a distance of 1010 km. I assumed that the first and last five kilometres would be limited to 50 km/h.

From this, it can be calculated that a non-stop journey would take four hours and 12 minutes - 4 hours for the 1000 km at 250 km/h and 12 minutes for the 10 km at 50 km/h.

Supposing that there is an airport 15km from the city. Those 15 km are limited to 100 km/h (5 km/h faster than Heathrow Express's average start-stop time). The station stop itself takes 2 minutes.

Under those assumptions, the 12 minutes would still be required for the terminal stations at each end. The line speed of 250 km/h could only be used for 985 km: the remaining 15 km would be done at 100 km/h. This gives a total journey time of 4 hours 19 minutes - 7 minutes longer than the non-stop journey.

The third calculation assumed that the airport was 30km from the city, and that those 30km are limited to 100 km/h.

Under those assumptions, the 12 minutes would still be required for the terminal stations at each end. The line speed of 250 km/h could only be used for 970 km: the remaining 30 km would be done at 100 km/h. This gives a total journey time of 4 hours 25 minutes - 13 minutes longer than the non-stop journey.

Finally, I tested the impact of reducing the overall distance - to 510 km rather than 1010.

This reduced the total journey time in all three scenarios by two hours, so the percentage incremental time caused by the airport stop is larger, but the absolute amount is not.

With the longer journey, the percentage increases in journey time are 3% and 5%: with the shorter one, they are under 6% and 10%.

So there is certainly a time penalty. Are its effects offset by the increased revenue from serving the airport directly? Is this greater than the reduced revenue caused by attracting fewer centre-centre passengers with the longer journey time?

Suppose a 10% increase in travel time leads to a 10% decrease in passenger numbers? Would that be offset by the revenue from airport-specific passengers? What percentage increase in passenger numbers would you get from a stop at the airport?

Only specific studies in specific localities will show this with any degree of accuracy. But it shows the kind of size of the effect we may be considering here.

The Port Authority of New York and New Jersey publish an annual Airport Trafic Report, a mine of information about its airports and their traffic.

One gem I noticed in the 2008 issue (published last May: the 2009 issue should be out soon) concerns inter-airport transfers.

Table 2.8.1, part of the Passenger Demographics section, shows the profile of departing passengers - including their trip origin.
17% of JFK's passengers and 17.4% of Newark's passengers said that this was "another airport".

The survey was done in the terminals, but it isn't clear whether it was landside or airside. I'm trying to find out - does anyone out there know?

If it was airside, then these passengers could be in transit. They could have started their journey at London Heathrow airport, and were caught by survey staff as they passed through Kennedy or Newark Airports where they were catching flights on to domestic US destinations.

If however it was landside, then it implies that about one in six of the passengers departing those two airports had arrived at another New York airport and made a surface connection.

This, of course, isn't impossible - it's just time-consuming and not particularly easy, especially if you don't like cabs!

So it's curious that so many people appeared willing to do it - and, presumably, for them it was the best way to make the trip they were making.

In 2008, JFK saw 48 million passengers and Newark 35 million: 17% of these is about 8 million and 6 million.

Given the nature of the traffic at the major airports in the region, it is likely that a significant number of these passengers were transferring from LaGuardia airport. Supposing the number really is 8 million a year between JFK and LGA - that's around 30,000 a day or over 800/hour (assuming an 18 hour day). That's a lot of people!

People sometimes suggest to IARO that there should be inter-airport rail links - Heathrow to Gatwick is the favourite. But it's always felt that relatively few people actually make that kind of trip. People dislike journeys involving a change of terminal, leave alone a change of airport!

But recently a proposal has surfaced in China for a high speed rail link betwen the airports of Hong Kong (mainly served by international flights) and Shenzhen (mainly domestic flights). The concept is that people would fly into Shenzhen from mainland China, take the airside train for the 50km to Hong Kong Airport, then catch an international flight from there. A recent IARO report highlighted some of the difficult issues needing resolution for that system to work - and one of these was forecasting the size of the market (see the publications page of the IARO web-site.)

So it's interesting to see these figures from the other side of the world!

