In the press and analyst Q&A session following the NC (well, mostly Solaris) launch for Sun the other day, I asked a simple question – what does Solaris need to compete with Linux on – cost or feature/function? Before I get to the responses, I should note that I believe Sun’s made fairly convincing arguments on both fronts in recent months, but feature/function is in my mind the far stronger position to take.
Linux does not have, as far as I’m aware, anything comparable to DTrace or ZFS, among other features that will be available in version 10. As an aside, before you Linux advocates flame me, understand that I’m not saying Solaris is better than Linux. These features are of more interest to some areas of the market than others, and Linux is not going anywhere but up. But while Linux is a great operating system – not to mention my chosen desktop platform – in terms of enterprise features, Solaris in my view has a few extras for the enterprise crowd that Linux lacks (for now).
The cost argument, however, is a different animal. It is predicated on a contention that Sun is trying exceptionally hard to drive home – for the enterprise, Linux = Red Hat. This is not just Sun FUD, however – for the domestic market at least, this assertion is very supportable (just look at the ISV rosters). And Red Hat has indeed irked some customers with its cost hikes. As a result, Solaris as cheaper than Linux does work, if you freely substitute – as Sun wants you to – Linux for Red Hat. Which brings me back to the responses.
Jonathan Schwartz answered my question with one of his own – “You mean Red Hat?” And that actually segued nicely into my real question – which is really a larger question for Linux supporters. What if Red Hat != Linux?
What if another distribution was tapped as a target for ISV support – say Debian stable? Scott McNealy scoffed at the idea saying “there’s no there there,” and Jonathan more or less shot it down. I can’t say as I really blame them. I’ll certainly grant that it’s a major reach. Supporting a new OS, as any ISV knows, is not a trivial exercise. It takes a Lot of money to make it happen, and a clear incentive from a cost/benefit standpoint.
But just for the sake of argument, let’s say IBM (or substitute your favorite major ISV with a big bet on Linux) sees Solaris making serious incursions into Linux territory by hammering Red Hat pricing. What’s their play? There are several, but none that look great, from where I’m sitting. The logical move – pushing the Novell/SuSE alternative – is certainly a viable option, but still a bit pricy. Might IBM be a bit more aggressive and consider targeting a free distribution like Debian with its application portfolio? Maybe, maybe not, but I’d sure give it some thought. It seems a much better option than, say, starting their own distro or buying Red Hat or Novell. You’d undercut one of the major pillars of Sun’s Solaris pitch – cost, albeit it by introducing significant cost component to your own business.
The other wildcard here is the emergence of open source support vendors like SourceLabs or SpikeSource. Could another distro (Debian, Gentoo, etc) be supportable, at a significantly lower pricepoint than Red Hat or Solaris? Perhaps.
Anyhow, this is just so much idle speculation at this point (and hence is making its way onto the blog rather than a research report), and who knows – Red Hat may well be the big kid on the Linux block for years to come. But I’m curious to see if any project/vendor combinations see the opportunity in attacking Red Hat from the bottom up on a cost control basis, and can attract enough ISV support to make it work.
A bunch of if’s, but an interesting question to consider, I think. In the meantime, I’m sure Sun’s more than happy with the status quo.
Alex Bosworth says:
November 18, 2004 at 12:48 pm
Interesting post, I think we here at SourceLabs would definitely be interested in supporting system configurations beyond Red Hat. The other day I saw a presentation from Feedster about their LAMP stack that is run on Gentoo, so Red Hat is certainly not the be all end all of serious Linux useage. That being said, SourceLabs is interested in supporting application stacks over just OS support.
November 18, 2004 at 2:24 pm
interesting points Alex. i certainly was not aware that Feedster had settled on Gentoo – that's interesting, and i'll have to drop them a line.
and i would certainly agree that RH = Linux is an oversimplification of the situation – clearly SuSE and the Debian's and Gentoo's of the world have their share of adherents. from an ISV perspective, however, the world does look something like the picture that Sun paints.
and it's the OS+application stack that's interesting, which is in turn why i find the work that you guys are doing intriguing. we should definitely chat.