Thursday, December 21, 2006

BAIA invited post on the Funambol model

The guys at BAIA - Business Association Italy America - asked me to write a post on their blog ("The Voice of the Italian & American Business Community") to explain the Funambol model and how I feel it could be replicated.
BAIA is an independent, nonprofit, member-based network whose mission is to create tangible business value by establishing a professional forum through which information, knowledge and opportunities can openly and effectively be exchanged between the business communities operating in the United States and Italian business interests. BAIA's new approach to business networking is based on an Open Source Governance model. BAIA aims to enjoy the support of anyone who shares the BAIA vision of an open exchange network.
I share the vision and I like the people running it (and I am not really a shy individual...), so here you have my post.

The Funambol model: US capital and Italian heart

Life as an emigrant is tough... You look back at your native country and you struggle between love and hate. Emigrants flee home because they go looking for something they could not find where they were born. They leave their heart in their home country, hoping to come back one day to find that everything has changed.

I left Italy for California in '99 and I go back home quite often. Not that much has changed, but I am not planning to give up that fast...

After founding a couple of companies in Italy and working for a public US company, I founded Funambol in 2002. Funambol is the mobile open source company, bringing BlackBerry-like capabilities to the masses. A Red Herring 100 company and the largest open source project in mobile with almost one million of downloads, Funambol has been funded mainly by US Venture Capitalist. The headquarter is in Silicon Valley, but the R&D is in Italy. In Pavia, close to the local university.

US capital and Italian heart.

Italy is a beautiful country, but in a state of crisis. The economy is suffering. Globalization is killing our small manufacturing companies. Competing with China in these markets is simply not doable. Protectionism attempt will fail. We need to move on.

Italy can compete on high tech and software in particular. Our labor cost in this sector is highly competitive. It might sound strange to many, but Italy has the lowest cost of software in Europe (35K/year Euro per engineer). That's extremely competitive with respect to Silicon Valley (the weather is still better here, sorry) and also with India or China, where wages keep appreciating every year.

In particular, if you compare the Intellectual Property protection Italy enjoys. I know of a few examples of outsourcing companies in India, where a group of employees took a product they were developing for a third party and started a new company... Outsourcing saves money and it is necessary, but can be risky in a country where the legal system does not support you. On top of it, Italy is the best with respect to loyalty of the employees (a key element in software, which is all about people)

When it comes to education, Italy ranks high. In my experience, the competence of Italian engineers is comparable if not better than American peers. Italy is the country of creativity. It is in our DNA. Software is creativity. Add some rigorous engineering and you have Ferrari, the most beautiful car on the planet, but also a technological jewel.

Software is great for Italy. It requires limited capital to start. If you use open source as a distribution model, it is even better. You can take advantage of globalization and reach markets anywhere on the planet.

When you are ready to go to the next step, however, risk capital is missing in Italy. The ability to manage risks of Silicon Valley VCs is unrivalled. All your connections are here, if you want to grow your company or find an exit via M&A. This is the place to be if you want to go big. Not just for cash, but for the mentality. Things move at a different speed in the Valley. And I do not believe it is going to change any time soon.

For now, I see a great opportunity for Italy to be a center of excellence for software outsourcing. Becoming a new Silicon Valley is the next step. We are not ready for that yet, but the Funambol model is a start. US capital and Italian heart. I hope many more companies will follow. It is simply the best of both worlds.

Monday, December 18, 2006

About Palm and buying back your own OS

I have tried to analyze the last move from Palm for a few days and I have found it an interesting challenge...

The story is quite complex, if you are not a Palm follower. Palm spun off PalmSource (their operating system) some time ago and launched a Treo with Windows Mobile. Then Access bought PalmSource for
$324 million and pushed it even further towards Mobile Linux. Now Palm is paying Access $44 million for the license of Palm OS Garnet (the old version 5, the one that was supposed to be a transition to the version 6 that never saw the light of day, nothing to do with Mobile Linux).

So... You own an OS that everybody loves. You throw it into a separate company. You embrace a different OS that everybody hates (but you still launch a lot of devices on Palm OS, like the recently announced 680). Now you buy back an OS that is pretty much dead...

It sounds stupid, however:
  1. $44M is a steal if you look at the $324M Access paid for PalmSource (THAT was crazy)
  2. It protects your investment on the Palm OS Treo (650 and 680) and buys you time to move to a full Mobile Linux (which is a must to survive)
  3. It puts you in a better position to negotiate with evil Microsoft if you are thinking to add another Windows Mobile device (please don't ;-)
Overall, I personally believe it was a smart move from Palm. However, the more I look at this and the more I am getting convinced that the need to get rid of Access means something: they can't work together. Therefore, I am ready to bet that Palm will announce a Mobile Linux device in the near future BUT it won't be based on ALP, the Access Linux Platform.

That's quite interesting. I look forward to seeing where they go with this...

Tuesday, December 12, 2006

The next SMS

I read an interesting article about text messaging today. For the lazy web link clickers out there (I subscribe to that list), here's a summary of the content:
Text messages sent to and from mobile phones will more than double over the next five years to 2.3 trillion messages sent by 2010, a survey said on Tuesday. [..] Total revenues from text messaging is forecast to grow to $72.5 billion in 2010 from $39.5 billion in 2005. [..] Gartner predicts the level of SMS messages will top 1.8 trillion in 2010" [adding that] "Wireless messaging is the most successful mainstream mobile data service to have emerged during the 30-year history of the cellular telecoms industry"
In a nutshell: mobile messaging rules when it comes to data and it will keep ruling for the next five years. Forget about mobile TV, games, picture sharing and such. Messaging is king.

I have a small question, though. If messaging is the driver of future data revenues, is SMS going to be adequate for the user needs? My answer is no...
  1. SMS is limited in size of messages. It is not a problem when sending (people send Short Messages... fast and long SMS-typers are rare, although amazing to watch) but more when receiving. Screens are getting bigger, reading is not a problem. You can easily read a message sent from a PC on a mobile phone. Not with the current SMS. Combining more than one SMS is a horrible solution.
  2. SMS just allows a single recipient. You cannot send a message to more than one person (and you pay for each message...). You cannot do reply-all (although there are some cool companies working on it). That's very limiting in this new age of social networking.
  3. SMS does not support attachments. I am not talking PowerPoint here. That's power users. I am talking pictures... Every phone now has a camera. People want to send the pictures out. MMS has been a failure for the cost, configuration and clumsiness (in this order). The concept is still very valid. And a clear need is there.
We need messaging and SMS is not adequate. But it is pre-installed AND a standard working on every phone. That's the beauty of it. If you are in Europe, you know that half of the phones in a carrier network were not bought through that carrier. You take a SIM card, put it in. What works right away? Voice and SMS. Pre-installation and standard-based messaging is necessary.

So... that leaves you with the only option: the next SMS. Push-messaging (yep, SMS is push) based on standards, pre-installed on devices, capable of supporting attachments, reply-all and receiving your email. Your mobile life integrated with your web life. It's a SyncML client, or p-Imap or Lemonade. I do not really care. But I am convinced that is the future of mobile messaging for the three billion mobile users out there.

Not really a novelty, I know, but
"Wireless messaging is the most successful mainstream mobile data service to have emerged during the 30-year history of the cellular telecoms industry". The funny thing is that it is not going to change any time soon...

Monday, December 04, 2006

It's the usability, stupid

In one of my past lives (I have plenty), I did a Ph.D. in usability. Since then, I can't resist from looking and playing with things, asking myself: why? Why are you asking me to do this? Why? Who designed this thing? How many real people (engineers belong to a peculiar species) ever looked at this stuff? Why?

A few weeks ago, my TiVo froze while playing. I called DirectTV and they made me re-format it. It froze again. They told me they were going to send me a new one for free.

They sent me the new DirecTV DVR instead.

