Friday, August 22, 2008

The Googlephone is out

Forget Android, here you have the real Googlephone. A General Electric device with a button to call 1-800-GOOG-411 directly. Who needs Android?

Ok, ok, it might not be what we were expecting, but it is still better than faking customers in line, like they did in Poland for the iPhone launch ;-)

Wednesday, August 20, 2008

Palm Treo Pro is out!!!

Ok, I really had to stretch my enthusiasm to create the title of this post... Today Palm has announced yet-another Windows Mobile Treo, called Treo Pro. They announced the Treo 800W a few weeks ago with Sprint, generating pretty much the same reaction from the media (none). The specs are what you would expect from a smartphone these days: wi-fi, GPS, keyboard, 3G (the 800W on CDMA, the PRO on GSM networks). All on WM 6.1.

It will be a well engineered device, targeted to the enterprise (at $549 a pop, it is hard to imagine otherwise). It will sell ok, another stop-gap for Palm towards the next big thing. Same as the Centro (which is consumer-focused), buying Palm time to the iPhone/BlackBerry/Android killer everybody knows they are working on (but nobody has seen).

Unfortunately, these days it is hard to generate buzz. Coming up with a new release of the same device does not do it, unless you are Apple (I am still amazed at how much noise the iPhone 3G has made, and I can't get those lines of customers out of my mind...). To make noise in this space, you need to come up with something totally new. And totally cool. Palm was not going for buzz (although they generated some). They just needed a bit more cash and some more time.

I am still a believer in Palm (and a long-time fan). I hope the stop-gap measures will work and give them enough time to launch that killer. I met some Palm executives recently and it is clear they have been recruiting top-notch people, therefore being optimistic is easy.

If the iPhone stays too close and Apple too greedy (quite likely), Android does not get a good dress (very probable, the device from HTC looked horrible) and my Windows Mobile continues to reboot (pretty much guaranteed), then there is room for a white-knight. An open, nice-looking, solid OS with a good brand and lots of developers, full control of OS and hardware from the vendor (what Google is lacking), and maybe some cloud thingy in the backend (like MobileMe but working).

That would do it for Palm. Lots of buzz guaranteed. It might be mobile nirvana.

Android video

Today I went from yawn to actually looking at the Android GUI. Google has released its latest SDK and, if you do not want to spend the time to install it and clog your computer, you can watch videos on the Internet. The one below is from Thoughts Media.



First impressions:
  1. iPhone is still miles away and it will impossible for Android to catch up, from a usability standpoint. Having control of the hardware and the software makes a huge difference. Also, having Apple's designers on board is a nice differentiator...
  2. The home page (or desktop) looks very flexible. Very cool that you can have virtual space to put all the icons where you want them (compared to the iPhone, where you are locked on a grid) and slide right and left. If they could just come up with nicer icons... I know, it does not take that much, but icons make a huge difference visually.
  3. Email is still not working and there is no calendar yet. I have not seen notes and tasks as well, sign of a more consumerish direction (or at least, priority).
  4. The idea of a checkbox to send a contact directly to voicemail, all the time, is novel. Apparently, Andy Rubin receives lots of calls from people he does not like ;-)
  5. The music interface looks not competitive with the one of the iPhone. Years of iPod experience become very visible...
  6. Android will be fully open. THAT is the differentiator from Apple (although it is shared with Windows Mobile, Symbian and others). I am not sure it will be enough. It all depends on how many Android phones will be sold worldwide.
  7. Yes, I do love PI.

Tuesday, August 19, 2008

Funambol Community Showcase #6: Citadel and GroupDAV connectors

August is a great month to catch up on things. I have been behind on Community Showcases and here I am presenting two in one single post ;-) Stefano talked to Mathew McBride and asked him a few question. Let me know if you find the Q&A structure more interesting or less.

Q: Your projects are two Funambol connectors: one for GroupDAV and one for Citadel groupware: what problems are solved by these connectors?

A: The first Funambol project I had was the GroupDAV connector. I initially built it for use with the Citadel groupware package, as it uses a common standard (GroupDAV) it works with other servers such as OpenGroupware and now eGroupware. It bridges these servers and the event, task and address database on SyncML clients.

The next project was the Citadel connector, designed to synchronize email between Citadel and a Funambol client. Funambol already has a email connector, however, in the context of a Citadel environment, my opinion is that its over-engineered. It has a core sync engine designed specifically for Citadel, and leverages the Funambol email connector to perform data formatting etc. I should note that Citadel has had native support for Funambol's push email notification system since January of last year, so this connector doesn't require an intermediary component (inbox-listener) to do polling.

