The Zune is the future of mobility

You are the largest software company in the world. You have built a device that nobody is buying. You feel the next big thing will be in the smartphone market (hello!? anybody home?). What do you do?

You take that device and you make it a smartphone, of course!

Ok, the news about Microsoft taking the Zune and making it a smartphone are too funny to be true.

However, I said it before and I repeat it. I believe Microsoft needs a smartphone to compete with the rest of the world. The dream of replicating the PC world and have everyone in the world use Windows Mobile is gone. It did not happen. The opposite is happening.

In mobile, it is all about user experience. If you control the hardware and the operating system, you have a chance to build a great experience. If you don't, it is tough. See all the bad HW implementation of Windows Mobile. Or the G1 story (good OS, bad HW).

The good stories are all coming from the combo OS+HW: Apple with the iPhone, RIM with its BlackBerries and so on.

Therefore, I am in favor of a Microsoft-branded smartphone (although I do not believe they will have one ready for MWC in Barcelona, no way).

But please, do not call it Zune...

iPhone 2.2 and the Google envy

I updated my iPhone to 2.2 during the weekend (still a laborious task, if you want to keep it unlocked and jailbroken, but worth the effort) and I found the new update quite nice. Not that much that I would use, but the Street View of Google Maps is spectacular (although the compass feature of the G1 is even more spectacular).

The thing that caught my attention a lot more, though, was the voice search application launched last Monday by Google. You raise the phone to your ear, it beeps, you say what you are looking for, it beeps again and it gives you the result. No finger needed. Quite awesome.

I tried saying Funambol as a Northern Italian would (yep, there are at least two different ways to say Funambol even in Italy) and the results were hilarious. From football (hey, do the people at Google know everything about me?) to Lambo (as in Lamborghini). Nice.

I tried saying Funambol as a Texan would (not that I know how a Texan speaks but...) and it came out perfectly right. Amazing. Where did they find anyone saying Funambol to match it? A Texan-based robot, I would guess.

If you recall, the app was announced with much fanfare the week before but it took a few days to appear on the App Store. Theories abound but I have my take on this one for a simple reason: I envy Google. Badly.

Why?

Because the Google voice search app uses a call to an API that is not publicly available... The app starts recording your voice when you tilt the phone and put close to your ear. The tilt is in the official API, but the "close to your ear" depends on the proximity sensor (the one that turns the screen off when you talk). You cannot control the proximity sensor with the published Apple SDK. No way.

Google uses an unpublished call. They submitted it to the App Store and Apple let them go through, even if they always said they would not allow any application in the App Store that would access unpublished APIs...

Beside explaining why the publication took a few days (Schmidt calling Jobs and saying pleeeease? Jobs calling Schmidt and saying whaaaaaat, show me the money!), it is still a reason for most of us to be upset.

Our community wants to have access to the iPhone calendar badly. But we can't. Even if the version for the jailbroken phones has been mostly developed by the community, I won't ever try to publish it to the App Store. I can't risk Apple to kick our contact sync app out, just because they decide to do so... And I am sure over-the-air calendar sync is an important feature for most iPhone users (who might not want to use MobileMe).

Well, the end result is just that I envy Google. What a novelty.

P.S. I am sure you noticed that the iPhone 2.2. upgrade was made available by Apple the day RIM launched the Storm... What a coincidence! Apparently, Jobs was scared about the Storm and wanted to make sure people remembered that the iPhone is getting better by the day. No need to change, says Apple. I have to agree on this one: if you have a BlackBerry and you want to upgrade, get a Bold... If you do not have a BlackBerry, well, you got a lot of choices these days and the Storm is probably not the best one. Nonetheless, Apple felt the need to fire back, which is a signal: they were worried. That said, I am ready to bet the iPhone will win the Christmas season for smartphones, hands down.

Sorry, the Storm is no iPhone

I had very high hopes for the BlackBerry Storm. I had the opportunity to play with one this morning, thanks to one of our great employees (and her husband) who stayed in line since 5 am today...

Unfortunately, I was disappointed. That happens when you have high expectations, I know. But I had super high expectations for the iPhone and it flew easily beyond them...

Simply, the key feature - the screen with tactile feedback - is just bad. That is the one that should have been its flagship one, the better-than-the-iPhone one... However, it is not convincing at all.

