Because of Android's relative openness, Aro is able to completely integrate its PIM solution onto the Android mobile platform. But now, as the company prepares to launch its iPhone version, compromises had to be made. This begs the question: can innovation around core apps even work on iPhone?

Aro Mobile, to catch you up, is a semantic technology software company led by CEO John Lazarus, a former Microsoft employee, now senior advisor to Allen's Vulcan Capital and board member at Evri, another notable semantic tech company, which recently went mobile too. Simply put, Aro's goal is to make our so-called "smartphones" much, much smarter by using machine intelligence to interconnect the apps we use the most, the core apps used in communications.

Semantic technology is a difficult concept to explain to mainstream users, but Aro demonstrates its potential by innovating on top of the PIM, the core communications applications that mobile users access daily. The software understands the language of mobile devices, including the way you chat via SMS, the way you email, even the way you Tweet. It can identify people, places and dates referenced in your emails, messages and calendar appointments and the importance of those items to you. Within Aro's graph of your social connections, it understands that there's a stronger connection between someone that you've emailed multiple times and someone who's only listed as an address book entry. That's the power of semantics, actually: the ability to understand.

We've seen semantic technology cropping up here and there lately, but mostly on Android devices. Keyboard replacement apps like SwiftKey and Siine, for example, are using machine intelligence to help you type more quickly with word suggestions - but not just suggestions based on a database of words in a given language, but those delivered through an understanding of how you - you personally - communicate. If you start typing "cy" on your phone, SwiftKey or Siine might suggest "cya!!!!," if that's the way you normally complete a text message conversation. Siine plans to take things a step further, with support for translation between languages and integration with other mobile applications, too.

Aro, along these same lines, wants to function as a conduit between the phone and the "cloud." Although it focuses on email for now, it will soon integrate with Facebook and Twitter to add another layer of data and understanding to your communications.

That's an explanation of Aro at a very high level, however. In practice, the app, or rather, the suite of apps installed function as your phone book, contact lists, calendar and email client. Even in its raw, early format, the app is smart. Very smart. It can identify a person, place or date, and then, when you tap on the item (identified objects are circled), a list of actions appear. If you tap a name, Aro may suggest that you edit the contact, add to your contacts, send an email, place a call, etc. And if you were to proceed, it knows what information to use - it knows the phone number to call or it could automatically fill in the contact details. The exchange of data between the apps is seamless - you don't have to copy and paste information from one app to another and you don't have to constantly toggle between apps, either. Everything talks to everything else.

For now, Aro serves as a useful personal assistant of sorts - in addition to its interoperable apps, you can use Aro search against your entire archive of data, the results custom-tailored to you. In a later version of the app, a location service will be added, allowing you to further personalize your search results. For example, a search for a "coffee" in a given city could return recommendations based on where you had been recently, as opposed to the sort of default list that appears in Google.

Design Challenges

As innovative as the technology is, the interface still feels too technical, as if built by engineers not designers. The objects Aro recognizes are circled so as to highlight them, like a teacher correcting a student's homework. Aro is training wheels for learning semantic technology. Here's a person, here's a date, it tells you. Tap this here. See what happens.

The text is circled because we, as Web users, can't quite grasp the concept of actionable data that's not highlighted in some way. We expect hyperlinks, colored and underlined, to direct our clicks. But links are going away, says Andrew Hickl, Aro's CTO. Semantic technology will eventually lead to their demise. In a decade or so, any object, any piece of data, anything that you can touch will be able to do anything and no one will need training to know that's the case - it will just be the way it is.

On iPhone, Limitations

Android was the perfect place for new, experimental technology like this to launch. Because Android isn't as tightly locked down as iPhone, there's room for software like this to take hold. There, Aro can behave more like individual applications and can more deeply integrate into various Android menus and functions.

As Aro prepares to launch on the iPhone, however, there are limitations. The iPhone Aro app when launched opens up the Aro suite of apps, the icons appearing as if in a folder. While the "apps" can still talk to each other, on iPhone they aren't really individual apps. The email "app" can't become the default mail app on the iPhone and the calendar "app" can't become the default calendar. This limits how deeply Aro's integration can be. You can't, for example, email a picture using Aro's email app - you still have to use the Apple-provided mail client.

For iPhone users, the level of control and consistency Apple provides is one of the key selling points of the device. Everything works as it should. But it's all Steve Jobs' vision - and while that's nothing to sniff at, not by any means - there's a reason why many early adopters are going to Android. The platform allows others outside of Apple to fully develop ideas of their own, too. And we get to watch them as they do, we get to test the apps in early alpha and beta formats, and we get to participate in the innovation that occurs.

