CTIA seemed smaller this year. I walked the floor and got a chance to look at pretty much every booth in about half a day. But at the same time the meeting spaces seemed to grow a bit. This was a conference for getting deals made, not as much for putting on flashy shows.

Below are some of my thoughts.

MICROSOFT
Microsoft really came correct this year. Not only did Steve Ballmer get the first advertisement…. er… keynote, but the Widows Mobile booth was packed to the brim every day all day. Their phone lineup was very impressive, and damned if I didn’t want an HTC Tilt after playing with it.

Get this - we were inside the Moscone Center, and the Tilt grabbed a GPS location accurate to within a few blocks, with a lock on time of a few seconds. It displayed the results in Windows Live Search. This was me messing with the device, not a sales rep doing a demo. My mind suddenly drifted to a wonderland of snapping photos and posting them, GPS-tagged, using Shozu as I walked the floor and the streets of San Francisco. This phone could do that - my iPhone had neither the GPS, nor the third-party software to.

But at the same time, Pocket IE sucks so badly, that all the joys I have using a slick browser on the iPhone on a high resolution screen would need to go away if I moved to WinMo. Microsoft - these devices have been using 320×240 screens for years now. I’ve had enough. Figure out how to stick a bigger screen on them. The devices are getting so small now anyway, you could get away with putting bigger batteries in them.

The WinMo devices made me long for third-party apps on my phone again. These iPhone web apps are slick and all, but in a conference packed with mobile phone data users clogging up the networks, waiting for your UI to load over a network before you even get to push a button is unacceptable. I don’t care how efficient your AJAX iPhone twitter app is, you’re not going to beat a native app with your last messages cached, only pushing a small bit of data over the network. And oh how I missed an always-on IM client. Meebo for iPhone is a wonder of modern AJAX, it really is. But it just can’t handle a flaky slow network.

NOKIA
I don’t want to give all the props to Microsoft. Nokia has been doing the third-party app, GPS, 3G thing with the N95 for a while now. But it doesn’t have a keyboard, so it doesn’t get my gadget lust. What’s this with keyboards only on E-series devices Nokia? The hip kids want keyboards too sometimes.

One N device that does have a keyboard is Nokia’s N810 tablet. It really is freaking amazing, but it’s not a phone. You need to hold one to understand how thin and sleek it is relative to what it does. It’s devices like these that get me thinking of moving back to a two-device configuration, with the phone being used as a bluetooth modem. But the whole bluetooth thing breaks down when you want always-on applications. Bluetooth isn’t meant to be on when a device is in hibernate mode, so forget about keeping a persistent connection to IM or brokering email in the background.

The new N95 8GB is killer as well - those N95 screens are really gorgeous, I can’t believe they’re the same resolution (QVGA) of the WinMo phones that I was complaining about earlier. Here’s the thing I want to try with an N95. It apparently supports UPnP. So I could theoretically have it connect to my wireless network when I’m home, and freely pull up videos on the phone with my PS3. I bet if I did that on battery, the phone would be dead in 10 mins, but what a great way to support a standard, and a great payoff in functionality for doing so. Anyone try something like this yet?

FACEBOOK
It’s good that Russ and Mike got to try out their initial FBML goes mobile offering. Yeah, Facebook rushed it. But I still believe in the idea. Dustin honestly seemed excited about mobile, and pushing this stuff out, no matter how alpha it is, shows some real enthusiasm for mobile as a company. So they get the props counter incremented for that move in my book.

Not many people are talking about Facebook’s love fest with RIM and integration with Blackberries. The integration is cool, yeah. But what has been almost COMPLETELY ignored by the media and bloggers is the way they get FB messages to the devices. They piggyback the push email signal! No polling, no extra communication from the device. Saves battery life, and makes any alert reception very very prompt. What an awesome backbone for alert delivery, and it’s the first time I’ve heard of this being used for a third-party app.

MOBITV
These guys keep getting stronger each year. For the last couple CTIA’s the question has been “will IPTV be able to compete with emerging tech like MediaFLO?” The answer this year seems to be “yes!” Browsing their booth, I saw a lot of very mediocre phones running their service very well (that’s a compliment - you can’t just be on high-end phones to win this game). Apparently, they’ve hit the 3 million subscriber mark, and have extended their contract with Sprint. They’re currently trying to get their channel changing time down to 2 seconds to compete with MediaFLO, but they say they’re on track to doing it.

It’s hard to cover all of my CTIA thoughts sitting here after the fact. A lot of the smaller but still cool things get lost in the data compression in my head that happens after every conference. I send a lot of my real-time thoughts to Twitter, but the character limit there is too restricting. Back to Shozu. I would have loved to post pics with thoughts from the show floor. Maybe in Jan, they’ll sneak some better third-party blogging apps onto the iPhone.