I decided to give it a try.
Although it was not built on open source as TiVo, all TiVo features were there. Actually, they even had a few more than TiVo. It was perfect from the requirements standpoint. The product manager checked all the boxes, QA did not find any bugs.

However, from a usability standpoint, it just sucks.

Nobody at DirectTV ever used it to watch a game. It is slow to respond to commands. It does not jump from one speed to another smoothly. You can just use it to watch a program as you would watch a tape on a VCR. Do not try to move around the program... All the features are there, but the consumer experience is just plainly bad.

What's the story here? The usability factor is key. Always key. Forever key. If you do not look at things with the eye of the user, you will build a perfect product that nobody will like to use. Emotions are everything.

When you look at mobile, the usability factor goes up an order of magnitude. You are dealing with a very small thing, that you carry around with you. Your attention span is measured in fraction of seconds, not minutes. You are not sitting on a couch or on a comfortable chair waiting. You are standing. You have people around you. It rains...

Still, so many products fail the usability test. Companies keep pushing the web paradigm to mobile users and complain about the network or the operators when their applications fail. They can't be used, they won't be used. Give me WAP a million times faster, I won't use it. Do not ask me to scroll and click. I won't. It should be simple but it is not. How many failures are we going to need for people to realize this?

I tried to send this message to DirecTV. I called them and told them their DVR sucked and I wanted my TiVo back. They said "Sorry, Sir, we do not have them anymore. And you just signed up for two
additional years with us because you requested a new DVR. If you want, you can return it." That's interesting... My TiVo breaks. They send me a fake replacement "for free" and I am signed up for two more years... Or I can go back to VCR.

I went online, found a hacker TiVo site. Ordered a replacement hard drive. Assembled it. Put the TiVo back with double capacity. Sent the DirecTV DVR back to them.

I am back in TiVo heaven, smiling at how quick it is to respond to my commands and how well thought it is from the consumer standpoint. And I am now ready to move away from DirecTV directly to their competitors...

So much for usability. Believe me, I am not alone.

Thursday, November 30, 2006

Bill Gates for President, good for Open Source?

I know some will be surprised, but I have always been a Bill Gates fan. It might sound bizarre, since I am an open source guy at heart, but if you look at the business Mr. Gates founded, it is hard not to admire his creation. A bit evil, for sure, but grandiose. Mostly late on innovation, but always using his unfair advantage at its best.

At the same time, I do not think there has ever been a moment in my life when I disliked Microsoft as now. Today, I would never buy anything Microsoft, not a game console, not an MP3 player, not software. I would avoid them with all my forces, from using Firefox to push the entire Funambol people to use OpenOffice. It was not like this in the past, although I never cheered for Microsoft in my life.

I asked myself why. Maybe because Apple looks so cool, maybe because the Zune sucks, maybe because of the Novell-Microsoft pact, maybe because of ActiveSync. I think I found the answer: Bill Gates is not there anymore. He is doing other stuff. Microsoft looks stale without him, including today the depressing launch of Vista (an Italian name BTW ;-)

What is Bill Gates doing these days? My younger brother works for the World Bank. His life, including his work at Unicef and WHO in Africa, has been dedicated to helping people in need. He told me once that what Bill Gates is doing for Africa with his foundation is incredible.
He has a real admiration for Bill Gates, which is hard to believe since I would not consider my brother a capitalist. So, Bill Gates is using his money (a lot) to help people (a lot). He even got a load more cash from Warren Buffett recently... I trust my brother: Bill Gates is clearly doing a great job there.

A few weeks ago, reading the Scott Adams blog (I know, I am a geek, but I met him once when I was at the HP Labs and I got hooked), he proposed Bill Gates for President. Today, he did it again. I always find it hard to disagree with Dilbert.

What benefit would open source get from President Bill Gates? I have no idea, but he will definitely be forced to balance between his evil past and his current job. Somehow, I feel Open Source will benefit since it is good for the people. Because of it, he will have to push it. Cut him loose from Microsoft and he will come to the light side.

I know it sounds bizarre, but it is an interesting idea. Too bad I can't vote in this country yet...

Tuesday, November 28, 2006

Mobile Linux going up the stack with Trolltech Greensuite

Today Trolltech announced the Greensuite Initiative with the support of some nice companies (among them, Funambol). The idea is simple: move Mobile Linux up the stack.

If you have been following Mobile Linux and the various initiatives, you know the bulk of the discussion is still on how to make the power management work... Trolltech moves the discussion a bit up the stack, integrating solutions from multiple vendors and providing what they call a "menu of technologies". In their words:
The Qtopia Greensuite initiative gives customers the ability to select from a menu of technology choices enabling them to build a device specific to their requirements and target market they wish to address. The Initiative brings the leading Linux-based application platform, Qtopia Phone Edition; a hardware development device, Qtopia Greenphone; and a set of partner technologies together giving customers a pre-integrated software solution.
The idea is smart. You take a Linux OS and you build a GUI around it, but you still missing some components that are key for operators (such as push email from Funambol, which the world is really in needs of ;-). The variety of demands from consumers in this market is unbelievable. You cannot cover all the bases. Therefore, you partner with software vendors and have them build something on top of your "client middleware".

The Greenphone and this initiative are not that far away from OpenMoko and our agreement with FIC. The difference is that Trolltech is not a device manufacturer, so they cannot build unlocked devices and flood the market. However, they have relationships with carriers and device manufacturer, which helps a lot.

In any way you look at it, it is another great step for open source in mobile. Mobile Linux is enjoying a lot of momentum lately. If you read the recent Vodafone announcement about standardizing on three operating systems (Linux, Symbian, Windows Mobile), which one will cover the extreme vast majority of their devices, which are consumer oriented?

I guessed so. Long life to Mobile Linux.

Wednesday, November 22, 2006

Palm and lessons from Apple

Today, I happened to be at the Stanford Shopping Center for lunch. I stopped by the Palm store to see the Palm 680. The Palm store in Palo Alto should be their flagship store, I guess. It is the birthplace of Silicon Valley, after all. It should be shiny and inviting.

Instead, it is such a depressing place... The front window displays only PDAs, not even one Treo. Who in the world is still buying PDAs??? You walk in, the colors are bleak. Few Treos on two tables. Nobody inside but a guy talking to his phone (at least, that was a Treo). Since there was no 680 around, I asked him about it and he said "we do not have it, but I can put you on the waiting list". On the waiting list?? You announced it on October 12th and today you announced it is available on Cingular!! You do not even have ONE phone on display??

I walked out and I thought "man, this sucks, how in the world can they attract consumers?". So I walked few steps and I entered the Apple store. No windows. Full of people playing with devices. I looked around a bit lost for three seconds and a friendly guy asked me if I needed help. Actually, I was looking for the new iPod shuffle. It was on the table. I could not see it because it was so small. But he had one attached to his t-shirt and he gave it to me... I played with it for a minute. I loved it. I will definitely buy one (or maybe way more, we'll see). I walked out smiling.

That's a consumer experience. Apple announces a product, the same morning you walk in any Apple store and you play with it. My daughter wants to go to the Apple store because she wants to play with the videogames they have there. I am forced there. I ended up buying an iPod and an iBook for my wife, although I have never been a Mac person. I might become one, though...

Funny enough, I came to the office and I saw an article on the Mercury News about Palm and Apple where the CEO of Palm says "What, me worry" about Apple? Naaahhh.

Well, I do not think Palm should be worried about Apple. They should simply learn from what they do. They do it simply great. I am on THAT waiting list, the one of the people waiting for the iPod phone. It will be an awesome device and it will change the way we look at mobile. I am ready to bet on it. They will not do a deal with a carrier. They will go around them, with an unlocked device or an Apple MVNO. It will have messaging, beyond SMS (something between IM and email). They will open this market up and shake it.

I am looking forward to it, so should be Palm, if they are really thinking about consumers.

Monday, November 20, 2006

Motorola buys Good: the Palm opportunity

Motorola buying Good has been a strange transaction.

