GUEST: NDC Provides 'No Discernible Contributions' For Travel Sellers

Christian Kameir, a former exec with travel management company Traveltrust and CEO of travel market startup noGDS, submitted this column on the International Air Transport Association's New Distribution Capability standard-making initiative.

Similar to "Web 2.0" the term "New Distribution Capability" is a marketing term wrapped around a new proposed standard and therefore more confusing than helpful to a technologist.

At its core, NDC is a set of schema using XML created by Farelogix, which, according to IATA at least, enhances the communications between airlines and travel agents. It further proclaims to be open to any third party, intermediary, IT provider or non-IATA member for implementation and use.

But is NDC useful and will it be adopted?

Dominated by three global distribution systems, current airline distribution uses a closed communication channel with proprietary language to connect travel agencies to hotel and airline inventory. So, even though a GDS may use the Internet to move bits around, it might as well still deploy hardwired terminals at a travel agent's office.

Many travel agents still use the cryptic query language specific to the GDS operated by their employers or host agencies. And, unlike most modern software-as-a-service applications, a GDS interface is often a client software using a VPN tunnel into the actual booking utility. That GDS interface closely resembles previous IBM 3270 terminal technology. If you lean over a travel agent's workstation virtually anywhere around the globe, you will likely see blue- or green-screen text interfaces reminiscent of the early days of personal computing.

There have been various announcements by each of the three GDSs touting NDC applications, mostly to sell ancillary services like premium seats and checked luggage.

Developers familiar with GDS API integrations report that these have little or no support and documentation and seem to be designed to be "backwards compatible" with the green-screen used by agents.

GDSs fancy themselves as leading technology providers. However, a quick look at these corporations' financial statements reveals a different picture: These companies make money from every transaction conducted through their channels__mostly from travel agents, and a little from OTAs. They aggregate buying power and use it as leverage.

Despite various announcements by GDSs, it is unlikely that Amadeus, Sabre or Travelport will embrace NDC/XML or any other Web standard that would enable travel agents to connect directly to airline inventory, as the GDS would lose control over the buying power it holds by aggregating agency purchases.

As a specific set of XML schema, NDC defines airline distribution via a set of rules for encoding data in a format that is readable by both humans and machines. The design goals of XML specifically emphasize simplicity, generality and usability across the Internet.

While XML has been the preferred data-exchange format of most programming interfaces in the past, more than 80 percent of APIs today use newer protocols, according to ProgrammableWeb. Google, Twitter and Facebook predominantly use JavaScript Object Notation, mostly for its higher efficiency.

It seems likely that by the time all parties have agreed upon NDC's XML specifications, the Internet will have moved on.

NDC was designed to give airlines inventory data control. However, the true limitations are rooted in the fact that buyers do not control client and traveler data. With a few exceptions, travel agencies do not control their own client data today, as it is stored within the GDS in a proprietary and mostly unstructured format. Any "new distribution capability" must first solve the problem of unstructured traveler data. This is currently not addressed by NDC.

To the degree that NDC is a set of defined XML schema, NDC provides some utility, such as allowing developers insights into airline ancillary structures, even without the support of the GDS. Potentially NDC might provide functions exceeding those of OpenTravel Alliance schema, which also is XML based. However, by not addressing more practical concerns of converting legacy data conversions, this format is unlikely to be adapted by professional sellers of travel.

Finally, NDC is still missing most hospitality-related requirements. Since airline ticket sales are not always self-contained purchases, IATA would be well advised to rejoin the efforts of the OpenTravel Alliance rather than trying to create a competing standard favoring airline control.