Oracle buying Sun: what does it mean for Open Source?
April 20th, 2009 | Published in java, oracle, oss | 6 Comments
So Oracle just bought Sun. At first look it might seem like closed source versus open source. Most of Oracle’s well known products are closed source, whereas most of Sun’s products are open source. But that’s not the full story. Oracle is also involved in Open Source. It just takes a different approach.
There are a couple of ways to be involved in open source:
- You can use it in your products. Oracle uses a lot of open source in their products. Java libraries, Linux, Xen. Basically it sells products that use Open Source. Smart, lowering costs, using accepted frameworks or products, charging for support and improved features.
- You can contribute to open source projects. Oracle, for example, has contributed a lot of code to Linux. You can find some of Oracle’s Open Source contributions here: Free and Open Source Software. Oracle also contributed code to glassfish, and toplink to the eclipse foundation.
- Run open source communities. This is the big difference between Oracle and Sun. Sun runs a lot of Open Source Communities: Netbeans, OpenOffice, MySQL, Glassfish, etc. Most of these project mostly live on sponsoring from Sun.
The first two approaches are cost effective: if successful they will lower costs. The last one is expensive: the company taking that approach puts a lot of money into open source, and has a hard time earning that money back. After all, open source products are usually cheap or free, and users are expected to pay for support.
So why did oracle buy sun? There’s couple of possibilities:
- Java credibility. Oracle uses a lot of Java in it’s products and tools. But they’re not very popular in the Java community. The Java community is mostly a 3GL, open source community, whereas Oracle is more about 4GL tooling, and talk about Open Standards. The problem with oracle is that it’s still very proprietary. For example: data binding. Oracle has created ADF, JBoss has created Seam. Seam is going to be the standard, and Oracle is the only one with an ADF implementation. Using ADF means using a proprietary solution: you cannot get a similar implementation from a different provider. Oracle is also very much about making bad programmers productive (4GL tooling which can be used by non programmers), whereas Open Source is about making good programmers more productive (using dependency injection, dynamic languages, functional languages, etc). Oracle and Java are targeting different developer communities. So i don’t think Oracle will fix it’s credibility problem.
- Products. Did Oracle buy Sun for it’s products? Regarding software, could be. Regarding hardware, likely. Oracle has competing products for most of what Sun has to offer, maybe except for OpenOffice. But feature wise, Suns SOA stack (application server, visual tooling, etc) is the most complete competing product against Oracles SOA suite. Everything that Oracle has to offer, both a lot cheaper. So Oracle now has a chance to kill/make money from it’s most important competitor. Other than that, i don’t think Oracle can convince Open Source users to switch to Oracle much more expensive solutions. In fact, we are seeing a lot of customers abandoning Oracle as it is becoming too expensive.
- Customers. As i said, it’ll be hard to convince Open Source customers to Oracle expensive licenses. And Sun already had a problem selling their hardware, as most people are moving to cheaper commodity hardware running Linux.
One more thing: OpenOffice. Many have complained about the OpenOffice community. That it’s not really a community, but mostly Sun developers. I think that this event shows that it’s extremely important that more organizations become involved in the OpenOffice community. Many governments are switching to (or trying to) OpenOffice, and if Oracle withdraws the sponsorship for OpenOffice, these governments will have serious issues. More companies and organizations will have to start contributing to OpenOffice.
Although it is very generous of companies like Sun to invest so much money in Open Source, it really creates an undesired effect for open source products. Open Source should be about community. Lots of people and companies collaborating on building software products. Open Source products run by a single company are too easily impacted when something happens to the sponsor. Good Open Source follows the example of Linux: many companies all contributing a little bit.
April 20th, 2009 at 5:23 pm (#)
Do you think Larry might try to use OpenOffice to finally attack Microsofts hold on the market? He’s been aiming at them for years. Now might be his chance!
April 20th, 2009 at 6:11 pm (#)
This acquisition puzzles me from a technology perspective – especially considering the recent BEA buy. There is so much overlap in the offerings acquired that we cannot expect an actual technical strategic view/perspective from Oracle any time soon. In my opinion this means yet more confusion for customers that are using Oracle as their prime infrastructure and for prospects that have to decide on their next infrastructure.
Oracle only recently explained their strategy how to include, use or kill off the various pieces of the BEA puzzle. Now – imagine having to advice a customer/prospect on what their next Oracle branded infrastructure should look like and be composed of…
Maybe it is best if your company is not depending on the Oracle Applications and/or database to rethink your alternatives, as long as there are any
.
In the end acquisitions like this one might well mean that the adoption of cloud/grid/’whatever’ computing – where you use infrastructure as a service/utility, and no longer have to make a guess about what to use – gains more traction.
April 20th, 2009 at 7:31 pm (#)
@Jan: completely agree with everything you said. Oracle usually promises lifetime support, and they keep this promise. But it is very confusing. And lifetime support doesn’t mean the product gets the attention it needs to be relevant in the future. Customers want to choose future-proof products. And future proof doesn’t mean support. It means new development, new features.
April 20th, 2009 at 7:37 pm (#)
@Mike: yes, i think Oracle will use OpenOffice to irritate Microsoft. I think attack is too much to expect.
But even if Oracle supports and funds the OpenOffice community, it’s still a dangerous situation for OpenOffice users. Any opensource product which relies on a single commercial vendor is a risk. Just like any closed source product which relies on a single vendor… The beauty of an opensource product like linux is that it’s supported and funded by a large number of vendors.
April 21st, 2009 at 8:18 am (#)
@Mike: I agree with Andrej. All opportunities to challenge a rival…
More on MySQL staying open source here.
December 7th, 2009 at 6:24 am (#)
I would really like that oracle would try to make the newly acquired java software products, namely glassfish, net beans and mysql work together with the old Jdeveloper, oc4j and adf faces (toplink ejb included).
Adf faces and toplink ejb 10g are already successfully deployed in glassfish 2.1.1., and I would hate to put that to waste.
Sun has so many good software products that it would be a shame to abandom them because they are not commercial!