Linux: registering demand

Last week's request that talkback contributors discuss what it would take to get Linux to displace Windows as the dominant desktop produced a large number of interesting ideas.

Among them, a comment by CobraA1 struck me as particularly important. Here's part of the post:

I'll make sure to get started - where should I go to demand better support for NTFS, 3D graphics, better WINE compatibility, etc?

In the meantime, I guess I'll be writing letters to software developers about their games, and I hope other people join me.

I've already written a letter to the author of Pegasus Mail & Mercury transport. Unfortunately, like so many authors, he seems set in his ways.

A demand, I can create, but I am only one small voice; I cannot create change by myself. I need others to participate!

Other posts talked about the usual suspects - games, licensing, interfacing, support, and unified marketing - but CobraA1 adds a new one: there's a real problem with how we understand and express the concept of demand.

In a related posting fellow zdnet blogger John Carroll, who works for Microsoft, called my suggestion that we should let Linux be Linux instead of pretending it's a second rate Windows a "recipe for the failure of Linux" and recommends that the Linux community try to succeed by "figuring out what those customers like about Windows, and riding that wave into Windows' users home."

So what's the thing customers like about Microsoft products? they're not cheaper, they're not better, they're merely more nearly ubiquitous. Walk into your local Best Buy to look at a digital camera and the clerk will assume you have a Microsoft PC. Try it: go to any consumer electronics store and ask for the Linux software that goes with whatever they're selling -you'll get nowhere unless you're a masochist who enjoys being condescended to by a snotty clerk.

It's the same at the office: ask someone to send you a document by email and most of the time you'll get a Microsoft Word document. Phone up your friendly local network services supplier to ask about a DNS failure, and they'll want to know which Windows release you've got.

And so it goes, the PC assumption rules -and that's what drives those customer decisions. It's a classic emperor's new clothes issue: nobody not involved in selling or supporting Wintel actually likes the products, but everybody assumes that everybody else uses them. Ask around among the non techies you know, and you'll find something shocking: many non techies can't get the thing to do half of what it's advertised to and blame themselves.

A big part of that problem arises because contrarian demand for Linux or MacOS X software isn't visible. You can ask as many store clerks as you like for Linux software, but requests like that aren't tracked and therefore don't exist.

And that's true across the board: Sony doesn't see demand for Linux camera software in its stores, Best Buy doesn't see demand for Linux games, Office Depot doesn't see demand for Linux accounting software.

So what to do?

Part of the answer may be to create a linux demand registry - a site where people could literally register demand: sign an "if you build it, I will try it" pledge. Get enough people signed up and you have verified demand of the kind that the financial press would have fun writing about - thus putting great pressure on the companies involved.

Publicity stunts would be fun too: get 100,000 names on a list asking Sony for Linux support on their cameras and PCs, and it would take 15 minutes to carry a 1,250 foot fan fold printout past TV cameras in front of Sony's US headquarters. Think they could ignore that?