Search This Blog

Tuesday, January 31, 2012

Create a Bootable Ubuntu USB Flash Drive the Easy Way - How-To Geek

Create a Bootable Ubuntu USB Flash Drive the Easy Way - How-To Geek: "UNetbootin "

Create a Bootable Ubuntu USB Flash Drive the Easy Way

We’ve already covered how to use an Ubuntu Live CD to backup files from your dead Windows computer, but using the boot cd can sometimes be a little slow. We can speed up the booting process by installing Ubuntu to a bootable USB flash drive instead.

To accomplish this, we’ll use a tiny software package called UNetbootin, which is designed to make the installation process simple and easy.

Create the Bootable Flash Drive

You’ll first need to download the UNetbootin software and save it somewhere useful, since there’s no installation required, just double-click to run.

I chose to use an already downloaded ISO image of the Ubuntu installation cd, and then chose my flash drive, and clicked the OK button. Yes, this step is as simple as that.


The process will extract the files from the ISO image (or download them), copy them to the flash drive and then install the bootloader. Depending on what you are installing, this really doesn’t take very long.

Once the process is completed, you’ll be prompted to reboot… which you don’t necessarily have to do unless you want to test booting the flash drive on the same machine you are using.

Otherwise you can hit the Exit button.

Make Sure the Partition is Active

If you get a boot device error when you try and boot from the flash device, it could be that your partition is not marked as active. What we’ll do is use the command line diskpart utility to fix this… if you are in Vista open an administrator mode command prompt by right-clicking and choosing Run as Administrator.

Now you’ll need to run this command to figure out the number of your flash drive:

list disk

This will show you the list of drives, and you will use the disk number in the “select disk” command:

select disk 1

select partition 1


The “active” command will actually mark the current partition as active, which is why you need to select the disk and then the partition. At this point you should be done.

Booting From the Flash Drive

Now that you are all finished, you can try and boot from the flash drive. Every BIOS is different, but most of them will have a message like “Press F12 for the Boot Menu”, which is highly suggested. The boot menu will allow you to select the USB drive as the boot device. (apologies for the simply horrible camera phone screenshots)

Instead of the regular Ubuntu boot menu that you might be used to seeing, you will see the UNetbootin menu, which has essentially the same options.

Useful Note

If you are having issues booting Ubuntu Hardy (8.04) on a Dell machine, you might want to switch back to using Ubuntu 7.10 instead.

I was able to boot all the way into the desktop in about 35 seconds using the flash drive… way faster than using the regular live cd.

Google Analytics Blog: Google Analytics Enhancements for Mobile Apps

Google Analytics Blog: Google Analytics Enhancements for Mobile Apps:

November was a busy time in Google Analytics. In particular, the Mobile App Tracking Team has a few things to announce.

  • EasyTracking Library - automatic session management, better integration with Google Analytics SDK
  • Updated Google Analytics SDK - More reliable method for sending hits, Android Market referral issue fixed, available via the Android SDK manager
  • More samples - new open source application aimed to help reduce the ramp up time for new developers who want to track their apps

EasyTracker Library
We’ve created EasyTracker libraries for both iOS and Android. The EasyTracker library will enable tracking of your application down to the Activity (or UIViewController for iOS) level with almost no coding required on your part. See the ReadMe file and source code for details. These Libraries are intended for use with the standard Google Analytics SDKs and should make it very easy to add standard tracking to your applications.

Another advantage to using the EasyTracker library is session management. As many developers know, it’s not always easy to determine whether your application is active and when to start a new session. The EasyTracker library handles this for you. It will determine when your application has been put into the background and will start a new session automatically.

The Android version of the Library not only provides for easy tracking, but also ensures that all calls to GoogleAnalyticsTracker are done off the main UI Thread. Using this library should address responsiveness issues some Android developers have seen using the Google Analytics SDK.

We’ve adapted the Android Notepad sample application to use the EasyTracker library, just to show you how easy it can be.

