The Amazing Race 11 (All-Star Edition), Episode 11

Macau SAR (China) - Andersen Air Force Base (U.S. Territory of Guam) - U.S. Naval Base Guam (U.S. Territory of Guam) - Fort Soledad (U.S. Territory of Guam)

The growing popularity of The Amazing Race and its continued status as the only travel show on primetime national commercial broadcast television in the USA has made the producers of the race increasingly the focus of pitches from government tourism promotion officials eager to see their country, region, or city portrayed favorably on television as a destination for tourists.

This week there was a new twist: instead of a "product placement " for a destination or a product, the bulk of the episode was a product placement for the U.S. military.

Guam remains a colony of the USA (a "territory" governed by a Congress and a President for whom Guamanians have no vote) primarily because of its militarily advantageous location. Much of the territory is off-limits to local people, tourists, or or any other civilians, unless they have permission from the military. So for most of this episode the racers had military drivers and escorts.

The "roadblock" ("a choice between two tasks, each with its own pros and cons") could have been a choice between "killing" and "dying". Instead, it was a choice between two tasks whose descriptions had in common only the word "care", and the military was repeatedly cast in the role of helpers and rescuers.

The racers weren't doing anything you could do if you go to Guam -- unless you get there by enlisting in the military. If there's any lesson in this for travellers, it's that there are as many differences between typical tourist experiences and the images in tourism marketing materials as there are between real life in the military and the pictures painted by military recruiters.

These differences between the race and reality are not as unusual as one might imagine. More native-born U.S. citizens have travelled outside the First World in the military than as tourists. In much of the world -- including some of the places with the largest numbers of people from the USA -- those Americans mostly live in all-American (or at least all-foreign) enclaves such as military bases or residential and employment "compounds" for oil workers and other expatriates. Often those areas are off-limits to local people or to tourists without an invitation and perhaps an escort.

Local people's opinions of America and Americans are often shaped, for better or worse, by their attitudes toward those who live within the walls they've built to close themselves off from the locals. So you can't ignore the presence or impact of these other foreigners. But your experience of a place like Guam or Nigeria as a tourist may be very different from that of someone who saw it from the inside of a military base or an oil-workers' compound.

Limitations of review of ICANN's transparency and accountability

Internet "governing" body ICANN continues to ignore me, but I haven't forgotten them, and I haven't given up on either my attempts to report on ICANN's mis-allocation of top-level Internet domains ".aero" and ".travel" to the travel industry -- to the exclusion and detrimaent of travellers -- or my attempts to hold ICANN to the standards of transparency and acocuntability in its own bylaws.

In its latest round of self-congratulations, ICANN has commissioned (under a secret contract) a review of ICANN's processes by an (unnamed) group of "independent" reviewers, meeting in secret.

My comments today on the limitations of this "independent review", and ICANN's continuing failure to follow any of its own rules, are here .

Identity Project comments on European Union CRS regulations

I've mentioned previously that the European Commission has been conducting a public consultation on whether to revise or entirely repeal the European Union's Code of Conduct for Computerized Reservation Systems (CRS's).

Because there are no similar rules in any other jursidiction, but all the major CRS's do business in the EU, the EU Code of Conduct for CRS's currently sets the global standard for legal protection of personal information contained in passenger name records (travel reservations).

Today the Identity Project filed comments with the European Commission, explaining why the privacy provisions of the Code of Conduct for CRS's should be retained, strengthened, and enforced -- not repealed.

I'll keep you posted as I learn of any developments. There's no publicly-disclosed calendar for the "consultation" or review of the regulations, and no particular deadline for the Commission to publish or act on the public comments.

The Amazing Race 11 (All-Star Edition), Episode 10

Hong Kong SAR (China) - Macau SAR (China)

This week's episode of The Amazing Race didn't cover much ground: it all took place within a small part of the Pearl River Delta region along the coast of China. It's an interesting region, with a hundred million or more people and at least three other cities (Guangzhou, Shenzhen, and Dongguan) as large as, or larger than, the area's best-known urban tourist attraction, Hong Kong.

Yet tourists as well as television producers have a difficult time knowing how to describe their path through the several Chinas of the Pearl River Delta.

