Luca Passani is the shadow behind WURFL (Wireless Universal Resource File), the open source repository for mobile device information which is key for mobile internet sites and content providers.
Luca Passani is an Italian software engineer with many years experience in Web and Mobile Internet development. Prior to joining AdMob, where he currently works, Luca has spent seven years with Openwave Systems and taken part in projects for Telecoms in the US and Europe. Luca is known to the community of developers for creating popular software tools such as WURFL and WALL. In addition to that, Luca has authored articles and co-authored books on Mobile development. Recently, Luca published the so-called GAP guidelines, Global Authoring Practices for the Mobile Web, which helps developers create mobile sites with minimal effort.
We are pleased to publish the interview of key questions on Luca’s thoughts on the mobile internet.
Q: Firstly what made you decide to create WURFL ?
Luca: that decision did not come overnight. I had been working with the problem of device fragmentation even before WURFL was born in 2002. In fact, as early as 1999, I was involved in the creation of the first WAP portal for a major operator in Norway. Soon enough, I realized that the WAP micro-browsers of the only two WAP devices on the market at the time, Nokia 7110 and Siemens C35, behaved differently when you tried to code for high usability (programming the softkeys, using keyboard accelerators, minimizing the number of clicks required of users to operate applications). It was already in 99 that I started wondering how, as a programmer, I could work around such ‘deficiencies’ and deliver a good user experience on both phones. Believe it or not, that was the beginning of a long journey that brought me to a basic intuition, the solution to mobile market fragmentation had to come from the developer community, since the industry was ‘genetically’ unable to provide device information and open-source programming APIs that the industry itself badly needed to take off.
Being the founder of WMLProgramming at Yahoo! Groups, a 3000-developer strong community, I had a great place to look for supporters. In early 2002, I created the WURFL schema and developers started pouring device information into the WURFL repository. Some offered APIs to access the repository in different languages (primarily, PHP and Java). More and more often discussions about development problems ended with the suggestion that WURFL was the way to go. The market was obviously there, but the tools were not. WURFL deserves credit for turning the mobile web into the development platform it is today.
Q: Are you forging partnerships with the device manufacturers ? Or are they not interested ?
Luca: I never really talked to device manufacturers. And they never came and talked to me. Manufacturers’ background is consumer electronics. Their business and culture is about keeping proprietary information confidential, go through strict legal loops when interaction with other entities is needed. My background, and the background of most developers in the WMLProgramming/WURFL community is different. Our background is the Internet and the web: discussing openly, sharing, open-sourcing…a very different story.
Anyway, to answer your question, I never really talked to manufacturers and they never really talked to us (with the possible exception of Nokia, who has made a substantial investment for developers through Forum Nokia). They already have UAProf after all. If manufacturers were interested in sharing, they would put their device information in UAprof. If they don’t, there is a reason for it. They do not want their competitor to know exactly what their devices can or cannot do. They prefer working behind closed doors. As with many different aspects of human activities, this can be traced back to culture and economics. WURFL tries to collect device information through channels that do not depend on the cooperation of device manufacturers (not carriers). Of course, if manufacturers or carriers wanted to talk to me about contributing device information to WURFL, I would be very willing to accept their contribution and to commend them for the choice. Finally, for sake of truth, when a new device leaves traces of its existence in someone’s logs, UAProf information is used to initialize a WURFL profile.
Q: There remains lots of holes in WURFL, have you thought about creating an API that allows developers the world over to update the database when they have the correct info ?
Luca: Yes. Will you ask me the same question three months later down the road? Things are moving in this area, but I don’t want to say too much just yet. I would like to make it simpler for people to add/fix WURFL information directly, thus maximizing developers capability to help one another with their common problem (device market fragmentation) and go back to what each and everyone of those developers knows best: their fantastic unique mobile idea.
I’ll use this opportunity to reaffirm a basic concept: WURFL will stay as open and free for everyone as it can be.
Q: How have the telcos reacted to WURFL ?
Luca: This is a great question. Operators have ignored WURFL initially. They figured they could keep track of their own devices with Excel spreadsheets and the cooperation of device manufacturers. Soon enough, GSM carriers realized this was not really working. They were losing track of devices on their networks and the situation is only getting worse. Today, also non-GSM carriers have the same problem: in spite of the closer cooperation with OEMs, it is not simple to figure out device capabilities for all the devices on their network. Today we are at the point where some operators have adopted WURFL in one form or the other, and endless more have evaluated it, even though they decided that WURFL did not cut it for them because they could not trust the information there and, more importantly, there was no company to call in case something went wrong.
Today all carriers are aware of WURFL and they are even recommending it more or less officially to their content partners who cannot afford the expensive commercial solutions in terms of money and flexibility (or, more precisely, lack of flexibility). Carriers are realizing of the big opportunity with mobile content and web, and they need to stimulate their ecosystem.
Q: Talk to us about Vodafone and the repercussions on the mobile internet. Do you think other telcos will do the same thing ?
