So when I say “Well I reckoned that $20.00 a share was going to be the magic number, but hey I was pretty close given it went for $19.38!” It’s ok to say “You lying little scamp, you’re just saying that…”

But I did. Honest!

/* I’ve now managed to find a back-up of the original thinkovation database, and have restored the original Oracle/BEA Post */

My wordpress incompetence aside…

I still think that this is a good thing… But there are plenty of challenges ahead of the Oracle and BEA teams. I know it’s ancient history, but I have painful memories of going to visit the NetDynamics and Netscape (Kiva) teams at Sun, soon after the “merged” app server was announced. I ended up meeting the teams separately – and got the impression that all the NetDynamics guys really wanted to talk about was how rubbish Kiva was, and the Kiva guys did the opposite..

Back to today… There are now two pretty rich and fully fledged middleware offerings in the mix;

• Oracle’s Fusion is strong but badly described, and hasn’t got much traction outside Oracle’s well understood and well sold-to customer base.
• BEA’s Aqualogic story is also strong… and better described

The problem now is …. how do you merge the two technology sets… Ellison is quoted in a Washington Post article as saying that while both companies have “numerous” middleware offerings, BEA’s line is nonetheless “overwhelmingly complementary,”.

I guess that I’d say that if I was Larry Ellison too – But it’s really not that accurate a statement…

As Micheal Meehan points out in his commentary if the deal goes through Oracle will be the proud owner of;
2 ESB products

2 BPM products

2 Portals

2 SOA stacks

2 Data integration platforms

2 Development tools

Even if you step back and let your eyes go out of focus… that’s still two pretty overlapping sets of technology.

In fairness, Oracle has a better track record than some (Sun for instance) where it comes to taking the tough decisions that are needed when merging products – and it’s going to have some tough decisions to make.

Meanwhile – IBM’s Websphere still has the #1 slot, by some margin. At best Oracle bumps up to #2 in the J2EE middleware space – And to number three if you talk about platform middleware and don’t exclude Microsoft’s .NET – Which by my (admittedly wet finger in the air) reckoning is grossly overlooked, or grossly under-estimated by most analysts.

Sun has acquired MySQL AB the Swedish software company behind the eponymous (well without the “AB”) open source database. for what the press release describes as “approximately$1 billion”.

This move is of real interest to me since I’m currently writing a “MySQL and VB.NET” how to guide (see my development blog for the first installment).

I think that on balance this is good news. Of course there are caveats – Sun has a mixed record when it comes to acquisitions – but provided that they keep faith with the user base (and I think that they will) and can keep a few of the key MySQL engineers (and I think that they can) MySQL will continue to prosper.

There are couple of other “wrinkles” two key database vendors own bits of technology that are pretty “core” to MySQL; Oracle owns InnoDB (one of the core database engines that MySQL uses), and IBM owns Solid (another storage engine). But both Oracle and IBM should be smart enough to know “not to go there” when it comes to horsing with something as popular with the community as MySQL.

Some people have asked me whether this makes MySQL a “contender” – in other words should Oracle and IBM be scared by the threat that MySQL poses to their databases? My answer is that MySQL was already a “contender” in many respects – Sure, banks haven’t flocked to use MySQL as the data store for their core systems – but gazillions of web-sites and “Web Two-do-Oh” businesses are built on MySQL.

Where it comes to existing applications, and the data that resides in them, people don’t just “swap out” one database for another – As a general rule, databases only get switched off when the apps that use them get switched off – And here’s the long-term “gotcha” for the existing DB vendors – As new applications emerge MySQL will increasingly be the database that runs them. Whether it’s SugarCRM, dotProject or WordPress… the next generation of CRM, Collaboration, and publishing applications all (more or less) use MySQL.

Some of my colleagues are still sniffy about MySQL – one past colleague even refused to call it a “Database” – insisting on referring to it as a “data-store” – And, indeed, I would have said the same right up until the end of 2005 when version 5.1 was released (which added vastly improved support for transactions, along with cursors, stored procedures, triggers and views).

MySQL is out there – in production. It sits behind millions of websites (including mine), and it’s increasingly finding itself used as the data store for custom and ISV applications.

I’m developing a client-server application for a client, they’re an all windows shop, and the client will be written in VB.NET (so that they can maintain it themselves) and I’ve chosen to use MySQL rather than add to my client’s cost by licensing SQLServer or another database.

Wouldn’t it be cool if Sun were to capitalise on this – Add a few simple API’s to make it easy for a third party application to administer a MySQL database (not that it’s that hard to do today) but with a small investment on embeddability – MySQL could be even more attractive to ISV’s.

Another commonly asked question relates to MySQL’s scalability. MySQL’s scalability isn’t really in doubt – it’s sitting behind some massive applications already – and the work on improving the performance of the different core “engines” that you can choose continues.

So.. MySQL’s adoption will grow, surely and steadily – and in a couple of years from now, some of the bigger database vendors will notice some eating away at the edges of their market share – Their choice will be to carry on competing in a market that will be essentially commodtised, or to carry on adding value over and above the core database – IBM didn’t buy Cognos for $5bn just for laughs, let’s face it.

DB2 and Oracle will retain their blue chip client-base – residing at the top of the market in terms of scale and complexity – They’ll just begin to notice that the number of “little apps” that are deployed on their technology will first stop growing as quickly as it once did and will then begin to slow.

The biggest potential loser, I reckon, is Microsoft’s SQLServer – SQLServer is a tremendously sophisticated, scalable and fast database and for 99.99% of applications it’s every bit as good as DB2 and Oracle… but Microsoft hasn’t been able to get past the “toy database” stigma – which is really very very unfair – and now there’s a free database that can also cater for the vast majority of applications – MySQL.

