Or you could just get a Mac. I used Freebsd a lot during the 90s dotcom boom, so its Junior Unix enough for me.

For playing around Linux, I would highly recommend Wubu - a Windows based installer for Ubuntu that configures Ubuntu to run side-by-side with Windows very easily:

http://wubi-installer.org/

This post was mentioned on Twitter by butaji: RT @karlseguin Blogged: You should know Linux - http://bit.ly/6E5dcD

While it is true that there are a lot of programming and development tools available for Linux distros, if those tools are open source then most of them are available for Windows as well. What makes learning to work with Linux special is if we can get under the hood.

Perhaps more importantly to become a better programmer, we need to participate in open source projects whether on Windows or Linux.

By the way, I am using Ubuntu and have never found a distro as good as this especially for first time Linux users.

I must admit I fail to see the argument your making about tools that "you need to know" which (supposedly?) require Linux. So my comment should be seen as a "please educate me" comment - because I would actually like to see it.

My first thoughts are that in-memory distributed caching and non-relational persistence options are not really required and maybe not even applicable for most of the systems that I build professionally.

The next thing I can't help think about is that there are versions of these kinds of tools that run on windows. There are open source projects (NCache express, Memcached, DB4O, CouchDB and mongoDB) already offering these things, and when referring to  in-memory distributed caching MS are working on the project named Velocity, which is just that.

No doubt several of the projects I listed originate from Linux, but if they run I do not see why I should spend time getting to know a new OS, when there is no real chance I'll end up doing "real" apps for these platforms any way.

Thanks for your clarification.

I certainly do pride myself in being a passionate dev. and my point is not that one should get stuck in the ms sandbox. Just that I think there are other ways than diving into Linux that are just as good if one wants to expand ones horizon.

There are lots and lots of tools and languages out there. That is actually the best thing for me about being a dev, because you can always improve and learn new stuff.

There is no need to buy anything. Just create a partition and dual boot, or go virtual.

Ahh...I was really hoping for more of a PS3 v. XBox360 religious flame war. :)

On a more serious note, comparing memcached and Velocity aren't really that accurate.  Sure, from a 10,000 ft view they can both be lumped under "Distributed Cache", but that's where their similarities end.  Distributed Caching apps have been availalbe for years on Windows, they just didn't come from Microsoft.

I guess my points is using MS OS doesn't mean you are limited to the MS Toolbox.  

I do agree with your main point though, even if it's simply for your Resume.  But in a broader sense, you are only hurting yourself by not at least opening your eyes to what else is out there.  That's why my new laptop is a a MacBookPro (helps to have a father-in-law who works for Apple), and now my new web projects are in Rails using MySql, something I never would have been open to if I stuck with only Windows (or at least a lot less likely).

A couple of small points first:

1. "I've long believed that knowledge of C is fundamental to becoming a great developer" - A programmer, maybe (though I disagree with that too), but not a *developer*. Knowing how to listen to and understand your customer - *that's* fundamental.

2. "...learn to be a better programmers" - Proof reading also helps.

I generally don't agree with most "programmers/developers should know X fundamentals" assertions for the simple reason I've seen too many different situations that provide too many exceptions to the rule.

Let's also first agree on what a "programmer" or "developer" is *then* we can talk about what we should know *in general*.

Consider: A person can build a fine career as a developer doing nothing more than writing Word, Excel and Access macros and automation for businesses. Do they really *need* to know ‘C’, memory management and pointer arithmetic?

This is a broad and varying industry. It is also a rapidly advancing industry. And in the same way as there are a lot of new advancements we need to keep up with there are many fundamentals that also become “obsolete” (at least, to some of us).

Everyone should at least try linux for it's incredible command line and remote administration capabilities.

In terms of remote configuration and administration of servers as well as use of server resources it is light years ahead of Windows.

It's a different mindset, that's for sure. And absolutely something that everyone should at least experience.

Personally, I'd far rather be administering a linux box.

Karl, to those aiming to learn C and discovering the power of pointers and interrupts :), would you recommend any specific book?

Nice post!

Another must have book is  "Advanced Programming in the UNIX Environment".

www.amazon.com/.../0201563177

Keeping  up on what is popular among open source tools is a good idea, but I don't know that learning to use Linux buys a developer all that much.  The technology in the Linux operating system is still mostly 1970s technology:  Everything is still based on text files and command lines, even the latest tools!

I've used Unix and Linux extensively through the past three decades, doing software develoment, network support, kernel development, you name it.  Undersanding the concepts of shell scripting , rsh, and daemons is nice, but they are just other names for what most developers already know.  The nuances of how a daemon is started vs a Windows service is interesting but not all that significant to developer productivity.

This is the big difference between learning C and learning Linux:  If you learn C you learn some fundamental concepts that computers are inherently built upon, namely pointers.  If you learn to use the Linux OS, you don't encounter any basic fundamental technologies like that.  You don't even encounter any new technologies - only old technologies that have largely been replaced with better techniques over the years.  It is only when you learn specific open-source tools and technologies built *on top of* Linux that you actually encounter some innovation.

I think the main problem you are addressing is that developers often are unaware of open source innovations until years after they happen.  This is a valid concern, but has nothing to do with what OS the developers may be using.  There are thousands of open source tools out there, so it is very easy to get lost and not notice something new and innovative among all the chaff.  Tracking what other developers are using tends to help somewhat, as does reading blogs and looking at what open source applications are using.

I would recommend people read a great variety of blogs from the open source world so they understand what is out there, then download and play with them on their current OS, whatever it be.  That way they focus their time on what is really important to a software developer's well-roundedness:  Being familiar and comfortable with a large variety of tools and techniques.  Learning outdated text file configuration techniques seems like a waste of time in comparison.

Personally I do not feel like its enough to tell developers to install a linux distro on a old PC. You have to give them something specific to DO with it! Once you have a linux box up and running, what's worth spending time on? For example, you could suggest that the developer tries their hand at web development with Rails. I think your post would add more value if it had more suggestions along these lines.

I'm agree with Paul Batum.

You can start with project like collaborator or creator preferably <a href="en.wikipedia.org/.../a>

.NET Multi-process C# app like Google Chrome Virtual Method Performance Penalty Performance Profiler