Monday, March 31, 2008

Google blocking AGPL in Google Code

I received an email this morning from a friend who is planning to put his open source project in Google Code, the open source repository hosted by Google.

He is concerned by this thread in Google groups. A developer asked, back in November:
Can I host a project on Google's Project Hosting, if it's licensed under AGPL? It's one of FSF's licenses. I really need AGPL instead of GPL, as my project is a server-side software.
Chris DiBona, Google master of open source, answered:
In fact we do not support the AGPL on code.google.com. We are actively trying to fight the proliferation of licenses that are considered open source and the AGPL both has very little market share and has not been certified as being open source by the OSI.
Now, that was November 2007... Later in the thread Chris writes:
We have no current plans to do so. If AGPL adoption is high enough to warrant it, we'll revisit. The question is this: Who will propose the AGPL to the OSI? :-)
Nice :-) Funambol did it. AGPL is OSI approved. When pushed again, Chris switched from AGPL not being OSI-approved to not being a popular license:
what really matters is that the license is adopted. So far, we haven't seen much of it.
So, first AGPL was not good enough for Google because it was not OSI-approved. That limited its popularity... Now it is OSI-approved. Still, it is not popular enough to be accepted in the Google closed open source hosting site?

And, by the way, why should people put their open source code in the hands of someone who likes open source only when it does good to its business (ehm, that could include me, but we are not talking about me, are we ;-) ?

C'mon Chris, give developers the ability of using AGPL for their own projects in Google Code. Your fight for no proliferation of licenses is something I subscribe to, but AGPL is the license of the future, no matter if Google likes it or not. And I can guarantee you it will become even more popular if it is accepted in Google Code...

Friday, March 28, 2008

Tmobile: when the pipe becomes too dumb

As you might know, I believe mobile operators are all destined to become pipes. They are a voice pipe today, after all, and they will be a data pipe shortly as well. They simply do not own content or any user content. The content owner will deliver the services, not the operators.

One example is in mobile email. Everyone thought the carrier would win, because they are the gate keeper. Wrong. The gate is not between you and your email provider. The gate is in the hands of your email provider. They have the content (your email). They can decide not to serve it to the mobile operator (or gouge them for it). Eventually, market forces will push mobile operators towards open data plans, and the power will shift to email providers (ISPs and portals).

Some days, however, I get surprised how a potential dumb pipe can become too dumb. In the last few days, we started getting reports that syncs on Tmobile stopped working. From people that bought a data plan. Same phone with an at&t SIM card worked. Put a Tmobile SIM card in and it would not work (actually, it was worst, sync with native SyncML clients were giving false positives...).

After some analysis, we found the issue. All of a sudden, Tmobile is chopping off the word jsessionid from the URLs. That is what native SyncML clients use to keep track of the session. If you chop it off, the sync can't work.

Now, there are two options:
  1. they did it on purpose, to block some application they do not like. Certainly it is not sync, but I just can't see what it could be (Google and Yahoo apps work... They are the enemy, right?)
  2. they did it by mistake, putting in the network a new gateway that has a bug or a feature they did not know about
I do not like conspiracy theories and I have been around mobile operators enough to believe option #2 is the right one. We spent three hours on the phone with Tmobile just to hear that "they do not support third party application", so they could do nothing for us. Bottom line: I believe they made a mistake, which will be fixed sooner or later.

If this is the case, we are facing a case of a dumb pipe which is a bit too dumb. That is a scary thought... Let me count on market forces to fix it. It always works like this...

Thursday, March 27, 2008

Funambol in my car

Yesterday, Dash finally launched its device. The first feedback from Walt Mossberg at the WSJ has been "I like it a lot". I can't agree more...

What is Dash Express? It is a car navigator. A device you put in your car and gives you directions. Ok, I know you might have one already. But this is different. Dash Express is connected to the net. Via GPRS or wifi (if you have an hotspot around, e.g. in your house when your car is parked).