First, you have visual feedback (box around the key turning blue) even when you do not click. So you think you clicked but you actually did not. Why they did not build the device to have the box turn blue when you actually click is beyond me...

Second, it take physical effort to press the key. You have to press the entire screen. It is heavy. I typed for a minute and my muscles were sore (ok, I am exaggerating, I am in good shape, I can handle a good five minutes of it :-) It should be mandatory for 24 hours fitness centers: train your thumbs...

Thanks, but I want my BlackBerry keyboard back.

On top of it, the scrolling is not smooth. If you accelerate the scrolling, like you would do on the iPhone, nothing changes. It always scrolls at the same speed. I do not expect the nice visual effect of bouncing at the end (although it is one of those that wow people on the iPhone), but it takes forever to scroll.

Thanks, but I want my BlackBerry navigation ball (or even the wheel) back.

When it comes to switching between portrait and landscape, the Storm is slow. A fraction of a second too slow. But still too slow.

Thanks, but I want my old BlackBerry back (or I would take an iPhone for it)

Oh, what about everything else? It looks good... but you can't go for the kill, miss it completely and then say "well, look at the rest". It just does not cut it.

BTW, for those interested, the Funambol sync client appears to work, but the email client needs a few touches. Too bad, but you know we'll get them working before you say "what?" ;-)

AOL is live with open source sync!

We made the announcement of AOL choosing Funambol a few months ago. After a friends and family test, they are now out officially in beta.

For us, this is a great achievement because the Carrier Edition is now deployed at AOL. Talk about scalability -- our mobile open source software can now benefit all AOL users, and you know they have a lot!

To set up AOL Sync:

1) Go to http://aolsync.aol.com/.
2) Click "Register your phone".
3) Enter your mobile number, mobile carrier and device manufacturer and click "Send Code".
4) When the verification code arrives on your mobile phone, enter it and click "Verify".
5) Select your phone model and click "Continue".
6) To sync additional calendars or auto added contacts, check them. Click "Configure Now".
7) For iPhone, Windows Mobile and BlackBerry devices, the next step is to download a sync client. For other devices, you'll receive instructions on how to configure your phone's built-in sync client.

Here's a YouTube screencam about AOL Sync powered by Funambol:



To read more about AOL Sync, check out the post on the AOL website.

Stay tuned -- this is just the beta. We'll keep you posted on its progress.

Is RIM trying too hard to be a consumer company?

I wrote before about RIM trying to be a consumer company, how hard it is and how risky it might be for them. Exactly like Nokia trying to be an enterprise company, attempt which failed miserably.

My analysis came from the way RIM was presenting the Storm. Much more as a consumer device than as an enterprise device.

I wrote (sorry for quoting myself, it looks cheesy):

Where can they shine? In the enterprise. They still have 98% of the enterprise to capture. They are entrenched. They have to make sure nobody gets in. Like the iPhone. Or the Android phones. Or the Windows Mobile ones. Or Nokia.

I feel the Storm is a smart move in this direction. You have to give enterprise people a little of coolness (music, pictures, some social networking), being strong on your basic values (email), and improve on the competition.

Today, I read an article on The Register about the Storm launch in the UK. In a nutshell, if you are a consumer and you buy the Storm, everything works. If you are an enterprise and you need Exchange support, you are screwed. Not only it costs more (about $25 more), but you simply cannot get it right now.

Quoting The Register (much less cheesy):

"The BES service books for the Storm are not quite ready ... chances are it still wouldn't work. Obviously we'd rather not take the chance just yet until we know everything is working 100 per cent on that side."

All this is part of a "planned, phased rollout", and Vodafone tells us that the Enterprise-ready version of the Storm is coming over the next few weeks.

What is clear is that RIM is trying to launch the Storm as a consumer device. Period. Even taking the risk of pissing off their core customers. I think it is a big mistake on their part. Chasing the iPhone is a good idea, but upsetting your core target is a very bad one.

And, by the way, if you are an Enterprise buying a Storm and you want to save the $25/month, take a look at the Funambol Community Edition. It is free free free ($0/month), it has an Exchange connector and a BlackBerry client. You can buy the Storm as a consumer and use it in the enterprise.

After all, RIM is pushing you to do it...

An Italian software story

I know the blog is about mobile and open source, but I bumped into this story of an Italian developer (on the New York Times...) and I could not avoid to write about it ;-)

The story is about Giacomo 'Peldi' Guilizzoni, a former Senior Software Engineering Lead at Adobe in San Francisco, who moved back to Italy and started a company called Balsamiq.

