% fortune -ae paul murphy

So where are we?

Let me try to summarise - and then ask for feedback on whether I get this right or wrong.

First: It seems to me that there is no strong disagreement that the end goal in Linux evangelism is ultimately to create rational markets in which people choose their operating systems and tools by trading off cost against functionality, the total cost of a decision to use something reflects that product's serviceability, there are clear business rewards for excellence, and any monopolies that get established are based on temporary technical advantages.

Second: I think it's pretty clear that the joint issues of patents and licensing are extraordinarily confusing with good arguments both for and against every major license now in use and the only real agreement being that current American patent law as it affects software is ridiculous while intellectual property theft in the rest of the world, whether organised by governments or practised by individuals, is inappropriate.

Third: Although lots of people disagree on what constitutes "support" most people agree that this is a critical determinant of product acceptability. There seem to be two extremes: people who think that internet access to information suffices, and others, (including me) who think it has to come from the guy next door to provide both tangible technical support and deep person to person emotional support, with most people breaking somewhere in between.

Fourth: There's no agreement on whether Linux should do its own thing, or imitate Windows in order to gain market share. Fundamentally this breaks down between those who argue that looking like Microsoft reduces barriers to acceptance and those who believe innovation pays. Most, however, seem to agree that Linux has to meet the ubiquity standard set by Microsoft if it is to compete on the desktop.

Here (with very minor editing) is how talkback contributor schoolfieldd summarised what I think of as the balanced view on this issue:

Linux doesn't need to look like Microsoft. It just needs to provide full functionality. You must be able to connect to our PDAs or other peripheral devices. Must be able to pass a file to Microsoft customers and they must be able to read and manipulate it. Must get vendor drivers for all devices. Must have software that works seamlessly with other platforms. And must eliminate the perception to the world that the shareware environment is a bunch of hackers who throw together solutions that are untested and incomplete; doesn't matter if the perception is true or not, the perception must be eliminated. People go with Microsoft because in their mind, they know it will be here tomorrow. Linux needs to be a leader, not a follower and must link up with software and hardware providers to close out all holes of support.

Fifth: Any discussion on who has, and who does not have, the skills needed to deliver Linux services efficiently seems to degenerate almost instantly into mutual name calling. A big part of the problem, of course, is that there's no common referent: no standard of efficient service delivery against which to compare our expectations or judge the performance we see from the people around us.

I notice, however, that there is something different about the people who think the typical MCSE isn't up to the job: there's a whiff of gunpowder around their posts - a been there, done that, got the scars kind of feeling.

I'm in that group, but recognise that experience doesn't make us right. On the other hand, efficient Unix service delivery may be something like understanding the utility of smart displays or what it really takes to raise kids: you may think you understand it, but if you haven't done it, you're just not in the game.


Paul Murphy wrote and published The Unix Guide to Defenestration. Murphy is a 25-year veteran of the I.T. consulting industry, specialising in Unix and Unix-related management issues.