August 25, 2002

Applications are still the biggest Linux adoption barrier

- By Robin "Roblimo" Miller -
Here's a common statement from sysadmins who have switched part of their corporate networks to Linux: "We'd like to go totally Linux and Open Source, but my [marketing people -- engineers -- accountants -- customer service people] must use [name of application], and there's nothing like it available for Linux."
Whenever something like this is said in an email list or public online forum, a whole bunch of people usually chime in with something like, "You don't really need proprietary SuperProDesigner 6.8. You can use GPLed GnuAmateurDesigner .02 instead. It has a lot of the same functionality."

"A lot" is in the eye of the beholder. The sysadmin's engineers do a lot of squiggle creation, and while creating a squiggle only takes one click in the program they're used to using, the squiggle function is hidden five menu layers deep in the GPL program and isn't mentioned anywhere in the documentation. Not only that, the proprietary program creates 180-degree squiggles, while the GPL one only creates 90-degree ones. So you can say that both programs create squiggles if you like, but for fast, practical, everyday squiggling, the proprietary program beats the GPL one hollow. The engineers simply aren't going to change.

Then there's the "yes, we have a Linux version, but our Windows version has more features" problem that is so common among proprietary software producers. Let's talk about a hypothetical company we'll called Undernational Business Machinery, "UBM" for short. This company can talk all day about how much it supports Linux, but people who want to use UBM's famous WebFootSphere program to build feature-rich WebFoot sites soon find that the latest Linux version is two full waddles behind the Windows version, and is missing much of the functionality and new, ease of use features that are primary selling points for the latest Windows WebFootSphere version.

Hope from the commercial software sector

The "programmers scratching their own itches" style of volunteer Open Source and Free Software development is not going to produce user-level, enterprise-capable applications software. Not many programmers have a great personal urge to produce software that will calculate shipping charges, sales taxes, and VATs for a company that does business in 62 countries. But companies like Oracle, SAP, BEA, and PeopleSoft have a huge motivation to write this sort of boring-but-necessary software -- and are highly motivated to port it to Linux as Linux becomes more popular among their Global 1000-level customers. Other, smaller software companies and independent developers are going to find themselves scratching an important itch, namely their desire to earn a living, by producing desktop software for Linux as Linux desktop use grows. And don't kid yourself: Desktop Linux use is growing slowly and steadily, individual by individual, company by company, month by month.

It would be nice to live in a world where all programmers work for the common good without expecting pay, just as it would be nice to live in a world where I could go to the gas station and pump 15 gallons into my car, then wave and drive off without laying out money or sticking my credit card into a slot. But right now, I pay for gas, and the gas station owner and the oil company both expect to profit from my purchase. There is no reason programmers who write software that helps the gas station owner and oil company earn money shouldn't get paid for their work, too.

Those programmers can call those earnings "support fees" if this nomenclature makes them feel feel all warm and GPL inside, but they are still writing software someone else wants to use, and getting paid to write that software.

We are going to see more commercial software for Linux. Some of it will be good, and some of it will be lousy. Some -- probably the vast majority over the next few years -- will be written with enterprise users in mind, but I expect to see a gradual increase in the number of commercial home/small office programs either written for Linux or ported from Windows versions. Perhaps some of the ports will be done by Wine-connected companies like Codeweavers. They are certainly pushing this side of their business hard.

In any case, get ready for more Linux commercial applications. We are inching closer and closer to the point where mass market software houses consider the desktop Linux market worth their time and money, and we have already reached that point for server-based, enterprise-level software packages.

The one thing commercial software publishers that expect to compete in the Linux marketplace must bear in mind is that there is still going to be plenty of volunteer-written Linux software out there, and it is going to improve steadily as the people who write it become more sophisticated and user-oriented. This means commercial software for Linux must offer substantial advantages over the "free competition" if it is going to be successful.

Some commercial software publishers will feel up to meeting this challenge, and some won't. Hopefully, those who take it up in the future will offer their best software, not half-assed, cut-down versions that insult Linux users' intelligence -- and fail to sell as a result of this stupidity -- as so many commercial software vendors have done in the past.

"Commentary" articles are contributed by and readers. The opinions they contain are strictly those held by their authors, and may not be the same as those held by OSDN management. We welcome "Commentary" contributions from anyone who deals with Linux and Open Source at any level, whether as a corporate officer; as a programmer or sysadmin; or as a home/office desktop user. If you would like to write one, please email with "Commentary" in the subject line.


  • Migration
Click Here!