His company is actually a "Micro-ISV" or a single-employee company. That is, a guy in a studio. A guy who wrote a product called "Balsamiq Mockups" which allows anyone to build a mockup of a GUI: the genius is that the mockup resembles one you would scribble on a piece of paper, rather than a computer one. It sells online for $79.

To make the story short, the product is fantastic and he grossed over $100k in five months. And the business is growing fast.

Why do I know? Because he made it all public in his blog. He promised he would do it. And you have all the measures of his business (and how much he is donating to non-profits).

The story is great because is the counter story of all Silicon Valley VC-backed startups. Get money from VC and you have a chance to be rich quick... Peldi is doing everything the opposite way, and he is becoming rich quick. A one-man shop. The anti-VC story, much appropriate for an economic downturn. With software built with creativity in Italy. By a guy who lived in Silicon Valley. I have to love this one!

I do not know Peldi, but I hope I will have the opportunity to meet him one day. I am sure he is having a great time (and food) in Bologna, but I feel the pain he will experience when the company grows. He has great advisers and he is going to do just fine.

An Italian software story. Just a reminder on how smart Italian engineers are (ok, I am biased - I know - but do not tell me you did not like it ;-)

It IS a garage door opener!

Nice to be right, once in a while. I received many comments on my first post about the G1, when I claimed it was a garage door opener (actually, my wife did...). Some people said I was too harsh on T-Mobile and Google.

Today, I was vindicated. IntoMobile reports the G1 is actually a garage door opener. And it is even open source!

Change we can't?

There is an article on The New York Times that talks about our new President-elect Barack Obama and his addiction to mobile email. He is the first candidate for president (worldwide, I have to say) that was going around always with a BlackBerry in his hands. A tremendous slogan for those who live on mobile email :-)

Now, the security services are saying that he has to relinquish his BlackBerry when entering the White House. No more mobile email for him. BlackBerry is not secure enough.

I have seen this debate before: BlackBerry can't be used by politicians in office. Period. All emails go through RIM servers in Canada. As much as they tell us it is all secure and encrypted, they are a corporation and no government in its right mind would trust them. You just never know. And they are our competitors!! :-))

Jokes apart, how can our next President keep in touch with the rest of the world? How can he have a reliable secure controlled untouchable mobile email system that nobody can hack?

Ok, you know where I am going... Open Source. Open Source is the answer.

First, you install Funambol in the White House (not in a corporation).

Second, you have a bunch of engineers look at every single line of code to make sure those evil Italians did not put any dangerous code to steal vital information: look for

if (user == "Barack Obama") send_email_to("Silvio Berlusconi")

Third, once the code is certified clean, apply your super secret security patch and you are good to go.

As a device to carry, may I suggest OpenMoko? Fully open source, including the device and its design.

Barack, we want you back in the mobile email world as an ambassador for mobile email in the world. You need it. You are addicted to it. You can't live without it. We gain from it.

Dump your BlackBerry, go for Open Source. You can have mobile email in the White House, yes you can.

Android for carriers

The conventional wisdom is that with Android you get Google. Therefore, if you are a mobile operator and you are looking at Android, you must take it all. The G1 comes with Google, you need to put in a Gmail login to turn it on. Android is Google. As a mobile operator, you must kill any hope to control your users and give them all to Google.

Well, that is not true. Android and Google look intertwined but it is a lot easier to take them apart than most people think. Actually, it is quite easy...

First, the OS is totally available and it is based on the Apache license. If you take that code and run away, you do not have to return anything back to the community. Not my preferred license, but... that's what Google chose. Good for them I guess.

Second, the Google part in the OS is really the Gmail client and the sync. Nothing else. Gmail is a visible app, and easy to remove (if you really want to). The sync is a bit deeper, but again, fairly easy to unplug. You just have to put something else there: email and PIM sync (guess who can provide you with that? ;-)

It is very easy, at least, for a company who has decided to package Android as an OS and sell it to mobile operators. A Red Hat of mobile. I am expecting a bunch of them to appear and grow fast. I met one yesterday...

What do you get if you are a mobile operator? A full phone, you can brand. Totally brand. From the HW to the SW, including the look and the applications on it. And you can put together a service to provide a MobileMe-like solution for your users. The full package, without Google.

