Submitted by “US software maker Oracle is considering launching a version of the Linux operating system and has looked at buying one of the two firms dominating the technology, the Financial Times newspaper reported on Monday. As part of a recent study of the open-source software market, Ellison told the newspaper, Oracle had considered buying Novell, which after Red Hat is the biggest distributor of Linux.”
Interesting. While I don’t see Oracle buying Novell, I do think this is great to see a large vendor looking into rolling out their own distro. When larger corporations get more and more involed in the development process, only good things can happen. Novell is not going to sell out to Oracle, so I wonder what distro Oracle may end up going after? This could be interesting to sit back and watch.
Oracle currently has about $7 billion cash on hand. Novell’s market cap is $3 billion. A hostile takeover and Novell is a subsidiary of Oracle.
Well, there are not too many major commercial distributions around. So if not Novell/Suse or Redhat what would be left for Oracle to buy? Mandriva, Xandros or Linspire…? Maybe Mandriva – with all its problems – would at least be relatively cheap for Oracle…?
Anyway, and as a side note, there could be a few smaller but interesting commercial distributions outside of North America and western Europe too (Turbolinux, Alt, ASP etc.) that might have international potential too. For example, the Russian distro Alt Linux focuses nicely both on security and on usability, has a good apt-rpm package management system etc. The problem for potential international customers is, however, that their documentation and support is mostly only for Russian speakers.
Ubuntu…
I don’t know whether I like the whole “every company builds its own distro” thing.
A distribution has a basic concept of how things are done so in my opinion we either get
– a generic distribution that all big vendors use and rebrand with their logo everywhere (so it would have been smarter to find a good distro and support it)
or
– we get shitloads of incompatible and different distros, one for every big piece of software which would lead to more fragmentation
This is true, the number of distros is both a pro and con at once. Lots of choices gives you a nice selection, but creates a nightmare for developers.
I would like to see them ‘piggy-back’ (is that the right word) something like:
Fedora Core 5 ‘Oracle Edition’ – replacing FC5 with your preferred plat.
This way, the development of the platform is left to the platform developers – and the application is left to the application developers.
And this is the basic problem with “Flavor of the Week” distributions. If Oracle created their own distro, it could lead to incompatibilities with the Oracle database for others.
Personally, when I see things fork off for no reason, I cringe; ala: ubuntu, kbuntu. There is no difference except packages?? That’s called software bundling.
Nah. All they need to do is come up with a common package management system that all the major corporate distros can agree on. The smaller distros can follow suit or not. Then the software vendors can develop apps for that system. Those distros that don’t get on board, include it in will just be left without. Some distros use only free as in speech apps anyway, and wouldn’t want anything to do with commerical apps anyway. That’s their choice. Vendors like Red Hat, Novell and Sun (if they still intend to offer Linux on the desktop) need to agree on that one package management system.
Ubuntu and Kubuntu are the same distribution. More specifically… your normal Ubuntu install is a combination of ubuntu-desktop and ubuntu-base.
Kubuntu is… kubuntu-desktop and ubuntu-base… Honestly though I think they should change the names some…. ubuntu-gnome/ubuntu-kde or something but thats really a different issue
They are meta-distributions. I would seriously like to see more of these meta-distributions show up… its just so much smarter then rolling a brand new distribution. Oracle could easily roll a oracle meta package… combine that with a ubuntu-server meta package and you have a ‘oracle server distribution’. While I could download and install my ‘oracle distribution’ I could just as easily add the oracle and server meta packages to get the same thing, but mixed into my current installation.
If Oracle does its own Distro, IBM will follow suit.
I think Redhat buying JBoss has made IBM and Oracle very very mad.
There are so many distros, hundreds, you get lost…
All current Oracle products are certified to run on both SuSE (Novell) and Red Hat Enterprise Linux. So if Oracle is going to buy a current Linux distro, it’s one of those two, for sure.
Of course, Oracle could roll it’s own distro. Most certainly they have the development staff to do it. Also, they would have a head start in that they could use all of the GPL and LGPL bits of both SuSE and Red Hat.
Of course, Oracle would be interested in the customer base and engineering talent of Novell or Red Hat as well. So it’s much more likely that Oracle would acquire one of those.
If Oracle bought Red Hat, they’d now get JBoss with the deal, who they were courting anyway.
It makes sense for Oracle to have it’s own Linux distro, for them to complete the enterprise software stack, and compete head to head with MS on everything.