Two things pop up, in my opinion:

1. It was announced on a Friday
2. It was announced with an undisclosed amount

Every PR agent on the planet will tell you that you do NOT announce anything on Friday. You are not going to get press or move the stock price. When you spend 500 millions on a company (as later reported), you are supposed to be proud of your acquisition and make a big deal out of it. Why would you not disclose the amount? Because you feel it was too expensive and you signed a bad deal?

I am not sure if we'll ever know, but it sounds to me like someone in Motorola was not really convinced about this acquisition. How would you? Good was getting crashed by Microsoft on the server side (zero dollars is hard to fight), with Palm OS disappearing (it was the only thing they had left to make money). They were about to disappear... And they got bought for 500 millions. Looks like a good deal for the investors in Good, which poured 200+ millions in it. They must be so relieved ;-)

The acquisition has been clearly a reaction to Nokia Enterprise buying Intellisync. Moto had to show something in the enterprise as well, they went with Symbol first and Good second. They were kinda forced to.

Motorola is not an enterprise company. The are a consumer company. Symbol and Good will not help them selling more RAZR. They will help them to have an enterprise story (a "story", not a business, since I will be surprised if they will get 10% of the revenues from enterprises compared to what they are getting from consumers...). Same for Nokia, where the acquisition of Intellisync has not really been a great success... The goal (for both)? Make sure Microsoft does not eat the entire smartphone market.

Here lies the only good reason for Motorola I could come up with: avoid Microsoft to control their destiny in the enterprise (Motorola is shipping the Q, with Windows Smartphone on it), to then protect their consumer business. Which is the only one they really care about.

Now, would you spend 500M for something you really do not care about? Yes, if you have a lot of cash ;-) But you will not make a big deal out of it. You keep it quiet as much as you can.

Next step, if they are smart: you make sure to avoid the mistake to move Symbol and Good into the consumer space. They do not get it. They will never get it. They are close protocol, closed source, close everything... All the people in those companies just get enterprises only. You better keep them separated from your core consumer business.

If Moto is thinking about using Good for a consumer push email and multimedia sync play, they are setting themselves up for a big failure. If they plan to migrate the Good client on Mobile Linux, I just hope they target it to enterprises only. The consumer market is going to be open. It must be open as in SMS open, as in Voice open. Carriers need it to make money... Interoperability is the only way to move data on mass market mobile devices to the next level. You do not do it with a proprietary approach. You do not do it with the Good protocol.

In any way you look at this market, you see four big silos appearing: RIM, Microsoft, Nokia/Intellisync, Motorola/Good. All proprietary. All closed. All enterprise focused. Anybody trying to go horizontally on the mass market (supporting a billion devices) with a proprietary approach will be crushed (and acquired). Honestly, I will be surprised if Seven and Visto will be around at the end of 2007. The only possible approach for the mass market is going with open standards and a leveled play field. And open source, if I might add ;-)

Now, what about Palm? They are not a silo. They do not have a server, no push-email, no multimedia over-the-air synching. They could have been RIM but they missed it, because they never developed a server component. Their entire story around push email was Good, now gone to a competitor... It must be tough to be on Palm's board these days...

However, this actually might be a great opportunity for Palm.

If I can suggest a strategy (I am a Palm fan, after all): go open source. From the device (with Mobile Linux: throw away the Treo 700w that is a piece of #!#@) to the server. Forget the enterprise market and focus on consumers, moving smartphones to the rest of us. When I look at the Treo 680, it points to the right direction. Cut a niche for yourself, move out from Palm OS to Mobile Linux. Embrace open source on the server side (no suggestions here ;-) and remember why you made it initially.

It was HotSync who made Palm. Synching your data with your PC via a cable was the reason for using a Palm. The next level is over-the-air synching. Your PC becomes a server in the cloud, powered by Palm (who once was a MVNO, if you remember...). You sync your email, PIM, pictures and music over-the-air, with multiple devices. It is your consumer mobile hub. Powered by Palm. Do it open, with SyncML, and you'll be a player. Stay the course, support Microsoft and proprietary approaches and you'll be another Symbol, bought by a silo.

I know I am biased, but this looks like a good opportunity to me...

An Ironman for open source

In my post about "How to create a successful open source company", I mentioned "Think IRONMAN" as one of the key elements:
Most people that went public with a proprietary company will tell you "it was a marathon". Open source companies are in the Ironman category. That is "SWIM 2.4 MILES! BIKE 112 MILES! RUN 26.2 MILES". The race ends with a marathon, but you have to swim a lot and bike a ton before you even start the marathon. Aggregating a large community takes years, there is nothing money can do to accelerate the process. It is a natural long process with phenomenal fruits at the end. But you have to be patient.
Clearly, I was thinking about Daniele, when I wrote this ;-) You might have met him at LinuxWorld, OSCON, 3GSM and other conferences (or online). He just represented Italy (and Funambol, and the open source movement in general) at the Ironman World Championship in Florida. He made it to the end. Strong and patient as an open source guy. You need stamina in this world...

Friday, November 17, 2006

Why open source: viral marketing and visibility

The second most asked question I get, when I talk about Funambol, is "why open source" (the first one being "how do you make money?"...).

I believe I have spent few posts writing about the value of a community-driven development and, in particular, QA effort. The former being the best way to innovate and keep up with the unbelievable speed of the mobile market (linked to the Code Sniper program). The latter being the crucial element for a successful company in mobile, because it is simply impossible to test every phone on the market, if you are a proprietary company (this one linked to the Phone Sniper program).

Speed of development and innovation, combined with the widest device compatibility, high quality and support are not the only positive of open source. The other one is marketing and visibility.

Open Source is viral. If you build an open protocol open platform product with open source code (a GOOD one) and you are honest with your community, you will have a large audience. The audience will be a mix of developers and users. They will enhance it, create derivative products or simply use it. In all cases, they will simply talk about it. Blog about it. Whisper about it.

Below you see the graph taken from Alexa (which measures web sites traffic rankings) about the Funambol site, compared to the one of our competitors (although some of them are not really competitors, since they are focused on enterprises). We are the blue line...

We are a small company of 40+ people, which raised 5M in Venture Capital. Some of those in the graph are public, some raised 250M+, all have been around for ages. How can we be that much more popular than them?

It is called viral marketing. It is a phenomenal added benefit of open source. It will never stop, as long as you remain true to your open source roots and do not screw up your community. You can be public and spend millions in marketing, but you will never get that. You get it if you create a successful open source project. And it will cost you zero dollars.

Believe me, it is not easy. But once you get there, it feels good to look at that graph every morning and at our balance sheet at night ;-)

Wednesday, November 15, 2006

NewsForge: "open source software makes business sense"

Nice article today by Nathan Willis on NewsForge on Funambol and our tight-rope walk. In particular, I like the closing statement:
But Funambol's efforts so far are impressive, and the availability of its free PIM synchronization server is a big win, particularly given how locked-down and vendor-controlled the mobile device market is today.
We are upgrading the Funambol Portal this weekend to v3 and we have great ideas for the next version in January. We are trying to make the Portal another good example on how a commercial open source company can return some of its revenues to the community. For us, it will be a free PIM and push email portal, for starters. A great service for the community, another QA, development and innovation tool for Funambol. I like win-win...

Monday, November 13, 2006

Sun opensourcing Java: it is all about mobile

I have a privileged point of view, when it comes to Sun and Motorola, since I often talk to both. It is interesting how the press really missed the point on Sun opensourcing Java and the reason for them to choose GPL.

It is all about mobile (open source).

Let me go back in time. At OSBC, in April 2005, I heard Jonathan Schwartz say that GPL was really bad and CDDL was the way to go. At Java One 2006, he said that they were ready to open source Java but set no timeframe. On October 23rd, Rich Green said that "
All of Java SE and all of Java ME should be completed, in terms of open source availability, by the end of the first quarter of calendar ’07". Two days later (two!), his boss Jonathan Schwartz said "by the end of the year". Three weeks later, it is all done, all open source and GPL. What prompted Sun to move that fast? And why GPL and not CDDL?