So why not LIMO or any other Linux variation? Where is the difference? Why Android?

Well, it is the applications. How many apps are out there for Android? Plenty. I mean, a lot... Check the Android Market. It is unbelievable. Developers drive applications. And they go where there are deployments. Deployments go where there are applications and developers. It is a positive loop.

Are there any LIMO apps? Nope... Will developers build for it? It depends, let's see if there are deployments. No deployments? No apps. No apps? No chance to make it a success... Google broke the cycle pushing apps even before the first phone came out. Their brand did the trick, now it is a snowball effect. Not for LIMO (yet).

If you are a mobile operator, do not look far. Take Android, strip out Google (or leave it somewhere, but not make it a Google phone and kill your brand) and have your own branded phone with a branded experience on the service side. Allow people to download their applications and customize their experience.

You need a couple of companies to make it happen (do not do it yourself, that is not your business...) and you have a BIG winner.

The source code matters more in a downturn

Today I was in a meeting with a future customer (hey, you have to be optimistic on a Friday night, stuck in an airport ;-) and we talked about commercial open source companies providing source code to its customers.

The source code has always been an important advantage for us. If you are buying Funambol, you get the source code. No need to resort to costly Escrow agreements (which are a mess, I did one when I was at Reuters). You simply get the source code. And a lot of people that know about it.

We call that "The Bus Factor". If I get hit by a bus, and so does the entire Funambol team, our customers still have the source code and a lot of people behind it (most likely, many in their own town). You can't get that with an Escrow agreement...

Marten called it once the airbag in the car. You do not want to use it, but when you need it... you are thankful it was there. And you make sure it is in the car when you buy it.

How does the economic downturn matters? Well, not for us, but for everyone of our competitors. The likelihood of them going belly up in the next twelve months is quite high. Most of them have high burn and little revenues. With no cash in the bank due to a recent significant fund raising. And zero chance to get more soon.

If you are a service provider or a device manufacturer and you license the code from one of our competitors, how high is the probability of you being stuck with a piece of unusable code in the next 18 months? Being pessimistic (see, I am not always optimistic...) I would say very high.

Having the source code is a much more important factor these days. The Bus is the economy. And when it hits you, you better have an airbag...

The iPhone with a keyboard

If there is one feature (or lack thereof) that people talk about is a keyboard for the iPhone.

When the iPhone was announced, everyone said "it will never work" because the keyboard was the most important feature for any BlackBerry user (it still is). Therefore, people said "the iPhone will never make it in the enterprise". Now, surprisingly for some, the iPhone is in the enterprise.

True, it is mostly a consumer device. But you hear few people complain about the lack of a keyboard these days. Mostly, because the iPhone is an upgrade from a phone without a keyboard... Therefore, it is better than zero.

However, in many cases people did move from a keyboard-full device to the iPhone. They bitched for a week, balancing the lack of a keyboard with the coolness of the device. Then they forgot about it. They learned. They became proficient at it.

Actually, when we did a test at the OSGR conference, iPhone users won the contest against BlackBerry users for fast typing. It was more than surprising, not only to me but to the iPhone users.

Now that I have used a G1, I was shocked that I found the iPhone virtual keyboard better... I still prefer the BlackBerry keyboard, but I can definitely live with the iPhone one.

When the BlackBerry Storm will hit the stores later this month, we'll see if the tactile feedback they added makes a big difference. I am ready to bet it does. And that it will make all mobile keyboards obsolete.

Unless you are someone that would buy the one in the video below. If you do, then please buy a netbook instead ;-)

Funambol Developer Conference

After a lot of requests, we gave up and organized the first ever Funambol Developer Conference (or FunCon ;-) It will happen on November 25, 2008, in Milan, Italy, at the Politecnico di Milano. In the open source spirit, we made it free to attend (free as in beer). More info below, directly from the website.

The conference's purpose is to provide a forum for Funambol community members and development partners to meet with the Funambol engineering team. The aim is to conduct an open discussion about Funambol technology, products, future plans and ways to improve community interaction.

The conference will provide an opportunity for Funambol to inform everyone about the state and future of our products. It will also be a chance for Funambol community members and development partners to inform the Funambol engineering team and each other about what they are doing with Funambol software, what they would like in the future, and how Funambol can help and improve the community. It is a chance for Funambol development partners to network with the Funambol engineering team and community, which may be helpful for their future endeavors.

