FSO Milestone II Photo Safari and rant

In the spirit of the ASU Photo Tour, I’ve decided to do a little collection of images for the FSO image which is built on the FreeSmartphone.Orgframework.

Let me first get some weight off my chest. I’ve been feeling burnt out by Openmoko since the end of June when some confusion was cleared up about the ASU, the FSO and in general, the direction of the software for Openmoko devices.

I’m going to toss around some acronyms and a new title to think about. The ASU was a proof-of-concept image that combined Qtopia, Enlightenment and GTK running on top of Xorg. This first image was built (I’m told) by rasterman by hand and that issues would be worked on until August, when a first “functional” image would be released. ASU, then, means April/August Software Update. Either month works just fine – the first image was released in April, the result slated for release in August.

To address the confusion, that software line is now Openmoko 2008.08 so “ASU” and 2008.08 are the same thing for the most part. With me? Good, let’s move on.

When the ASU was announced, two other things happened. The GTK based 2007.02 line was obsoleted (or some might argue deprecated since it’s now getting attention from the community) and the FSO Framework was announced. I’ll focus on the framework, because it’s where my complaints and burnout centered around.

The ASU and Framework were announced at roughly the same time. To the community it appeared as if Sean Moss-Pultz had pulled a decision out of his ass to abandon the software that people knew and go with Qtopia instead. This decision appeared to be arbitrary and sudden, leaving a lot of questions unanswered. Frankly, it left a lot of questions unasked, because this move was so radical and sudden. The biggest unanswered question was I want to develop an app for Freerunner, what should I start with? This is when the Framework was revealed by Mickey Lauer.

Perhaps this was too big of a leap, but from then on it appeared as if the ASU and the Framework were complimentary projects within Openmoko. The end result of both projects would be to develop dbus based services and signaling and adapt the Qtopia applications to use this system. As I reasoned, ASU focused on the userspace apps and the Framework would focus on the back-end and the merger of those two would be the “mass market” software Openmoko would ship on average-Joe ready devices. The future looked promising, and I was among one of the most vocal people in regard to this.

On 28 June 2008, Mickey Lauer posted again, making it pretty clear my assumptions were wrong. The Framework would not be merged into the ASU line to be shipped to customers. If the two were to merge, it would be by community support.

Frankly, after the disappointing abandonment of the Neo 1973 device, the sudden shift from one development line to another, the horrible communication about the aim and directive of the “official” and general all around cluster-fuckery that is Openmoko I began to loose faith in the project. I know I’m not alone in that regard, because within 24 hours of feeling I’d been hoodwinked, the SHR project was announced. While it may not have been the “fork heard ’round the world” it was a significant point, because I think there the camel’s back snapped.

Allow me to take a moment to give my impression about the other side too. I mentioned above that the general opinion was that Sean Moss-Pultz issued a directive to begin adopting Qtopia and that it was contrary to the aims of Openmoko. Now, I don’t know since I’m not in regular communication with him (the times we’ve exchanged e-mails have been polite and to the point) but I suspect the issue is deeper than that. From reading Sean’s blog posts and mailing list messages, I think the man is committed to open platforms and more than that, the power that leveraging those platforms gives to users. I think he truly cares about Openmoko’s goals. On the flip side, he’s also the CEO of a company and in that respect, his job is to ensure that Openmoko turns a profit. Furthermore, while he wears the title “Chief Executive Officer” there’s nothing chief about his position if it’s anything like other companies. The financiers, investors, venture capitalists and shareholders (if it’s public) are the real chiefs. Speaking bluntly, I suspect that those people started getting freaked out by the Freerunner “delay”. The media had long moved on and the term “vaporware” began floating around in regard to Openmoko and I’m sure this made the investor squeamish.

Sean had a choice then. In order to see his goal happen, he needs to compromise. To continue building and selling open platforms to people, Openmoko would need to make their first platform profitable to please the investors. That platform might not see the light of day unless they could be convinced that Openmoko could begin turning profit quickly. With the GTK based software showing problems, some architecturally, the decision was to either rebuild a boat mid-voyage or hop on that shiny ocean liner next to you.

I don’t fault Sean for making that call. I fault the investors who decided to dump money into a project and then didn’t allow the vision to be pushed forward. If they wanted to back an open platform, they should let the people making open platforms do what they need to meet that goal. If they wanted to back a company shipping phones that were nothing special, they should have done that.

It wasn’t the one decision alone that hurt the project though. Openmoko is open. The first time this hurt Openmoko was with Freerunner. When Freerunner was conceptualized and discussed, the media (even open source centric media) stuck to their traditions and speculated that “Freerunner will be an iPhone killer”. The problem with talking about your products is that nobody is used to companies talking about their products before they’re in warehouses, awaiting some magical signal to roll out to customers. So while it was pretty clear to someone who thinks critically, a lot of people saw the Freerunner as being “delayed”. There was also the “developer” and “mass market” confusion – even as the Freerunner was shipping, people didn’t understand that there were two markets, and two release dates, for the device. Of course, I get the impression that even this might actually have changed, instead to focus a smaller run on “niche markets” like universities and specialty vendors and aim the next product more broadly, but this isn’t confirmed for me. But I digress.

