Linux Today: Linux News On Internet Time.

More on LinuxToday

Editor's Note: Linux and Too Many Choices

Sep 17, 2010, 23:02 (48 Talkback[s])
(Other stories by Carla Schroder)


Desktop-as-a-Service Designed for Any Cloud ? Nutanix Frame

by Carla Schroder
Managing Editor

A perennial whinge is "Linux and FOSS have too many choices! It's confusing and scary!" So what's the answer, a single global dictator?

It must be the season for recycled anti-Linux whinges, because in the past few weeks I have had the pleasure of wading through a flurry of stories about Linux has too many choices, Linux is not ready for prime time, Linux is too expensive just like proprietary software, and FOSS is amateur hour and all insecure. We've heard it all before.

The one that is worth a bit of discussion is "Linux has too many choices." I rather like that the Linux/FOSS ecosystem is huge, messy, and highly productive. I understand that standing before such a vast colorful feast can be overwhelming. But there is one key point that has not been addressed: how could any kind of simplification be achieved? Think about it-- how would this work? All I can think of is some kind of central clearinghouse run by an iron-fisted tyrant who approves or disapproves everything. It's absurd. FOSS is a giant wonderful cat herd. There is no single turtlenecked dictator. By design it is decentralized and distributed. Anyone can play, and the only entry requirements are ability and desire to learn.

There are times I have thought it would be nice to see more energy going into existing projects instead of creating new ones. How many Twitter clients, FTP servers, instant messengers, window managers, text editors, command shells, and Tetris clones do we need?

But that concept depends on a number of assumptions. One, that new contributors can waltz into existing projects and find a welcome and a need for whatever they can contribute. Two, that one size fits all; that a limited of number of choices will do the job for everyone.

There is a lot of value in launching a new software project because it is a tremendous learning experience. If it's just one or two people then they have to learn how to do everything-- coding, version control, managing releases, bug tracking, interacting with users, artwork, documentation, distribution, packaging, and so on. Bigger teams have to figure out how to divide up the work and get along. Ultimately it succeeds or fails, and if it goes nowhere it's still been a valuable experience, and the pool of skilled contributors is that much richer. For end users failure is greatly preferable to the proprietary habit of keeping inferior products in service and forcing customers to use them. Exhibit A: Microsoft Windows. In a competitive market it would have been laughed out of existence long ago. Or possibly have become less awful.

At any rate, we all know what's behind this "too many choices" guff. Most of it is deliberate misinformation, or quick n cheap recycling to fill column inches from people too lazy to write something worthwhile. It's hot air. We don't need gatekeepers-- FOSS is working fine, the way it is supposed to.