For people that are new to Funambol, this will be a chance to learn about Funambol Sniper Projects (Code, Phone and Lion) and get a head start with participating to the first mobile open source community.

The day will feature presentations by the Funambol engineering staff and community. If you are interested in attending and/or presenting, please contact Stefano Maffulli, Funambol Community Manager. The conference is free and open to everyone, although we do require that people register in advance so that we know how many people are planning to attend.

Great opportunity for all you out there in our community. Did I mention it is free? ;-)

When Android beats the iPhone

Today I discovered two reasons why Android might beat the iPhone, eventually. One, it is open source (ok, I am kidding, I did know that before ;-) Two, it has OTA firmware update.

As you might recall, I complained about the email client on the phone. Not the Gmail one, the other one. The POP/IMAP client. Apparently, I was not alone. Pretty much everyone has been complaining about the connection error I have with POP. The Tmobile support forum currently shows 229 messages on this topic...

Bottom line: T-Mobile botched this one. Google could not care less (Gmail works nice) but the mobile operator should have been more careful. People are returning the device to the store, because there has not been a fix (yet).

Why does open source help? Well, on the Android Market a new email client appeared a few days ago. It is called K9. It is a fork of the original Android client (which is open source). It has more features than the original one and, not surprisingly, it fixes the issue above.

This is great. A week or so ago Google was alone in developing Android. They made it open source and boom, an internal application is getting perfected by the community. End user benefit from it immediately, because they can download it for free from the Android Market (which is such an easy task to do). The OS allows you to define a default email client, so you are good to go.

All this, without any intervention of Google or T-Mobile. The community fixed the issue for T-Mobile... You have to love this one if you are a mobile operator.

What would have happened with the iPhone? Probably, Apple would not have released an application that buggy in the first place, but... you would have had to wait for the next firmware update. Hoping the fix would be added (and, believe me, I have been frustrated by the lack of copy&paste in the iPhone for 18 months and there is no fix in sight...). You do not have to rely on Apple. You can rely on the community as well.

When it comes to firmware update, here is a good one. Android does it over-the-air (OTA). You receive a message, click on it, wait a few minutes and your are done. No iTunes, no cable. Pushed to your phone. At the moment, it is a bit random (some got RC29 already, some did not) and they should make the process more transparent. But it is a great innovation over the iPhone.

Nice to see Apple has to catch up, once in a while...

Time to be aggressive, if you can

The state of the economy is scaring everyone, rightly so. I do not think it is going to get better soon, actually I feel it is going to get worse. A couple of weeks ago, I opened a management meeting with the Sequoia slides, just to set the tone... This is serious stuff. Companies that do not have cash flowing in and have 6/9 months of cash in the bank are toasted. If they do not cut 50% of the workforce now and give themselves at least 18 months, they are gone. If they do, their product and sales stop. I feel we will see a devastation in 2009.

Mobile is not immune, but it will be affected much less. People might buy less phones (although they still do, like they buy new clothes), but they definitely lose or break the old ones... This market is too hot to slow down too much. It will slow down, but significant growth will be still there.

Open source will be a relative safe heaven. When everyone is cutting cost, do you think an enterprise will go for a $50/month BlackBerry Enterprise solution or will they download an open source project, and run it for free? Well... You know my answer ;-)

Now, what do you do if you are in mobile, you have cash flowing in and you just raised 12.5M dollars? On top of it, you are open source and you built a company to be run lean and mean?

First, you make it leaner and meaner ;-) Out fly the free jelly beans for the employees...

Second, you become aggressive.

I am ready to bet a lot of our competitors will be out of business or acquired in 2009. Many - some are public so it is easy to see - have net burn rate of millions a quarter, with the need of raising cash in 2009. Good luck with that.

Therefore, if you are on top of your game, this is the time to go for the kill. If we do not do it, who can?