I’ll get right to it. On mobiles, discovery of off-deck (and sometimes even on-deck) content remains tough.

Clearsping is enabling content owners to link their mobile content with Web widgets, creating a unified distribution push giving content greater viral sharing capabilities. Check out the product page.

Users can transition to mobile content directly from Web widgets through SMS. Additionally, Clearspring provides a WAP-based sharing service that lets users send mobile content to a friend, or place associated Web widgets on a number of social networking and aggregation sites right from the phone.

We’ve created a content and platform-agnostic, round-trip syndication platform that shares content from Web-to-mobile, mobile-to-mobile, and mobile-to-Web, with integrated analytics. No matter if you have a mobile application, a WAP site, or a mobile widget written in one or more of the mobile widget platforms coming onto the scene, you can create a unified campaign between your Web widget and all of your mobile content formats with Clearspring’s Mobile Sharing Service.

TRY IT OUT
Check out the (admittedly trivial) widget below. Click “Get & Share”, click on the phone, and send it to your mobile. Once at the mobile content, click “Get & Share” again and use the WAP sharing service to send it to a friend or get the Web widget.

We’re working with some big partners and will have some major web to mobile content soon.

Below I get a little promotey. (Is that a word?) But if you want to learn the space a bit more and see how we fit in, keep reading. Please keep this in mind - beeing in the mobile industry for a while - both developing user-facing apps and infrastructure components - I know the frustration of having some awesome content or service and having ineffective or prohibitively expensive options for getting the word out. I helped Clearspring develop this product because I believe widgets are a powerful way to blend the online world of desktops with connected mobile devices, and I’m excited about it as a mobile enthusiast, and a content owner.

We’re just getting started, so I’d love to hear your suggestions. What mobile pain points can we help solve with widgets, viral syndication, and awesome analytics? I’m all ears, and I’ll be at CTIA next week hoping to have some great conversations around this.

OK. On to the details…

THE WAVE OF WIDGETS
On the desktop, widgets have taken the Web by storm, and suddenly users are bringing content to themselves rather than going to the content. With the rise of social networking and content aggregation sites like Facebook, iGoogle, MySpace, etc. many people are logging on and staying within a single domain for the entire duration.

Whether or not any of the current community sites continue to thrive, the trend is moving towards content aggregation rather than disperse content seeking. Widgets play an enormous role in this, and many companies have adopted widget strategies to set their content free. Widgetized content is able to be virally shared through social networks across all of these aggregation sites, and implant themselves in places where interested users will view it very frequently.

EMPOWERING USERS
Widgetized content is personal, it is bite-sized, it is perfect for phones, and a user should be able to send a widget to their phone as easily as they can move it to another blog, aggregator, or social network page.

EMPOWERING CONTENT OWNERS
Content owners should start thinking about their content pushes as being unified. So frequently, I speak to media companies and other content owners who completely separate mobile from their online content efforts. Many times they even draw upon the same backend for their data, but they segment the teams, and they segment the business decisions and treat them like completely different efforts.

Sure, platforms have differing challenges, and it is important to create the right presentation on mobiles, but the overall efforts of getting content to users should be unified, and mobile shouldn’t be treated as a second-class citizen. We’re hoping that our unified syndication and analytics will help promote that kind of thinking.

That said, I’m very excited by the attitudes of our initial content partners that are going mobile with us. Though I can’t say their names just yet, working with huge media players that have such forward thinking attitudes about digital content on desktop and mobile devices is really refreshing.

WHO SHOULD USE THIS?
If you have content that you are currently considering a mobile push for, consider a Web widget push as well and create a unified syndication and tracking mechanism with Clearspring.

If you are making a new mobile service offering, like one of the many mobile-only social networks popping up recently. You need to link users to your service from where they are online. Use widgets to create a connection to users on the Web, and allow word of your service to spread virally. Sure, you can use mobile ads, and buy a shortcode and put posters up in the subway to get your shortcode out there. But widgets will connect existing users to your service wherever they are, while at the same time promote it. No other mobile discovery mechanism can do that.

Are you making a mobile game? Make a trial version for the Web, turn it into a widget, and syndicate with Clearspring. People will play it online and want to get it for their phone. Plus you can link out to the mobile sharing service from your game and let it spread virally from phone to phone. Beats begging for space on a carrier deck, no?