At the end of March, the UK Department for Transport issued its "Transport Carbon Reduction Delivery Plan".

A major problem I have with it is the downbeat attitude to freight transfer from road to more benign modes. This is typified by the use of statistics which, while no doubt true and accurate, are downright misleading. Moreover they've been known to be misleading since the mid 1970s.

The report, on page 64, says that, "68% of all road freight movements (measured by tonnes lifted) are within the same region and so are unlikely to have a viable mode shift option".

Let's set aside the fact that historically some of the heaviest freight flows in the country - by tonnes lifted - have been within the same region: these are flows of coal for power stations.

But let's look at two important factors.

First, we're talking carbon reduction. And that is a function, not of tonnes lifted, but of tonne-kilometers moved. 100 tonnes moved 20 kilometres use a lot less carbon than 100 tonnes moved 200 kilometres: you need the distance factor to measure the potential importance. Tonnes lifted is irrelevant.

Second, in 1974 a book called "Changing Directions" was published (and I remember, as a post-graduate student at Birmingham University, meeting some of the team who wrote it). This showed the the importance of distance very vividly on page 229.

The then Department of the Environment had calculated that the maximum amount likely to be transferred from road to other means was only 1.5%, based on the average length of haul of road freight travelling over 25 miles.

However, "Changing Directions" made the point that, if there was a mode transfer away from road of 60% of the road freight in the 150 miles and over range, 40% in the 100 - 149 mile range and 30% in the 50-99 mile range, the figure would be very different. About a quarter of all road tonne miles would be removed to other modes.

"Those who cannot remember the past are condemned to repeat it" (George Santayana).

The Malaysian Transport Minister, Datuk Seri Ong Tee Keat, is discussing with ERL, operator of Kuala Lumpur's Airport Express, the planned extension of the rail service to the new Low Cost Carrier Terminal (LCCT).

He is reported to have said that financial implications as well as the overall operation and technicalities needed to be studied.

Malaysia Airport Holdings, the airport authority, is to start construction of the RM2bn terminal in the middle of the year, for completion in 2011. It will be some 1.5km from the main terminal.

The Transport Minister said he was proud of ERL for its 99.7% punctuality record over the last 6 years.

This is all eminently sensible.

The present LCCT is very difficult to access from the main terminal and the train service. This leads to major congestion in parts of the city - in particular, at Sentral Station - as buses struggle to cope with demand. The new LCCT will have a capacity of 30 million passengers: a rail link is certainly justified.

Moreover, as the Minister said, ERL has clearly demonstrated its competence over the last six years of operation: KLIA Ekspres and its associated commuter operation have always shown excellence in their operations and their commercial policies.

A report from Dubai notes that the waiting list for driving lessons has now disappeared - it was 4-8 weeks in the past.

The two main causes are thought to be the financial situation and the much-improved public transport system.

The report is on the web-site of the Dubai Metro. This notes that the pass rate for driving tests in the country was low - so perhaps the improved public transport will save people the time and expense ofhaving to take multiple tests.

Bus services are improving, with modern air-conditioned waiting areas on streets and equally modern air-conditioned buses in service. In addition, the excellent Dubai Metro is opening in phases - the main part of the Red Line opened last year (on 09-09-09), with more stations opening at the end of April and on 15 May. The Green Line is to open in August 2011.

The station serving Terminal 3 at the airport opened last September: the one serving Terminal 1 was among those opening on 30 April.

For more information about rail connections to Dubai Airport, look at IARO's website, the airportrailwaysoftheworld website.

There are a few tram systems around the world where, in some streets, overhead wires are not used in order to preserve the historic street-scape.

These systems tend to use either underground current collection systems (and since this was developed in France, it is referred to by the French term, alimentation par sol) or some kind of battery.

The T2 line being built in the suburbs of Paris uses ultracapacitors as a storage system. They are charged when the tram brakes: this is just like the regenerative braking system used on a number of electrified railways. The project is part of a major test by Alstom and Paris transit operator RATP.

The really novel thing about this particular installation is that pantographs are lowered automatically, without driver intervention: they are activated by the on-board GPS system.