You can find the libraries and sample applications at Check the downloads section for the libraries. The source for the libraries is available in subversion as well. Drill down intotrunk/src/tracking/mobile/android/EasyTracker for Android andtrunk/src/tracking/mobile/ios/EasyTracker for iOS. The Notepad sample application is there as well. We’ve released them as open source and contributions to making them better are welcome.

Check the ReadMe files in the libraries themselves for more information on how to use them.

New Versions of Google Analytics SDK for Android and iOS
We’ve released version 1.4 for iOS and version 1.4.2 for Android. The iOS version of the SDK has one new feature. Both versions contain several bug fixes as well. Read on for details.

We’ve added a new method, dispatchSynchronously, that blocks while it dispatches hits. It won’t return until the hits sent have either been acknowledged by the Google Analytics servers or the specified timeout period has elapsed. This provides a more reliable method for sending hits before your application terminates or goes into the background.

We’ve also addressed several memory leaks and crashes reported against the SDK.

More details on the new version of the SDK can be found at

The Android SDK will now handle referrals from the Android Market properly. This applies to autotagging as well.

We’ve fixed several other bugs in the Android SDK. Check out the details at

Google Analytics SDK now available via the Android SDK Manager
We’ve added the Google Analytics SDK to the Android SDK Manager. You can download the latest versions using the Android SDK Manager instead of checking the website for updates.

Of course, this only applies to the Android version of the SDK.

We’ve released an open source application for both iOS and Android that exercises all the APIs for Google Analytics that are available to Mobile Application developers. You can find them at trunk/src/tracking/mobile.

New Home for the Mobile Tracking Documentation
The Mobile Tracking documentation has moved. It now resides with the rest of the Google Analytics tracking documentation. Check it out at

Reporting Problems and Feature Requests
We really value your feedback. If you are having problems with the SDKs, let us know by posting them on the Google Analytics issues website at Use the component MobileTracking when entering an issue or looking through the list for issues already reported against the Mobile App SDKs.

Please stay tuned for more exciting news regarding Mobile Application Tracking with Google Analytics.

Games Decreasing In Popularity On Android, Entertainment Apps On The Rise | TechCrunch

Games Decreasing In Popularity On Android, Entertainment Apps On The Rise | TechCrunch:

Games Decreasing In Popularity On Android, Entertainment Apps On The Rise

Sunday, January 29th, 2012

The mobile app ecosystem is growing so quickly – exponentially, even – that sometimes it’s hard to see the larger trends occurring in the space. To really gain insight, it helps to look back over a longer period of time, like a year for example, in order to take stock of the changes taking place.

To that end, I reached out to mobile app search company Chomp, who kindly obligated my curiousity by packaging together its first ever annual app search analytics report. The report digs into the data from Chomp’s 1 million app searches per month, to reveal trends in app prices, category share, top apps and more.

Note: The data below is conclusive for 2011, but Chomp’s Android app was only released in February. Chomp felt the graphs looked better starting from May, when data stabilized on both platforms.

To begin, Chomp took a deep dive in category share trends. The very first finding was a bit surprising. Chomp found that games are now seeing a decreasing number of downloads on Android. That’s interesting because we’ve come to think of the mobile gaming market as this booming space, which is defining itself as the top category driving mobile app download growth. But, based on Chomp’s data at least, this is only true for iOS. Over the course of the year, games became an increasingly large percentage of iTunes downloads, but have been decreasing on Android.

In December, for example, games were 36.1% of iTunes downloads and 22% of Android downloads.

So if not games, what’s growing in popularity on Android? I would have guessed Utilities but would have been wrong. It’s Entertainment apps. This includes apps like Netflix, Talking Tom, Crackle and Flixster, to give you an idea.

Meanwhile, on iTunes, music and fitness apps have been seeing increased popularity, and utilities are decreasing. Again, this goes contrary to popular belief that normal users (the non-technical/non-early adopters) are looking for more mobile tools – things that help them get a specific task done. Clearly, they’re also enjoying apps Spotify, Mog, RunKeeper and Nike Fitness, too.