Comparatively to iPhone's own core apps, Aro may not be as pretty, but its underlying technology needs a place to flourish and develop. Like other semantic apps, there needs to be a real-world playground where people can try these new things. For now, that playground is Android.

Aro is only one example of this. Siine and SwiftKey (mentioned above) are another. ON, a new company with an innovative take on the address book is yet a third. ON lets you maintain different profiles for different groups of users - one status message for friends, another for colleagues; one voicemail message for the boss, another for the spouse, and so forth. Like Aro, ON launched on Android, then dumbed itself down for iPhone. SwiftkKey hasn't even bothered with iPhone app. It's unclear how Siine will manage.

iPhone users wishing to get a taste of what's possible on Android can try Aro, when the app becomes available in Q1 2011. A lot of the functionality will still be present. But for iPhone users, trying Aro may feel like toying around with a demo. For full-speed immersion, try the Android version instead.

Earlier this month, OSAF announced that Mitch Kapor was stepping down from the OSAF board and will stop funding Chandler by the end of 2008. This was viewed in some quarters as signaling the death of Chandler.

But is Chandler really dead in the water? Why is it taking sooooo long to get to Version 1.0. They're still on 0.7, after 7 years!

I think Hank hit the nail on the head when he said that "Chandler was hurt by a shift from desktop applications to web applications." He mentions that competing apps such as iCal, Google Calendar, and dozens of others "beat Chandler to the punch." I would add Zimbra to that list - it managed to build up a solid email/calendaring app in less than 2 years, which eventually got acquired by Yahoo late last year.

But having said that, there is a Chandler product in preview that people can check out. Lifehacker did a good review in late December '07, noting that "while it's still got kinks to work out, it looks as promising as its vision." Lifehacker describes Chandler as "essentially a combination calendar, email inbox and task list for the individual, as well as a collaboration tool for small groups."

Chandler desktop version circa Jan 08

The promise of Chandler has always been that it will be an alternative to the costly Microsoft Outlook/Exchange combo, although as noted below OSAF is now attempting to downgrade that expectation and make Chandler more of a middleware or add-on product. Whatever the case, the Lifehacker review hinted that the reality is a buggy and not-ready-for-primetime software. One commenter confirms this suspicion:

"I've tried Chandler many times throughout it's development and each time I'm am disappointed. They have a lofty promise and goals, but I think they are still far far away from really achieving that. As much as I would like to try and find a worthy open source / free alternative, I just don't believe there is (or will be soon) anything that comes close to the Outlook/Exchange environment."

It's also troubling, after all these years, to see Chandler reaching for new product strategies and still defining what the core product offers. In another blog post, OSAF's Mimi Yin remarks that they want to "extend Chandler to reach into places where there are already users", such as Thunderbird plug-in, SMS to/from mobile devices, Facebook app, iGoogle widget. The Thunderbird connection is a sensible idea though; and one I raised with Mitchell Baker, Chair of the Mozilla Foundation, in September when Mozilla launched a new organization to develop Internet mail and communications software. Mitchell was non-committal on it, but surely it makes sense for the two open source organizations to work closer together to solve the email/calendaring conundrum that OSAF is struggling so much with?

But going back to the OSAF blog post from Mimi Yin, which also noted that OSAF must "converge on a crisp articulation of Chandler’s core offering. In particular, who is Chandler intended for, what problems does it solve and how does it solve them?"

Honestly, I can't believe they are still asking that question nearly 7 years after the Chandler project first began. Another post from Yin tries to flesh out more of the product strategy, but it is a painful read. For example: "Chandler is meant to live in the middle of all these tools as a way to pull all the disparate bits and pieces of information we receive out-of-context, into a contextualized, personal and shared ’source of truth’."

Conclusion

I was a big fan of Chandler when it first started and avidly followed the mailing lists and announcements. I tried it out a few times too, in its early incarnations. I lost interest though when web 2.0 came along and apps like Google Calendar and Zimbra actually delivered working products that did pretty much what Chandler was promising, albeit on a smaller scale.

But perhaps now is the time to check out the desktop and Web versions of Chandler, again. I'm no longer holding my breath though - it would be a major surprise if Chandler ever did amount to an Outlook/Exchange competitor, nearly 7 years after starting and with a 1.0 release seemingly still years away.

So can it survive the year, especially after Kapor's money stops coming in? Given that the current 0.7 product is buggy, the product strategy is muddled, and the project is nearly 7 years old.... well it would be a brave person to bet on Chandler's survival post Kapor. Everyone hopes it will, being open source and cross platform - but at some point reality has to set in. That point will be the end of 2008, if it still hasn't reached 1.0 or at least a workable beta.