In his voiceover, host Phil Keoghan described the racers as travelling "from Hong Kong to Macau, in mainland China". The term "mainland" in this context is a term of art, not of geography. But the TV show's characterization is inaccurate in either sense. Geographically, portions of both the Macau and Hong Kong are on the mainland, and portions on islands. The ferry terminal in Hong Kong, from which the racers depart, is on the Kowloon or mainland side of Hong Kong harbor, not the Hong Kong Island side. Politically, both the former Portuguese colony of Macau and the former British colony of Hong Kong are now "Special Administrative Regions" of China, with the same relationship to the sovereignty of China.

This may seem pedantic nit-picking, and perhaps it is. Nonetheless, it's important for travellers in many regions to be attentive to the ways that local people, and the governments that wield power, in the places we visit define themselves and their nations or communities. It won't help your chances for being granted permission to visit what its government calls the "People's Republic of China" if you say you want to go to "Red China".

Should you say you want to visit what Pakistanis call "Azad [Free] Kashmir" or what Indians call "Pakistan-Occupied Kashmir" (POK)? That depends on the identity of the people to whom you are speaking, or whose forms you are filling out, just as it can matter whether the map you want to bring into Argentina shows the islands off its coast as the "Falklands" or the "Malvinas", and whether the map depicts them as part of Argentina. It's almost impossible to draw a map of Kashmir acceptable to the governments of both India and Pakistan, a problem which bedevils publishers of international newspapers and magazines. The application for for a visa to Syria asks if you "have ever visited Occupied Palestine", which others would call "Israel".

In the Chinese example, Taiwan describes itself as the "Republic of China" (ROC). In the PRC, which calls the island across the straits "Taiwan Province", it's best to refer to it simply as Taiwan rather than as the "ROC", a "republic", or worst of all "independent".

"Mainland China" is a relatively neutral term that is generally understood to exclude Taiwan, Hong Kong, and Macau. "PRC" or simply "China" is more ambiguous, particularly when it comes to entry, exit, and border-crossing formalities. Is crossing the (fenced and fortified) "frontier" between the Hong Kong Special Administrative Region of the PRC, and the rest of the PRC, considered an "entry" or "exit" from the PRC for visa purposes? At present, it is, even though this in an "internal" border within the PRC -- but there's no way you would know that from the visa application forms, the visa itself, or any of the signs at the crossing points. The only way to find out is to ask the locals.

The first time I entered Hong Kong, I was coming from and was going to be leaving to other countries outside any part of China. When the immigration inspector at Hong Kong airport asked me, "Are you going to China?", I assumed that their point was to make sure I understood that Hong Kong was now part of China. I answered, very carefully, "Yes, but only to Hong Kong, not to the mainland".

As it happens, that was the wrong answer: Under the "one country, two systems" formula, even government officers in Hong Kong resist identifying their Special Administrative Region as "part of China". The acceptable answer would have been, "No, I'm not going to China, just to Hong Kong". And in practical terms, that stay in Hong Kong only was not considered one of the specified number of "entries" allowed by my visa to the PRC, whereas crosssing between Hong Kong and the PRC was.

{Update: The article above has been corrected to clarify the counting of entries and exits for PRC visas.]

"Bottom Line Personal" on travel insurance

There's an interview with me on "Travel Insurance: What You Need -- and Don't Need" in the May 2007 issue of Bottom Line Personal magazine. If you are looking for more detailed advice and suggested providers of insurance for travellers, see my FAQ on Travel Insurance and the chapter and references on Safety and Health" in The Practical Nomad: How To Travel Around the World .

What's wrong with "Open Skies"?

"Open" is good, right? And what could be wrong with a new civil aviation treaty between the USA and the European Union that would "liberate" trans-Atlantic flights from the bogeyman of government regulation?

Actually, quite a lot is wrong with the proposed Open Skies treaty, although you wouldn't guess that from the laudatory press releases from the U.S. Department of State and the European Commission when the negotiations were concluded last month. Most news reports have been equally uncritical, even when they haven't gotten their facts so badly wrong that they had to correct them (as the Wall Street Journal did in response to my letters).

The proposed treaty has now been approved by the European Union, and is due to be signed in Washington at the end of this month. The draft treaty will then go to the U.S. Senate for a vote on ratification, as the final step in its adoption and entry into force.

