To view parent comment, click here.
To read all comments associated with this story, please click here.
I would actually nuance your comments by saying that what geeks adopt at home they eventually drag in to work. Most businesses in the early 1980's had to be dgragged kicking and screaming off their green-screens to PC's. Their IT people were to busy fiddling with their Wangs and mainframes. Just as in the 1990's I started seeing "unofficial" machines pop up in data centers used for various tasks - running Linux.
I whole heartedly agree that they have to figure out how to economically support Linux. Some companies support Linux, but what they mean is they're still supporting RH9, AS 3, and SuSE 9. However, I think it would be worth their while to crack this nut. Of the people that buy computers every couple of years - I imagine quite a number are geeks. It would be nice to see the $999 laptop with a Linux compatible wirless chipset and video card.
I disagree. Many corporate computing environments make it nearly impossible for all but the most die-hard Linux gurus to run their preferred OS at work. Between Exchange, Citrix, and <insert homegrown accounting/sales software here>, using Linux at work is often a fruitless struggle.
Even at IBM it wasn't easy to get everything working until they semi-recently rolled out the Open Client internally. Before this there was an internal community project that maintained a Loki installer for the Lotus Notes client via Wine, which worked sometimes (for sufficiently broad definitions of "worked"). Access to certain VM mainframes wasn't fully compliant with telnet, so they had their own Windows-only remote session client for that. And who can forget the time your presentation looked great in Impress, only to discover (during your presentation for high-level managers) that your bullets were replaced by ASCII rectangles and your titles were hidden behind the background when displayed via PowerPoint?
On the other hand, GAIM with the Meanwhile plugin for Sametime messaging has always been far superior to the Lotus Sametime client. I even use it on Windows!
At home I run Linux for many reasons, among them is that I like having a choice. At work, I don't have a choice. I either play nice and work with their infrastructure, or I forfeit my meal ticket. Thankfully the Open Client is here and Linux at work is a pleasant reality for me. But this is not the case for many less fortunate Linux users.
Edited 2007-02-25 05:10






Member since:
2005-07-13
As you know, Windows is sold to businesses (volume licenses) and to OEM for preloading computers. Thus, if Linux is to gain any significant marketshare, it must come from preloading by OEMs. If this happens, more people will want to run Linux at work, leading to the business licenses.
Historically, it's business use that drives home use, not the other way around. Popular home computers were steamrolled by the trickle-down adoption of Windows, whereas Apple's popularity in the consumer market is barely registering anything more than a statistical anomaly in terms of enterprise use. Besides, anyone that has dealt with a corporate IT department knows that corporate IT types show little concern for what their users want, unless, and only if, it happens to bisect nicely with their own objectives.
Dell simply doesn't have the infrastructure to support home users buying Dimensions and Inspirons preloaded with Ubuntu et al. And they will be calling Dell when they need support, otherwise they wouldn't be buying Dell. Without major subsidization like they receive from Microsoft, or significant economies of scale to balance fixed costs, the investment required in building the engineering, sales, marketing and support infrastructures for a linux distro would likely be prohibitive when they consider the cost-recovery that has to be passed on to the final price the consumer pays. In other words, Ubuntu loaded machines would have to cost more than otherwise identical Windows loaded machines. That is the part that often gets overlooked in discussions about trying to mesh an OSS community-driven model with an established consumer business model.
Dell knows that the Joe Average Dell customer simply will not settle for being provided with a url for community-based forum support when he can't figure out why sound won't work with a certain application. Disgruntled customers can damage brand value, so the big companies rarely gamble.
The only alternative is to engineer the hardware to ensure that it meets common linux compatibility criteria (ie. drivers supported by the current kernel), and then allow the consumer to choose their distro and install it themselves; I don't think that's as big an obstacle as it seems. A modern distro installs painlessly on a platform with supported hardware.
Novell is probably the one company with the infrastructure and resources to provide the backend support that organizations like Dell would require to provide any sort of quasi-official support for linux at the desktop level, but even then that effort is better tailored towards corporate clients than home users.
I'll be happy enough when the OEMs start using hardware components that are supported by more than just Windows, since that is probably the single biggest obstacle to adoption, one that I think is an even bigger issue than application availability. So even if it's a baby step in the overall scheme of things, it's still an important move.
Just my 2c...