Ada and Scala developers appear to be seven times more likely than C# developers to be interested in Clojure.  While the small sample size and overall selection bias makes the Clojure survey not entirely authoritative, I found the process of compiling the ratio to be interesting.

Sources:

• Survey of 487 Clojurians
• TIOBE programming index for August 2010

Related articles by Zemanta

• Leiningen ? Clojure (slideshare.net)
• How to setup Clojure from scratch (programmingzen.com)
• Clojure gets sponsored (thoughts.contegix.com)
• Thoughts on Clojure (reddit.com)

Best wishes and heartfelt thanks to Susan Scrupski, Jeff Dachis, and the many helpful members that make the Council such a worthwhile group.

Related articles by Zemanta

• The 2.0 Adoption Council is now part of The Dachis Group (itsinsider.com)
• Peter Kim Brainfood Store Presentation 240210 (slideshare.net)
• Social Business Summit 2010 Sydney (slideshare.net)
• Dachis Group welcomes The 2.0 Adoption Council (beingpeterkim.com)

Image by kenleyneufeld via Flickr

Microsoft SharePoint and Jive SBS are the two biggest players I know of in the one-size-fits-all enterprise collaboration world.  Thanks to the two year lag between consumer web product innovation and enterprise collaboration followup, we now see microblogs gaining recognition as a must-have feature for the biggest players in this market.

This is great news but there’s still a worry that the future of office microblogging will be born on top of some half-baked implementations if MS and Jive get it wrong.  My fear is that an integration of microblogging into Jive and SharePoint will wind up being less enjoyable to use than the standalone microblog networks such as Twitter and Yammer.  I’d really like to see Friendfeed-style realtime activity streams included rather than a me-too microblog that looks good on a brochure.

More important than being a twitteralike:

• Real time update syndication (PubSubHubbub, rsscloud).  Serendipitous Yammer conversations happen when two-plus people are online and reading/responding in real time.  Standard RSS can’t meet this demand.
• Integration of external content. See Facebook news streams for examples.  Automatically slurping in activity on other busy sites like YouTube and Twitter add a lot to your average Facebook news feed.  It actually adds *too much*, which leads to the next need…
• Granular filtering and presentation.  When you get around to allowing the syndication of any and all external content, then you *must* provide filtering and occlusion.  The person who tweets 50 times a day must not ruin the experience of the other 10,000 users of your system.  Users need granular filtering and control if they’re going to get any value out of your system. If we were to overlay our Yammer stream onto our Jive “recent activity” stream right now all of the long-form content would disappear in a sea of low-value Yams.  This isn’t to say that either mode of communication is more valuable than the other, only that we need multiple views: one for only micromessages, one for only long-form content, and one view for everything.
• Multiple modes of access.  Yammer and Twitter have standalone microblogging clients for desktop, smartphone, and web.  A Jive client that rolls microblogging into a one stop solution is unlikely to be as fast or as pleasant as the standalone microblogging solutions.  Ideally I’d like to see your microblog have its own first-class client that runs alongside the standard web+iphone Jive clients you already have.
• Easy interaction with items within an activity stream.  The baseline here is Like + Comment.  Facebook,  Yammer, and Friendfeed do this.

Bottom line

What I’d most like to see out of Jive/Microsoft is a way to bring in near-realtime items from other feeds.  Whether or not you implement your own “microblog” is immaterial as long as I can see Tweets and Yams and similar content brought in immediately via PubSubHubbub and then I can like/comment/share them within the walled garden of our Jive community.

Related articles by Zemanta

• Enterprise Microblogging Startup Yammer Raises $10 Million (paidcontent.org)
• Jive Software Grows Up (bits.blogs.nytimes.com)
• Yammer: Will Viral Work in the Enterprise? (radar.oreilly.com)
• Socialtext 4.0 Launches With Groups, Better Search, And Activity Stream Filtering (techcrunch.com)

Image via Wikipedia

If you work in a multinational corporation like mine, you regularly work with people who don’t live in your area.  If you’re a contractor or self-employed, this will be even more familiar.  You’ll see them on mailing lists and dialed in to meetings from time to time.  We might not realize it, but often we leave these dial-in employees out of important team functions.

Maybe you’re having parties and meetings without informing the remote teammates.  Maybe your meetings feature photocopied handouts that the dialed-in folks never see.