Today, in an announcement that PR people call "momentum release" we are telling the world that:

1. We are selling like crazy... We had another record quarter. This is the best news of all. Actually, the word I am hearing from prospects these days is "It is URGENT". I never heard this before... It is a very nice indication of how badly service providers need to launch a MobileWe solution.
2. We won some more awards. I am not a big believer in awards (I prefer sales ;-) but I am a competitive person and winning is always a pleasure. Coming #1 as the best mobile consumer email solution in a contest is a good thing (and thanks for voting me a visionary for the third year in a row... I am not sure what that means but I am told it is good :-)
   
3. We are expanding in markets where we did not have a presence: Beijing and Dubai. It is a smart move, I believe. We hedge the risk of one of the world markets going down. We have presence now in the US, Europe, Asia and the Middle East. If the US comes to a halt, China might not. This is the time to expand, not retreat.
   

Bottom line: when things are down, it is time to look up. We are in a tunnel but there is light at the end for those that have the means to get there. If you are one of the few that can make it, this might be the best time to build a company to last and grab market share. GO for it.

Motorola focusing on Android: why it is big news

Today, the Wall Street Journal pretty much reported that Motorola will cut any other operating system to focus on Android. It was the co-CEO talking about it, not some Joe (the Motorola) on the street.

I wrote "pretty much" because they also said they will keep support for Windows Mobile and P2K (their legacy platform for low end devices). Moving away from about half dozen operating systems is already big news. However, if you look closely, he also said they will "outsource" the development on Windows Mobile. A nice way to say "we are not going to focus on it anymore but we still have some business there, so we will not kill it". What about P2K? Well, same as above. It is for low end devices and Motorola has many of them (in particular, in emerging markets), but Android can support middle and high end devices, and that is the future of mobility. Low end devices will disappear quite quickly.

Motorola is focusing all its mobile future on Android. This is the big news.

It is big news for Motorola (and very sad news for the tons of employees that will get a lay-off tomorrow when they announce their numbers), but it is even bigger for Android - and mobile open source in general.

I have been playing with Android. I am still trying to use the G1, but I am about to give up. The hardware is horrible, but the software is very good. And open. Including the source code of all the internal apps being available out there (I browsed through it, it is quite a remarkable amount of code).

What is missing with the G1, to make it a killer device? The device itself... The internal is good, the external is bad.

What does Android need? An hardware vendor that focuses on it. Someone with a big brand. Someone that can come up with a very cool device.

Motorola is that company.

It is easy to discount Motorola. However, look at how strong their brand still is (hey, NFL coaches have the Motorola headset...). And remember they came up with the best selling mobile device ever, the RAZR. It has been the killer device for years. Still the most sold device around.

Motorola can do it again. With focus completely on mobile open source and Android, they can rise again. This is what Android needed (badly). This is what could have prevented Android from becoming a big player in the mobile market.

Not anymore. It is another good day for mobile open source.

Funambol on the Android Market (and the G1)

I spent some quality time last week with Carlo Codega, the developer in our community who built the Android client. He is in Italy and I have the G1, so it made sense for us to work together and build a version ready for the G1 (and the Android Market).

This morning, as soon as the Market opened, the Funambol client has been uploaded on the Android Market. If you have a G1, you can check it out right now. It allows to sync your address book with a SyncML server (such as the free my.funambol.com), which gives you over-the-air sync with Outlook or Thunderbird or any other client we support. Also, if you have a new G1, it allows you to migrate your existing contacts from your phone to the G1 (sync first with your old phone, then with the G1). Your old phone could be a dumb phone, but also an iPhone, BlackBerry or Windows Mobile. We support all of them.

I learned a lot from this process, as usual.

First, developing for Android is not as easy as one would like it to be. Intent and Activity are concepts that make sense, but it takes a moment to get used to them. Our original version had the Action for Settings which was a bit too generic: you clicked on Settings and all the apps that registered for the View action would pop up... When we made it application specific, everything worked. You could not tell it from the emulator, since there you have just one application (yours)... In general, I found the process of compiling and getting the apps on the phone - including debugging over the cable - quite easy. Good job there by Google.

Second, the Gmail sync is very transparent but it has bugs. Some contacts get duplicated in the process. Weird things happen around "Suggested Contacts", the people you replied to from Gmail. On Thursday, they were appearing on the phone as "hidden" (but screwing up our sync), then they became totally visible. I am sure many will complain. If you use Gmail, you might have hundreds of emails as suggested contacts. I replied to a mailing list of my neighborhood (where the people were put in CC) once, and I ended up with hundreds of neighbors email addresses on the G1. I love them but I do not want them on my phone ;-)

Third, posting on the Android Market was a breeze. Past the $25 I had to pay (does Google really need that money??), I had to fill up a form, upload the app and voila' it showed up on my phone. No questions asked. No Apple police to check it out. It felt, well, open.. And I like open ;-)

Lastly, the security environment for Android seems quite strong. I browsed around the file system, but everything was shut down. Every application is its own user (in the Unix sense) and has its own subdirectory to play with. No permissions to go around and screw up the system. I am still missing a file browser and a tool to send files out (you can't attach them to emails), but I am sure they will come shortly.

