The APIs themselves are quite powerful and easy to work with.  Complications arose when it came time to convert the temporary JSAPI tokens to their oAuth equivalents.

https://developer.linkedin.com/documents/exchange-jsapi-tokens-rest-api-oauth-tokens

First things first, LinkedIn only passes the JS API oAuth 2.0 token over an SSL connection.  Slighly annoying from a development perspective, as I wasn’t normally running SSL on my local Mac. That being said, getting nginx setup through homebrew was straight forward. It works fine with a self-signed certificate as well.

Once you have SSL setup

<script type="IN/Login" data-onAuth="onAuth" data-onLogout="onLogout"></script>

function onAuth() {            

	$.post("https://" + window.location.hostname + "/my-app/token-exchange");

}

After a successful LinkedIn authentication, the onAuth() function will be invoked and a POST made to your backend resource over SSL.

def token_exchange(request):

 oauth_token = None oauth_secret = None

 if request.session.get('linkedin_oauth_token'):        

	oauth_token = request.session.get('linkedin_oauth_token')        

	oauth_secret = request.session.get('linkedin_oauth_secret')    

 else:        

 	oauth_token = request.COOKIES.get('linkedin_oauth_%s' % settings.LINKEDIN_API_KEY)        

	oauth_token = urllib.unquote(oauth_token)        

	oauth_token = json.loads(oauth_token)

 consumer_key = settings.LINKEDIN_API_KEY        

 consumer_secret = settings.LINKEDIN_SECRET_KEY        

 access_token_url = 'https://api.linkedin.com/uas/oauth/accessToken'

 access_token = oauth_token.get('access_token') consumer = oauth2.Consumer(             

	key=consumer_key,             

	secret=consumer_secret

 )

 client = oauth2.Client(consumer) resp, content = client.request(access_token_url, "POST", body='xoauth_oauth2_access_token=%s' % access_token)        

 request_token = dict(urlparse.parse_qsl(content))        

 oauth_secret = request_token.get('oauth_token_secret')        

 oauth_token = request_token.get('oauth_token') request.session['linkedin_oauth_token'] = oauth_token        

 request.session['linkedin_oauth_secret'] = oauth_secret

The code snippet above is a django view that requires oauth2.

And that’s it.  You’ve now successfully converted JS API tokens to oAuth tokens in a Python/Django environment.

My latest forays into Akka came as a result of some investigation I was doing for a new platform at work.  Basically, I was looking for something to handle job/task execution across a number of different physical machines.

One option would have been to rig something up with Groovy, a simple daemon exposing a RESTful API (or fetched jobs from a message queue) that used GPars behind the scenes to manage concurrency on each individual machine.  Job state would be bubbling up to an overarching service that could manage scheduling across a few machines.

We’re already using Groovy/Java on a couple other newish projects and ideally the solution to this problem would also fit in that box.  Scala, as interesting a language as it is wouldn’t be fair to thrust on a team who’s language of choice has historically been Perl (but is moving towards the JVM).

And so I set forth to survey the landscape for Akka + Java/Groovy examples.  In particular, I was looking for something that combined Remote Actors + HTTP/REST, perferably being built in Maven (but I’d take some Gradle).  I found a few, the chat server ones looked relevant but a little bit dated.  I ended up putting something together myself and pushing it to github.

https://github.com/ajordens/akka-job-scheduler-example

It uses Akka 1.3 and has a server component (runnable via mvn jetty:run) that hosts remote actors and a simple REST resource.  There is also a client service that can be run on multiple machines, it polls the server for jobs, executes them and fires a response back.

The example is trivialized but it does tie a number of different pieces together.  One bit of initial grief I had was around trying to get akka-http running in Jetty 8.0.4.  It wouldn’t fly, and I ended up having to fall back to 7.4.0.

Enjoy.

You know, the little step-by-step walkthroughs of what’s new/changed.

In the interest of being lean, I’ve been focusing on getting a simple foundation in-place and building it out based on personal needs and feedback from the community.

The ONE Feature Introduction:  http://onesecondshy.com/one-feature-introductions

The ONE Feature Introduction walks a user through particular features or aspects of your site.

Think of it as a guided tour that helps ensure your users can effectively use your application.

At this point, I’m looking to broaden my reach and gather the next round of feedback as well as getting it rolled out and active on additional sites.

From the development side, it’s a fairly simple low-touch integration.  That being said, I’m more than willing to jump in and answer any questions / ensure that the integration remains as simple as possible.

So if you build web applications, or run a site with regular content updates, I’d love to talk more.

Happy Holidays!

Everyone around me had upgraded, at _least_ to a 3GS, but my unwillingness to give the Canadian carriers any more money (early upgrade fees) kept me dated.

That is until yesterday when an unlocked 32GB 4S showed up on my door step, and I must say that the difference is remarkable… unbelievable really.

So it’s with pleasure that I can now dust off instagram, yelp, rdio, twitter, runkeeper and facebook on the phone. Constant crashes on the 3G basically forced me to use the phone as… a phone, for calls and text messages.

On the iOS 5 front, it’s definitely breathed new life into my trusty iPad (mind you I was running 4.2 previously).

Split keyboard, tabbed browsing, wifi sync and personal wifi hotspots to the rescue.

That’s it for now.

My underlying motivation is to try and help developers make better sense of the app store (different categories, countries, competitors, etc.). It’s also scratching a personal itch of mine around application discovery by providing a historical performance context around each application.