An interesting news hit two weeks ago. Few talked about it but it was really big. On November 1st, Motorola announced
plans to build an open-source version of Java Mobile adopting the Apache License. They simply dropped a bomb inside Sun...

The amount of revenues Sun gets from Java ME is absolutely significant. I would bet it is the #1 software revenue stream at Sun. Motorola moving to its own open source Java ME means taking out 23% of those revenues (that's Moto market share on device shipments). If Nokia follows suit, there it goes another 34% of those revenues... That's almost 60% of a significant revenue stream down the drain, thanks to someone else doing what Sun was supposed to do months (years?) ago.

Sun moved fast. It might be too late but they really stepped up to the plate. I fully appreciate it. They had to send a shock and they went with GPL, not CDDL. Great choice, in my opinion. Where does GPL matters? In mobile, of course.

Java ME GPL means dual licensing. It triggers when embedded (who embeds Java SE?). It is the same model of Funambol, MySQL, Sleepycat and others (in alphabetical order, of course). The only possible commercial open source model, in my opinion, after the devastation of the services-only model by Oracle. With CDDL, they would have lost the virality of GPL. They would have lost the chance to still make money licensing their VM, while attracting the largest possible community. It is a smart move, and another testament of dual licensing. It is the model that works, in particular in mobile.

In summary, Sun opensourcing Java is all driven by mobile. The timing came from mobile. The license is due to mobile. Motorola, in my opinion, was the target, not IBM. I am a Java fan and I always will be. They were clearly late but maybe not too late. Let's see what happens next. This market is moving so fast, it will be interesting to watch... Once again, though, one thing is clear to me: mobile open source is king and it is gaining momentum every day.

Sunday, November 12, 2006

My Mobile 2.0 Manifesto

Last week I was invited to speak at the Mobile2.0 event in San Francisco. It had all the right elements for a starting point of a revolution: a not-too-perfect organization but with the right people at the right time. I feel it will define the start of a new era. Turning points are defined by simple things, for example a mobile conference that happens in a basement of a hotel, where there is no cellular reception... People that cannot show demos. Just talk. And talk. Guys on the stage with a gazillion boring PowerPoint presentations and everybody that just wait for the discussion to start. Everything happening in the coffee break.

I know Mike did not do it on purpose (see his comment after the conference to believe me), but the result was just perfect. Read all the blogs that it generated. Check Brian's comment in particular. Something is boiling here.

As Steve Bratt noticed in the keynote, this looks like 1994. That was the year I founded Internet Graffiti. Connection to the web was slow and difficult (via modem, paying per minutes). It was all proprietary with some few good standards emerging (HTML and HTTP). Web clients were crappy and you had to develop the same page for each of them (remember Mozilla 2 and IE 2?). There were walled gardens everywhere. I launched Internet Graffiti because I noticed one simple thing: the world needed the web badly. Therefore, I thought, it will happen. Nothing could stop it. Walled gardens collapses, technology improves, standards emerge, speed goes up and cost goes down.

Now, it is just the same in mobile. Everybody is talking about Web 2.0, but they are missing the real revolution. Mobile is the next big thing. The web is just evolving and maturing. The Web was 1.0 in 1994. It is probably at version 7.0 now, following the versions of IE... Mobile is really moving into 2.0 now. And it will represent the biggest paradigm shift of this decade, as the web in the nineties.

Now, I have seen just one attempt to define what Mobile 2.0 means, from Dan (the other organizer of the conference). Although my friend Peter had the best comment during the conference: "It is Mobile 2.0 because we say so", let me try to give a concise summary of what I expect Mobile 2.0 to be.

My Mobile 2.0 Manifesto

1. Mobile 2.0 is NOT a mobile version of Web 2.0. We made that mistake before, please let's avoid it this time. WAP was a failure because the hypertext paradigm (with links and clicks) requires a mouse. No mouse in mobile. Forget the web. Let's look at mobile apps on devices, that are available when you turn the device on, that store local data, that react to push messages. Call them mobile widgets and have them use HTTP to communicate, or AJAX. But let's forget the browser paradigm or we'll be targeting a WAP 2.0 failure.

2. Mobile 2.0 is all about open standards and open platforms. Same as Web 1.0. It all happens when standards get into the mainstream. Let's forget ActiveSync, BlackBerry, Good and the like. Standards are here and will make this big. It is SyncML and others. They are on 800,000,000 phones today.

Mobile 2.0 is driven by open source. Open source is the center element of Mobile 2.0. Developers drive it. It is an unstoppable force. Look at what we are doing with OpenMoko and Mobile Linux. Look at Java ME going open source today or the announcement of Motorola a couple of weeks ago. We are pushing big companies to change and move towards open source. It is an unstoppable process.

4. Mobile 2.0 happens with flat fee billing. It is the same as Web 1.0, which exploded with flat fees after starting with modems. Same here. People using Mobile 2.0 will come from enterprises first (as in Web 1.0), then will move mainstream when flat fee billing will be available and cheap enough for everybody. Let's all push the carriers to understand this. They might risk to become a dumb pipe, but they will make money with it. And it is inevitable, with Wi-Max and the like coming into the picture. Better to move now than when it is too late.

5. Mobile 2.0 is centered on content and messaging. This one is easy. The phone is a device you react to, when content is pushed to you (your email, RSS feeds, soccer results, music, ...) or you push it to someone else (when you take a picture or a video, and you send it around). It is not about standalone games.

That's it, five rules I will follow working with our Funambol community. I know we'll represent a big element of Mobile 2.0 and I am excited about it. If we are at 1994 now, the next five years will be a lot of fun.

Wednesday, November 08, 2006

After election day...

Thanks to everyone who voted for this blog and made it in the Top 20 Wireless Industry Blogs ("that certainly represent the crème de la crème in industry coverage", yeah right). Being a Fierce Favorite "after thousands of votes" is nice, although I know I made it only because some of you hackers managed to write a script to circumvent the sum trick and voted a thousand times. Nice to have the hackers on my side. Let's all get together and crack the BlackBerry protocol now ;-)

Tuesday, November 07, 2006

OpenMoko: how you change the game in mobile

At the Open Source in Mobile conference, today Sean finally announced OpenMoko. He was the "mystery speaker" at the conference. Slashdot picked up the news, which usually means it is big.
It is a truly mobile open source phone, from the bottom up. As you can see from the press release, Funambol happens to be the thing powering the top of the stack, push email and PIM for starters, so I am superexcited.
We are not talking about an academic effort here. We are talking about the largest ODM on the planet (FIC, the Taiwanese company that makes the phone, a sister company of HTC) putting its weight behind a mobile open source phone. They can crank up millions of phones... And the largest open source community in mobile (that's us ;-) is already on it. I am expecting our developers to start building any sort of app on that phone, from enterprise verticals to consumer stuff. Super super super cool.
As the Inquirer wrote:
"This is the first phone in a long time to get us really interested in what it is, what it isn't, and the philosophy behind it. The philosophy is the thing that makes Linux great... it is really open."
The world is changing... Mobile Open Source is the way to go. Join the revolution.

Sunday, November 05, 2006

I am a Linus Fonero

A few days ago, it was Freedom Friday in San Francisco. The FON guys were giving away their routers for free. Good marketing stunt. At least, it attracted my attention... What is FON?

FON is the largest WiFi community in the world. Our members share their wireless Internet access at home and, in return, enjoy free WiFi wherever they find another Fonero’s Access Point.

It all started as a simple idea. Why should you pay for Internet access on the go when you have already paid for it at home? Exactly, you shouldn’t. So we decided to help create a community of people who get more out of their connection through sharing.