Overall, it has been a good experience. I have to say I am pleased with the OS, environment for developers and market. The hardware is a different story, but it is not tight to the OS. We'll see better ones (I hope, for Google!).

Few tips if you are planning to use the 0.1 Funambol Sync Client for Android:
1. there are some bugs left (that is why Carlo calls it 0.1 ;-) For example, it does not support foreign characters (your contact will be displayed with its phone number) and contacts with First Name but no Last Name will show up as ",FirstName". Use it at your own risk.
2. if you do not want Gmail automatic sync to screw things up, uncheck the automatic sync for contacts (Settings -> Data Synchronization -> Contacts).

Enjoy and let Carlo know how things are looking. He is busy preparing for an exam (Wednesday), so... better leave him alone for the next two days ;-)

Because open source people like a good debate...

There is a discussion going on in one of our forums. The topic is AGPLv3.

For those that do not remember what AGPLv3 is (shame on you! I have a bet with Mark Radcliffe to win ;-)... AGPL is the GPLv3 variation that adds distribution of Software as a Service as one of the triggers for returning the code to the community. In practical terms, it makes distribution of SaaS as any other distribution (like in a floppy). If you host AGPL code and make changes to it, you have to return the changes to the community. Exactly as if you were distributing that code in a floppy.

AGPLv3 has been created by the Free Software Foundation (same people who did GPLv3) and approved by OSI as an "official" open source license (thanks to the submission of Funambol).

What is the discussion about? A crazy detail that does not make any sense. But we open source people like crazy details...

The problem is due to us releasing our clients with AGPL as well (same as for the server). It does not make sense, because the clients are not used in a server, so the A in AGPL is never triggered. We could have released them as GPLv3. However, we thought it would be dumb not to give it an A, with the idea that someone might reuse that code to build a gateway or a server (see... we are forward thinking ;-)

Clearly, the A in AGPL (section 13) applies if the code is used in a SaaS mode. If you use the code in a client and you do not offer it as a service, it is exactly as GPLv3. Section 13 does not apply. Period.

Thankfully, the Free Software Foundation has added now a comment in the FAQ that kills the debate (they agree with me, so... I won the debate!).

Sometimes, being too forward looking might be a bad idea. But if you like debates as we do in open source, then it is not ;-)

The G in G1 stands for Garage

My fearless office manager circled around her local T-Mobile store last night. First at midnight (no lines), then at 3 am (no lines), then finally at 7 am. She was fourth in line and came back to the office sleepless but with a prized T-Mobile Android G1 phone for me. I actually asked her only to be there at 7 am to get me one, but the people at Funambol always go 101% to get the job done (and I am very proud of them).

So I have a G1 in my hands. I played around with it extensively today. Tonight I am planning to go deeper into the file system and play with applications outside the Android Market (that has only 50 apps, the rest will start appearing on the 27th, when the market will actually open).

The first impression must be on the device itself, not what is inside. I could give you my perspective, but I feel my fashion-conscious Italian wife summarized it all today. She saw I had a new gadget and asked me "what's that, a new garage door opener?". Yep, it is that ugly.

That said, it is not just the look... The sliding keyboard is hard to use, due to the bulky thingy at the end (the one with the buttons to open the garage). If you try to type while charging the device, it is almost impossible. I never thought I would miss the virtual keyboard of the iPhone...

I have to admit I never liked the idea of a sliding keyboard. Changing mode is a no no in usability. And it shows. You are browsing with the vertical screen and you need to type a URL? Go landscape, type the URL, then come back vertical. Worst: you are in your car and need to call someone? In order to type their name, you have to slide the keyboard out, type, click than go vertical again (without crashing the car, knowing you are also violating the law in California). It is quite bad.

The inside is much nicer. Not just because it is open source (that, you can't see). The GUI is smooth, the home page looks good, the Google integration is well done.