Openmoko is trying to walk the balance between being open and being effective. How do you talk about plans about future dazzling products and (a year or 18 months later) ship a product that STILL dazzles people? How do you market a product as meeting people’s needs when the product is designed to not assume to know the user’s needs and be flexible enough to still fill them? Openmoko hasn’t mastered it and it’s caused some backlash, but nobody else is doing the same thing better. In some ways, it’s a gigantic experiment.

As I sat down to write all that, I planned on it being a small post and then jumping right into the pictures of the FSO, but it turned out filling up two (going on three) pages in OpenOffice Writer, but it feels good to have put the feelings into words. With that, let’s move onward, and look at what the FSO Framework has to offer.

Meet Zhone.

Zhone is a very basic program designed to demonstrate the capabilities of the Framework backend systems. As an added bonus it looks spiffy. I don’t have a working SIM inserted in this image, so you can’t see the network I’m connected to or signal strength but these features normally exist and can be seen in later images.

For a moment, we’ll ignore the irony of the icon selection. I’m a married man with a professional career and other than ones in use for the purpose of having a rotary phone in use, I’ve not seen one in the wild. Yet in my hands to take this screenshot was a device that has substantially more computing power than a Game Boy Advance, which itself has substantially more computing power than all of NASA Mission Control when they placed a human being on the moon. Let’s ignore this and click the phone icon, shall we?

At the top you can see two bars, one blue and one green. The blue bar represents cellular signal strength and above it displays the carrier message. This SIM works, but is not active, so can only be used for emergency calls so displays “SOS Only”.

To the right is a green bar, displaying battery strength. In all honesty, I’m not sure what the letters above that bar indicate, but I’m sure that I’ll get that information soon after this post is public. If you’re terribly interested, check back later or leave a comment with an e-mail address (doesn’t require registration, that bugs the hell out of me!) and I’ll send you a message once I know for sure.

Got it already. 🙂 They stand for Bluetooth, GSM, WLAN and GPS, respectively. Small letters indicate “turned off”, capital letters are for “turned on”. Thanks for pointing this out wurp!

Below that is the dialer display and keypad. The back arrow deletes and the big green check enters the number. The smaller arrows, as well as the bottom row of icons do nothing, except for the “X” which exits the dialer.

The next section well look at include the messaging features.

The forward, delete and reply buttons simply drop me back at the main screen. I’m assuming this is because I have no SMS messages but it could be unimplemented features. I have a strong hunch that they work. In the Milestone 1 release, unimplemented features were clearly marked as such.

Clicking on the “Send” button brings up a selection screen where you pick your contact, touch the bar for that contact and are presented with a message compose screen.

I had to censor the first image because it contained phone numbers. The nifty thing about this is that it actually imported them from the SIM card automagically.

Contacts management is the next feature of Zhone to explore. Like the SMS section, you’re presented with a menu to select your action, including “Edit Name”, “Edit Number” and “new”. I elected to do each of those for illustrative purposes and use random keypad presses for the number.

The final portion of Zhone is something GPS related. This does nothing for me as of yet, but I wanted to show it anyway.

Exiting Zhone is easy since Illume is installed and functional. First is the ASU launcher with a light blue hue as icons are selected. The next image is a screenshot of the screenshot application to show that GTK works but looks ugly unthemed, but I’m sure that’s not really in dispute.

It’s been a while since I’ve seen the keyboard in action, and it’s improved quite a bit. I’m not sure if this is a FSO improvement or done to Illume directly, but it makes the Freerunner quite a bit more functional. There’s several modes now, including a full QWERTY mode which can be selected. The great news is that it’s actually terminal capable!

A few new images that didn’t make it onto this page can be found in my Openmoko Image Gallery including a possible bug (the keyboard selection menu doesn’t close if the keyboard does) but all told the “interesting” stuff made it’s way here.

What can’t be put in a photo collection is stability. None, and I do repeat none, of the apps crashed. Suspend and resume are actually working reliably on the Freerunner with a press of the power button for each. The AUX button puts the device into lock mode. My only complaint is that the lock mode still accepts screen presses, meaning that to put the device in a purse or a pocket it needs to be fully suspended else it runs the risk of powering up and down the backlite as keys and spare change collide with it.

All in all, there’s promising progress from all different fronts and it’s rapidly rekindling my desire to get involved and take control of my phone. Now… Where do I find a good media player…

Comments are Disabled