Keep your remote friends in mind.

Be sure every meeting involving a call-in has a legitimate online component.  Share out your agenda and meeting documents online.  Take advantage of  screen sharing and video conferencing when you can.  Be sure that remote attendees can get their questions answered during the meeting, not just after.

Take care of your remote contributors.  In many cases you’re paying them as much as you pay your onsite employees, but you’re not empowering them to contribute and participate to their full potential.  The modern workplace becoming more and more decentralized.  You will be a remote contributor yourself at times, perhaps even permanently.

Take advantage of the opportunities you have now to get better at working with people in other places.

It helps them and it helps you.  You’ll need the skills soon.  Claire Flanagan, the most successful enterprise collaboration practitioner of 2009, honed her unique skills and insights by telecommuting.  One of the most creative and well-known E2.0 advocates has a similar story to tell:

I have been a remote employee, working from my home office, for over six years now and I absolutely love the experience. I probably wouldn’t even change it for anything else. And I suppose that would apply as well to the over 50% of IBM remote employees who work away from a traditional office. And I bet that would apply to most of you folks out there as well who have been working remotely for a while now. – Luis Suarez

Related articles by Zemanta

• Affordable and Usable Video Conferencing? (ask.slashdot.org)
• Procter & Gamble tries a bring-your-laptop-to-work program (computerworld.com)
• Pa. district accused of student spying puts IT workers on leave (computerworld.com)
• At1 All Things Wan Introductory Presentation (slideshare.net)

I have been quiet on the enterprise collaboration front for the past few weeks. Lately I’ve been brushing up on public-facing web app development using PHP (and sometimes Ruby/Rails).

I’ve been blogging in quick spurts about the sort of tasks I’m putting myself through over on my Posterous mailblog.

Recent topics on the mailblog:

• Test driving Munin service statistics on a LAMP virtual machine
• Learning more about CakePHP and git rollbacks
• Get up and running with a Turnkey LAMP server

Image via Wikipedia

Reward programs gone horribly wrong

I think we’ve all seen examples of reward systems leading to counterproductive behavior.  My favorite example is working for a fast food joint when I was in college  and having a “seventy second drive-through time” contest.  Our goal was to keep the average time between customers leaving the drive-through menu (presumably having just placed an order) and customers leaving the drive through window (presumably having just received their food) below one minute and ten seconds – for an entire week.  The reward for meeting this goal was a week’s vacation in Hawaii for the GM and his wife.

We had a clock above the drive through window that let us know what our average drive through time was for the day.  That clock was the center of our world for a week, and we were led through some strange strategies to beat it.  The eye-popping cheats my team employed to meet this goal were as follows:

1. Assistant GM wearing headphones sets up an unsecured cash drawer on a makeshift table outside in front of the drive through menu.  Customers placed orders with her directly and paid.  She asked them to wait until she gave them the go ahead to drive forward to get their food.
2. AGM receives word via her headset that the customer’s order is nearly complete and she waves them through.  Customer drives over the sensor, starting the seventy second drive through clock at 0.
3. Customer arrives at the window seconds later, receives the food through the window, and drives off.   Total automatically measured drive through time: less than ten seconds.
4. In case of an unusual order, the customer was asked to pull all the way through and wait in the front parking lot to have the order delivered when it was finished.

Why people buy in to flawed reward systems

I can only really defend #4 as sound practice for operating a drive through.  If a customer’s complex order is going to slow up the rest of the business, it makes sense to pull them aside and let simpler orders keep going through.  Similarly, a table for two is always quicker to get at a restaurant than a table for ten.  The rest of these things were unsafe and dishonest and completely against the spirit of a program to minimize drive-through time, but I believe the GM received his trip to Hawaii all the same.  The only reason the rest of us were complicit in these ill-advised plans was because we wanted to keep our jobs.

So – make sure your employees’ individual incentives are aligned with your organizational goals.  If your new systems and processes (such as Enterprise 2.0) are going to require realignment of your employee’s daily work, be sure their goals and rewards are similarly recalibrated.

