It's a moot point: there is little argument that the database that drives the internet is MySQL, but is it better as an open source project or in-house at a major software company?

And it's an important question for Sun has agreed to pay USD1,000 million for the product.

But the changes may not be quite as radical as they would be if the same course of action were to be taken with other open source product.

MySQL is produced by MySQL AB, a Swedish company - but its headquarters has been split between California and Sweden - with a gradual move west.

MySQL is phenomenally popular. The world's biggest internet companies use it - Facebook, Google, Nokia, Baidu and China Mobile. So do some of the smallest - like us, for example: all the servers in our group of companies run on MySQL as part of a Linux, Apache, MySQL, PhP (or "LAMP") stack.

But although MySQL is open source, and is free for the vast majority of users, it is not entirely free. And it's here that Sun may be seeing some value. MySQL has generally not enforced that part of its licence agreement that says that any developer that incorporates MySQL into his product, or requires a client to install MySQL as a condition of running his application must pay for a commercial licence.

MySQL has largely relied on the honesty of the developers - and to be frank, not a lot of them have been very honest.

The condition is not easy to interpret - seemingly, if a company hosts an application and sells access to it, that's not chargeable. But if the company sells exactly the same product to a client and says "you need MySQL on your servers to run this," that is chargeable.

But Sun is making a bid for the de facto takeover of the net: already it owns Java - which it mostly gives away. It also owns OpenOffice.Org - to which organisations around the world are migrating in the face of ever-increasing costs of ownership for Microsoft Office.

Despite critical acclaim in 2005, Solaris 10, Sun's own Unix-based operating system, failed to grab market share - even on Sun's own servers. That was just two years after Sun abandoned production and distribution of its own flavour of Linux in favour of a distribution deal with Red Hat Linux and SUSE (which is now part of Novell). Since then, Sun has also added support for Ubuntu, which has been developed especially for developing markets.

Also in 2005, Sun released "Open Solaris" - and not only was the code open (but protected) the product is free.

The MySQL purchase fits with Sun's overall strategy: free versions and paid for versions of the same product. The principle difference is the level of support. With free versions, you have to pay for it straight away - although one of the benefits of the open source movement in general is that it's often quicker to look up an answer from someone who has had the same problem as it is to get a reply out of a support desk that only works office hours in a different part of the world to where you are.

With paid for versions, you get installation help, set up assistance, problem solving with applications installation.

At least, you do with paid for most open source products.

And here's the interesting thing: the level of support from proprietary companies is often no better. And just because you've bought the software, don't expect e.g. Microsoft to help you with it if you get into trouble unless you send them even more money.

So, if Sun's strategy is to sell commercial licences to that small class of people that want to buy them, and provide support to the hundreds of millions of users who run into trouble occasionally, then this might just be a good idea.

If there is one concern from the user's point of view it is this: Sun is gathering a huge concentration of the internet's core technologies. One of the reasons Microsoft is do despised and also why IBM, ultimately, lost its direction was because they aimed for total market domination. What if Sun decides that it wants to exercise greater commercial exploitation over the products?

And that is a very real threat. Every year, at the company's annual general meeting, in discussions with market analysts, someone is going to demand to know why the company is not making short term profits by charging even USD10 for a copy of OpenOffice.Org and USD50 for an installation of MySQL, and the same each of the various bits of Java.

Given the short-termism of the markets, someone will instruct a class action lawyer to sue the company for failing to maximise the returns on investment.

And then the whole internet - and open source community - will be in trouble.

If it's not immediate profit maximisation, then what's in the deal for Sun?

Actually, quite a lot: Sun has been making a tidy sum from selling support services for the (distant) MySQL rival PostGre. Being able to rake in revenues from right across the world will improve the returns on that investment.

Also, Sun software engineers will, undoubtedly, be identifying what server each support client is running - and that opens sales channels to grab business from the likes of HP, Dell and - yes - IBM. That information gathering exercise would cost an enormous amount - and produce a far less accurate and comprehensive picture - if it had to be done by survey.

But if Sun go after the Apache Foundation and / or PhP, then things will begin to look extremely disturbing. For then, they will own the vertical market that is the technology underpinning the internet.

And that concentration of power is doomed to be abused at some point in the future.