I've rolled these and some auxiliary modules into a special "Groupware sync server" distribution for users of Citadel, eGroupware and OpenGroupware, so administrators can install and use this software quickly and easily.

Q: How did you come up with the idea of creating this software?

A: Years ago I had become an enthusiastic PalmOS PDA user. I had upgraded through several PDA's (second hand Palm III's up to a Treo 600). Then I decided it would be neat if I can have a central mail server at home which could store my PIM data as well - which is where Citadel came in. There wasn't a good way to sync between any sort of PDA and Citadel (or some other open source groupware) at the time. Eventually I became aware of Funambol.

Q: What challenges did you encounter and what achievements are you proud of?

A: Programming is still (mostly) a hobby for me, so my expertise has matured greatly over the course of my connector projects. First was an uphill battle to understand the basics of SyncML, and then the Funambol API's.

Eventually people started to try it out and give feedback, it's quite interesting to find how people use the software. I've had some 'brown paper bag' (as in Linus Trovalds) moments when I get bug reports. Once I had a user send me their PIM databases so I could get all the issues they had resolved in one hit - there was quite a few. When it came to reviewing and rewriting the GroupDAV connector last year, I made a big push for quality control - used all the feedback I got and spent quite a bit of time finding and squashing bugs. It's definitely made it a better experience for users.

Q: How would you summarize your experience developing Funambol?

A: I started out back in 2005 where I had a bit of trouble understanding how Funambol worked, at the same time the community has grown and there's a good knowledge base. My understanding of the Funambol system has improved during time, I've also seen efforts from Funambol to improve developer documentation etc. which has helped.

Q: What are the steps that you would suggest to a newbie interested in developing a Funambol connector? What are the mistakes you made in earlier stages of development that you wouldn't repeat, knowing what you do now?

A: See how other people have attempted to solve similar problems, reading over the source of the included Funambol 'foundation' connector can be daunting for a developer starting off, now we have other connectors, from me and others which all have to use the same interface. You can learn a lot from other people, which is the beauty of open source.

Q: What is the roadmap for future versions of the Funambol connector for Citadel?

A: I believe that the connectors are feature complete - as I see them, future versions are driven by user feedback, there will always be something to improve. I'm definitely excited about new phones which are becoming more ubiquitous (iPhone etc.) and flowing into new environments which brings more users with different requirements. There is still functionality such as notes sync which could still be added as well.

Thanks Mathew for your feedback and contribution to our community!!

If you want to know more, here are the links to the project:

Monday, August 18, 2008

Bye bye iPhone push

One of the features clearly missing from the iPhone SDK (together with calendar and notes access...) is push. There is no push in iPhone, besides MobileMe and ActiveSync. If you are a developer and want to push something to the phone, tough luck. There is no way.

In the last keynote, however, the Apple guys showed us the way. A push mechanism completely managed by Apple, the big brother. You want your application to receive something from the network? You send it to Apple, they send it to your iPhone. It will be called The Apple Push Notification Service.

The idea they explained in the keynote is that you can actually send only the notification, there is no process in background. Your application does not really wake up. You either hear a noise (custom sound), get the icon of your app to change with a number (icon badge) or see a text message (same as when an SMS arrives).

It is not perfect, there is big brother in the middle but I can live with it. In the keynote, they said it will be here by September.

Apparently, they put it in a beta of 2.1 for developers then they pull it from the latest beta... If you were waiting for push in September, too bad. Push is not going to be there, at least until 2.2 is out (they have to give developers some time to play with it).

I suspect the issue is with the MobileMe meltdown. I feel they were planning to use the same channel for it and they decided they were not ready. Or maybe they just decided competition from developers is bad. Better to keep the thing closed for a few more months, while they fix their own product. We are open as long as it does not hurt us... I heard that before...

Is the iPhone 3G ever going to work?

As everyone knows, the iPhone 3G does not work very well as a phone. Not a big deal, unless you realize it is actually a phone ;-) On a phone, nobody wants their calls to be dropped. Is it so old century.

Everyone I talked to says the 3G reception is bad. Some friends are in the US and have no way to test the same SIM card on a different 3G phone. Some are in Italy and have been using 3G for a long time. I can confirm it just does not work well... It is not a network-only issue. There is something in the phone.

First, AT&T was under pressure (I even read the problem was visible only in major cities, just because it is where people are more spoiled ;-), then it was Apple. Apparently, the 3G chip does not work too well.