Current management theory on rewards and engagement

I had a conversation this morning with a coworker about career advancement in the form of promotions and raises and how (or if) they aligned with internal collaboration tools like SharePoint. I was thinking of Freakonomics when discussing this but I just got around to reading Venkatesh Rao’s review of “Drive” by Dan Pink (drive in this context appears to be mean employee engagement) and this part jumped out at me:

[W]e get a speculative list of reasons why carrots and sticks don’t work for creative work. The book calls this list the “Seven Deadly Flaws”

1. They can extinguish intrinsic motivation
2. They can diminish performance
3. They can crush creativity
4. They can crowd out good behavior
5. They can encourage cheating, shortcuts, and unethical behavior
6. They can become addictive
7. They can foster short-term thinking

In-the-flow versus “sure we’ll get to it the day after we run out of real work”

This fits right in with the accepted idea that the way to make enterprise collaboration work is to use it “in the flow” of our everyday work rather than using it “above the flow”.  In the flow means you’re actually getting your job done by using these tools as opposed to doing your job once and then going to another platform to write about it.

At the end of the year when your corporation looks over its employees’ performance and decides who’s helping the company out to the degree they need to be reinforced and who isn’t, are your reward systems structured to account for team-wide successes or individual stars?  It is easy enough to compare one person’s productivity against another’s as long as you are only measuring their direct outputs and not those of the people they work with and facilitate on a daily basis.  If you want your collaborative systems to succeed, be sure employees are incentivized to share rather than hoard.

Related articles by Zemanta

• Jack Covert Selects – Drive (800ceoread.com)
• Can an organization not be ‘ready’ for Enterprise 2.0? (fastforwardblog.com)
• Do Engaged Employees Make for Profitable Companys? Try the Gallup Q12 (questionpro.com)
• Incentives AND Recognition – Forbes Article AND Some Thoughts (i2i-align.com)
• I’m On A Mission to Ban Motivation Programs – Join the Movement (i2i-align.com)

Facebook co-founder Dustin Moskovitz is still working on his own business collaboration startup, Asana

Today’s announcment of Salesforce “Chatter” as a SaaSy enterprise Facebook product adds another option to the red-hot market for enterprise social platforms modeled on Facebook.  Regular readers will remember that I called out  ”Internal Facebooks” as the #1 collaboration trend for 2009 in a post last month.

We’re still waiting to hear from Asana, the business collaboration company started by Facebook cofounder Dustin Moskovitz after he resigned from Facebook thirteen months ago.

Salesforce’s “Chatter” further validates the internal Facebook trend

The company, in a statement, says Chatter will “revolutionize the workplace” by leveraging social networking models that have become popular among the mainstream, notably Facebook, and bringing them to a secure and private cloud where people, content and applications will have profiles feeds and groups.

As you can probably imagine, Chatter incorporates things like status updates (I’m on a conference call now or maybe in a meeting), groups (which helps a project team stay updated), sharing (which allows users to determine who sees what) and, of course, Twitter, with the most relevant tweets being filtered into Chatter. – ZDNet

Cisco’s “Show and Share” video-based social network

Cisco’s announcement of a “Show and Share” platform came nine days ago via Burton Group analyst Mike Gotta (subscribe to his blog by the way, it’s excellent!).  There’s also a dryly named Cisco Enterprise Collaboration Platform:

Cisco Enterprise Collaboration Platform is an enterprise-class social software portal that features a corporate directory with social networking capabilities. It allows users to create team spaces and community environments ‘on the fly’ and also offers a customizable framework for integration of legacy business applications and web 2.0 content. Unlike today’s document-centric portals, the Cisco Enterprise Collaboration Platform is people-centric, facilitating real-time voice and video communication to connect people, communities, and information to make faster business decisions. – Mike Gotta

Novell’s new Wave-enabled “Pulse” complements their more orthodox “Teaming“offering

Pulse (announced November 5) adds Google Wave capabilities to the existing group-based collaboration of Teaming: Pulse, announced November 5 shares some similar characteristics:

Pulse is among a new wave of enterprise collaboration suites that aim to bring features from popular consumer social networking sites like Facebook and Twitter to the workplace, with the added security and IT management that organizations require.

Some players in this space include IBM’s Lotus, Socialtext, Jive Software and NewsGator, which provides an enterprise social networking component to Microsoft’s SharePoint. – Computer World

Who’s next in the enterprise Facebook market?

I fully expect to hear an announcement out of Asana in the next 6 to 9 months.  Newsgator is already making a niche for themselves by Facebookifying Microsoft’s comparatively slow-moving SharePoint product.  What else is out there? Please let me know with a comment or a tweet!