The details:

The latest draft approved by the EU is widely viewed as a victory for U.S. diplomatic bullying and intransigence, on issues mainly related to the economic interests of USA-based versus EU-based airlines. (As usual, the interests of travellers have been poorly represented in the negotiations.) Unless there's a new and loud public hue and cry, debate in the Senate is likely to be perfunctory. It's crucial that the travelling public understands the implications of this treaty, and that Senators hear from their constituents about these concerns.

The best thing about the proposed "Open Skies" treaty -- and pretty much the only thing that has gotten any notice -- is that it would repeal the authority of national governments to regulate which EU or USA airlines are allowed to fly between the EU and the USA, and which cities in the USA they serve (as long as they do so nonstop to and from Europe).

Leaving aside any debate on the relative merits of government regulations and "market forces" (which of course would still be grossly distorted by government subsidies to airlines and airports) as means of allocating popular routes, gates, and landing and takeoff slots, the obvious problem is that this deregulation would be neither reciprocal nor fair as between the USA and the EU: Airlines from the USA would continue to be able to carry passengers within the EU, while EU airlines would continue to be prohibited from carrying passengers within the USA.

The proposed treaty is nominally reciprocal: USA-based airlines would be prohibited from carrying domestic passengers within EU member countries, just as EU airlines are prohibited from carrying domestic traffic within the USA. But because the vast majority of air traffic within the EU is international, not domestic, the result is grossly inequitable, as European airlines and governments have said throughout the negotiations.

The continued losers would be travellers in the USA denied the benefits of competition on domestic routes from European or other foreign-owned airlines.

As I took off from San Francisco for Amsterdam last month, I looked down out the right side of the plane just after wheels-up from Runway 28 at SFO to see a row of at least three parked Airbuses in the red-and-white livery of "Virgin America", the would-be counterpart in the USA to Virgin Atlantic, Virgin Blue, Virgin Nigeria, and the former Virgin Express.

No matter how many hoops Virgin Group and Richard Branson jump through to satisfy protectionist rules limiting foreign ownership of airlines flying within the USA, the Department of Transportation (at the behest of lobbyists for USA-based airlines) keeps moving the goalposts and withholding Virgin America's operating license. So the planes sit idle on the ramp, the old international terminal (now "Terminal 2") at SFO sits vacant instead of being used as the Virgin America hub, and travellers within the USA continue to be denied the chance to see what Virgin has to offer.

But that's not the worst of the proposed "Open Skies" treaty. As I described them in my statement in Brussels last month, the key clauses are buried deep in the document in Article 8 on "security". All of Article 8 is problematic, but the worst parts are sections 3 and 4:

Article 8

Security

... 3. The Parties shall, in their mutual relations, act in conformity with the aviation security standards and appropriate recommended practices established by the International Civil Aviation Organization and designated as Annexes to the Convention; they shall require that operators of aircraft of their registries; operators of aircraft who have their principal place of business or permanent residence in their territory; and the operators of airports in their territory act in conformity with such aviation security provisions.

4. ... Each Party agrees that the security provisions required by the other Party for departure from and while within the territory of that other Party must be observed....

As I described it in my statement in Brussels last month,

The latest draft of the so-called "Open Skies" agreement on air transport between the E.U. and the U.S. fails to respect the rights of travellers. It would override portions of the PNR agreement [between the EU and the USA] and create new mechanisms by which other provisions could be overturned without adequate public participation in those future decisions.

The "Open Skies" agreement incorporates obligations under law enforcement treaties [Article 8, Section 1], but fails to mention any of those treaties limiting the powers of law enforcement or protecting the rights of travellers. In particular, it fails to mention the International Covenant on Civil and Political Rights, which guarantees freedom of movement -- including international movement -- under its Article 12. The "Open Skies" agreement thus might be interpreted to take precedence over the ICCPR or any restrictions in the PNR agreement, especially if the "Open Skies" agreement is ratified as a treaty, while the PNR agreement is not, but remains a non-binding undertaking not enforceable through any judicial process in the U.S.

The "Open Skies" agreement [Article 8, Section 3] requires compliance with all "recommended practices" of the International Civil Aviation Organization (ICAO). By making ICAO recommendations mandatory, the "Open Skies" agreement effectively delegates to ICAO the legislative power of the E.U. and the US. This is especially problematic because national delegations to ICAO have never included data protection, civil liberties, or human rights authorities.

The "Open Skies" agreement requires compliance with "security measures" adopted by the parties to the agreement, but sets no standards for the manner in which they should be adopted, or can be reviewed. The "Open Skies" agreement thus requires compliance with even secret, unreviewable, orders adopted without due process or democratic decision-making procedures.