I’m looking for feedback and to engage with more developers.

Key Features:

• 30 day trailing performance for any ranked application (actually have > 12 months worth of data but am only actively exposing a months worth)
• top paid / free / grossing chart positions vs. pricing (measure the actual impact of a price adjustment on gross position)
• ratings per app per version (what impact does releasing new versions have on chart positions)
• apps per developer (how are all the apps from a particular developer currently or historically performing)
• iPhone/iPad/Mac app tracking

If you’ve published an application or are actively developing one, I’d like to hear your thoughts. I’m quite passionate about helping developers better understand their target market/competitors, and this is my first attempt at addressing some immediate needs.

If you’re looking for an alternative way to discover quality applications, I’d encourage you to check it out as well. I’ve found the historical performance of an application and it’s developer to be a good indicator of it’s quality. We all know that dropping your price certainly doesn’t improve the quality of your app and yet it often can boost your ranking.

Feel free to follow us on Twitter: @appiity

Identifying trends on the iTunes App Store and making it easier to find great apps!

At that time, it was the cumulation of over a years worth of data from the iTunes Marketplace.  When launched, it officially supported US, Canada and UK app stores.

I’m happy to say that since then, the platform has been expanded to include:

• Austrailia
• China
• Germany
• France
• Hong Kong
• Isreal
• Korea
• Japan
• Sweden

… with additional countries actively being brought online.

It’s been interesting to look at charts (both historical and present), there is actually a lot of variability on charts country-to-country.  Check it out for yourself!  Only the last 30 days is being exposed via the web interface, but there is a lot more data available.  Contact me if you’re interested in learning more.

Further to that, I’m also reaching out to developers with the goal of understanding the pain they’re experiencing when building apps.  If you have five minutes, I would appreciate your feedback on this survey.

Lastly, I’m proud to announce the first public preview of our application search.  It’s live and links into detailed application pages that show, in detail, an applications performance over the past 30 days.  Want to understand the impact of an Electronic Arts sale?  We have that data.

Take some time to explore the site, I’d love your feedback.

That’s all for this week!

He’s a scam. Do not pay him anything. He will not deliver.

I unfortunately made the mistake and this is my warning to anyone else in a similar situation.

• iPad Chart and Data
• iPhone Chart and Data

There’s still some actual analysis to be done but you can clearly see the state of EA’s applications before, during and after the sale.

It was definitely a winning move for EA.

At the height of the sale, EA had more than 20 iPhone apps appearing in the Top 100.  Prior to the sale starting, there were only 10.  Today, 4 or 5 days after the sale ended, there are 12.

The true impact is better measured on the Top Grossing charts.

For a better idea of what those impacts were, checkout Electronic Arts on Appiity.

The follow data has been pulled from the Top Paid US iPhone and iPad charts.

Prior to the sale, EA had a total of five ranked iPad applications (Tiger Woods, Boogle, Scrabble, Max and the Magic Marker, and Monopoly).

The highest ranking iPad application was Max and the Magic Marker at 31st spot.

With regards to the iPhone, EA had a total of ten ranked applications (NBA Jam, The Sims, Scrabble, Battlefield, NBA Elite, Tetris, Tiger Woods, Mortal Kombat, The Game of Life, and Monopoly).

The highest ranking iPhone application was NBA Elite at 28th spot.

In terms of pricing, the average prices for EA iPhone and iPad applications pre-sale was > $2.99.  During the sale, most applications were dropped to 99c. The few apps that didn’t receive a price cut maintained their chart positions but didn’t make any gains.

As of today (2011-04-25), Electronic Arts has fifteen iPad applications and twenty-two iPhone applications on the top 100 charts.

Many of these applications had fallen off the charts in the weeks and months prior to this sale.

10 of the top 20 paid and 3 of the top 20 grossing applications (iPhone) now belong to EA.

Looking at the iPad, it’s even more exaggerated with EA controlling 7 of the top 10 paid and 4 of the top 20 grossing applications.

This is big business and Electronic Arts are one of the masters when it comes to timely price manipulations.

We, as consumers, are extremely price sensitive.

The raw data underlying this study was extracted from appiity.com and is available below:

iPhone

iPad

About a year ago, I started collecting the data to drive the site that has eventually morphed into Appiity.

If you haven’t seen it, Appiity is an application that enhances the iTunes App Store charts with historical information about each application.  It answers the questions: ‘Does this app have staying power or is it a one hit wonder?’ or ‘Does this vendor have a history of releasing quality applications or do they just manipulate their prices enough to maintain a particular rank?’.

Over the past month or so, I’ve been looking to make application discovery a little more personal.  I believe that finding great applications is an inherently social activity. We’re more likely to try an application recommended by friends than one with a significant advertising budget.

With that, I’m proud to announce that my latest experiment is ready for a bit more public attention.

Appiity Discover

It’s still very much a work in progress, but for those of you wanting to help improve application discovery, check it out.

Appiity Discover aims to help you find better applications through the development of a personal profile. This profile defines who you are and the types of applications you have indicated a preference for.

We realize that this goal cannot be accomplished overnight and hope that this first incarnation will be a stepping stone towards realizing our goals.

In the meantime, we hope you’re willing to help us out. In return, we’ll do our best to help you find that golden nugget of an app that would have otherwise been missed.

Follow us on Twitter: @appiity