Related articles by Zemanta

• Salesforce.com Launches Social Enterprise Application at Dreamforce (readwriteweb.com)
• What’s in Store for 2010? A Few Predictions (redmonk.com)
• Salesforce Chatter: A Real-Time Social Network For The Enterprise (techcrunch.com)
• Hot collaboration trends in 2009: Internal Facebooks, SharePoint microblogging, expert location, and homegrown wikis (sharingatwork.com)
• Salesforce Announces a “Facebook for the Enterprise” (mashable.com)
• With Chatter, Salesforce Takes a Facebook Approach to Collaboration (gigaom.com)

@BlakePalmer presenting for @Lunaweb at last year's BarCamp Memphis

The second annual BarCamp Memphis is finally here!

Tomorrow I’ll be heading to the MASE building to listen in to soome good presentations, panels, and wide-open discussions (“core conversations”) about technology development, new media, and entrepreneurship.

My two scheduled sessions are:

• “Social Tools in the Enterprise” with Ryan Hinricher.  Ryan and I are going to get into the how and why of using collaborative software to improve team work.  Ryan’s perspective is that of a realtor using SaaS tools like Salesforce and Basecamp to connect with his associates across the country.  Mine is that of an enterprise software developer using collaboration platforms such as SharePoint and Jive to work with local teammates and with our branch offices.
• “It’s a Small URL After All” with Bob Hazlett.  Bob and I walk through the arguments for and installation of a personalized URL shortening service of your very own using  Yourls.

@skippytpe Macbooking at the barbecue lunch during BarCamp Memphis 2008

I’ve also got a few presentation ideas in mind for optional afternoon sessions if I can drum up some interest.  I’d be happy to review several of my recent blog topics in a discussion format, such as

• Get started building web apps with your own Ruby on Rails virtual development server
• Found: “What problems does Google Wave Solve?”
• Hot collaboration trends in 2009: Internal Facebooks, SharePoint microblogging, expert location, and homegrown wikis
• Presentation: New Wave Collaboration And Enterprise 2.0

Check out the full list of scheduled BarCamp Memphis topics.  Any topics you want to lead or suggest for discussion can be written on the flip chart tomorrow morning and voted on by your fellow attendees.  Proposals with the most votes will be slotted in for afternoon sessions! See you there!

Many thanks to platinum sponsors Social Media Expedition and MpactMemphis!

Related articles by Zemanta

• Novell integrates to Google wave (wgadgets.net)
• Microsoft Enterprise Search Blog on SharePoint Search (arnoldit.com)
• Lotus Connections now with Microblogging and Wikis (cmswatch.com)
• Glasscubes Brings Small Enterprises Slick Cloud Based CRM (thenextweb.com)

David Heinemeier Hansson, creator of Ruby on Rails

Digging deeper into Rails programming styles

The last Sharing at Work post showed how to set up a Rails development environment using a Ubuntu Linux virtual machine.  This week, I’m in the midst of developing my first ground-up public-facing web application using Ruby on Rails.

Learning a new way to dynamically generate HTML

My Rails development has already been aided in several places by kind folks in the #rubyonrails IRC channel on Freenode.  One interesting thing they’ve directed me to is an alternate templating language called Haml.  Rails uses something called “embedded Ruby” by default – .erb files that describe HTML output with some necessary Ruby interwoven to generate the dynamic parts of a page.  haml does much the same but it takes a minimalist approach to syntax.

Nesting is handled by indentation (two spaces per level of nesting).  Tags are not closed, the closer is simply inferred by the order in which they are used and the associated indentation.

To illustrate the goodness of haml I’ve borrowed some code snippets from a longer article on Hhaml at Rail Spikes.

Embedded Ruby example; 20 lines and 547 bytes:

This shows a stock offering div containing a name, a symbol, a the latest price and date of the latest trade, and some dynamically generated links to buy and sell the stock in question.

The same code using Haml; 13 lines and 365 bytes:

Click through to the original Sharing at Work post if these code snippets aren’t showing up on your screen.

What’s Haml doing for us?

You can see right away that the Haml code provides a lot less for the developer to read through and its appearance nicely dovetails with the CSS models we use to do page layout. To quote the Rail Spikes article linked above, “Hampton says he came up with Haml by starting with a fragment like this and seeing what he could take out without losing anything. I’ll follow his example.” I like it and I’m going to try to use this throughout my project. Maybe it will be useful to you too.