This is the key point – So what if MySQL can’t, yet, scale to a gazillion writes/second… the number of applications that require that kind of scale is so low that it can almost be written off as a statistical error. Steve O’Grady at Redmonk blogged on this topic this morning and produced a quote that I’ve never heard before – Apparently someone at Oracle once said “MySQL is a Toyota Corolla to our 747″ in order to disparage MySQL… what an utterly stupid thing to say – As Steve points out – there are many many more Corollas kicking about than there are 747’s and very very few of us would choose a 747 to nip down to the shops or to deliver goods to our clients – And just as the corner shop is far more likely to choose a Corolla over a 747 for the delivery – it seems that the budding e-commerce and social networking sites are choosing MySQL’s Corolla over Oracle’s 747 to build their businesses on.

The Commission’s examination will therefore focus on all these areas, including the question whether Microsoft’s new file format Office Open XML, as implemented in Office, is sufficiently interoperable with competitors’ products.

The second complaint, in which the browser company Opera is cited as the company making the allegations, Microsoft is (once more) being investigated for “tying” the explorer browser to its operating system. The Commission goes on;

In addition, allegations of tying of other separate software products by Microsoft, including desktop search and Windows Live have been brought to the Commission’s attention.

No doubt this announcement will bring more joy to Microsoft’s opponents – but that’s not really the point of the European Commission, the key question is whether the commission’s actions will actually result in greater choice, better products and lower prices for consumers.

And given the Commission’s record thus far…. I’m not sure that it will.

In the interests of disclosure, I should say that I’ve been very critical of both Microsoft and the EU throughout my coverage of the anti-trust saga – One one hand, I think that it was right to require Microsoft to open up its protocols (and I’ve spent hours trying to explain to Microsoft how doing so would actually benefit them). On the other, I think that the EU’s management of the process has been shamefully shambolic and arrogant. Like another analyst – the wonderful James G I feel obliged to add a picture of a donkey whenever the EU is mentioned (Mr G got there first with the donkey pic I should add).

I should also disclose that Microsoft has been a client in the past, and may well be one in the future.

The two issuesAs is often the case, the really interesting “bit” is never mentioned in the headline – Both cases raise issues where it comes to the “oh and we’re going to look at this as well…” bits in the statement.

Both issues need to be looked at separately -

The first issue – Disclosure of interoperability information

The first is related to the EU’s “decision” of March 2004 in which Microsoft was required to document key client-server protocols within the Windows O/S – specifically Microsoft Workgroup Server. It should be noted that the formal text does not mention the .NET framework – So while “related” the addition of .NET to the mix would represent an extension of the decision. The interesting tid bit comes at the end – where the EU adds Microsoft’s Office Open XML (OOXML for short) to the mix.

The inclusion of the .NET framework would increase the stakes significantly but would also potentially create all sorts of pandora’s box effects – Not the least of which would be the raising of some difficult questions concerning the fact that J2EE hasn’t exactly kept its early promise to become the de-facto framework for application development. As it happens, I think the main beneficiary of an “Open .NET” would be Microsoft, and for years I’ve tried to convince the company that it should give up its addiction to the OS kernel and focus on the value-added layers. If .NET were fully ported to Linux, there’s no doubt it would give Linux a lift (not least because writing GUI software for Linux is way beyond the experience – if not the ability – of most programmers) – but if VB developers could develop full-blown apps that make complete use of the .NET framework and which run on Windows or Linux you’d find a lot of developers using VB to write Linux apps. The Mono project has already made considerable progress in creating a rich port (I say “rich” because it’s not quite a 100% port – but it can be used today to develop proper cross-platform .NET applications).

The addition of OOXML to the investigation, almost as if it were an afterthought, is interesting, because I think that OOXML is by far the most important issue in the context of “levelling the playing field”. It is such an important issue, that I think it should be the subject of an entirely separate investigation. Microsoft has done a full-on, professional, job of getting OOXML onto the agenda as a prospective OSI standard. They lobbied, invested, cajoled and co-erced OOXML through the early OSI processes, and it seemed almost inevitable that OOXML would be fast-tracked through the OSI process – so it was something of a pleasant surprise when it stalled. Not because I want OOXML to fail necessarily, but because it needs to be subject to a much higher level of scrutiny and assessment than the OSI fast-track process would allow.

The EU, and the US DOJ, should both investigate the extent to which the world actually needs two differing document formatting standards – I happen to think that one is enough.

The second issue – Browser Tie-in

My first reaction on seeing that the company making the complaint was Opera was, to be honest, to laugh out loud. While the Opera browser has all sorts of wonderful properties but it’s a big fat loser in the desktop browser wars. Any claims about it being difficult to compete with Microsoft have to be off-set by the fact that Firefox has been doing rather well (Although gushing statements that FF has 25% share are tough to prove – I think that at close to 14% Firefox has made amazing progress).

The “thing” in this though, is the extension..

In addition, allegations of tying of other separate software products by Microsoft, including desktop search and Windows Live have been brought to the Commission’s attention.

As it happens, the tie-in between Windows Live and IE is irritating – The only time I ever use IE is when I’m checking a legacy hotmail account – And I really would like Microsoft to disentangle the two so I can use FF. But… as Mr Governor rightly points out in his blog on this topic – “Market pressure is where it’s at”.

The world is changing – faster than a bumbling bureaucracy like the EU can chase it – and potentially faster than Microsoft can…

[EDITED - to correct terminology. I'd used "Open XML" to describe Microsoft's "Office Open XML" and as Hoàng Đức Hiếu  has pointed out, this isn't accurate]