I love communities and this one looked interesting. I just need La Fonera router in my house, I offer Wi-Fi to people passing by (I got plenty of bandwidth I do not use, and you can limit the public one) and I get free access to any other Fonera on the planet (here you have the map, check your town, there are even five in Pavia, Italy). I have been preaching to my daughter about the value of sharing. THIS is sharing. And I have the feeling I am getting back way more than I give, which is my favorite sharing :-)

How do they make money? If you are just passing by and you do not own and share a Fonero router, you pay. Just 3 dollars per day. FON takes it, or shares with you as a Fonero owner. Yep, you can make money with this... There are three kind of Foneros: Alien (do not own a router, just suck your network paying), Linus (sharing the router and accessing every other router for free) and Bill (making money from sharing their router).

What do you guess did I choose between Linus and Bill?

Correct answer ;-) I bought the router online (for five dollars five), plugged in and I am good to go. I have two networks, one private and one public. I can even give free access to five friends (even if they are not Foneros).
Move fast, because the price will raise to 29.95 on Wednesday.

This is a nice community, backed by good money (sorry, I live in Silicon Valley, communities are cool but when Index Ventures and Sequoia are involved, they grow faster. See Skype, Google, YouTube and so on). It is a big play and very challenging, but if it takes off, it might be huge.

One issue for them to solve: how do you get access to the FON Maps when you are traveling and you do not have access to your computer (since you first need to find another Fonero...)? You need it on your mobile, possibly with GPS and directions on where the closest active one is. I am sure our community can help here... FONambol anyone?

Friday, November 03, 2006

Thin and Fat (clients)

An article caught my attention today. It talks about Jonathan Schwartz, Sun CEO. I already expressed my admiration for him in a previous post (and now that he added also when to if and how Java will be open source, I even like him more). In the article, one answer - in my opinion - stands out:
Schwartz panned thin clients as an end-all, be-all solution: "I just don't believe in thin clients. Right there, I said it." Even iPods are thick clients, he said.
It looks like a simple comment, but it is simply going against the entire world of Web 2.0 ;-) That is like saying "a web browser is only meant to browse the web". Unheard of. Blasphemy.

Well, I agree with Jonathan. I do not buy the idea of Web 2.0 around doing Word Processing, Email and stuff inside a browser. Maybe because, when I started Internet Graffiti in 1994, I was forced to develop sites thinking about the concept of the hypertext, the links and the Back and Forward buttons. The last three items are the basis for web browsing. If you run an app inside a browser and you screw everything up clicking on Back or Forward, why are you running it inside the browser?? Why don't you have a downloaded app on your PC, that updates automatically? Just because Google say so??

Ok, enough with the PC Web 2.0. When it comes to mobile (not just the iPod hardware as a thick client, but also the software inside the iPod) I just do not see the thin client paradigm work at all. I thought about Mobile Ajax, but I do not seem to get convinced about it. I want a mobile widget pre-installed (the music player). I want the data there when I turn it on (my music stored locally). I want an iPod (actually, I like the new shuffle, in case you do not know which present to buy me for Christmas).

T-shirt idea of the month:


BTW, if you happen to be at Mobile2.0 in San Francisco on Monday, I'll be there giving a demo of our Phone Sniper tool during the Launch Pad session ("showcasing cutting edge start-ups in the mobile space"). If you are around, give me a buzz.

Thursday, November 02, 2006

It is raining...

Today it is raining in Silicon Valley. For us living here, it is a major disappointment. We pay very high rents to stay here. We pay for the sun. When it rains, I just want my money back.

On a different note, we won another award today. It is raining awards, lately, but these are not disappointing. In particular, I like this one a lot. It is the Editor's Choice 2006, by the Linux Journal. The category is "Mobile Device".
Funambol isn't actually a mobile device, but we chose to give it the Editors' Choice if for no other reason than to avoid plugging the Nokia 770 yet again. Funambol is an open-source SyncML server that acts as a middleware between groupware servers and mobile devices. It supports the most popular PDAs and commodity mobile phones. It's great, and the community is finally coming up with a solution that rivals the best commercial competition.
What amazes me is the list of the other products there: Firefox, OpenOffice, Asterisk, Apache, Eclipse, Ubuntu, Postgres, Ruby, Quake (that's my favorite)... Not bad. I am starting to believe we are onto something here... Let the rain continue!

Tuesday, October 31, 2006

Tales from a user

I was browsing around sites and I found one comment titled "How to do over the air sync with your own server" that matches nicely my stance on ActiveSync and Microsoft. I thought you might enjoy it ;-)
Thanks to someone here, I took a look at funambol (an open source SyncML server with associated client plugins for Outlook, Smartphones etc).

I have to say that so far it RULES. I am going to delete active sync from my machines. Ding-dong the witch is dead !

Funambol just works (except for encryption, presumably a bug
in the version I'm using). It syncs notes, tasks, contacts and
calendar over TCP/IP. It will use either WiFi or CDMA on the
VX6700. You install a server (I have it on our linux mail server),
and clients for all your devices (I use two PCs with outlook,
an iPAQ PDA and the XV6700). Each client pushes and pulls
its changes to the server. They all stay in sync, all done over
the network.

This is exactly the product that Microsoft should have made
but did not. Activesync is the spawn of the devil by comparison.

This makes the VX6700 so much nicer to use because I can
keep it my pocket at all times (never forget it when I go
to a meeting, plugged into the cradle). But yet it can keep in

Sunday, October 29, 2006

Phone Sniper launched: "You give them cash???"

A few weeks ago, we launched a program called Code Sniper. The goal was to accelerate development of open source code from the community, giving a cash incentive (1,000 to 3,000 dollars). Last week we launched Phone Sniper. The goal here is to formalize the unbelievable amount of input we get every day about phones working with our platform and the development of "Synclets", small pieces of code that make a particular phone (with a certain firmware version, on a specific carrier on the planet) work with Funambol. If you do it and share it, we pay you 25 dollars.
Now the question: wait, are you giving them cash??? Isn't this open source, where everybody works for free, you love each other and just want to topple Micro$oft because it is evil?
Yes, this is open source. The source is open. Everybody benefits from it. We love each other because of this.
Nope, not everybody works for free. Some need food on the table for their kids. And we do not care about Microsoft, we are just building the best platform possible for mobile. Actually, we would love to work with Microsoft (although they might not like our open standard approach, because they might actually trying to be the evil empire afterall...)
With our honest dual licensing approach, we are not upselling to our community. We are building the best platform possible, together. The community is us. Many make money bundling the solution and giving it to their customers, with the professional services around it. Some do not, they are single individuals just helping for the sake of it. Funambol though keeps making money with the Carrier Edition, selling it to mobile operators. Our customers are giving us cash (not Playstations). Why should we keep it all for ourselves or give the community yet-another-gadget?
Therefore, yes, we are giving them cash. There is nothing evil in that. It helps us making cash. Some comes back. The software improves even more. We make more cash. More comes back. Is that simple. There is nothing dirty in cash. Cash is great.
And, by the way, it works phenomenally. Code Sniper attracted developers now building a Google Calendar and Address Book connector, a Yahoo one, a Symbian one, a DB4O integration and so on. Phone Sniper got to 50 phones in two days. Good luck to any proprietary company trying to do this. Open source rules.
In case you are interested to hear more, I wrote about Balancing Open Source and Commerce in the Enterprise Open Source Magazine which is out this week (I believe). It is a long article, I warn you. But it might give you the entire story of the tight-rope walk in a single piece. So you could stop follow my blog and save some time ;-)

Wednesday, October 25, 2006

Red Hat, Oracle and (again) dual licensing

A week ago I was writing about RedHat... Quoting myself (I feel can do it, since just a week passed, otherwise I would look like a old grumpy man ;-) "They [Red Hat] do not control it [Linux]. There is nothing they can do about it. They then take it and wrap it around as Red Hat Enterprise. It is still GPL (they have no chances, they must release it as GPL, they do not own the IP of Linux) and people can take it even for free and build an identical product (see Centos). The only way Red Hat can defend itself is by enforcing its copyright on the Red Hat logo (and they are really aggressive on it, rightly so). Pretty weak but it seems to be working well for them - for now."

