Blog homeAs I mentioned before, I am not that excited by any of the new meta-search travel websites that are out there. The reason for this is that they tend to just have data from the top hotel distribution companies, a few car hire operators and the top airlines. They are also interested in making money from meta-search - which - although a sensible idea for a business (!) - results in a less than perfect meta-search solution.
Just to be clear - I define meta-search differently to price-comparison. Most price-comparison sites are also meta-search sites - but not all meta-search sites have price comparison functionality (Google is an example of the latter)
My objectives for a perfect meta-search
- A distributed technology based on data standards (no single point of failure or control)
- No single commercial entity responsible for its management
- Anyone with travel products should be able to join in - either as a product provider - or as a meta-search consumer
What is my big idea?
The principle would be that all travel websites (providers & suppliers) would have, in a standard location on their website, a data file that describes their products, prices, availability and other commercial data. This data standard could be based on the Open Travel standards - however frankly their standards are just too bulky and complex for a less skilled developer to generate. They are great for highly paid developers found in the leading travel companies - but not those affordable by smaller companies. The standards need to be as clear as the RSS XML standard.
What you could now have is e.g. 10,000 product description & availability files dispersed around the web.
(It is possible to simplify the data standards if they only need to contain sufficient information for marketing purposes - the problem with the Open Travel standards is that they contain sufficient information for a sale to take place - which is why the data burden becomes too heavy)
How would a meta-search work
A meta search company is unlikely to want to work with every travel provider - they could choose a subset of the 10,000. They would import that data and create whatever search they want with it (meta-search, price comparison, inspiration based search etc). Another meta-search company would choose a different subset and so on (A ski meta-search would take ski products etc)
Consumer desktop tools could be created where a consumer could say “I am going to Egypt” - and they would, for a couple of months, pull in the Egyptian data feeds.
Think RSS / web feeds - you have an entire industry of aggregators and software providers. This is what the travel industry needs. It would make best use of the what the web can offer us….
This is the semantic web for travel.
Benefits
- Make the playing field more open for smaller travel companies
- Would deliver technical stability to the online travel industry and ensure that no single large travel company could take full control of the web (which is the status quo larger companies want to maintain)
- Provides a platform for other ideas that no one has even thought of yet
This idea is a chicken and egg problem. Step 1 would be to convince Open Travel that they need to produce data standards that developers like me can actually develop against and that focus on marketing rather than sales…..
If you want to be notified next time something is published sign up for email alerts or subscribe to the RSS feed. Thank you for reading!
There is something happening regarding tourism and semantc web in Austria:
http://www.ebsemantics.net/
You are spot on with your take on meta-search. When we built our own product we faced the same predicament in terms of trying to develop the system on a standardized xml framework for product data and availability syndication. As much as I like the idea of the OTA standards, they are really cumbersome (even for leading developers). We ended up developing our own XML API which we used as the basis for our front-end but when compared to the OTA standard, it is much simpler. After having worked with a number of different APIs from a variety of suppliers, it is very obvious that some kind of simple standard needs to be developed.
Hi Stephen,
We recently *unpublished* the TourCMS standard XML for integration with 3rd parties - because it wasn’t really being used. [Some are still using it, just we don’t explain it any more in order to put off new users while we work out what is best for the future]
I know this is a direction you want to go with your product.
Open Travel do have a new standard coming out in a few weeks time for the kind of products that you and we service. Some of your and our competitors have been involved with creating it - Adventure Central, Adventure Engine etc. I have a draft copy of it - thankfully it isn’t too far away from what we could support - but it is still complex.
I will show you mine if you show me yours…. (XML standard that is!)
Alex
@Harald - thanks - I will show to one of my colleagues who actually speaks German!
I knew they were both involved. We had considered joining in order to contribute, but the cost of $2,250 to join seems a little steep. I realize I can’t complain about the standard if I decided not to participate in the development of the standard, but with a limited budget, justifying an annual membership fee like that has to have some decent return.
I reviewed the 2008A publication and focused specifically on the Tour components. You’re right, we could probably support most of it, but it does have a very heavy leaning towards adventure, multi-day, product. The Destination Activity components are a little closer to the product types we support. I’m not sure if AC and AE had any involvement in that schema, do you know?
With regards to the showing of each other’s XML… I’m into that. It seems to me that AE and AC are very consumer focused (selling adventure product to consumers) and acting as aggregators for product where as you and I are focused on providing technology tools for suppliers and helping the supplier to sell their products better. Not to discount them as competitors, but AC received a significant VC investment ($3.5 Million) in 2007. I don’t know about you, but our pockets are no where near as deep. I’ll post our XML API and post the link for you as soon as the latest release revisions have been made.
Great ideia, good principles. I have always been working in IT, and on the recent years I have been more and more devoted to the tourism industry, and I have never seen an industry so needy to open up and, at the same time, so afraid or uncapable to do so?
There is still a long long long way to go to reach Web 3.0 in tourism.
As a crude example, do you know that, for an example, the Kayak API is now deprecated in favour of their widgets? Pfffff….
regarding ebsemantics. One of their projects is http://www.openevents.at/
And you will find the specifications at:
http://www.openevents.at/docs/OpenEvents_Spezifikation.pdf
They are also in German, but the charts and tables may give you an impression.
I’m not involved with ebsemantics but also think that we will need something like a semantic web for the tourism industry!
Being closely involved with the OpenTravel standards (both creating and vetting schemas) I agree that on first glance they appear cumbersome and complicated, but that is commonplace among standards which are created to be widely used.
The advantage of the OpenTravel xml schemas is that it is expected that implementors will create their own sub-set of a schema (a usage profile) to suit the requirements of their specific trading partnership which will still be vaild for the standard. Because they are re-using common data structures they will be able to leverage the first implementation to replicate it more cost-effectively to new partnerships and potentially also join existing OpenTravel-trading partnerships with less start-up costs than writing stand-alone apis from scratch.
Another aspect of the OpenTravel philosophy is that everyone (not just members) can submit a request to enhance a schema (http://www.opentravel.org/Specifications/CommentOnSpec.aspx) and all comments are treated equally and responded to. There is also an independent Implementors Forum (http://groups.google.com/group/OTA-Impl-Forum).
The viral approach is successful in other areas; there is no reason it couldn’t catch on with the OpenTravel schemas - there is a growing number of small travel organisations (from one-man bands upwards) taking the plunge and it only needs a few more small ‘converts’ to spread the word.
Re. Stephen’s question about Adventure Central and Adventure Engine: they were not involved in the development of the destination activity schemas - that was Viator, who also participated in the development of the tour schemas.
Tony