Luca: As you know, I was really mad about what Vodafone did by forcing reformatting on all of its users. Let me take one step back. The fact that operators want to control their network, and find new ways to monetize it, is understandable and legitimate to some extent. But only as long as they are a good citizens in the mobile ecosystem and do not threaten the existence of everyone else. The discussion around the removal of the user-agent string from HTTP requests might seem academic and way too technical, but it is the lynchpin to demonstrating that VodafoneUK is trying to take possession of the mobile net in ways that would be unthinkable on the regular Internet.
By forcing the Novarra reformatting proxy on all of its regular subscribers, VodafoneUK has escalated the tension between carriers and the rest of the world to levels which were unheard of before. If we discard the ideas that Vodafone wants to kill the mobile web or that they are simply totally inept at what they are doing, the only explanation left is that we are in front of a re-edition of the walled-garden in the worst possible form: enlarge the list of partners a bit, but, at the same time, empower the network and turn HTTP-clients into the stupid thin-clients that the telephony world has always wanted them to be.
Vodafone’s move is a declaration of war against smaller players in the mobile ecosystem. If the first battle is lost, winning the next ones will be harder. Other carriers will feel entitled to follow suit. It will be increasingly hard for developers and smaller companies to sell content and services independently of operators.
On the other hand, I do believe that developers and companies have a chance to react and make their voice heard. And there are a lot of voices out there. With my Vodafone rant, I tried to catalyze this process and bring the “chorus” to sing the same hymn: “take your hands off our HTTP requests”.
The biggest problem here is that the companies impacted by the introduction of Novarra are typically too small to react to Vodafone. So small that they cannot engage Voda either at the business or legal level. Individual companies have tried to react, but their voice was to feeble to be heard and respected at the corporate level. Interestingly, Google, Microsoft and Yahoo have remained suspiciously silent in all of this. Either they have not realized what is going on, or they have their own hidden agenda to squeeze smaller players out of the industry in cooperation with carriers. Even entities such as W3C and dotMobi, which one would expect to go balistic on VodafoneUK’s attempt to disrupt HTTP, only came up with a mild reaction (irresponsably mild, I would say). It is obvious to me that while everyone has been very good at filling their mouths with words like “developers”, “content” and “open-up”, when it comes to defending legitimate developer needs, all of this support disappears and everyone prefers to be friend with the big bad guy.
This being the situation, I and the WURFL community are first in line to pick up the developer flag and protect the foundation of the mobile Internet. History will show that this is the right thing to do. If we successfully protect everyone’s chance to offer services on the net, this will be to everyone’s advantage. Not only small players, but also carriers which will see their customer-base enlarge as a consequence of the plethora of content and services available on the wide open Mobile Internet.
If we lose, other carriers will follow VodaUK’s example and the Mobile Internet will be set a few years back.
Q: What do you see as the future for the off deck mobile internet ? Where do you see this space going ? Do you think that the web 2.0 actors are going steamroll the mobile players?
Luca: I don’t have a crystal ball, so I am not comfortable making projections. What I know is that I was there in 94, when the web was available to students in Universities and not many others. At the time, I did not believe that the Internet in general and the web in particular would become the medium of the new millennium. In fact, I wasn’t even considering the notion at the time. The internet seemed just meant for the academic world.
Yet, creating web content was so simple and the concepts so powerful that the Internet took the world by storm within a few years.
If I had to compare the mobile web with big web in 1994, I would say that there are a lot of analogies and that there is a decent chance that mobile web content will be so simple to produce and consume, that it will take the world by storm. More than that, I think this is exactly what is going to happen eventually, with the biggest questions being how many years this is going to take.
Having said this, there are also important differences between the mobile web and the web. When the web was born companies were not even noticing, and making money with the web wasn’t being considered by those who created the first generation of web content. With the mobile web it’s a different situation. Carriers and other companies are thinking of monetizing the mobile web even before the mobile web is an established consumer reality. If the industry prevents human greed from creating insurmountable obstacles to content providers (as Vodafone and Novarra are doing), there is obviously space for on-deck and off-deck content complementing each other, with both worlds thriving in their respective spaces. Carriers make a large investment in setting up and managing the network. Their investments sometimes involve placing expensive devices in the hands of users at a fraction of their real price (subsidizing). Carriers obviously deserve a share of the pie because of this. Smaller players will also invest money and time in setting up new services. They also deserve a share of the pie when successful. If this vision of mine is shared by people who can take a decision, carriers and off-deck portals will co-exist for a long time to come. If the two realities will go to war with one another, this will likely create obstacles that damage everyone along the mobile value chain.
Q: Can you tell us something about your move to join AdMob?
Luca: Absolutely. AdMob embodies the spirit of innovation in the mobile internet which I feared to be lost for good back in 2001. AdMob has an honest (yet solid) business model that thrives on the satisfaction of everyone involved in delivering great content to end-users. Advertisers get a great new channel to place ads. Publishers get a simple way to monetize their applications. When I met Omar (AdMob’s CEO) at 3GSM in Barcellona earlier this year, he illustrated me some very simple facts: AdMob uses WURFL and so does a large chunk of their publishers. If I joined AdMob, I would be using a lot of my paid time to support WURFL and the mobile ecosystem. To me this was an offer I couldn’t refuse…
Thank you Luca for those pertinent comments and for your time.