Well, pretty weak it was a week ago... Working well - for now - a week ago...
Today Centos became Oracle... Red Hat price collapsed (their market cap is half of what it was six months ago). Oracle has a bigger support organization than Red Hat... They will support the same exact identical precise code. At half the price (!). Now it is very weak and not working well... That's scary stuff if you are an investor in Red Hat.

Honestly, I am not convinced Oracle can really do it. Taking a product from someone else and support it is not an easy task (although this is what Red Hat is doing with Linux...). My bet is that Mr. Ellison is going to let Red Hat stock price slip a little bit more, than buy it with another hostile takeover. That will make him quasi-Microsoft, which has always been his goal. And he will buy JBoss in it, which would be a nice revenge.

Anyway, my point is that - once again - dual licensing is the only viable option to build a defensible business with open source. Anybody can take Funambol Community Edition, but I can give mobile operators my Carrier Edition and be sure that nobody will hijack my business. The mobile operators will get the source code, not my competitors. It is licensing, not just support.

As Marten once said (I guess it was him, I have a veneration for Marten so anything smart must come from him), my mobile operators want the source code of my product as much as they want the airbag in their car. It must be there (who would buy a car without airbag?), but they hope they will never use it. What's the point to give them the instructions to build the car? Just give them the airbag, which is what they want. And your competitor will not benefit from the instructions to build the car...

Let me just be clear. That does NOT apply for the Community Edition. There, the source code is everything. We are building the city car together, the one that everybody needs, cheap but reliable. On top of it, we are taking that design and we are building a race car for the mobile operators. That's what they need. They race cars. They do not build them. They do not care about building them. They want the fastest, most affordable, most secure, most reliable race car. They will pay top dollars for it. Our community is different from our customers. That's my honest dual licensing. Not weak and working well (hopefully for a while ;-)

Walking the tight-rope walk in open source is everything. Building a defensible business model is the only way to get open source companies to the famous billion dollars in revenues. Being disruptive is cool, but being smartly disruptive is better. That is, I guess, what Mr. Ellison is all about.

Sunday, October 22, 2006

Microsoft to dominate mobile? They do not think so

I was looking at an article today titled "Microsoft sees fast growth in Windows phones: report" and I started to wonder a bit about "fast growth" and control of the market by Microsoft.

Some people - just the ones that do not know/understand mobile ;-) - are telling me that Microsoft will dominate this market as they dominated the PC market (BTW, I just bought the third Mac laptop for my employees because they asked for it and I could not find one single reason to say no...).

I do not think Microsoft will dominate mobile, sorry. They do not think it either...

Check the numbers: they had 6M devices last year with Windows Mobile. They plan to grow 100% this year. It looks like a big number... That's 12M devices. Maybe they will do 100% next year and the following one. That's 48M devices at the end of 2008. Wow.

In Q3 of this year, there were 245M phones shipped worldwide. We'll have 1B (one billion) phones shipped in 2006, still growing in the next two years.

If Microsoft is right and they make their numbers, they will have about a whopping 3% of the mobile market by the end of 2008. Apple is at 7% today on the PC market... Is Apple dominating the PC market??

Friday, October 20, 2006

Mobile TV and the inevitable lack of bandwidth

As you might remember, I am now a fan of mobile TV. No, I do not believe I will spend hours watching TV on my phone, but I feel I occasionally can spend few minutes watching something I taped before on my TiVo (if it is free). My SlingBox has been helpful for my father too, when we were visiting my younger brother in Washington DC and he could watch live soccer while I was driving from the airport (until my daughter at home started watching Sesame Street, but that's a different story...).
Two stories crossed my path today. The first one: "Video-hungry users could push Net to brink: Nortel". The second one: "TV on Mobile Phones: Heat or Hype?". They made me think (out loud, as usual).
If the net, that has been around for ages, cannot support the YouTube crowd, what makes we think Mobile TV will be different? The bandwidth is scarce... So scarce that, as soon as four people in my network cell have a SlingPlayer going on their mobile phone, the data traffic collapses (and I have EVDO). Unfortunately, the success of mobile TV might simply kill it. And I do not think networks will have more bandwidth that fast. That puts me in the Hype category, for now... I am grateful to be in the low bandwidth world of push email and synchronization ;-)
Please, do not get a SlingBox and, mostly, try to stay away from my network cell. I want to be the only one watching TV on my mobile phone, as long as I can. Thank you.

Thursday, October 19, 2006

I want to thank my mom again (but I need you this time)

Ok, you win something and you get greedy. It happens. I want to win the World Cup again, for example, but I have to wait...
I just found out my blog has been nominated in the top 50 by Fierce Wireless "based on editorial quality, site usability and penetration among wireless industry executives". Yeah, right.
Anyway, thanks for reading this. If you want to vote for this blog you can click on the icon on the right, chose "Mobile Open Source" and do your math (that is quite funny, BTW). I do not think the rules allow me to give away t-shirts to everybody voting for me, sorry. But I am Italian, therefore flexible by nature :-))

Wednesday, October 18, 2006

I want to thank my mom...

Funambol won big at the Mobile Star Awards this week. We won Gold Star (that's the #1 prize ;-) for both our consumer email and our PIM solution. That's very nice and much appreciated. Thanks to all the people that voted for us. And all the crazy ones out there that voted for me and made me a Bright visionary in consumer mobile software. You should not have but I will definitely take it. To show my appreciation, I will spare you the acceptance speech :-)

Tuesday, October 17, 2006

MySQL, dual licensing and Red Hat

As you might have noticed, MySQL today announced a new offering. In a nutshell, they now have a Community Edition and an Enterprise Edition. The first is free and has external contributions. The second is something you pay for and fully owned by MySQL. That looks a lot like Red Hat, but - in reality - it is still really different. I have been exchanging emails with Marten in the last few days about the topic and I am fully convinced this is a smart move. It actually creates a model which is an improvement over Red Hat, in my opinion.
On the surface, it looks like MySQL is moving into the Red Hat model. The difference is that the source of Red Hat is Linux. They do not control it. There is nothing they can do about it. They then take it and wrap it around as Red Hat Enterprise. It is still GPL (they have no chances, they must release it as GPL, they do not own the IP of Linux) and people can take it even for free and build an identical product (see Centos). The only way Red Hat can defend itself is by enforcing its copyright on the Red Hat logo (and they are really aggressive on it, rightly so). Pretty weak but it seems to be working well for them - for now.
In July, I wrote a post about my way of doing dual licensing. That is working nicely for Funambol, with our Community Edition and our Carrier Edition. Nice separation of church and state. Everybody is happy.
MySQL is doing something similar. MySQL controls and owns its code. They are now opening the Community Edition to the world. They will still control that, though. And decide what goes into the Enterprise edition. That is the product they will sell. They can license it, not just get subscription for it. This can still generate revenues of a different scale (I am still a fan of software licensing and its margins, sorry), way beyond Red Hat (which is a nice business, do not get me wrong). It is a defensible strategy with full control of their future (they can change it when and if they want). The risk is that the move might screw up their ability to do dual licensing, if they are not careful. They won't control the full IP anymore on the Community Edition, so they will have to cherry pick for the Enterprise Edition (or buy out developers, but it did not work out that well with InnoDB...). It is called tight-rope walking (or funambulism), but Marten and Zach are two of the smartest people in the open source world and I know they will make it happen, no matter what.
Go dual licensing. Go MySQL. We just want to see you public before the end of 2007 ;-)

Saturday, October 14, 2006

The Holy Grail of Synchronization

Engtech has written a very popular post on his blog, called The Holy Grail of Synchronization: How to synchronize Microsoft Outlook (multiple locations), Google Calendar, Gmail, iPod, and mobile phone with Funambol / ScheduleWorld. If you are using Funambol - or thinking about using Funambol - I would definitely recommend it.