The only negative? It is just Google for now. You have to put in your Google credentials when you start the phone for the first time. It loads Gmail, contacts and calendar. Nice and smooth.

Do you want to use any other email? Forget it. The email client (the second one, they have two... one Gmail and one Email) is plainly horrible.

I tried setting my work IMAP email. It took an hour to download all my folders... Also, I am so unlucky to have called the company Funambol, and have hundreds of folders beneath it: Inbox starts with I and comes after F of Funambol. I never got to see the Inbox. When it showed up, it then disappeared again. Just unusable. I gave up after trying for at least an hour (I swear. I care about mobile email...).

So I tried POP. Much better (it gets only the Inbox) but the scheduled POP never worked. For the whole day, I was never notified of a new email...

On top of it, the client has no attachment support. A very disappointing experience.

Then I tested the Market: it worked flawlessly. I downloaded a Barcode Scanner app which allows you to take a picture of a barcode and sends you to the Google Product page of it. Very interesting. You can check a book in a bookstore and see how much it actually cost to buy it online. The end of brick and mortar store, dot come style.

GPS: average. Wifi: pretty good. Calls (hey, it is a phone as well, you know): no issues. Maps: very nice, with the cool Street View feature that you will never use once in your life (but it is still cool so it is worth it). Browser: vastly inferior to the iPhone. Music: same as before. Camera: not bad.

I also tried to download some apps online (outside the Market). If you have an .apk file somewhere and you download it with the browser, it gets installed. I tried the Funambol client, it installed perfectly but it needs some work (Carlo is working on it, I believe). I tried a ssh client, same experience.

My brain works in a strange way (surprise!): I never have expectations. I just experience life as it comes and that has served me reasonably well along the years. Very hard to be disappointed when you did not have high expectations of something.

This is the case for the G1 as well. I did not have any expectation for this device. I disliked the hardware, but I liked the software (although it needs some more work). Nothing compared to the reaction I had when I opened the box of the iPhone, the first day it went on sale. It was an emotional experience. It still is (when I look at it in the charger, I just want to pick it up and play with it). The book "Emotional Design" (a must buy) describes why I might feel this way. Definitely, no emotions on the G1 for me...

Overall, it looks like a good thing for Google (since the software is ok), but not a smash hit. I will be very very surprised if they sell a ton of G1s. Developers will buy it. Some Google maniac will buy it. But it is going to fade very fast.

Unless they can add a garage opener application on the Market. For that purpose, it looks great.

Lion Sniper (roarrr ;-)

Today we announced a sequel to our two very successful sniper programs: the Lion Sniper. The program focuses on translating Funambol software, online help and the Funambol Community Forge into many foreign languages.

Localization (or L10n, if you wondered where the name came from...) has always been a key contribution by open source communities. Due to the geographical distribution and local needs, one of the first community effort has always been localization.

A company wants to support an open source project in a country? First thing they do is translating the product in the local language. And they return it back to the community. Very simple and effective. Also for the open source project.

After having a community member translating the Funambol Forge into Chinese earlier this year, we have seen our adoption rates in China skyrocket. That translates into sales, so it is a nice positive feedback.

So much that we are pushing this forward, setting bounties for $250 for those that are willing to translate the community stuff in their language. If you are interested, check it out.

Apparently, Savio at InfoWorld agrees (thanks!).

Open Source in Mobile Challenges

Today I bumped into an interesting opinion piece called "Open Source in Mobile Phones: Challenges for Software Vendors" by Adam Leach, principal analyst at Ovum.

Adam analyzes the current mobile environment, the emerging growth of open source and how ISVs need to evolve to meet new customer expectations.

In particular:

"ISVs need to be familiar with open source environments to be able to respond to demands to integrate customers' solutions into those environments. In addition, OEMs and operators will start to expect similar levels of transparency and collaboration on key areas of interest for them. As a result, ISVs will need to evolve their business practices to meet these new expectations, in some cases embracing an open source business model."

So... you better familiarize yourself with open source because it is here to stay. Then you have to find a way to make money, which is even tougher (but there are ways ;-)

Lastly, he gives tips on how to integrate an open source software into an ISV offering and what to look for: features, license, support and indemnification. All of them are usually provided by commercial open source companies based on a dual licensing model. If you go outside it, you risk to be screwed (but not always, so just be careful).

I believe it is a must read, if you are looking at embedding open source into your product - mobile or not.