It's not just that ICAO makes no attempt to consider human rights, or that the USA and some of its allies such as the UK have a track record of using ICAO for policy laundering . Worse, because the U.S. Constitution makes treaties part of the "supreme law of the land", below the Constitution itself but higher than any state or Federal laws, the requirement for compliance with ICAO recommendations, once agreed to and ratified as part of a treaty, would tie the hands of Congress and could only be modified or repealed by a new treaty. And as a more recent treaty, it could be held to take precedence over any earlier treaty such as the ICCPR.

This clause of the proposed "Open Skies" treaty should be recognized, and denounced, for what it is: a surreptitious back-door attempt, bundled with a treaty that would otherwise face little opposition, to remove air travel from the protection of the most fundamental international principles of human rights.

Open skies? Or closed? Tell your Senators, now, not to ratify the proposed "Open Skies" treaty with the EU, and to hold hearings and debate on the proposed treaty that specifically include witnesses on the implications of these provisions for the human rights of travellers.

On Internet radio Wednesday; in New York next Wednesday

I'll be talking about similar issues of privacy and travel in an Internet radio interview this Wednesday morning (18 April) and in a guest lecture at New York University next Wednesday afternoon (25 April):

• Wednesday, 18 April 2007: Guest on the inaugural broadcast of "Uncovering the Truth" with Katherine Albrecht on the We the People Radio Network . 7-9 a.m. Pacific Daylight Time, 10 a.m.-noon EDT, 14:00-16:00 GMT/UTC. Realtime streaming audio and archived podcast at WTPRN.com .
  
  
• Wednesday, 25 April 2007: New York University, guest lecturer discussing, "Privacy and human rights issues posed by computerized travel records", Asst. Prof. of Computer Science Evan Korth's class in Computers and Society . 3:30-4.45 p.m. CIWW, 251 Mercer St. (Warren Weaver Hall), Room 101. Free and open to the public.
  
  

In both these presentations, I'll be giving an introduction to the privacy, civil liberties, and human right issues posed by automated records of our travels, including airline reservations and machine-readable passports.

How do travel records map (literally) our movements? What data is contained in these records? How are they used? How could they be (or are they being) misused?

Is the issue a "balancing" between security and civil liberties, privacy and safety? Or is there some other agenda and set of interests at stake?

In particular, I'll be talking about some of the specific issues that are the subject of current debate, including passenger name records and RFID passports .

It's a great honor for me to be invited to both venues. Katherine Albrecht is the author of Spychips and the world's foremost independent expert, critic and muckraker on the subject of RFID chips. Tomorrow's broadcast kicks off her new daily radio show. In Evan Korth's class at NYU, I'm the relatively undistinguished tail end of a lineup of guest speakers for the semester that has included such luminaries as Richard Stallman, Cory Doctorow, and Jimmy Wales.

Your questions -- on the phone on the radio, and in person in New York -- will be welcomed.

The following week, as usual, I'll be at the conference on Computers, Freedom, and Privacy , this year in Montréal where I'll be on a panel on "Your Reputation Precedes You: The Transfer of European Union Passenger Name Records to U.S. and Canada".

Details, as always, are on my events page, where you can also find the tentative itinerary for my upcoming trip around the world. Let me know if I'm planning to be near you, and you'd like to organize an event while I'm passing through.

[Update: While I'm at NYU, I'll be giving an additional, more technical presentation at 7 p.m. on Tuesday, April 24th, on "The network of computerized travel reservation systems." It's sponsored by the NYU chapters of the Association for Computing Machinery (ACM) and Women and Computing, and will be at CIWW, 251 Mercer St. (Warren Weaver Hall), Room 1314.]

The Amazing Race 11 (All-Star Edition), Episode 9

Kuala Lumpur (Malaysia) - Hong Kong SAR (China)

Once again in the latest episode of the "reality" television series about travel, the deciding factor in The Amazing Race 11 was which teams got on the fastest flights.

With six nonstop flights a day on three airlines from Kuala Lumpur to Hong Kong, it wasn't a question of finding the best connection. And all the teams figured out that the first flight was neither on a Malaysian nor a Hong Kong-based airline, but on China Airlines, which has the rights to carry local traffic on the daily nonstop legs between Hong Kong and K.L. (as well as those between Hong Kong and Singapore and Hong Kong and Bangkok) of its through flights to and from Taipei and Kaohsiung, Taiwan.

