Open Source: Irreconcilable Differences - Systems Management News On The Web

Sort By :DateRelevance

DEPARTMENTS

HOME

TOP STORIES

ON THE WEB

BZ MEDIA

Printable version

Open Source: Irreconcilable Differences
By Kenneth Hess

August 15, 2008 — There’s a mastodon in the room and no one’s discussing it. What they are discussing is how my free, open-source operating system needs to be compatible with a certain commercial, proprietary operating system whose identity I’ll leave up to you to decipher. The conversation is taking a bad turn to such topics as portability, comprehensive inter-operating system compatibility, cooperative development and how I have to buy the next round of beer.

In my article, “Last Year’s Penguin,” I had a similar attitude against the notion of combining OpenSolaris and Linux into a single entity. I’m not implying that anyone is attempting to create a Windows/Linux hybrid—nothing of the sort. What I am saying is that too much compatibility is just as bad.

The mastodon to which I refer is the overwhelming sufficiency of existing interoperability. Allow me to clarify that little tongue twister for you. Shouldn’t interoperability between the two operating systems be enough? Why must there be compatibility between them? After all, the relationship between the two is not a match made in heaven. And, in this relationship, why must it always be my operating system that has to conform to the other? I know in any relationship there's a giver and a taker, but must the Linux community be the perennial giver for every situation?

Just once, I’d like to hear the other guys say, “Hey, maybe we should create our applications, filesystems and documents to be compatible with the open-source ones.” Instead, they develop their own standards to which we all must happily conform. Now that’s what I call a harsh mistress!

But when did any level of compatibility between the two operating systems become desirable? And where do we draw the line? If the two systems become too compatible, what we’ll have then is an ugly offspring—a Devil’s Spawn hybrid that exists between worlds but comfortable in neither. We already have an operating system that fits that description: Mac OS X. There’s enough ugly in the world without purposefully adding more.

Compatibility between applications, though, has its upside, I admit. Some of the resistance to open-source conversion is the idea that mainstream compatibility doesn’t exist. Productivity applications such as office suites, money managers, e-mail, project management and document management exist for both systems, and most of them are interoperable while a few are even somewhat compatible. By somewhat compatible, I mean that you can, with varying degrees of effort and success, interchange documents and data between similar applications.

The open-source equivalent to Microsoft Office, OpenOffice.org, is very compatible with Microsoft Office versions 1997 through 2003. It’s just as compatible with those versions as the new Office 2007 is. Evolution, an e-mail client available for almost every operating system, is compatible with Novell GroupWise and Microsoft Exchange. Additionally, if a user really needs to run Microsoft Office on a Linux computer, it’s possible through a program called Wine. Wine provides an environment in which you can install and run some Windows applications.

For long-term usability and compatibility between similar applications, there needs to be a concerted effort on the part of all software vendors to create a pathway for compatibility. Right now, that pathway doesn’t exist. If it ever does exist, I expect that it will be paved with dollars and not good intentions. The first version of Microsoft Office was for the Apple Macintosh and, at some point, I expect there will be a Linux version. But I won’t hold my breath waiting for it. The problem with getting a Linux version of Microsoft Office is twofold: market share and market share.

First, the development time, effort and expense is too great to create a commercial office suite for Linux when it doesn’t have sufficient market share currently to support it. Desktop Linux only enjoys a very small percentage of the available desktop space globally, although that is changing in small increments. Second, if Microsoft Office were to become available for Linux, there would be fewer barriers to conversion to it as a viable desktop OS, which would increase its market share thereby decreasing the market share of other operating systems.

For me, Linux is just fine as an independent operating system. And through the magic of virtualization, there’s no real need to work on operating system-level compatibility with any other operating system. I know I have nothing to fear from that Redmond-based software company, as they won’t expend any energy creating applications to run on Linux anytime soon, and for that, I’m grateful. The bottom line is that if you like Windows enough to want to turn Linux into Windows, why not just use Windows in the first place? After buying the latest round of beer, I know why no one’s discussing that mastodon.

Ken Hess is a freelance technical writer who writes on a variety of subjects including Linux, MySQL, SQLite, PHP, and Apache. Reach him at kenhess.com.

Related Search Term(s): Linux, open source, Windows, Microsoft

Share this link: http://www.sysmannews.com/link/32698

EMAIL THIS ARTICLE

SEND FEEDBACK

MORE COLUMNS