In the next post I’ll share what I’ve learned from Rails.  Consumer web app development seems to be evolving faster than the SAP/Business Objects toolset I’ve been using at work.

Related articles by Zemanta

• Techmosis Rails (slideshare.net)
• Testing Rails Applications (timbauerblog.wordpress.com)
• JavaScript MVC (alistapart.com)
• SharePoint 2010 For Developers (slideshare.net)

Image via Wikipedia

Today we’re going to take a detour into web programming using Ruby on Rails.  The hardest part for me has been getting the development environment up and running.  Some guides will walk you through setting up Ruby and then Rails directly on your Windows machine, but I abandoned that pretty quickly.  You’ll soon grow tired of having to translate 99% of the Rails articles you read from Linux to Windows.  I’ve found that running a development server in a virtual machine (more on that later) is a better way to go.  It was a somewhat hairy process so we’ll walk through the steps it took to make it work.

What in the world is a “virtual development environment”?

Our goal today is to create our own web server to build and test web applications on.  We could go out and buy a web hosting package for $10/mo. somewhere and do our development directly against a public web site but that’s not necessary for introductory development.  Until you are ready to publish your application for all to see you can do just fine running a virtual server that looks and feels just like the public server you might pay someone else to set up for you.  Click through to see the basics of how this is going to work:

Fig. 1. When you see this welcome screen you'll know you've gotten Ruby on Rails running on your virtual server.

First you’ll set up a virtual machine (VM). This VM is a program that runs on your computer – that pretends to be another computer altogether.  This allows you to leverage the benefits of having multiple computers in the same room without actually having to buy extra machines and wire them up.  The two components of a virtual machine that you need to be concerned with are the VM software itself and the operating system you’ll be running inside the VM.  Once you get the operating system up and running in your VM it will act just like a second computer on your network.  You’ll be able to log into the VM, configure it, and run new programs on it that interact with your other computer(s).  Today we’re going to install a slimmed down Linux distribution with Ruby on Rails built in and ready for you to work on.  When it’s all finished you’ll be able to write Ruby code either on your real machine or on the virtual machine and then transfer it to the virtual Ubuntu system so that you can test it.  Testing will work just like any other web application – you’ll open a web browser on your host machine (Windows) and view the application at a URL like http://localhost:3000.

Get VirtualBox

VirtualBox is a free (GPL) virtualization tool from Sun that will allow you to run a “guest” operating system without having to mess with your computer’s existing “host” operating system.  For the purposes of this article we’ll be using Windows as the host and Ubuntu Linux as the guest.  Download the appropriate VirtualBox installer for your operating system here.  I installed VirtualBox using all of the defaults on both Windows Vista and XP laptops.

Fig. 2. Create a virtual machine named Rails. Select 'Linux' as your operating system.

Create a virtual Linux machine with VirtualBox

Once VirtualBox is up and running you’ll be prompted to create your first virtual machine.  Name it ‘Rails’ or whatever you’d like to remember your machine as and be sure to tell it you’ll be using Ubuntu as your guest operating system.  See Figure 2 to make sure you’re going down the right path.  Click through all the defaults on the rest of the pages until the machine is configured.  Now we’ve got to download Linux and install it into our currently empty virtual machine named Rails.

Fig. 3. Tell VirtualBox to let your virtual machine access this Turnkey Rails image you just downloaded.

Head over to Turnkey Linux to pick up a preinstalled Linux server running Ruby on Rails.  Click on the Download link and you’ll get a ~170MB .iso file containing the entire preconfigured Rails environment.  You’re going to wind up with a file on your hard drive named something like turnkey-rails-2009.03-hardy-x86.iso.  Once the file finishes downloading go back to VirtualBox, select your Rails virtual machine, and then click Settings.  Click on Network on the side panel and then change the Attached To dropdown from its default of NAT over to Bridged Adapter.  This will cause the virtual machine to connect to your local network’s router and demand its own unique IP address.  If you’ve got a consumer network (e.g. a Linksys router or anything you’d find at an electronics store) then this should work first try.  If that doesn’t work, you can go back to the default of NAT but there will be some extra configuration required to make the virtual server accessible to your host computer.