Instead, it was a question of which of the teams actually got on the earliest flights, after they were all told those flights were "fully booked".

As is routine in such a situation, each team was offered the chance to have their names "placed on the waiting list". That's a misleading term, since most airlines -- certainly including the ones involved in this incident, China Airlines and Malaysian Airlines (MAS), have multiple waiting lists for each flight.

The racers were told their chances were "very slim". That might have been an accurate statement of the airline staff's assessment of the situation, or it may have been at attempt to dissuade them from bothering to try. Airlines want their flights full, so they would prefer to have a few people standing by to take the place of any last-minute no-shows. But waiting lists are decidedly unwanted extra work for the ground staff, especially with impatient and demanding waitlisted passengers in their face at the ticket counter or in their office.

Each of the teams attempted -- with varying degrees and styles of rudeness and cultural insensitivity -- to plead their cases and argue with the staff about who was "first on the list".

But their tactics portrayed a fundamental misunderstanding of waiting lists and how they work.

All airlines overbook their coach/economy (third class) cabin. Some also overbook first and business/club class. Airlines expect a certain number of no-shows and unticketed reservations. Airlines can't always tell if tickets have been issued, and don't want to cancel reservations that might be ticketed. So airlines confirm more reservations than there are seats.

When they say that a flight is "fully booked", that means the airline has confirmed as many reservations as they are willing to confirm for that flight, at that time, in that booking class. It doesn't mean the flight will actually be full. It's normal for last-minute "walk-up" passengers to be told that they will have to stand by until the cutoff time when confirmed passengers forfeit their seats as no-shows. It's equally normal for walk-up passengers to be boarded at that time, if they are willing to pay full fare.

If they are next on the proper waiting list, a passenger can be confirmed from a waiting list as a result of any three actions:

1. A confirmed reservation is cancelled. Either someone deliberately cancels their reservation; or a time limit expires without the airline receiving notice that tickets have been issued, and the airline (automatically or manually) cancels the reservation; or someone fails to re-confirm and their reservations are cancelled 72 hours before scheduled departure; or they no-show and their reservations are cancelled (typically between 30 minutes and 2 hours before the flight, depending on the airline and whether it is a domestic or an international flight).
   
   
2. The airline decides to allow more confirmations in that booking class on that flight. That could be because of changes in how many no-shows or unticketed, uncancelled reservations the airline expects (airlines use a combination of past experience and patterns, human experts, and expensive, extremely sophisticated software to make these predictions); because a sufficiently high level of supervisor decides to authorize a higher level of overbooking in order to be able to confirm a specific passenger or group; or because the airline reallocates its confirmation limits from one booking class to another (for example, to make more seats available for sale in a lower-fare booking class if it doesn't expect to sell them for full fare).
   
   

Different airlines have different procedures for making these decisions. Is "capacity management" highly automated or manual? Is it centralized at the airline's headquarters, or delegated to the airline's office at the city or airport where the flight originates? Or does that depend on how far in the future the flight is scheduled to depart? Are local decisions made by the station manager or by lower-level sales or operations supervisors? Is the airline more influenced in these decisions by financial analysis (which potential passenger will generate the most profit?), personal relationships and loyalties (to individual passengers and/or travel agencies and agents) -- or bribes (rare, actually).

Once the flight is as far overbooked as the yield management system is programmed to allow, manual decisions are typically made on a case-by-case basis. It's not a question of, "Should we override the system to confirm X number of additional people on this flight?" but "Should we override the system to confirm this specific party?" Different levels of airline staff have the authority to authorize different degrees of overbooking. Typically, the flight is placed under "airport control", meaning that authority to make these decisions is transferred from headquarters to the local office, sometime between 24 and 12 hours before scheduled departure.

Only someone who works for the airline, or a travel agent with a lot of experience working with that airline, is likely to know which strategy is most likely to get a particular waitlisted reservation confirmed. Is is better for the passenger to make the request directly, or have it made by a travel agent? Is it better to plead your case politely in person at a ticket office, call the telephone reservation center, have the agency phone their contact at the airline's sales office, or have the agency send an electronic confirmation request through their CRS ?

Most people assume that a waiting list is a first-come, first-served queue. That's usually true, but usually irrelevant, because there are usually many waiting lists for any one flight.