Being connected to the net has few major advantages:
  1. every car tells the Dash system where it is and how fast is going. That gives them an incredible accuracy on knowing how the traffic is doing and giving you suggestions on where to go to avoid jams. Choosing 280 over 101 can save you hours in Silicon Valley...
  2. you can browse for places to go to in the car, directly searching in Yahoo Local (which has every possible place you want to go to). You are not limited by the old and out-of-date CD database.
  3. you can highlight an address on your browser (when you are in front of your computer) and send it to your car. Once you get in the car, the address is there and you can just follow the instructions to get to your destination.
  4. the device updates itself over-the-air, including maps and features and bug fixes.
I have been a beta user of Dash Express. I received my first unit at the end of 2006, over a year ago. That device had an external GPS antenna and some quirks in the navigation: for example, it gave me instructions to go into a road, violating the traffic code, or suggested me a never ending loop to go home ;-) Now it is bug-free and pretty much perfect.

Actually, it just became a dream for me. I developed an integration with myFUNAMBOL, which now pushes to the device my calendar events. That is, if I put an appointment on my Outlook and I add the location in it, it shows up in the car. I mean, I get in the car and I have all my appointments there. One click and I get routed to the place (with the fastest route, avoiding traffic jams). No worries.

The way I got addicted to Dash + Funambol Calendar sync equals my addiction to TiVo. I do not believe I found a device that changed my life as TiVo did, giving me back my freedom. Time is the most precious thing I have. TiVo saves it. Dash is very very close to it. I do not think I could live without it now...

We are demoing the calendar sync on the Dash Express at CTIA next week in Vegas. If you want to take a look, stop by our booth (#3800). We are making this available to the myFUNAMBOL public - for free and open source as usual - in a couple of weeks. Ahh, and if you just can't wait, just go and buy your Dash Express today. They sell it on Amazon as well.

Tuesday, March 25, 2008

451 on AGPL

Jay Layman of the 451 group wrote a post on AGPL, which I found quite interesting (I am biased, of course ;-)

His question is if AGPL is half empty or half full. In particular, he wrote:
Google’s license choices and glass-half-empty approach may end up costing the vendor some of its dominance in the future
This ties nicely with my suggested business model for the future... I believe Google should rethink its strategy around open source, be more open and gain from the power of a community, also for its core.

Jay adds:
Similar to how open source software has historically been among the only effective weapons to fight market and monopoly powers such as Microsoft, the AGPLv3 may elevate the more collaborative, more transparent and productive communities above those that continue to leverage old GPL loopholes. In other words, there are more advantages to GPL-style collaboration, community and cachet (glass half-full) than there are disadvantages to having to share code.
Can't agree more :-)

Monday, March 24, 2008

The (open source) business model

These are the days of OSBC in San Francisco. A conference that is still quite valid and interesting, despite open source being now everywhere. My hope is that we are done with talking about licenses (enough, we have GPL and AGPL v3, that's all we need) and we'll focus on business models. Here, I am not sure we are done.

I have been thinking a lot about it lately (maybe because I was on a beach on vacation...) and I came to the conclusion that, eventually, there will be one prevalent business model around open source. I am not talking 2009, maybe 2015.

I have never been a fan of business plans based on support or consulting or professional services. They do not scale. Good to make a living (a very respectable thing, do not get me wrong), not to build a billion dollar company. Licensing it the answer for hyper-growth.

Now, with the world moving towards Software as a Service (SaaS), and everything being fed through the Net, how do you make it all work?

First, you need open source. There is no question about it.

Second, you need a market for your open source product. Where you will not make money. But that will give you all the benefits of a big community. That market - in my opinion - is the enterprise. Not every enterprise, those companies that value the source code in open source. That want to touch it. Those that want to deploy open source in house. To them, you give your software for free. You do not make money. It is your community. You do not upsell to them.

Third, If someone is hosting your software, they pay. You "license" the code to them. The end user might still be an enterprise, who has pushed that application in outsourcing. That is, enterprise that use SaaS software pay for it, even if it is originally open source. They won't see the code or touch it, but they do not want to do it. That's why they chose outsourcing. You can be the hoster or you can sell to hosters (your call).

Fourth, If the end user does not want to pay, you subsidize the price with advertising. The SaaS offering will be free, but you will make money anyway.

That's it. Simple as it goes.

Who is doing it? Well, Funambol is... That was too easy :-)

What about SugarCRM? They are close to this model. They have Sugar On-Demand. They could change the license, dump all the features in the community edition for deployments within the enterprise, and focus entirely on their hosting business. Maybe they will, one day... I would be curious to check with John how the revenue split is today and what the trend is (I might actually do it tomorrow on the boat ;-)

What about others, non-open source company? How different is Google from hosted open source with advertising? Should they just open source the code they have built so that people could deploy it in house, protecting it with AGPL? Why not? What about Salesforce.com? Why not do the same? Why protecting your code if you can defend it from competitors and get the advance of having a community working with you?

