Tue Sep 7 11:11:51 2004 The future of
Linux is getting uglier
Posted by Tony Lawrence
Search Keys: business and consulting
Referencing: /Blog/B1062.html
Right now, Linux is ascendant, and Microsoft (and a lot of other people) are worried about its effect on their income.
I quite seriously wonder how long it can last. The oft-discussed "succession" problem is part of my concern, but it goes deeper than that. Not that succession isn't important to Linux; it certainly is. Linus Torvalds may insist it has already been dealt with: (from http://zdnet.com.com/2100-11-526907.html?legacy=zdnn)
I think that a lot of the talk about the 'succession' is due to this--people
see the project growing, and see other people having a large impact, and don't
realize that it's already long since grown past being just 'Linus.
but that's really not the case at all. Linus is still the political head and no one below him has the fame and following to take over. Others have commented on this: (from http://www.business2.com/b2/web/articles/0,17863,528386,00.html)
A normal corporation is constituted with a board of directors that makes sure
solid succession plans are in place. No question, a breakthrough such as Linux
would never have happened had Torvalds gone by the rules of a normal
corporation. Still, it's possible that the Linux movement, and hence further
development of the software, could be thrown into disarray in the absence of
a smooth transition plan for new leadership. It's also possible - though
unlikely - that the Linux movement could splinter into factions in the face of
a leadership vacuum.
I don't know why they say "unlikely". I think they forget that the corporate interests (RedHat, IBM, et al.) are involved in all of this. If Linus gets hit by a bus, "taking over" will definitely be discussed in corporate boardrooms.
But I think there is already danger afoot. This http://www.msnbc.msn.com/id/5907194/ article tells how companies with a strong interest in Linux internals curry favor by hiring people in or close to the Linux "inner circle":
Hewlett-Packard, which has hired two maintainers and 100 lower-level Linux
developers, tries to hire only people who are within reach of Torvalds' inner
circle.
"I call it 'counting the hops,' which means we figure out how many steps
removed this person is from a top maintainer. I try to keep it under two
hops,"says Martin Fink, vice president of Linux at HP. "The way to get
stuff done in the Linux community is to hire the right people."
While those who are hired by HP and the other companies mentioned may insist that their employment doesn't affect anything, it's obvious that it does: (from http://www.msnbc.msn.com/id/5907194/ )
The payoff is real. Linux now contains bits of code written by
NetApp's programmers.
Linux flunks the Purity Test.
Yes, that's unavoidable at some level. But surely the public perception of Linux isn't that companies with enough money get code that benefits their apps and otherwise.. well, tough. If you aren't shoveling money at someone close to Linus, you can just step to the end of a very long line.
That's a situation ripe for abuse. In the case of the NetApp contributions, these may very well be just legitimate and needed improvements to NFS code. But who ensures that? What happens when NetApp or some other company playing this game sees an opportunity to affect a competitor by way of kernel code? There doesn't even have to be complicity on the part of the Linus confidant - laziness and inattention will suffice.
A scandal here could damage Linux. I'm sure that most of the people involved are aware of this and perhaps even believe that they can remain totally honest - that they can keep their Linux contributions pure. History and even the most basic understanding of human nature says otherwise.
Enter your email address for automatic notification of new posts here
(be sure to whitelist 'feedburner.com' if you use spam filtering)

| Views for this page | ||||
|---|---|---|---|---|
| Today | This Week | This Month | This Year | Overall |
| 4 | 4 | 68 | 1,259 | 5,779 |
Have you tried Searching this site?
Unix/Linux/Mac OS X support by phone, email or on-site: Support Rates
This is a Unix/Linux resource website. It contains technical articles about Unix, Linux and general computing related subjects, opinion, news, help files, how-to's, tutorials and more. We appreciate comments and article submissions.
Publish your articles, comments, book reviews or opinions here!
CommentsBlog1084 :
---September 7, 2004
Then I suppose it's a good thing Linux is only a kernel. After all 90% of the software I use can run just as well on any other kernel or system.
FreeBSD, OpenBSD, NetBSD, Dragonfly BSD, maybe even OpenSolaris (if that will ever exist), or even GNU's Hurd.
If you look at almost all major projects such as FreeDesktop.org for intance, then specificly aim to make it as OS independent as possible. Wine will run on FreeBSD, Mono will provide .NET support even for OS X. Hell a large percentage of Apache servers run on FreeBSD or even Windows operating systems!
Take a look at my personal OS of choice: Debian unstable. They have 3 major non-linux ports, and more are possible. Every thing is kept as independant as any other peice of software.
And forks aren't nessicarially bad. Look at what happenned from the EGCS fork from GCC, how was that anything but good for the FOSS software movement, and GCC is hell of a lot more important then just the Linux kernel itself. It's the compiler of choice even for Apple and OS X!
Your right to worry about HP and IBM and them getting all nazi about the kernel, but it's also self interest of other competing companies/projects that will help sustain the free software movement in case all of a sudden Linux croaked a violent death.
But then again I could be confused... I suppose it would help to differenciate between Linux as a kernel and the generic market-speak driven term: The Linux OS. hehe.
Linux kernel itself is realy cool though, and I'd hate to see it dissapeer. That would suck. Especially with the newer 2.6 series kernels it's realy ahead of other kernels in terms of performance and capabilities.
And HP definately has good reason to get on the good side of mainline kernel contributers. Here is a link to a IBM document that says that thru their various stress tests that they use to test Linux kernel changes they've found a 600% increase (at least in static web page performance benchmarks) in kernel performance in a 8-way Xeon-based IBM machine by simply switching from a vanilla 2.4.x kernel to a 2.6.0 kernel.
http://www-106.ibm.com/developerworks/linux/library/l-web26/index.html
Who wouldn't want those developers as employees?
--Drag
---September 8, 2004
But when the REASON you want those people is specifically so that you can influence kernel development, the aroma of fish is not far way.
As I said, most, may even all of these people may be straight up, honest as the day is long folk. But in any other situation, we'd call this a conflict of interest. It is potentially dangerous.
Note Linus's comment from the article: he doesn't trust companies and won't take code directly from them. He's dead right about that: he knows that sooner or later they'll be tempted to feather their own nest and that he can't guarantee that he'll spot it. So he delegates the "keep 'em honest" to someone he trusts. But that person now works there - is collecting paychecks and other perks. That's way too close to being exactly what Linus is afraid of.
Sooner or later something ugly will come of this.
--TonyLawrence
---September 8, 2004
Add your comments