There are separate waiting lists for each booking class. People willing to pay full fare, or already holding full-fare tickets, will be confirmed before passengers in low-fare booking classes. Waitlisted or standby non-revenue passengers, such as airline staff and holders of frequent-flyer tickets, will be confirmed or boarded only after all paying passengers.

For each booking class, there are typically three or more waiting lists. You can get on the lowest-priority waiting list just for the asking. An airline supervisor, sales manager, or revenue manager's authorization is usually needed to get you on the priority waiting list. The highest priority waiting list is reserved for those authorized by the station manager or other high-level supervisors. People who haven't yet bought tickets are less likely to be placed on the highest priority waiting than people who already have tickets, but whose reservations were cancelled (perhaps in error), who were involuntarily bumped from other flights, or who need to change their reservations due to truly dire emergencies.

The racers spent a lot of time arguing about their places in line, but they would have been better advised to focus on which line they were in. Where you are on a particular waiting list is usually much less important than which waiting list you are on. Unless a very high-level supervisor intervenes, everyone on the highest-priority waiting list for the highest booking class will be confirmed before anyone on any of the other waiting lists. The question to ask is not, "How many names are on the waiting list ahead of me?" but "Can you put me on a higher-priority waiting list, please?" If the answer is, "I don't have the authority to do that", ask, "Can you please send a message to ask that I be put on a higher-priority waiting list."

Travellers often want to know, "If I put my name on the waiting list for this flight at this fare, will I eventually be confirmed?" A travel agent who sells a lot of tickets on that specific flight, and has been doing so for years, may be able to make a pretty good guess. But nobody knows for sure. If you are waitlisted at the airport on the day of the flight, like the racers, you'll find out when it departs if you got on. Further in advance, it's hard to judge how long to wait for your waitlisted reservations to be confirmed, and when to give up and make some other reservations that you can confirm. I've had customers confirmed after months on waiting lists, but I wouldn't generally wait more than a couple of weeks on a waiting list, even for a flight many months in the future, before making alternate arrangements.

Privacy advice for travel managers

I'm featured in the cover story on How to protect travelers' personal data on pages 28-32 of the April 2007 issue of T & E [Travel and Entertainment] Magazine , a leading trade journal for corporate travel managers. (If you don't like the funky page viewer on the magazine's Web site, you can use it to download a selected range of pages as a PDF, at least in some browsers.)

The story focuses particularly on the data in PNR's , and includes sidebars with my recommendations on the contractual terms on data privacy that travel managers should negotiate for in their contracts with travel suppliers, and the lack of compliance with European Union privacy rules by USA-based travel companies, especially CRS's .

There's a lot more, of course, that could have been said, or that I did say but that didn't make it into the story. In particular, the "horror story" at the beginning of the article could have been even worse: As I've reported previosuly , it's easy for a stalker or identity thief to get your entire itinerary (including where you are staying, when you are returning, etc.) from a discarded luggage tag, boarding pass stub, or in other ways.

It's a positive sign of the (belated) attention being paid to the issue, and the growing recognition of the dangers of secret and/or nonconsensual "sharing" of reservation data. And it's especially timely with the European Union CRS regulations, including their privacy provisions, currently under review .

If you're a corporate traveller, show this article to your company's travel manager, and ask what they are doing to get clauses like these into their contracts, to protect your privacy and the company's trade secrets.

The Amazing Race 11 (All-Star Edition), Episode 8

Ojcow National Park (Poland) - Kuala Lumpur (Malaysia) - Batu Caves, Selangor (Malaysia) - Kuala Lumpur (Malaysia)

After the fact, it's easy to criticize Joyce and Uchenna on The Amazing Race for booking flights from Krakow to Kuala Lumpur with a only an hour to change planes in Frankfurt, and losing out when their flight to Frankfurt arrived too late to make the connection.

But they were only doing what most other travellers choose to do. As a travel agent, I constantly found myself having to warn customers about tight connections (often the connections which, because they provided the shortest total journey time, were listed first by CRS's and airline Web sites), or talk them out of insisting on choosing connections which, while they satisfied airlines' recommendations for "minimum connecting time" (MCT), left no margin for routine delays of the incoming flight.

It's tempting, I know. If there's a choice of a "legal" one-hour connection, or a six-hour connection, which would you choose?