In terms of app pricing, iTunes has seen purchase prices rise by 50% while Android prices have dropped. During December, the average iTunes app purchase price was $0.67 while the average Android app purchase price was only $0.09.

But here’s another interesting twist: the average purchase price for just paid apps is higher on Android ($3.17) than on iTunes ($2.41). This is due to iTunes’ large number of $0.99 downloads, however, which throw off the average.

It also has to be noted that Android sees a really low number of paid downloads (5%) compared with iTunes (22%), according to Chomp.

Finally, Chomp revealed its top 10 iTunes and Android apps of the year. These are different lists than what the app store itself would show, as they’re based on Chomp’s searches, not raw download numbers.

On Android, the top apps were:

  1. Pandora
  2. Facebook
  3. Lookout Mobile Security
  4. Zedge Ringtones
  5. Advanced Task Killer
  6. Handcent SMS
  7. Dolphin Browser
  8. The Weather Channel
  9. TuneIn
  10. GasBuddy

On iOS, they were:

  1. Angry Birds
  2. Facebook
  3. Dropbox
  4. IMDb
  5. Pandora
  6. Angry Birds Seasons
  7. SoundHound
  8. Netflix
  9. Angry Birds Rio
  10. Instagram (also Apple’s pick for “app of the year”)

These lists really help to showcase the differences between the two mobile platforms. Although Entertainment apps may be on the rise, Android users are still focused on tools and utilities, from mobile anti-virus programs and task killers, to things that will enhance or replace stock apps, like Handcent SMS or the Dolphin web browser. It will be interesting to see if, by next year, these trends will have changed at all, allowing the increasingly popular Entertainment-focused apps to emerge on the top Android apps list.

On iOS, there’s definitely more focus on having fun, with three versions of Angry Birds, music and photo apps, and Netflix, making the cut. The only remotely “utility-like” app on iOS’s like would be Dropbox.

Overall, the most interesting thing about this report was how it helped reframe some common misconceptions – that Android users only like utilities, for instance, or how cheap the paid Android apps are. They may be few and far between, but they’re not necessarily cheap.

Chomp’s full report includes its own Staff Picks section and a look into Chomp’s traffic search trends, too. (Contact the company if you’re in need of a copy). And may I just say: more data like this, please.

Piwik best alternative to Urchin! Web Analytics via Log files import - Piwik

Piwik best alternative to Urchin! Web Analytics via Log files import - Piwik:

Urchin development and support will be discontinued by Google as of March 2012. Urchin was a Log Analysis softwarebought by Google in 2005. They have used this software as a base for Google Analytics and have now announced they will focus exclusively on Google Analytics. We have since received a number of emails from Urchin users, asking if Piwik could be setup to do the Log Analysis the same way Urchin was doing it, and import all past logs in a Piwik server.

We are happy to say that we have been developing a powerful, simple to use script that will analyse your webserver log files (Apache, Nginx, IIS, Akamai, etc.) and will import visits, page views in Piwik.

We hope that in the next few months, Piwik will become the best alternative to Urchin and AWStats (and others).

Piwik Features when used to Import Log files

Piwik normally uses Javascript code to track visits and pages. This new script will also make it easy to track visits by importing one or many web server log files in Piwik. This is useful if you are not able to add the JS code to the websites, or if you wish to import large amount of historical data at once, or if you are looking for a software that does the same thing as Urchin, AWStats, Webanalyzer or Webtrends.

