Anyone have the HHI numbers?

HHI?

Herfindahl-Hirschman Index

It's a measure of the concentration of a market. If we had those numbers, we could calculate by how much this acquisition would increase the HHI and thus how much anticompetitive effect it would have.

Antitrust policy is very fact-intensive and I don't think the article alone allows one to make a judgment pro or con.

I don't think I can take a fact-intensive judgement of this acquisition. :-) People complain about Microsoft but Oracle is much, much worse imo

Also I imagine it's a hard one to analyze by conventional means as you would a couple of car manufacturers. Oracle isn't worried Sun is going to out compete them, if the market value of Sun comes into the equation I think you'll get warped results.
Plus the market is very, very homogenous, we couldn't move away from MySQL if we wanted to, and that's the story for the vast majority of hosting providers out there. The only alternative is PostgreSQL, but in MySQL's niche MySQL is the standard

It is weird that Oracle has such a tiny share of the market compared to MySQL yet it can practically destroy it just by threatening to buy it

Well, look at it this way. This isn't an antitrust suit against Oracle. It's an objection to a proposed business combination. You want to compare the degree by which this transaction would decrease the competitiveness of the market. If Oracle's actions are already running afoul of the antitrust laws, then the proper remedy is an antitrust suit. In the very limited context we have here, the inquiry has to be about how much worse things will get, not how bad they are.

Market value isn't what the HHI quantifies, but market share. This is vital in antitrust analysis because market share and market power are closely correlated. So the question is this: Will the absorption of Sun by Oracle substantially increase the concentration of the relevant market?

Where one firm has 100% of the market, it's easy to see the connection among market share, market power, and anticompetitive dangers. But even well below that level, antitrust regulators are concerned with business combinations that concentrate the market, for two reasons. First, a firm may be "merging to monopoly" in an attempt to acquire enough market share to exert illicit market power (with all the attendant disadvantages of monopoly). Second, a more concentrated market with fewer firms makes it easier for firms to coordinate, either because of actual cartels or simply because of oligopolistic pricing.

Now, like I said, the question isn't how bad it is now but how much worse it will get. If the change in HHI is very small, the danger of monopolization or some kind of anticompetitive coordination is minimal. These numbers are important indicia of antitrust danger because the numbers represent the types of behaviors with which antitrust law is concerned. When someone thinks an acquisition won't be good for the market, the numbers allow us to judge how bad it will be, if at all, in such a way that we can compare the effect with the amount we're willing to tolerate. I should point out that the amount we're willing to tolerate has to be considered together with the cost of investigation, the cost of stopping an acquisition, and the risk that the regulators are simply wrong (in other words, that they put a halt to a transaction that was actually procompetitive).

what are these numbers based on?

and also it may be bad for the for the open source market is Oracle buys Sun, depending on what Oracle does with their new acquisition.

Market share, slugger. Try to keep up.

so it depends on the number of sales, no are mysql is opensource that would be the number of users, right? I do think the opensource thing rather complicates matters....

Your grammar baffles me.

sorry, i may be a little drunk.

your market share is just your companies sales figures, compared to the total sales in that market.

on the other had if mysql isn't sold, merely used under the GPL license then market share is the number of users using your software divided by the total number of users.

Am i mistaken?

You've got a good grasp on it.

In my experience, PostrgreSQL is much, much faster than MySQL; if you can switch, I suppose now is the best time.

There are several alternatives including PostrgreSQL and SQLite but I find this trend of large companies swallowing successful open source projects to be very disturbing and the only thing I'd like better than seeing Oracle's acquisition bounced is to see Germany bounced as he tries to enter the North Sea this turn. :)

if the large company keeps the opensource project open then i don't see it as a major issue.