SHORTCODES, SHORTCODES
To send links to mobile users (Web to mobile and mobile to mobile), we have a dedicated US shortcode and an international SMS service. Don’t want to use our shortcode? You can use your own, as we have a flexible SMS api that can be linked to your SMS provider.

But if you want the geek cred, you might want to consider using our shortcode. We might be biased, but we think it’s the 31337-est shortcode on the planet. Ahem.

One of the best things about the Danger Sidekick platform, in my opinion, is its ability to sync everything to the web. You can see your pics, manage your contacts.. tons of stuff - and it’s a perfect backup solution for the everyday user.

Dashwire seems to be aiming at that functionality for many other devices.

Check out the demo video over at Dashwire.

One of the big bonuses - many users have no idea how to get pictures off their phones. The carriers certainly don’t make it easy, unless they’re monetizing it somehow. Dashwire bypasses all that, with no real work on the part of the user. Awesome.

Sign up for a demo account and check out the support list. They really have a good amount of devices for a startup. The problem is that they’re all Windows Mobile and S60, which are platforms that promote sideloading and offloading. People with these platforms usually have many ways of getting pics off their phones already. What we really need is integration with featurephones, and that’s not going to happen without OEM participation.

I wish these guys luck. It’s a great product, but they need to get themselves into the mid-range phones to achieve the use-cases this service is ideal for.

From James Gosling, Sun’s VP (via CNet)

“We’re trying to converge everything to the Java SE specification. Cell phones and TV set-top boxes are growing up,” Gosling said at a Java media event here Wednesday. “That convergence is going to take years.”

OK. I’m with you so far. through your big example of JavaFX is only a consideration for OEM and carrier apps right now, not for 3rd-party development. Lots of bizdev needs to happen before that bears fruit.

This is where I disagree:

“For Java ME, there are a large number of these JSRs for various features. That posed a challenge to Java’s original tagline, “write once, run anywhere.”
…
“Java SE has a much richer basic set of abilities, so using it instead of Java ME could at least in principle restore some of Java’s promise of software portability.”

Let’s list a few of these JSR’s:

• 180 SIP API for J2ME
• 179 Location API for J2ME
• 135 Mobile Media API
• 281 IMS Services API

Tell me, which one of these is solved by a standard J2SE implementation? Ok sure, you’ll probably get some help with JSR280 - XML API for JavaTM ME, but JSR’s like that weren’t the problem.

The biggest problem I found with J2ME was Sun’s substandard certification for virtual machines running on different handsets. Vendors create these VM’s for their platforms and get them Sun-certified. But so many things slipped through the cracks. Some famous ones I’ve run across have been color palette corruption depending on how you load bitmap assets (I’m looking at you SonyEricsson), to UI lockups if you do any networking in a background thread.

The only true way Sun could even these bumps out would be to produce the VM’s themselves - or at least have a much much better certification program. My question now is if JavaFX implementations are being developed by Sun or OEMs? They’re working hard to get OEMs to buy in right now but I can’t find who’s actually doing the integration work. Anyone know?

Well I got the official email this morning:

Wonderful Jaiku users,

Exciting news, Jaiku is joining Google!

While its too soon to comment on specific plans, we look forward to working with our new friends at Google over the coming months to expand in ways we hope you’ll find interesting and useful. Our engineers are excited to be working together and enthusiastic developers lead to great innovation. We look forward to accomplishing great things together.

In order to focus on innovation instead of scaling, we have decided to close new user sign-ups for now. But fear not! All our Jaiku services will stay running the way you are used to and you will continue to be able to invite your friends to Jaiku.

We have put together a quick Q&A about the acquisition at http://jaiku.com/help/google

Jyri Engestrom and Petteri Koponen, Jaiku Founders

I signed up for Jaiku before Twitter, due to the fact that it has tumblelog functionality with made my microblog stream richer. I also had a Nokia N73 at the time and liked the idea of having a native app. But then I realized I knew no one on the service, and that it’s completely useless without people. So I kept my account open but moved to the Twit.

Turns out, I switch between phone platforms too frequently to care about their single platform app. Twitter works fine using PocketTweets, and I find that for updates, SMS is quicker than anything else. Also, a bunch of Web 2.0 apps support Twitter and not Jaiku, like Remember the Milk. So although I like Jaiku better as a platform, Twitter beats them at community and added services. Now reliability is another matter…

So what’s Google going to do? Are they going to buy and park them like Dodgeball? I think the Jaiku guys are smart enough to stick some clause in the agreement to stop that from happening, so hopefully not.

