Red Hat Nears $1 Billion In Revenues, Closing Door On Clones 201
darthcamaro writes "Red Hat is almost at its goal of being the first pure-play open source vendor to hit $1 billion in Revenues. Red Hat reported its fiscal 2011 revenues this week which hit $909 million. Going forward, Red Hat has already taken steps to protect its business by changing the way it packages the Red Hat Enterprise Linux 6 kernel, making it harder for Oracle to clone. 'We are the top commercial contributor to most of the components of the Linux kernel and we think we have a lot of value and we want to make sure that, that value is recognized,' Red Hat CEO Jim Whitehurst said. 'In terms of competition, I don't think we necessarily saw anything different from before but I'd say better to close the barn door before the horses leave than afterwards.'"
Re:Here goes last supporter of open-source (Score:5, Insightful)
Goatse, again.
Man, you are hilarious. No one in history has ever done that before. And you've created a couple accounts today just for that. When you look back on your life, I'm sure you'll feel content and fulfilled.
Do not panic (Score:5, Insightful)
I believe they have no beef against CentOS, actually I've seen at least one Red Hat employee encouraging the use of CentOS, since Red Had is the "de facto upgrade path" (not the exact words, but something along this way). So you freely enlarge the customer base, which will go to Red Hat when they need higher level commercial support. And for the free ones, even Microsoft has recognized they cannot sell to students, and are giving away the software anyways.
However Oracle is another deal. They just slap Oracle logo on Red Hat, do not acknowledge the source, and sell is as "unbreakable Linux". This would make a regular person ashamed of himself. They benefit a lot from open source but not giving back much in return. Do not start me with what they're doing to Solaris, Java, and OpenOffice...
So I'm with Red Hat on this one, at least until they do something directly bad to CentOS.
Re:Well that's ominous (Score:5, Insightful)
As far as I can tell RH does give back. They give back a lot. And it must get kind'of annoying for them that other companies - some of whom give back nothing - copy RH Linux and significantly undermine RH's ability to earn revenue from their distro.
There are tons of companies out there that violate the GPL, give nothing back, or even actively undermine open source. I would suggest that your disapproval is better directed at them.
Re:Diff? (Score:5, Insightful)
That only gives you the patch broken up in space, not in time.
The idea is that Red Hat has patches for specific issues that are developed at different points in time. These patches may modify the same files as previous ones, or even the same blocks of code. By having all patches applied at once, the singular diff does not tell you which component of the patch fixed which issue.
This is really only relevant for providing commercial support. Previously, by having patches associated with known issues applied sequentially, it was much easier for another company to say "Oh you're having Issue X? Well Patch Y will fix it." Now their options are to reverse-engineer the monolithic .diff to find the part that fixes a specific issue, or tell their customers they have to apply the entire patch. Again, that's not something you'd care about if you're a desktop end-user, but in a corporate IT environment it makes a difference.
Re:Well that's ominous (Score:5, Insightful)
I don't really see it being that good for them, either. Oracle isn't going to have much trouble reverse-engineering the patches back out, but RedHat now ends up in a more difficult position: fewer of their patches will be incorporated upstream, so they have to spend more work porting them into each new release; they'll have less community review and bugfixes in their patches; and they're going to alienate the community.
I very much doubt Red Hat has any plans to change the way they work on the kernel master branch. This seems to be about their cherrypicking and backporting of patches to RHEL kernels. They want other distros - particularly Oracle it seems - to either do that work themselves or admit they are just rebranding Red Hat's work. For example in that big mega-patch they can simply add a few whitespace changes, if the same changes show up in Unbreakable Linux you know they started with the Red Hat kernel and worked from there. To be honest, I'm somewhat ambivalent about the whole thing. Making it a bit harder to cooperate is bad but making sure credit goes where credit is due is important so that people do the "invisible" work too.
Scummy is taking without giving back (Score:5, Insightful)
If you want scummy, look to companies like Oracle which just take, repackage, and rarely give back. They're the real problem, not RedHat.
RedHat's patches still get submitted upstream for inclusion in the main kernel, which very often does happen.
I have no sympathy whatsoever for leeches that were taking RedHat patches and rolling their own distributions without contributing enough back on their own.
I fail to see how this affects seperate distros like Debian, which aren't based on RedHat-patched source in the first place.
Not good for the future of Linux (Score:4, Insightful)
If every distro is doing the same thing, this is not going to be very good for the future of Linux. Engineers at every distro are going to waste a lot time trying to figure what other distros had been patching, which part of the code had been changed while a specific issue was fixed, etc. Everyone is going to end up wasting a lot of time, and creating a lot of confusion.
Even though Linux distros are quite fragmented, but the current kernel development has been working quite well, because every distro is playing by the rule (more or less), which is quite transparent. Now, with this kind of one time big change by RH, even though you can still diff on all the source codes, it's not going to be easy to figure what has bee done (and why). And I think it's going to trigger other distros to behave similarly.
And it will be even harder for the users. As a user, if we have in-house-built applications that rely on specific version of a library or module, we might not want to have a giant patch on basically everything, we probably want only small, concise, specific patch for some critical security problems. I'm starting to wonder how are we going to manage that.