Some said a firmware update will fix all the issues. However, the 2.0.2 update that came out today does not seem to do it. I bet the solution is a combo of hardware, software and network. But hardware is a major part.

Unfortunately, that means changing the chip of every iPhone 3G and I do not think it is ever going to work for Apple. If you bought an iPhone 3G, you can hope for a slight improvement. Then hope something else breaks to return it and get a newer version that will work better. Too bad you bought the first version. Luckily for me, I have not :-)

Friday, August 15, 2008

Android yawn

An Android phone is not going to be here before the end of the year. No, wait, it is. No, end of the year. No, later. No, earlier! No, never... Wait, is the iPhone running Android? Are Symbian and Google joining forces to make a single operating system?

I have been hearing news about Android pretty much every day. Somehow, I was entertained by the iPhone news (and I still am) but the ones on Android are making me yawn. Strange effect, it might be the Apple's aura...

The last news in the Android saga comes from the New York Times, as reported in this Reuters article (I spent some good years working for Reuters, so I usually trust them ;-) :
T-Mobile USA will be the first carrier to offer a mobile phone based on Google Inc's Android software [..] The phone is expected to go on sale in the U.S. before Christmas and perhaps as early as October
I have to say I reached a point where I won't trust anyone until I really see one device, not even Reuters. It might be because it is summer (hey, I am working and today is Ferragosto in Italy, I should be on a beach somewhere...) but I am starting to get sick about announcements of products years in advance. Where are the good times of Apple showing a device that nobody expected, and having it ready for sale the same day?

Monday, August 11, 2008

Steve is ready to destroy your phone

I have been an advocate for open since I teared apart my first toy, just to see what was inside. I liked that, although I never found out a way to put all the parts together. Seeing inside is important, gives you confidence and, if you know your stuff, you can even improve it and show it to others (for them to improve it further: it is called open source).

Now, giving the keys of my toy to the owner of the toy store and letting him destroy it if he wanted... nah, I do not think I would have liked that. Growing up, I still feel I would not give the builder of my house the key to get in and destroy it, without my permission. Even if he promised me he would give me a new one.

If you have an iPhone, you did just that. Steve Jobs today confirmed he can kill your phone if he wants. He can press the big red button in his office and destroy all you have on your device. Blow it apart (virtually), together with your emails, address books, friends, pictures of your newborn baby and/or mistress. He has the superpower.

Gee, I do not think even Microsoft got to this levels. I do not believe they can destroy a Windows installation, even if they would love it. Where are the regulators when you need them? Time to pick up your phone and call them. We are giving a one guy - although a very cool one - a bit too much power, in my personal and open-biased opinion.

Saturday, August 09, 2008

T-Mobile going for an App Store

The iPhone might be closed, the App Store might prevent open source application to be distributed, Apple might be closing critical APIs like Calendar, but one thing is clear: it is forcing everyone else to open up :-)

The last confirmation comes from an article on the Washington Post about T-Mobile. Apparently, T-Mobile will launch an App Store this fall. They will ditch the on-deck strategy (about time, did they really need the iPhone to see it was not working??) and will go with an 'open' deck. Open might just means you will be able to upload applications without lobbying within the company, looking for someone to listen to you. But it will be a great step forward.

Thanks Apple, I have a hard time liking what you do but - despite doing most of the things wrong - you are doing something very good to this industry.

Friday, August 08, 2008

No calendar sync on the iPhone?

I took a look at some of the reviews of our iPhone client and the main complaint is that it syncs only contacts, and it does not sync calendar. Beside the fact that the description of the app is "address book synchronization"... the issue is simple: Apple does not offer developers access to calendar on the iPhone. Period. There is nothing anyone can do to have a legit application on the App Store that is able to access calendar. Apple does not allow it.

Now, we have an application that works on calendar and notes as well, but it needs the phone to be jailbroken. And it will never see the light of day on the App Store (Apple checks every application manually). Therefore, it is useless for the mass market, which is the target of the 3G. But it proves that - had Apple decided to open something - we would have had a solution...

Why is Apple not allowing access to calendar, when they allow access to contacts?