I am really curious to see how this will play out. And to listen to what people say at OSBC. BTW, I am speaking there at 2 pm on Wednesday, the panel is "Out of the Server, Into the Network" chaired by the mythical Larry Augustin. If you are around, give me a holler.

Wednesday, March 19, 2008

Iphone vs. Android

Today at AJAXWorld 2008 in New York, Andrea Gazzaniga (our own Software Developer Manager at Funambol) delivered a very interesting talk about "developing Open Source Mobile Messaging and Sync Apps for iPhone vs. Android".

The topic is quite hot. Apple just released the iPhone SDK and Android has been around for enough time for developers to figure it out. Not completely (for both), but enough to have a few ideas. And to get pissed at some choices, in particular on the Apple side ;-)

The Funambol community built on both platforms, so Gazza is one of the few on the planet that can talk about both with competence. I have my opinions, that might differ from his ;-)

Let me start with Android: it is OPEN. I like it. I feel Google is trying to re-design a developer paradigm, which is a big risk. If they succeed (it depends on devices being sold, not developers...) then they might have a lock in the operating system of the next 20 years. Yep, Android spans from mobile phones to many other devices. In my view, it might go back to sub-notebooks, which are the desktop of the future (and near present). That is, everything that is not a server. Not a small piece of the market.

The iPhone SDK is a real disappointment. Sometimes, I feel we should have kept the world split between those that get it (and have a jailbroken phone) and those that don't (and are happy with a browser). The SDK - as it is today - is an half-assed effort (pardon my French). It is just too limited, too full of traps, to be really useful. And gives the false idea of an open platform.

Take our push email and PIM sync client. Apple announced ActiveSync support. Nice, that means an integration from Exchange to the native email, calendar and contacts (I assume). So you would think you can do the same with the SDK.

Wrong.

First, there is no calendar API. What? Nope, no calendar API. No way to interact with calendar. But they do it with ActiveSync. It is a private API. Not an open API...

Second, there is no email API. OK, I was expecting this, but it still bothers me. Same as above.

Third, there is no way to do push, since the application die when they are closed (e.g. when your phone rings). You cannot fire an application in the background... You cannot notify the phone where there is a new email. Very useful.

Fourth, you cannot talk about the SDK. You must sign an NDA to download the SDK. What?? An open SDK I cannot talk about? What is in the word open that Apple does not get?

Fifth, there are "rules" that Apple will enforce. One is that you cannot run an application that is capable of executing code, which is not Objective-C. That means, no Java Virtual Machine (despite Sun saying they wanted to do it... ooops, bad marketing attempt) and no Flash (although Adobe just said the same).

Sixth (but there is more), no developer - that I know - has been accepted in the program to "sign" applications for inclusion in the Apple Store. Everyone was rejected. How are people going to spend time to develop apps that might never get into the Apple Store, since the developer does not have the ability to sign them??

Bottom line: Android is good and open. The iPhone SDK is crap and closed. We'll build on both for now, but Apple needs to get a grip on openness or we will all move to Android. You do not screw up with developers too much. They are easy to get upset and do not like to be pushed around. Apple be warned.

Thursday, March 13, 2008

AGPL is OSI approved. Sweet victory.

I just found out that the OSI (Open Source Initiative, the stewards of the Open Source Definition (OSD) and the community-recognized body for reviewing and approving licenses as OSD-conformant) has approved AGPL v3.

Yay!!

For those who don't follow this blog (hey, why are you reading it now, then? ;-) the fight against the ASP loophole in GPL has been one of my long-standing battles. In GPL v2, those who ran open source software in a Software-as-a-Service (SaaS) environment, and modified the open source code, were not required to return the changes back to the community (just because of a technicality, since the clause of
copyleft applied to physical distribution of the software, not distribution of software as a service). For me, this has always been one of the worst risks for open source oblivion. If you can take and you do not give back, defeating the copyleft concept, you kill open source. The ASP loophole is the cancer of open source.