Saturday, October 07, 2006

What's on my Treo 650

One of Russ best post I remember was his "What's on my Nokia". I think I got more applications to download for my Nokia from that post than from any possible visit to any off-deck site or any Google search.

After so many years spent using a Palm OS, I am about to give up my Treo 650 for good. At Funambol, I had to make the sad choice to bet on different operating systems (as did PalmSource, now moving into my own space with
Access Linux Platform --> that's mobile open source ;-) In all these years, I collected a number of applications. I used some for few days and a small subset every day. Those few represent for me the minimum toolkit for anybody using a Palm OS device. Here they are.

1. A PIM sync client
First of all, a smartphone without PIM sync is useless. You need your contacts, calendar, task and notes from your standard email reader (mine is Outlook). There are many choices in this space. First of all, you can sync over cable or bluetooth. That's suboptimal if you have a Treo, but the only choice if you have an old PDA (what for? Get a smartphone! ;-). I have been using Chapura PocketMirror Professional and it never let me down. I use subfolders in Outlook to separate contacts and it worked great on that. If you want over-the-air (OTA) sync, SyncML is your best choice. I know I am biased. You should use the Funambol free portal and a SyncML client. There are many out there. Beside ours, I would recommend the Synthesis SyncML Palm Client.

2. A mobile email client
Once you have synched your address book, you are ready to send emails. As I wrote before, do not attempt to use mobile email without an address book or you will give up fast. Once again, there are many email clients for Palm. Some feel Versamail (which is usually preinstalled) is good enough. I don't. SnapperMail does it for me. I like how you can configure how to get mail. When I am in the US and I have a flat data plan, I do it every 20 mins and I download the entire message (but no attachments). When I am in Europe, I do it manually, I download just 12k and I set "disconnect afterwards". I save money. I like the ability to clean up my folder at night, since my email would fill up my Treo in three days. IMAP support is nice and support for multiple accounts is key for me.

3. A directory assistant
I have to thank Matt for this, since I can't live without this application. Alone, it gives me a reason to defend my Treo with my wife (who is always about to throw it in the toilet, since it takes a bit too much of my attention). Directory Assistant is a beautiful tool, that allows you to search for a restaurant or a shop (or even one of your friends, since it searches for Residential numbers as well), get the result via the web (fast), call with one click and reserve your table or get directions. It is a beautiful simple application, which uses data from the Yellow Pages web site. Thanks Rick Whitt for saving my day many times.

4. A converter
Ok, this might look a bit too high in the list, but I am a European and I will never ever get used to inches and feet, or fluid ounces. Who is the genius that invented the US standards? How can you create a system where there are a bunch of inches in a feet, a bunch of feet in a yard, a bunch of yards in a mile? None with the same metric? How can you possibly know how long is 3 inches and 5 eights? Sorry for the rant... Kilos, meters and liters are so easy to understand that it kills me on a daily basis (and BTW, zero is freezing, 100 is boiling. It is that simple, I swear). Anyway, I use Matt Marsh's Converter. Easy, simple, free. You can customize it for the units you use more often. This one saved my brain many times, at Home Depot in particular. On a similar note, sometimes I use a stupid app I wrote long time ago, called Spell It!, which converts a word to its NATO alphabet representation. That is: Fabrizio to foxtrot-alfa-bravo-... If you have my first and last name and live in the US, you might see why it could be helpful, in particular opening accounts on the phone.

5. A mobile phone calls redirector
Most likely another one that would not make the top list for many normal people. But I do not like to speak on my cell phone when I am home or in the office. I have a normal phone there (cordless, btw). The quality of the line is much better and it does not get hot while I talk. I use Call Director because you can set phone numbers (such as Home and Office) and it gets activated automatically when you plug your Treo. I come back home, plug the Treo and my mobile calls are redirected to my home phone. I unplug it to go out and my mobile calls are back on my mobile. You can even set times for office vs. home vs. anything else. When I am flying outside the US, I forward it to my SkypeIn. I know, I am weird.

6. An instant messaging client
I know, this should be much higher in the list but I do not use it as often as the converter... VeriChat is a fantastic tool. It supports Yahoo!, ICQ, MSN & AOL chat (I am on YM). It works like a charm. It allows you to transfer files (I used to send pictures taken on the phone in real time). It is full of Bots, small apps that can get interesting data for you, like weather forecast, stocks and so on. What I call mobile widgets. If I could choose, I would not put them inside an IM client, but on the top menu of my device. However, the idea is the same and it works. Highly recommended.

7. A tool to receive up-to-date sport results
If you do not like soccer, you can substitute it with baseball or whatever sport you like. There are few important things in life: sport is one of them. When four years ago I had to choose the first application to use the Funambol server on, I chose the World Cup (there is only one, it is soccer, it is the most viewed event in history, stop asking "of what?"). We had 20,000 downloads in two weeks. Not because I wrote the Palm client, but because the sync capability was a killer. This year (man, what a year!!!) we left someone else write it and Tiny Stocks got money from Palm to do it. Football 2006 (I guess they could not say "World Cup" or FIFA would have called them ;-) was just perfect. Over-the-air updates. Dynamic charts. I would love to have the same for the Italian league, but this year I am on soccer sabbatical. Some biased judge took soccer away from me. I am going to watch the World Cup games over and over, for 12 months. No soccer for me. Just Formula 1 and MotoGP.

8. Some games
I never have time to play games because I am a CEO of a startup and I work very hard (yeah, right). However, sometimes during my long travels for work (yeah, right) I have some spare time and I fill a very limited amount of minutes with games. I am not big on mobile games (this is true) because I feel the limitation of the keyboard and the screen. However, there are some classics that I always have on my Palm, to kill the minutes before boarding on a plane. Minesweeper, tetris, chess, backgammon, pac-man and galaga make my short list. If I just could have some more spare time... Yeah, right.

9. Some random geeky tools
There are some tools on Palm I could not live without. One is FileZ, an open source application to manage files and databases on your Palm. It is a must, if you know what you are trying to accomplish. If you don't, just do not try or you are going to screw up your Palm forever. Another must for me (I would say until I hired a Director Operations) was TuSSH, an SSH client. I was able to jump on any of our machines from my Treo and fix stuff. Not something you do every day but you are really thankful that moment when you really need it (and again, your wife will be chasing you trying to throw the damn thing in the toilet).

10. A way to use your phone as a modem
This should be very easy to do on a smartphone. It has bluetooth, so does your computer. It connects to the Internet, so you would think you can do it easily from your computer, pairing the two. Well, it ain't easy. But you can do it. I spent a few hours to find the right combination to use my Treo with Tmobile (in the US) as a modem for my laptop. I could not find anything online but I managed to make it work (btw, the phone number to dial is *99***1#) and I use my 4.99/month Tzone to download email on my laptop with the phone in my pocket (magic!). If you do not want to spend your life trying the right combination, I suggest PdaNet. No effort, it works really well. Your mobile operator might not like it, but do you really care?

That's it folks. I probably tried one hundred more but they did not manage to stick around and I cannot find them on my Palm anymore. There is usually a reason for that. As there is a reason for me letting Palm OS go. It breaks my heart, but when Palm runs Windows Mobile, Macs have Intel chips and Microsoft shares its source, you have to accept the world is changing and you should not look behind.

Tuesday, October 03, 2006

Mobile Messaging Webinar tomorrow

I am speaking at a mobile messaging webinar tomorrow (Oct 4th at 8:30 am Pacific).
The topic is quite interesting (if you are a geek :-) : "LEMONADE and SyncML: Key Standards for Mobile Messaging".
Questions Addressed by This Webinar Include:
 - What are LEMONADE and SyncML?
 - Who will use mobile messaging based on these protocols?
 - What are the advantages and disadvantages of the technologies vs. RIM and Windows Mobile?
It is FREE as in beer. So... you should not miss it.