Some features of the Piwik Log Import script include:

  • Great performance, we have tested to track several millions of log lines per day with success. See Piwik for high trafficcheck list.
  • Bot traffic is automatically excluded: to keep your Web Analytics report clean and useful, and increased performance.
  • Piwik can track websites with the standard Javascript code, and other websites could be tracked by importing the access logs. For example Javascript tagging for website 1 and 3, and Log import for site 2 and 4. We expect these hybrid Piwik servers to become a common configuration among the community.
  • File downloads appearing in the logs will be automatically tracked as "Downloads" in Piwik
  • Because Logs will be imported via the Tracking API, all Piwik features will be supported (Goal tracking based on URL, IP Anonymization, Visitor log, etc.)
  • Some reports will have no data because the log data is more limited that data obtained via Javascript. For example: screen resolutions, Supported Browser plugins, Custom variables, Ecommerce Analytics will not work.
  • This script will effectively replace Apache2Piwik, the new tool providing more features and better performance.
  • In later versions we are planning to support Log reprocessing, Error code tracking, Search engine & spam Bot tracking, Feature to use the logs to enhance existing JS tracked pages, and more (based on user popularity and feedback).

This script will be written in Python and will be released under the GPL license, for Free (just like Piwik!)

Perfect for Web hosts company and Web agencies, but also for a one-off log import

The script will have 2 modes:

  1. Web Host – Web Analytics Provider user
    This mode is ideal for Web hosts, where new websites are often added in the access logs, but the Piwik admin does not wish to create manually each website. The script will automatically detect the Piwik website ID to track based on the URL being parsed: it will look for any Piwik website registered with a URL or "Alias URL" set to this page view host. If a website with the hostname doesn't exist, a new website is automatically created for this URL.
    A summary is then emailed to the Piwik Super User so he/she knows which websites are automatically created by the Log import script, so you can create users or assign permissions to view these new websites.
  2. Simple Log Import for one or a few websites only
    This mode is ideal if you import only a small number of websites or if you wish to control exactly in which websites requests are tracked.
    When a line contains a URL to an unknown Piwik website, Piwik can either ignore all these pageviews, or you can choose to record these unknown pages in a specified "catchall" Website id, to double check they are not legitimate pageviews.
    If these unknown URLs turn out to be legitimate pageviews, you can either create a new website manually, or add an Alias URL to an existing website, so the page URLs are directly tracked in this website the next time you import similar logs.

Join the beta testing group

To be part of our beta testing group, please email us at and mention the Testing of the Urchin/Awstats log import script. Please also mention the number of websites to track, how many pages per day, and if you are willing to test the script and report bugs or feedback.

This work is sponsored by Alwaysdata, a French web hosting company. They provide Piwik as the Web Analytics package of choice, deprecating AWStats, for thousands of their users. They have been using Piwik for a few years and we are finally integrating this Log Import Analytics key feature in Piwik, and ensuring good performance for the script. We want to make it easy for Web hosts and large Web Agencies to use it as their Web Analytics platform.

Goodbye Urchin + Scale of Google Analytics in 2012

The Google Analytics team have decided to focus on the privately hosted Google Analytics (GA) service and discontinue the Log Analysis version (Urchin). At Piwik we are quite simply amazed at the scale and reach of Google Analytics in 2012: GA is used by over 55% of all Internet websites (source). At least 15 million websites use Google Analytics! (source). In comparison Piwik is used on 1% of the Internet (cheers!) and 250k+ websites.

Million of pings (page views) are tracked by GA per SECOND. It is hard to imagine such scale which would make any software developer speechless. We can only congratulate Google engineers and product designers for the work they are doing to track and aggregate so much data, while allowing to slice this data in real time across dozens of dimensions. This is an amazing technical milestone. We also hope that Google users privacy will be respected and privacy standards will improve in the future.

Regarding the end of Urchin, we at Piwik will do our best to provide to existing Urchin users a good experience to upgrade to Piwik, and try the leading Free software platform. If you are a Urchin user and would like to try Piwik, send us an email us with your current setup. We will help and check similar functionnality is do-able with Piwik and the log import script.

Privacy & Security implications of self hosting your Web Analytics data

Ensuring the full control over your customers Log files, Piwik database are important requirements if you are a Web agency or a Web host providing web analytics to hundreds or thousands of users.

The tips in the Privacy page will help ensure that you make the changes to data collection and data retention required by your Privacy Policy. We also focus on Code security and recommend to all Piwik users to spend some time securing the Piwik server.

