March 13, 2003

Michael Robertson: the Steve Case of Linux

- By Robin 'Roblimo' Miller -
AOL sucks. There are many better, lower-cost ISPs. And Lindows sucks. There are many better, lower-cost Linux distributions available. These are articles of faith among sophisticated Internet and Linux users. But the Internet as a whole owes a lot to AOL and Steve Case, and Lindows is doing as much for Linux as AOL has done for the Internet, whether you like it or not.

About five ago, after looking at Internet user statistics (and after receiving what seemed like hundreds of AOL floppies and CDs in the mail and having them shoved at me everywhere from electronics stores to supermarkets), I realized that nearly all U.S. Internet usage increases could be traced to AOL marketing. AOL was "the" entry-level ISP. Others -- including (then) #2 Earthlink, aimed most of their sales pitches at AOL subscribers, a tactic you'll notice MSN has made the core of their latest marketing campaign.

AOL PR people were happy to give me glowing signup numbers and talk my ear off about how their total number of subscribers was going up, up, up (a situation that has changed since then), but when I asked about their "churn rate" they clammed up.

"Churn rate" is industry slang for the percentage of subscribers who leave vs. new ones who subscribe. Think of it as similar to employee turnover. If the company has one million subscribers, signs up 100,000 new ones, and loses 100,000 during the same period, the net increase in subscriptions is zero, and the churn rate is ten percent. There is always some churn. You need to market constantly to stay even; to replace customers you lose in the normal course of events, and you only grow when you manage to get more new customers than you lose. (And even the most stable, wonderful service in the world will lose customers due to death or other non-business circumstances.)

AOL chugged on, recruiting new customers, then losing them. Yes, they could have offered better service, and yes, they could have lowered rates, but they didn't. They accepted the fact that they would win some and lose some, and seemed to be content as long as they won more than they lost.

So despite the fact that many hard-core geeks disdained AOL, I felt Steve Case and AOL deserved our gratitude since such a high percentage of software developers and others in the IT industry (and journalists) were getting paychecks from Internet services in one way or another, and AOL was doing more to evangelize the Internet than everyone else in the ISP business added together.

I got flamed hard for that article.

Now on to Lindows and Michael Robertson. We're looking at a similar situation. Lindows is for babies, Lindows isn't really Linux, Lindows is insecure, Lindows violates the GPL, Lindows is... you've seen all the complaints.

And that awful Robertson! He's offended "the community" over and over, in so many ways we can hardly count them all. He even managed to turn a potentially important event originally marketed as a vendor-neutral "Linux on the Desktop" conference into a Lindows marketing event, and did it so crudely that a number of significant vendors pulled out, and many media outlets (including NewsForge) killed their plans to attend.

And yet, Robertson is doing more to evangelize Linux on the desktop (and now, laptop) than the rest of the world added together. He deserves plenty of praise for this, way more than he is getting.

Lindows is to Linux as AOL is to the Internet: a cut-down, simplified version with a proprietary interface. Robertson, like Case, realizes that his market is not sophisticated users, but those who are just starting out -- in this case with Linux rather than the Internet.

In case you get all your news from NewsForge and other Linux-oriented media, you may not have noticed that there are still 50 to 100 times as many Linux desktop non-users as users out there. But this is the reality, and aiming for people who have never used Linux gives Lindows a far larger potential market than it would have if it tried to convert all the world's current Linux users to Robertson's vision of what Linux should be.

Perhaps Lycoris or Mandrake or Xandros or SuSE could do a better job of bringing in new Linux users and hooking up with hardware vendors and working with mass retailera like Wal~Mart if they had Robertson's money available, and you can argue that Red Hat should have been doing this all along and that it's sad that they haven't bothered, just as back when proprietary online services and local dialup bulletin boards were just turning into ISPs, you could have argued that any one of a number of them that were more technically competent than AOL could have used AOL's consumer products approach and made itself number one.

But most ISPs didn't have enough cash on hand for major marketing blitzes, and those that did weren't interested enough to actually do it.

This left the field clear for Case and AOL.

Q: How was this different from what we're seeing today with Linux and Lindows?

A: It's the same old same old deja vu all over again.

So a whole bunch of people are buying Lindows and Lindows-powered desktops and notebooks who might otherwise not have tried Linux. Despite its imperfections, Lindows is not a bad operating system. Software for it is expensive only in the Linux/Free Software context. Compared to Windows and Windows apps (or their Mac equivalents), Lindows and its applications look like a near-giveaway.

It's a little early for Lindows users to start exploring other flavors of Linux -- there aren't all that many of them yet -- but the day will come when we start to see them trickling into online Linux help forums and onto LUG email lists, and I don't think that day will be very far in the future.

We must be courteous to Lindows users who want to make the switch to more hard-core distributions, just as we were all (ahem!) polite to AOL users who decided to jump onto the "real" Internet in its early days.

And for every Lindows user who decides to move deeper into Linux, we should give Michael Robertson at least one "attaboy" no matter what we think of him personally or the way his company does business.

In the end, all Linux promotion is good, just as all Internet promotion was good even when the much-maligned Steve Case seemed to be doing most of it.

Click Here!