Monday, October 02, 2006

Release by aging

After a good seven months as a beta, Funambol v3 finally last week became General Availability. It is a nice milestone for Funambol and we are all very proud of it. Now the question: isn't that too long? Why did it take seven months?
That's a good question, thanks for asking ;-) The answer lies with the open source nature of our core code.
If you are a proprietary old boring company, you would have a product in beta, give it to a few customers who will give you a very limited feedback (close to zero), then a month later your marketing folks will force you to slap a GA tag on it and you will be shipping and praying your real customers will not find too many bugs. Inevitably - if you ever get customers - they will find a ton of bugs, you will scramble to build version x.z.1 and .2 and .3 and so on, until every customer will be happy (or gone for good). That will take you months and months, while you will not be able to build your next release and your product roadmap will keep slipping.
Open source is different. You throw the code out and you wait for feedback. You get a ton of feedback, sometimes beyond your ability to handle it. Thankfully, many of your beta testers fix the bugs they find themselves (that would be nice, Mr. Proprietary Company, uh?). You are able to release only when the community will stop finding and fixing bugs. There is no marketing GA tag in open source. It is community driven.
I call it release by aging: you leave the product out until it is mature. You cannot force it. It happens by itself. One day you discover that the amount of P1 bugs is dying off, because your community and your internal QA are not finding any. They are all gone, for good. And your customers are still there because they were part of the process.
For Funambol v3, we had over 300,000 downloads between the beta and the GA. That's a heck of a beta testing group. Maybe Microsoft can compete (maybe). Nobody else get close. Nobody else will ever have a product with the same quality that open source can provide. Period. It is the people factor. We are all working together and, as a community, we represent more QA people than any proprietary company can ever dream of. In mobile, that means we can test every phone on the planet (one billion today, going for two billion fast). That we can really address the mass market everybody is talking about. Nobody else can (sorry).
Release by aging. The key for high quality products, happier customers and healthier balance sheets.

Thursday, September 21, 2006


The Mobile Email conference in London has been great. I should have brought the entire company and a chunk of our 600k downloads there. Lots to learn. I will post about it later.

I landed in Italy last night. I am speaking at "
Congresso AICA 2006" today. Very interesting event (for Italian people only...) where we will talk about research, being competitive in the market and high tech. They even have a track on open source ("risk and opportunities of open source"), which tells me something. The Italian Minister of Innovation will be there, so I better put on a tie for once :-)

Sunday, September 17, 2006


I am flying to old Europe tonight. I will be speaking at the Informa Telecoms & Media Mobile E-mail 2006 in London on Wednesday at 14:45 (going with the European standard here ;-)

The topic is "Responding to Consumers Anxieties and Requirements in Order to Accelerate Mass Market Adoption". Are consumers really anxious about having mobile email on their devices?? I do not think so... I believe they are anxiously waiting for an easy-to-use low-cost technology... Anyway, I guess we'll find it out soon.

As usual, if you live and/or are around London, just give me a buzz.

Saturday, September 16, 2006

HPL version 1.0 is here

A month ago, I proposed a new license, called the Honest Public License (HPL). It is GPL v2 with an additional paragraph, clarifying that distribution of software as a service is exactly the same as distribution of software in a floppy. Therefore, GPL applies also when the software is used to provide a service to the public.

I left the license out for review for a month. I received a ton of comments on this blog and in many blogs and articles around the net. Evidently, I touched a nerve ;-) The feedback has been universally positive (and I have to admit I have been quite surprised). The only real negative comment I heard has been "yet another license". It is true, it is yet another license. But it changes the most important element of computing that was not covered in GPL, because it was written in 1991 when software was distributed in floppies and nobody knew what SaaS meant. For everything else, GPL is just fine. SaaS is the future of computing: HPL is just another incredibly important license ;-)

Anyway, here you find HPL 1.0 and here you find the diff between HPL and GPL v2. As I wrote a month ago, our next step is to have HPL disappear within GPL v3. That process is ongoing and I have no control on, so let's wait and see.

Regarding Funambol, we have decided to leave the clients on GPL v2 (no reason to change them to HPL, since there is no SaaS on clients). HPL will clearly end up in the incompatibility list of GPL, as AGPL before it (a license that tried to fix the same issue, but in a pretty "strong" way), so we would prefer people to be able to link clients based on GPL code. On the server side, to build SyncSources, you will have to use HPL. HPL affects every SyncSource, but it does not affect the product you are connecting the SyncSource to (if you talk to it with a separate protocol). Therefore, the impact should be minimal. Because of dual licensing, Funambol could also grant a special waiver for open source projects that cannot switch the SyncSource to HPL. Just give us a buzz. More to come in the new site after the launch of v 3.0 GA (very soon!).

I want to thank Patrick, Markus, Tomasz and Kari for the precious help in the review process. I just do not know what I would do without a community of people working on the project (wait, I know, I would be doing something else like opening a restaurant in Maui, probably called Funambol Italian Restaurant. THAT would be fun :-)

Sunday, September 10, 2006

CTIA - Los Angeles

For those that might be going there, I am flying tomorrow to CTIA in Los Angeles. I will be speaking at the SmartPhone Summit on Monday at 4 pm, track Mobile Messaging and Gaming. I'll be around on Tuesday, flying back on Tuesday night. My schedule is pretty much overbooked, but if you are around and you want to say hi, just drop me a note and I'll find some time. See you in the city of angels.

Saturday, September 09, 2006

The road to push email goes through address book sync

It is funny how small things sometimes prevent big changes from happening. There is one that prevents the potential explosion of push email on mobile devices: the lack of address book synchronization.
If you have a cell phone, you have received and sent at least a message in your life. A short text message or SMS, but still a message. You are ready for push email. Today. You are ready to receive messages a bit longer pushed to your device and to send short messages (no, I am not a believer of a long messages typed with a keyboard installed on every phone on the planet, sorry. I like dumb phones because they look cooler and - guess what - the phone is mostly a fashion item, trust an Italian on that).
When you send a SMS today, you send it to a phone number. That's what you have in your phone address book today.
If you want to send an email, you have to send it to an email address. Would you type on your cell phone? No. Believe someone with a long first name and last name... You will not do it.
So... you will not use push email unless you have my email address on your phone. Since you are not going to type it, it must get on your phone from somewhere else (your Outlook, your Yahoo address book, your Skype address book and so on). You need address book synchronization, possibly pushed on your device when something changes.
Without address book sync, there is no push email. It is a small thing, but it is small blocking thing.
Thankfully, SyncML solves that, if paired with the right solution on the server side that actually works with your phone (warning: small advertisement here ;-)
The road to push email goes through address book sync.

Thursday, September 07, 2006

The Community Code Sniper Program

Jason, our smart and creative Community Manager, just launched a new great program, called Community Code Sniper.
The idea is pretty simple: we asked the community which clients or connectors they wanted with a survey, then we looked at our resources and we realized we could do only some internally. Others would be left out waiting for us to get more resources. Or we could ask the community to contribute to the effort. The tightrope walker in us suggested we should contribute back with cash, giving back to the community something we received because of them (thanks). All the code developed with the bounties will go back in open source for everyone to share, which will be very cool.
Some people picked up the news, and I really liked Dana blog on ZDNet (in particular, because Funambol comes before Red Hat :-)
Anyway, this is a great opportunity for all the people out there that are sending us resumes: if you think you are good, just prove it. You will even make money. And it will be fun. Having your address book synched between your Nokia or RAZR phone, your iPod, your Gmail or Yahoo address, Skype will be very interesting for millions of people (not just you). If you have a Mac, just take our client API and build an iSync plug-in, so we finally can sync our cell phones over-the-air with our Mac (and yes, I believe Steve Jobs will announce the iPod phone on Tuesday...).
(yep, I know, we are having fun at Funambol... "Code Sniper"... What a name... It shows we have a new VP Marketing :-))