Good question... One can only guess. Let me try to virtually ask Apple and see what they might say:
  1. Calendar is so complicated that we could not make it. Contacts was simple so the API was easy to make. And we have not enough QA people and we are trying to make things right (like MobileMe, for example). One day, when we will have enough QA people, we will support it. It is a promise.
  2. Nobody uses calendar, so why bother? You should use paper instead for your events. Electronic calendar is just for geeks and geeks are so uncool. Get a life.
  3. Open? Who said the iPhone was going to be open? We are opening what the heck we like. And we do not like Calendar. Suck it up and do not bother us. This is not open source, go ask Android if you want calendar, or Windows Mobile for that matter (they are more open than we are but their CEO is like the PC in our ads).
  4. We are trying to defend MobileMe from any external attack, like your open source thingy. MobileMe works so well that, if your are lucky, your will see the Calendar API in 2020. For the time being, check ActiveStink out, they told me it kinda works.
Although all answers seem quite appropriate to explain the reasons behind Apple's choice (...), I tend to believe the right one is #4. MobileMe is too hot of a product for Apple to allow anyone else to create a competitor. You can have competition, as long as you do not do it on the iPhone... Open is not what Apple is. I just feel they will pay for it one day, but it is working quite well for them so far... Why change?

Tuesday, August 05, 2008

Steve Jobs on MobileMe

Today, an internal memo by Steve Jobs around the MobileMe disaster surfaced on the Internet:
Team,
The launch of MobileMe was not our finest hour. There are several things we could have done better:
- MobileMe was simply not up to Apple's standards - it clearly needed more time and testing.
- Rather than launch MobileMe as a monolithic service, we could have launched over-the-air syncing with iPhone to begin with, followed by the web applications one by one - Mail first, followed 30 days later (if things went well with Mail) by Calendar, then 30 days later by Contacts.
- It was a mistake to launch MobileMe at the same time as iPhone 3G, iPhone 2.0 software and the App Store. We all had more than enough to do, and MobileMe could have been delayed without consequence.

We are taking many steps to learn from this experience so that we can grow MobileMe into a service that our customers will love. One step that I can share with you today is that the MobileMe team will now report to Eddy Cue, who will lead all of our internet services - iTunes, the App Store and, starting today, MobileMe. Eddy's new title will be Vice President, Internet Services and he will now report directly to me.

The MobileMe launch clearly demonstrates that we have more to learn about Internet services. And learn we will. The vision of MobileMe is both exciting and ambitious, and we will press on to make it a service we are all proud of by the end of this year.

Steve
That must have been hard to write for Jobs... But I subscribe to the vision being both exciting and ambitious. My suggestion would have been to outsource it to someone that knows how tough it is to build a sync platform. It looks easy from the outside... but it takes years to get it right. It is probably too late for that (if not, we are here to help...). In any case, I am confident they will make it eventually, and I am very happy they showed the world why such a service makes sense. Everyone else now wants to build a MobileMe that works on other phones and email accounts that are not just @me.com. And consumers now realize they need it. Our phones are ringing day and night... Thanks Steve, we love you ;-)

Funambol V7 is GA!

Pop the champagne, Funambol V7 is out ;-)

Wow, we are at version seven, pretty much one every year since 2001. It has been a long way in the open source world. I am amazed at how open the mobile space has become, since we started pushing the concept. I heard that one of the key theme for the next 3GSM conference will be Mobile Open Source... Can you believe it? When we went there for the first time with our booth there and an enormous sign on top (FUNAMBOL: Mobile Open Source) people were stopping by and asking: "What?? Are you crazy?? How do you put together Mobile AND Open Source??". Man, time changes everything ;-)

Enough with the memories, let's look at what is new in Funambol V7 (from the press release, Release Notes are here).
Funambol v7's new capabilities include open source software for BlackBerry push email and PIM sync, an iPhone plug-in for contacts sync, an enhanced Java ME push email client for hundreds of millions of feature phones, Yahoo! and Gmail contact import into the Funambol Portal, server syncing of tasks and notes via native SyncML clients, an enhanced SDK and greater scalability.

Funambol Client Improvements

  • BlackBerry push email client and PIM sync plug-in
  • iPhone plug-in for contacts sync
  • Windows Mobile and Outlook plug-ins sync notes and briefcase files
  • Java ME push email client supports Samsung and LG devices; TCP/IP push email; receiving & viewing attachments via mobile browser; send photo (selected devices)

Portal Improvements (Carrier Edition)

  • Yahoo! & Gmail contacts import
  • Support for mobile phones without a native SyncML client
  • Improved sign-up process security

Server Improvements

  • Sync of tasks and notes using native SyncML clients
  • Improved connector compatibility
  • Improved Funambol trailer for outgoing email

System Management and Admin Improvements

  • Database scalability and high availability (Carrier Edition)
  • Improved log searchability and tracking
  • Simplified system configuration structure