Learn from their mistake: If you are making connections to an infrequent or heavily booked flight that you can't easily afford to miss, make the extra effort to book flights with more connecting time. You may not have a million dollars at stake like the racers, but unless you are certain there is (A) another later flight (B) on which seats will be available © that you would be willing to wait for, at your own expense, at the connection point, you should choose another option.

Airlines and travel agents sometimes speak of a set of flights that satisfies the recommended MCT as a "legal connection". But that means only that airline rules allow them to issue through tickets for those connecting flights without being required to give the customer any special warnings about the risk of a missed connection.

But a "legal" connection is never a "guaranteed" connection. Even if the connections are recommended by the airline, and ticketed at a through fare, the airline has no responsibility except to put you on the next flight with seats available on the same route -- whenever that might be. Most long-haul flights are no more frequent than daily, so a missed international connection usually means at least a full day's delay, perhaps in some expensive international hub like Tokyo or Frankfurt. (See the analysis of the cost of missed flights on pages 22-27 of my comments on the USA's proposals for "vetting" of international passengers.) If the next flight on the same airline isn't for two or three days, and the next available seats aren't for two or three weeks, the airline has no obligation to reimburse any of your costs for accommodations, food, etc., during the delay.

Minimum connecting times are specified by the airlines, and vary substantially by airport, terminal, airline, and sometimes even specific flight number. I've seen them be as short as 25 minutes, and as long as 4 hours (6 hours if the connection involves a change of airports).

Most airline ticket Web sites don't allow you to see what the recommended minimum connecting time is between two specific flights, or even the defaults for the airport (which the airline might have overridden, or to specify that you prefer longer connection time. Typically, you're only shown the shortest connections satisfying the MCT. It's one of my pet peeves about the way travel Web sites are "personalized" to sell you more stuff, rather than to serve you better. To make reservations with a longer margin for delay of the inbound flight, you may have to phone the airline, go to a ticket counter in person, or contact an offline travel agency.

As a general rule, I would allow at least one hour more than the recommended minimum connecting time for any international connection, connection to the last flight of the day, or connection at a time (such as a holiday) when later flights might be fully booked. The longer and more prone to delay the incoming flight, and the less frequent the onward flight, the more time I would allow. Making connections to a once a week flight (at various times I've booked a lot of people on what were then weekly flights between San Francisco and Khabarovsk on Aeroflot, London and Asmara on Ethiopian Airlines, Delhi and Athens on Bangladesh Biman, and Mumbai and Cairo on Egypt Air), I'd plan to arrive at the transfer point least 4-6 hours before the scheduled weekly departure.

That may sound unpleasant. But the alternative -- getting stranded for a week or even "just" overnight -- is worse.

The Amazing Race 11 (All-Star Edition), Episode 7

Stone Town, Zanzibar (Tanzania) - Warsaw (Poland) - Auschwitz-Birkenau (Poland) - Krakow (Poland) - Ojcow National Park (Poland)

My first travel job was with an agency specializing in airline tickets to and from Africa (among other complex international tickets). And one of the first lessons I learned was that finding flights to, from, and within Africa is different from, and more difficult than, finding flights elsewhere in the world. Because few Africans can afford to fly, airline capacity is extremely limited relative to the populations of many African countries. Even between national capitals and other major cities in Africa, direct flights are relatively uncommon, and may operate only once or twice a week, rather than every day. Flights between regions of Africa (such as between West and east Africa) are especially rare. Only a couple of airlines (preeminently Ethiopian Airlines, and more recently South African Airlines), operate extensive intra-African international networks, so journeys between regions often require using multiple airlines.

So it was typical of African air travel that the racers' journeys from Maputo to Zanzibar, and then from Zanzibar to Warsaw (a city without direct flights from anywhere in Africa) all involved complex routings, multiple changes of planes, and long layovers.

Joyce and Uchenna went to the South African Airways ticket office. Not surprisingly, they weren't told of the faster routings (not involving SAA flights) through Nairobi, but were sent 2500 km (1500 miles) in the wrong direction through SAA's hub in Johannesburg.

Interline connections via Nairobi would have been the best options for the racers, both going to and coming from Zanzibar. Some of the teams were told about connections on the same airline (Air Malawi) from Jo'burg to Zanzibar via Lilongwe, but apparently none were told about the more frequent interline connections from Jo'burg to Zanzibar via Nairobi.

