Thursday, August 23, 2012

Free but not Gratis: A call for Open Source for everyone!

The term Open Source is not as old as you may think, and the concept actually predates the name. Initially the keyword was Free not Open, but Free is here in the sense of Freedom not in the sense "without cost", and this conflict in the English term "Free" was one of the big reasons that Free really wasn't a good word here. Which all in all doesn't mean that Free isn't still used to describe the Open Source movement, like in FSF (Free Software Foundation).

And Free as in Freedom, not Free as in "without cost", is an important distinction. What the deal was, in my interpretation at least but there are many different views here, was that the software should be available for use by anyone and for any purpose as long as they followed the rules. And the rules was there for a number of purposes, two important ones being:
  • To ensure that the software in question remained free and open.
  • To encourage the development of more free and open software.
Now, these are points I fully agree with, for a number of reasons. But I also think that that, and this is where I am not so politically correct anymore (if that is something I ever was), that there are markets for:
  • Making a living on Open Source software. And no one said this shouldn't be possible, but here I am thinking beyond master programmers like Linus, RMS and Ken Thompson, I'm thinking about your everyday coder, building applications for a customer, providing support or packaging software. Yes, there are many good developers out there without the ability to be master programmers like Linus or Alan Cox, but which are still needed and has competences beyond those guys, like Domain Knowledge.
  • Proprietary software. Yes, I am aware that this is something you probably didn't want to hear, but I do think that proprietary software will still be around and is still needed in many places. Let me tell you why I think so,and the main reason is that there are areas in Software that just isn't viable as Open Source. Like small specialized areas where it's not possible for someone to create good Open Source Software. This is not to say that specialized software cannot be Open Source in meaning that the source is open, but not Open Source in the meaning that there is a community development effort, with people around the globe finding and fixing issues adding features.
 Yes, I do believe that a developer who has significant domain knowledge of something, say car sales, can bring significant contributions to an Open Source project, even though that developer might not be more than an average programmer. And if the project in question is an Open Source system for, say, car rental, then his input might well be more important than any input from, say, Linus Torvalds.

Another thing to realize is that the higher up the ladder of software you go, from operating systems and device drivers, up through databases and middleware, past generic office applications and to domain specific applications adopted for a very specific purpose, the more non-IT specific knowledge is needed. To create an Operating System, like Linux, you need to be really good at, well this is a surprise I guess, but I do believe that being an Ace on Operating Systems is a good thing. And Operating Systems is something you can learn at University and the theories and operations of an OS are generic IT stuff. To a large extent, the same goes for databases. But for a system for, say, car rental, that is something you do not learn at University. And here are some issues with the Open Source movement, if you ask me:
  • The domain of Open Source is IT, i.e. computers, software etc, and the users of it are also presumed to be IT folks (which doesn't mean that Open Source isn't used by everywhere, it's just that Open Source it isn't a conscious choice by your average user at Joe Bloggs Car and Trunck Rental).
  • Open Source doesn't in and of itself bring a lot of advantages to those guys at Joe Bloggs Car and Truck Rental. Which is not to say that Open Source doesn't push prices down or isn't the most cost effective solution, but the Open nature of, say, Linux, means little beyond "lower price" to them,
That Linux is not as strong on the Desktop as on the Server side I think is largely based on this. It's not that the end users don't appreciate Linux and Openness, it's just that it's not easy to see that advantages in the case of an Operating System or even in the case of, say, Open Office, beyond that fact that the cost is less. But really, did want Freedom just to be cheap? There has to be more to it?

I think there is a market for much more domain specific Open Source software. Really, I do! But I think we (us IT folks) must start to be much more open towards things like end user needs and application specifics. In many cases, the way something domain specific works has been determined by some ancient IT-system built ages ago, so that the effective business logic is defined by how this old IT system works, not the other way around (flight booking anyone). I also firmly beleive that there is a big market for something like that out there, this is an area where the non-as-smart-as-RMS developers (like myself) can contribute, where we have some domain specific knowledge that can be turned into a useful and powerful application. But the Open Source movement needs to find a way to accommodate that, to understand that just because someone is not a good developer for writing SCSI device drivers, might not mean that he or she doesn't posses valuable knowledge, and to support those domain specific effort. Yes, to write an OS or a Database System, you can surely do that based on knowledge you picked up at University, but that is not the case for many, if not most, end user applications. Hey, maybe we can learn something interesting domain specific while contributing something IT specific to a project.

/Karlsson
Feeling Free and Open

1 comment:

VladimirRomanovsky said...

Sure, you are right the domain-specific programming is a niche where Open Source has a huge potential, personally I am dreaming to apply it to (1)medicine(2)metallurgy and possibly some other areas of interest for me and my friends. Moreover , I really think its priciples may and should be applied to the world of actual material manufacturing !
http://opensourceforrealworld.blogspot.com/