I’m no fan of Larry Ellison, but it would be fun to watch the Larry vs Bill/Steve deathmatch. Yes there already is one between Oracle and MS, but Oracle adding an OS to it’s offerings would send the deathmatch intensity through the stratosphere.
Yes probably Dead Rat, I mean Red Hat. It’s got the O.S that people use for the enterprise, and Oracle has an enterprise-grade DB, so Red Hat Enterprise Linux + Oracle 10g = Good team(TM). Also, Oracle would benefit from RH customer base, they’re wealthy customers.
I don’t see them doing it from scratch. Oracle Linux will be based on RHEL, most probably.
I agree with your thinking though it leads me to wonder how Red Hat or Novell would respond to being acquired by Oracle. Red Hat seems the better fit but I would think they would be reluctant to go quietly whereas Novell has still not stabilized since buying SUSE Linux but may be more agreeable to being brought. I think Larry Ellison is just testing the water to see how the financial markets respond to his comments. Already because of their company name being mentioned Novell stock has moved up and Red Hats has gone down for the day.
More info at this article at TG Daily:
http://www.tgdaily.com/2006/04/17/oracle_confirms_sights_set_on_nov…
Very amusing article:
“I’m not going to spend $5 billion, or $6 billion, for something that can just be so completely wiped off the map.”
I think all of this hints at just how worried Larry Ellison is. I think he’s got some right to be.
When Red Hat get their JBoss and Red Hat database (i.e. Postgres) act together I’d be more worried about Oracle being wiped off the map. Who in their right mind is going to buy expensive database and Java stuff from Oracle when they can get the same stuff and the whole smack from Red Hat far, far cheaper? The amusing thing is that Red Hat will never become another Microsoft simply because when the time comes other companies will take the open source software Red Hat uses and create competitors. This will have a pleasing effect on Red Hat because it will ensure that they always have competitors, and that’s what a company needs to last for a very long time. It’s impossible to survive as a monopoly on your own stuff for any length of time.
I think Larry has some pretty severe delusions about Oracle being a multi-billion software company that has a right to exist. There have been many such examples littered through history. Look at the big companies that existed a hundred, fifty or even twenty five years ago. The vast majority have been wiped out of existence. IBM are not so much under threat because they do many, many other things, but unless Oracle shock us all there’s going to be a lot of blood on the carpet over the next few years.
I don’t see it coming off, as Larry tends to have these wild ideas that don’t happen. If it does happen then it will certainly be a really big long nail in the coffin for both companies.
Oracle simply don’t have a clue what’s involved in producing an operating system and demonstrated that with those jokes called Network Computers. Producing and maintaining a Linux distribution is a also huge logistical exercise and there’s the inevitable conflict between open source software and the proprietary software, which exists even now at Novell. I doubt whether Oracle even know those conflicts exist let alone how to solve them. Oracle are a truly inept company in these areas, as Novell are also proving.
I don’t see why Oracle or IBM should be annoyed by Red Hat’s purchase of JBoss. IBM and Oracle sell ludicrously expensive and awful software in this area, and they’re going to get what’s coming to them. The business and industrial world is all about change, and I would advise them to get used to it.
Edited 2006-04-17 23:44
If Oracle enters the Linux market, you will see Linux fragment the way Unix did. You will also see Microsoft laugh its way to the bank, as the number of different “enterprise” Linux solutions increases from 2 to 20, and companies try to avoid the fiasco.
If Oracle enters the Linux market, you will see Linux fragment the way Unix did.
Oh, so it isn’t fragmented now? That won’t happen with Linux in the same way as Unix, because Linux is not a proprietary world. If Oracle fails and creates a huge mess, which they will do if anything happens, then someone else will just pick up the gauntlet. The world just doesn’t revolve around Novell (although they act like it)or Oracle regarding Linux, and Red Hat will never be another Microsoft. That’s the real problem for Microsoft.
“Oh, so it isn’t fragmented now? That won’t happen with Linux in the same way as Unix, because Linux is not a proprietary world. If Oracle fails and creates a huge mess, which they will do if anything happens, then someone else will just pick up the gauntlet. The world just doesn’t revolve around Novell (although they act like it)or Oracle regarding Linux, and Red Hat will never be another Microsoft. That’s the real problem for Microsoft.”