One company benefiting from it is Google, abusing open source software for their benefit (aren't you interested in seeing the modification to the Linux file system they did to run their gazillion of servers? I am ;-) In fact, they pushed for the ASP loophole to be ratified in GPL v3 and they submitted it to OSI for approval. They love the loophole. They made a business around it (and what a business!!). They got GPL v3 OSI approved...

Funambol submitted AGPL v3 to OSI. AGPL is the exact same as GPL v3, with a sentence that closes the ASP loophole. If you run AGPL code as a service, you need to return the changes to the community. Distribution of software as a service is distribution of software. Simple as that.

Google wasn't thrilled, obviously. Chris DiBona, the Open Source Programs Manager at Google recently wrote on the discussion list that AGPL needed improvements... Although I understand why he would say so ;-) I strongly disagree, so did the OSI and Eben Moglen with the FSF.

Eben wrote:
“AGPLv3 is a useful tool for developers of, for example, web applications who are seeking to ensure perpetual respect for end-users’ freedoms. SFLC believes there are many development projects that would benefit from considering the use of AGPLv3, and we welcome the OSI’s decision“
Amen. AGPL v3 is here. It is OSI approved. The fight is over.
Sweet victory. The loophole is closed forever. Let's open some Italian spumante.

Now we just need developers to understand that using GPL v3 instead of AGPL v3 is just dumb. Your software is going to be used as a service, if not today, in a few years. Yes, YOUR software. Everything is going to be used as a service, even word processors... They can take it and do not give any changes back to your community. Wavemaker is smart and they are now using AGPL. Many will follow.

Don't be dumb, give it an A.

Tuesday, March 11, 2008

Open Source in Mobile - tomorrow

I just got back from a quick European trip and I will be speaking tomorrow at the Open Source in Mobile (OSiM) conference in San Francisco.

I am chairing a panel at 2:15 pm and the title is "The Importance of Open Source Software in Mobile 2.0".

Topics:
* What is Mobile 2.0?
* The transition from Mobile 1.0 to Mobile 2.0
* What Mobile 2.0 means for users and all parts of the ecosystem
Panelists:
Thomas Jakob, Vice President, Innovation, Swisscom
Tony Wasserman, Executive Director, Center for Open Source Investigation, Carnegie Mellon West
Roberto Galoppini, Commercial Open Source Expert and Consultant for Microsoft
If you are at the conference (you might, this blog is Mobile Open Source and the conference is Open Source in Mobile ;-) I look forward to talking to you.

Wednesday, March 05, 2008

Richard Boly for President

Richard Boly is, in my opinion, the person that has done more for young Italian entrepreneurs than any Italian politician. He happens not to be Italian, just passing by. But he is leaving a mark that will last.

I found this interview on the web today, that describes his work pushing a US mentality in my native country. Something we desperately need and that I am trying to inject there every day (everyone must have a pet project and a long-term goal...).

Richard does everything with passion (including discussing with me about baseball, although we disagree on the best team) and he has been involved in almost any smart initiative in the last few years...

I am not sure if he wants to run for President here (the race seems a bit too tight), but I am definitely ready to support him if he runs for President in Italy (the race seems wide open).

He is just what Italy needs.

Monday, March 03, 2008

SDForum about building communities (tonight)

For those who live in the Valley, I am speaking tonight at the SDForum Open Source SIG. The topic is "Building Vibrant & Sustainable Communities". You just have to show up at the door, if you are interested (and it is only $15).

The panel includes me and John Mark Walker, who is the Community Manager of Hyperic. I saw his slides beforehand and they are awesome. You should not miss the opportunity to listen to him (I am optional).

What: SDForum Open Soruce SIG: "
Building Vibrant & Sustainable Communities"

Where
: DLA Piper Rudnick Gray Cary LLP. 2000 University Avenue (note: These are the tall buildings on the corner of University Ave. and Woodland, west of Hwy. 101) East Palo Alto

Description:
Building a healthy and productive open source community is essential for the success of any open source project or business.

Topics will include strategy for building a cohesive community, continuously creating value to members, managing communities effectively, and real world examples of success and failures. The format for these well-timed subjects will be both presentation and open forum so everyone will have a chance to participate, ask questions and provide answers. This is a wonderful opportunity to get expert insight on one of the most important aspects of successful open source…building strong and effective communities.
Agenda:
6:30 - 7:00 p.m. Registration/Networking
7:00 - 8:30 p.m. Presentation
How much: $15.00 for non-SDForum members and free for SDForum members