Click over to CD/DVD-ROM on the VirtualBox settings sidebar and make sure Mount CD/DVD-Rom is checked.  Select the ISO image file radio button and then browse to the Turnkey Rails .iso file you downloaded earlier.  Once you’re done it will look something like Figure 3.

Install Turnkey Rails

Now your Linux machine should be all “wired” up and ready to start up.  Click on the Start button in VirtualBox and you’ll see the “first run wizard” asking you where to find the installation media for your guest OS.  Select Image File at the bottom and be sure it’s pointing to the Turnkey Rails iso you’ve just downloaded.  Click Next and Finish and then you’ll see the Turnkey setup screen.  Select the second option “Install to Hard Disk” and then wait for your Ubuntu Linux installer to load.  Select all of the defaults :

• Guided – use entire disk
• Yes, write the changes to disks.
• Create passwords for your Linux and MySQL root users.
• Yes, restart now.

Fig 4. These are the Web and SSH ports you'll need to access your new Rails environment

When the machine reboots you’ll want to choose “Boot from First Hard Disk” at the boot menu (You can also go to the VirtualBox settings and disconnect your Turnkey Linux .iso file whenever you get tired of having to manually choose the local disk at boot time).  Your Turnkey Linux box will now boot up and you’ll see a blue configuration screen showing you the server’s IP address and the ports you’ll need to use to connect to the server.  You’re going to want to do all you can to work across the network so you can get practice using this as a remote development server rather than a second computer sitting right next to you.

If you can’t connect via a web browser using the address shown in the blue Turnkey console (see Figure 4), you’ve probably got something wrong with your VirtualBox network settings.  Close the window containing your virtual machine and then get to work on fixing the settings.  If you’re in a hurry, try the Bridged network as it should be the easiest to make work in a home / small office environment.  If you really must use NAT, yofu’ll need to open up a windows shell (cmd.exe) and set up port forwarding so that your virtual Rails box will be accessible through forwarded ports on your host machine.  Follow the steps in this quick  guide I wrote up on port forwarding using VirtualBox.

The nitty gritty: final setup from the Linux command line

Hopefully at this point you’ve gotten Turnkey Rails installed using all of the defaults and you’re able to connect to it across the network.  Now all that’s left is to update our Linux, Ruby, and Rails software packages to the latest versions and then add a few more pieces that the Turnkey .iso image doesn’t include.  Connect to your new machine using an SSH terminal (port 22 if you’re using network bridging; port 2222 if you used the port forwarding guide linked in the previous section).  Log in using the root password you specified in the Linux installation step above.

First stop: Update Linux and Ruby using package handlers apt and gem.  Enter in the following commands to update all of your Linux and Ruby packages to the latest available versions:

apt-get update
gem update --system
gem update

Once these complete there are two optional steps that will probably help you out.  First is to install sqlite.   sqlite is the database that Rails uses by default.  While you can use the MySQL server that Turnkey Rails supplies you will find yourself having to mentally translate a lot of the Rails guides you read from the default sqlite commands to equivalent MySQL commands.  It’s easy to install sqlite alongside MySQL, so that’s what we’re going to do:

apt-get install libsqlite3-dev sqlite3
gem install sqlite3-ruby

The final optional component is a WYSIWYG text editor. Turnkey Rails comes with Linuxy text editors like vim that require some reading to jump into.  If you’re anything like me and you’d rather jump right into coding without having to learn a new text editor UI then you will want to try diakonos (or joe for that matter). Install diakonos using apt:

apt-get install diakonos

Now if you find yourself needing to tweak a config file on your Turnkey Rails server you won’t have to open it up in a file transfer client – you can just fix it on the server.

Next step: Write a Rails application!

Fig. 5. My first Rails app - created using the "First Rails App Walkthrough".

Now you’ve got your system all set up and you’ll want to test it by building a simple Rails application.  I recommend the First Rails App Walkthrough on the official Rails wiki.  Remember, if the walkthrough mentions URLs like http://localhost:3000/users you may need to translate those to the name and port number of your new Turnkey Rails  server.

Good luck! I’m sure there are a few gaps and inaccuracies in this writeup and I’d love to hear your feedback and corrections in the comments below.

Related articles by Zemanta

• How To: Virtualize Any OS For Free [How To] (gizmodo.com)
• How to: Create and manage snapshots in VirtualBox (en.onsoftware.com)
• What OS is on Your Netbook? (jkontherun.com)