No, it isn’t fragmented now. In the enterprise space, you have two Linux distributions, SuSE and RedHat. IBM, Sun, Oracle, HP, and every other developer targets these two distributions. Now what would happen when Oracle starts pushing their distribution? IBM will counter with their own distribution to counter Oracle in the database and application server space. And then all of a sudden, you will have HP, CA, Dell, SAP, etc. and everyone else pushing their own “Enterprise” solution that does not help their competitor.
Now what would happen when Oracle starts pushing their distribution? IBM will counter with their own distribution to counter Oracle in the database and application server space. And then all of a sudden, you will have HP, CA, Dell, SAP, etc. and everyone else pushing their own “Enterprise” solution that does not help their competitor.
No actually. People will continue to target the market leader, Red Hat, and Oracle could even take Red Hat’s software and create their own compatible distribution. Although, most Linux distributions tend to be fairly compatible anyway.
There’s a big difference between creating your own distribution and people actually using it though.
Instead of making another disto (like we don’t have 5 million already) Oracle should start paying more attention to virtualize their database software. Also providing a cheaper method for smaller companies to use their product would be better than just creating their own distro. No one wants to reinstall an OS just to use an application.
Edited 2006-04-18 02:13
I predict MS will start giving away SQL Server with windows soon afterwards.
For my money as a decision maker on which technology to implement I need to have clearer and more concise information. I need answers that matter.
It seems now days there are so many distributions where the ‘features’ are more or less variations of packages.
I don’t really see the arguement of ‘greater choice’ as valid unless the distribution actually offers something substantially different to put it out on its own.
If Oracle were to produce a distribution optimised to offer greater scope and performance for its database suite then that would probably make it worth investigation, but if it’s merely to in effect rebadge another flavour of Linux then it’s really not worth much of my time.
No, it isn’t fragmented now. In the enterprise space, you have two Linux distributions, SuSE and RedHat. IBM, Sun, Oracle, HP, and every other developer targets these two distributions. Now what would happen when Oracle starts pushing their distribution? IBM will counter with their own distribution to counter Oracle in the database and application server space. And then all of a sudden, you will have HP, CA, Dell, SAP, etc. and everyone else pushing their own “Enterprise” solution that does not help their competitor.
Depends on how things work out; both Novell and Red Hat have their ‘community distributions’ in which their software is based upon.
Why would Oracle go out, purchase Novell (dying in the ass as we speak) or Red Hat (massively over valued) when they can easily embrace a community distro; OpenSuSE or Fedora Core, take you pick; Oracalise it, and bundle it as part of their linux version of Oracle?
Can you just imagine it? “Oracle database, with a free operating system!” – we can all imagine the number of CIO’s creaming themselves when they see Oracle provide something like that – it’ll also give Oracle the ability to tweak aspects of the operating system, and move it in a direction that directly benefits them rather than having to work within the ‘system’ which software vendors use when communicating with partners.
1. Oraclenux
2. Elissonux
3. Oraclinux
4. Oralinux
5. Orax
6. Ellix
7. Ellinix
…
PS: Soo stupid…
Maybe Ellison is kicking himself for not spotting all this sooner. Red Hat is now capitalized at, I think, around $5 billion and Novell at around $3 billion, but since they offer rarity value and quite a few buyers are out there, taking them over would probably cost quite a bit more than that. Wall Street would hold out for top money considerably above the current share prices. All of a sudden, Oracle’s cash mountain of $7 billion doesn’t look quite so large. Worse, the actually Linux turnover from either outfit is a fraction of their market value so, inititally anyway, a buyer would be paying hugely over the odds for a fairly modest income stream.
Personally, I think we need both a strong Red Hat and a strong SuSE. Otherwise, competition will suffer and we will all be the losers. Novell is an especial problem for a buyer, since you would be paying for all the netware legacy stuff and other bits and pieces that even Novell are trying to get rid of. They do have a very nice line in indentity management, though.
IBM must surely be favourites for getting SuSE. I don’t see how they can stand by and let the only two serious commercial Linux distros be acquired by competitors.
There’s a big difference between creating your own distribution and people actually using it though.
Just reading back, hasn’t Oracle already tried this one, claiming, ‘you don’t need no stinking operating system!’ and made their own very stripped down basic operating system with Oracle planted ontop?
If Oracle want to do something, why not purchase SUN? they’re undervalue, poorly managed, but at the same time, a great treasure trove of IP, talented engineers and a focus on quality hardware design.
im stating to feel like linux will give us a world where one black box runs the database, another runs the webserver, a third runs something else and so on…