Additional Improvements

  • Funambol SDK includes new connector testing framework
  • Improved documentation
  • Hundreds of other enhancements and bug fixes

Monday, August 04, 2008

Funambol Community Showcase #5: BlackBerry PIM plug-in

Here you have the fifth community showcase of the series. What I like about this one is that the community contribution does not come from a single individual, but by an entire company. Mailtrust felt they needed to contribute back to the community and empowered Mike Taczak to lead a team that is developing a ton of code in Funambol (Mike even flew to Pavia to work with the core team and eat risotto). The largest contribution from them is the BlackBerry plug-in, which provides PIM sync (contacts, calendar, tasks and notes). Let's hear from Mike.

Overview

At Mailtrust, we've built our synchronization solution based on Funambol software. We've found the flexibility of Funambol software invaluable in offering this service to our customers. After the initial server-side development -- connecting Funambol to our Webmail databases -- we tackled the Outlook plug-in, extending its functionality to meet our needs.

Next on the list was the BlackBerry platform, which is extremely popular for business purposes. We wanted to offer the same functionality on this platform as a user could find on a PC running Outlook.

What's different from Funambol BlackBerry plug-in 3.x?

We liked Funambol's existing BlackBerry plug-in (v3.x), but we wanted to include tasks and notes sync. We also wanted to bring the plug-in up-to-date with the most recent code in the J2ME (JAM) client API. In the process, we chose to remove the email sync functionality and focus on making the PIM sync features as rich and seamless as possible.

Development issues

Developing for the BlackBerry offered some unique challenges. Because the platform itself was so rich to begin with, we faced many issues fully integrating with the device to offer a seamless BlackBerry experience. RIM's API gave us the ability to access all of the user's PIM data on the device, but there are some hidden rules with respect to creating and managing that data.

For example, setting event fields in the wrong order can result in a recurrence being saved incorrectly, and the data being displayed differently in the day view versus the week or month view. All-day events were the biggest challenge, as it seemed that every view chose how to display this information based on different criteria. It was a long process of trial and error to make it appear as correctly as possible in the most views. In the end, however, some data still does not display perfectly in some views, although we have tried very hard to minimize this effect.

Another good example is the way BlackBerry handles event invitations. When an event is created and saved programmatically with attendees, an email is automatically generated and sent to invite those guests to the event. The same is true for deletions generating cancellation emails, and modifications generating update emails. To solve the problem of unwanted (and numerous) event emails, we integrated tightly with the email system to block and delete any email generated due to a sync. Now our users can have all their data on their BlackBerry so they always know what meeting they're walking into, and they will never know that even after an event is saved, the plug-in continues to work to make sure they have the best experience possible.

Future Plans for BlackBerry

Since Funambol decided to adopt the BlackBerry plug-in, we continue to work with them to improve and extend the functionality of it, integrating even more with features specific to the platform. We hope to improve the speed and reliability of the plug-in, and reduce the time spent configuring the plug-in for various carriers. Funambol has plans to integrate the email sync features in their Java email client with the PIM plug-in we've returned to them. We look forward to this integration and the day mobile email is easier for all our users. We've also had requests for additional functionality.

Future Plans for Mailtrust

We have some interesting and exciting projects based on the Funambol framework planned for the future. Currently we're working with them on the iPhone plug-in to sync contacts. We hope to continue to work with them closely on other projects as time goes on.

References

If you want to try the Funambol BlackBerry plug-in, the best way is to sign up to myFUNAMBOL and then download the PIM Sync plug-in.

We got momentum

I read Hal's recent press release about Fusemail (another smart email provider who realized there is ton of money to be made in mobilizing their customers, behind the carriers) and the last sentence hit me:
FuseMail selected Funambol's software during the second calendar quarter of 2008, the most successful quarter in Funambol's history. Funambol completed a company-record ten sales transactions with innovative providers of email and PIM sync around the world, including several well-known service providers, online portals, such as AOL, and mobile operators. In the second quarter, the company also closed a Series B round of $12.5M in venture capital and significantly strengthened its solution by introducing important new support for BlackBerry and the iPhone 3G.
I know, I should be aware of how well we are doing, but time flies and you need your VP Marketing sometimes to remind you on how many things you did in the last quarter... Now that I put it all together (I am preparing the slides for the next board meeting), I realize it has been the best quarter of my entire business life. Neat.

As my wife usually says "Great, but let me guess: the next quarter is the really though one...". Yep, she is right. Welcome to the startup world ;-)