Bill and Joe, who were eliminated from the first season of the race after they missed an opportunity to take a more direct set of flights, were eliminated once again this season after they (and several of the other teams) waited around at a travel agency in Zanzibar for 7 hours, rather than taking the first flight to the regional hub of Nairobi (as Mirna and Charla had done the week before when they took the first flight out of Maputo to Johannesburg) where there would have been more chances of getting on some flight toward their ultimate destination.

But as I've mentioned during previous seasons of the race, complex interline routings are the ones that are hardest to find, and that airlines -- or, as the racers found out in Zanzibar, most travel agencies -- are least likely to suggest.

Ironically, the same week that The Amazing Race thus spotlighted the importance of through ticketing and other "interline" arrangements between airlines, especially for smaller local and regional airlines, the USA Department of Transportation issued a long-pending ruling that will make it yet more difficult to find the best connections on routes like these, and more likely that travellers will be routed further out of their way through the hubs of larger airlines that belong to the dominant global marketing alliances.

Normally, agreements between competitors to set joint prices would violate antitrust law in the USA. But because it was obviously impossible to get to most international destinations without using multiple airlines, and because of the benefits to travellers of being able to buy a single ticket instead of having to buy separate tickets for each connecting flight, the DOT has long granted the International Air Transportation Association (IATA) a limited exemption from USA antitrust law in order to set interline procedures and joint "industry" fares.

Airlines are free (subject to the provisions of the aviation treaties between the countries served) to set their own fares for "online" service, but for complex journeys with multiple stopovers, or connections between multiple airlines, the IATA fares are often cheaper, more flexible, and better value for travellers than a series of separate tickets on each different airline.

More recently, the DOT has granted case-by-case antitrust waivers to major airline marketing alliances (Oneworld, Star Alliance, Skyteam, etc.) and code-sharing partnerships to "coordinate" fares, routes, and schedules between their member airlines.

In July 2006, the DOT proposed to eliminate the antitrust exemption for IATA's "traffic conferences" to set prices on trans-Atlantic routes (including between the USA and Africa, where there are almost no direct flights).

As I explained at that time, the DOT's proposal was based on a fundamental misunderstanding of the IATA interline fares, by whom they are used, and their benefits for consumers in those cases. The DOT's proposal was also based on an unquestioning acceptance of the claims of the largest airlines that a dozen or so airlines can over a worldwide route network and frequency of service comparable to that of IATA's 200-plus members.

On 30 March 2007, the DOT announced a final order identical to its original proposal. Although the DOT claims to have "considered" the public comments, it appears that the DOT's conclusion was pre-ordained, and the public comment period a mere formality.

The current IATA interline procedures and industry fares remain in place, but IATA is forbidden to discuss or adopt any changes to them. As they become obsolete, the likely result is that more and more airlines will withdraw from participation in industry fares. At the same time, the antirust exemptions for the largest alliances, and their member airlines in the USA, remain.

The DOT's ruling is a triumph for the lobbyists of the largest USA-based alliance-member airlines over the interests of smaller unaffiliated airlines elsewhere in the world, and a substantial blow to the interests of travellers between the USA and places without direct or same-airline service (such as most places in Africa) or on less well-travelled routes within and between other parts of the world.

Travellers are clearly the losers in the DOT's ruling -- not surprisingly, since mine were the only comments in the docket from any consumer advocate or any individual or organization other than airlines. In its final order, the DOT acknowledged an purported to "respond" to my comments, but failed to provide any basis for the claim that a ticket valid only on the limited routes of a few airlines is "equivalent" in value to a ticket valid on any of the routes of hundreds of airlines.

For complex journeys, it will be even harder than ever to find the most direct or efficient routings, if they involve multiple airlines that aren't part of the same alliance. (In a further irony, given the travails of the racers in getting in and out of Tanzania last week and this week, the example one airline gave me of how "useless" interline agreements are was of their agreement, which was being terminated, with Air Tanzania.) And more journeys will require separate tickets for each leg, at a higher total price. More travellers will have to claim and re-check their baggage at connecting points, instead of being able to avail themselves of interline agreements for through baggage checking.

And rather than offering new services to fill the gaps left by withdrawals from interline agreements, the big alliances are likely simply to add more misleading codeshare designations to existing flights, making it appear as though there are more flights and more seats available than there really are.