But aside from that, assuming that Google isn’t interested in buying their user base (Twitter’s is much more compelling) what tech does Jaiku bring to the table that Google couldn’t build relatively quickly? Jaiku has more IM adapters, AFAIK Google only has their Jabber clone. But you can get that functionality quickly with open source. Google has an SMS backend already, and they probably have a ton of RSS libraries floating around from Google Reader alone. Building out a Jaiku-style back end and presentation layer given all those core components is easy enough that I’d expect them to build rather than buy.

Who knows. Maybe they wanted a bootstrapped community more than anything, and Twitter asked for too much money.

I’ll stop speculating now. All I know is Twitter better step up their game. Google is probably going to come out with integration of Jaiku to all of their apps - post to Jaiku from Google Reader, view your Jaikus in Gmail, restrict your searches to microblog posts. I’m happy to stay with Twitter, but they won’t be the cool kids on the block forever, so I don’t think they can assume all the third-party apps will keep writing to their API without them lifting a finger and doing some more proactive integration.

RSS aggrigator taking over your life? Yeah, I know the feeling.

FeedHub was recently released and aims at trying to trim the fat off the torrent of posts, news and data flowing down your feeds. I’m all for that, but being in such fine control over my sources and where my attention focuses, I’m began using it wary of giving up my control. Let’s see how it turned out.

I’ll cut right to the chase. FeedHub actually does a fairly nice job of dwindling down my feeds, but only after I trained it via its memes interface for a bit. As FeedHub scans your feeds, it detects memes, which it then places into a simple interface where you can define their relative importance to you. You can also create memes by keyword and build up an importance profile. FeedHub uses this profile to decide which articles to include in your feed.

Speaking about the resultant feed, you have a lot of control over how you want it trimmed. You can trim to a specific number of posts per day, a percentage of your original feed, or have it just knock off the least important stuff or only send the most important. Very nice options. If you’re feeling overwhelmed, this gives you a knob you can turn to dial down your information flow, which would make that backlog of data during your vacation disappear to only the important things. You can get caught up in a fraction of the time when you get back to work. Awesome.

Training is relatively simple from the feed as well. FeedHub sticks a control at the top of each item where you can slap it on the wrist for including it, or jump over to your preferences and do some more detailed training.

For daily reading, however, my brain has been trained to handle the rush of data, so giving it up is tough. The biggest issue I have is that in the end FeedHub is only a binary filter. You either get a given item or don’t get it. For example, the other day there were a lot of articles about Nokia becoming a broadly-scoped software company, and FeedHub neither put them in the memes interface, nor showed them to me. I could go and create a specefic meme for it, but what about future articles with unknown memes that I really would like to see? FeedHub currently doesn’t have a concept of meme relationships - or at least hasn’t exposed it to the user. So I have no reason to believe it will make the right choices in prioritizing new memes.

But back to training. To solve this binary filter problem, I’m wondering if a periodic RSS item could be inserted that includes at least some of the top articles that were removed. Even if it’s a single item that includes a list of 20 or so titles of filtered posts. None of them should be in my “don’t care” classification, but it should include ones with completely new memes, or that have been filtered because they fell slightly below an importance threshold. That list could include links to raise or lower their importance or completely filter them. This would allow me significantly more detailed in-situ training and with minimum extra items to dig through.

My ideal situation would be integration with my aggregator (Google Reader). I’d love to see the most important things filtered to the top, and a gradation of color indicating I’m getting to less important items. Training it would be a matter of hitting a button to shift its importance up or down relative to where it is. That, combined with a rule to discard any unread messages under a certain importance threshold after a certain time would really bring the RSS experience to a new level.

But FeedHub isn’t an aggregator and it has no relationship to an aggregator (yet). So given the situation, I think they do a great job.

My uses for FeedHub at this stage of its development are the following.

1. Keep a dwindled feed next to my normal feeds that I can train. When I feel that it’s a good representation of my important news, I can declare RSS bankruptcy at any time with my raw feeds and just get the important stuff from it. That alone is an awesome use case for this technology.

2. Keep a severely dwindled feed (20 items per day) that I can use on my phone. In normal circumstances, I need to have my mobile RSS reading synced exactly with my web aggrigator. That’s why I use Google Reader mobile. But in this situation, I could have a small separate mobile feed that keeps me in the loop while I’m traveling for the day. When I get home, I can view the dwindled feed in my aggrigator (from #1 above) and zap the rest of the raw stuff.

To help with this mobile feed, I’d like to see a meme to filter out posts with media in them such as YouTube videos, or anything that’s not viewable on mobile devices. Mobile-specific memes would help the mobile filtering situation a lot.

All in all, FeedHub has made a great first step. Hopefully using it in parallel with my normal feeds for now will allow me to build up a training profile that I trust as it continues to add features and integrate more.