Piwik also an alternative to AWStats, Webalizer, : modern UI, better performance, and more!

We hope that Piwik will become the leading alternative to Urchin and to AWStats. AWStats was a great tool but we hope to modernize the Log analysis open source software world and make use of all the great Piwik features and capabilities in terms of data analysis and graphing. Users in 2012 and beyond will need a modern interface to access the data gathered from their Web server Access logs.

Monday, January 30, 2012

Smartphone App Usage in India: Google Products Lead the App Race -

Smartphone App Usage in India: Google Products Lead the App Race -

Smartphone App Usage in India: Google Products Lead the App Race
28 January 2012, Team PI

In India, smartphone users spend more than 40 percent of total active time on their smartphones on data-centric activities. A full 25 percent of users’ time with apps is spent on those developed by Google, with the Android Market, YouTube and Gmail leading the way.


Google has the #1 app in terms of reach for 5 out of top 7 genres, while Facebook dominates social networking

Google properties occupy the top spot in five of the top seven genres, by reach. Facebook and Nimbuzz lead the social networking and chat genres respectively.


Time spent with the top app categories
94% of time spent on streaming video goes to YouTube, while 86% of time spent on GPS is with Google. However, when it comes to social networking and chat applications, Google struggles to obtain more than a few percentage points of time spent. [Via: Neilsen report]


Wednesday, January 25, 2012

Nginx/ReverseProxy - Community Ubuntu Documentation

Nginx/ReverseProxy - Community Ubuntu Documentation:

Nginx & Apache

Working with virtual servers (like OpenVZ containers or XEN dom's) you might need a load balancing or a proxy solution to be able to run services from inside those machines through your host's single public IP address.

Here you will find how to solve such a situation using Nginx and Apache. (Apache already has mod_proxy module which can be used instead of Nginx, but the second one is much lighter and scalable, so I recommend you using it.)

Note: This tutorial assumes you already have a working environment like the one described above.

Nginx Installation

Just use your favorite package manager and install nginx package:

sudo apt-get install nginx

/!\ Warning Hardy users! In order to have SSL support please refer to the solution described in bug #395637.

Nginx configuration for proxy behavior

Nginx uses the following directives (as an example) in it's configuration files to behave like a proxy:

proxy_redirect          off; proxy_set_header        Host            $host; proxy_set_header        X-Real-IP       $remote_addr; proxy_set_header        X-Forwarded-For $proxy_add_x_forwarded_for; client_max_body_size    10m; client_body_buffer_size 128k; proxy_connect_timeout   90; proxy_send_timeout      90; proxy_read_timeout      90; proxy_buffers           32 4k;

Create the file /etc/nginx/proxy.conf containing all the informations above. We will include this file in every nginx vhost we need to be a proxy.

Nginx vhost proxy example

Create an nginx vhost by creating a file inside /etc/nginx/sites-enabled/ like this:

server {         listen   80;         server_name;          access_log  /var/log/nginx/access.log;           location / {                 proxy_pass;                 include         /etc/nginx/proxy.conf;         } }

The example above will be a proxy for every domain listed with server_name, and the server which will be used for proxy is underproxy_pass directive.

Done with nginx. Reload it:

invoke-rc.d nginx reload

Apache configuration

This assumes you are working on a machine behind the host that uses nginx as a proxy ( an OpenVZ container, XEN dom or a machine inside your network which is connected to the host we installed Nginx before)

Apache to use proxy

Apache will doesn't know who connects to it except the host ip address we set up nginx. To repair this, we need to install Apache's mod_rpaf. Use your favorite package manager to install libapache2-mod-rpaf:

sudo apt-get install libapache2-mod-rpaf

mod_rpaf requires the following directives inside a vhost to work:

 RPAFenable On RPAFsethostname On RPAFproxy_ips 

Where RPAFproxy_ips can be multiple IP addresses, one of which is our Nginx proxy private IP address.