# Sun To Acquire MySQL

Sun Microsystems announced this morning that it has agreed to acquire open source database leader MySQL AB for $1 billion in cash and assumed stock options. (Disclosure: I am on the board of directors of MySQL, and O’Reilly co-produces the MySQL User Conference with MySQL. In addition, O’Reilly produces the java.net community site for Sun.) This seems to me to be a great deal both for Sun and for MySQL. Anyone who follows this blog or has heard my talks will have seen me say “Data is the Intel Inside” of the next generation of internet applications, the very heart of Web 2.0. And of course, most of those Web 2.0 applications are built on the LAMP stack, where M stands for MySQL, far and away the leading open source database. Years ago, John Gage, Sun’s chief scientist, made the provocative statement “the network is the computer.” And bit by bit, the industry has been realizing that dream. What we didn’t understand when we first started thinking about that emerging network operating system was just how much it would be a data-oriented system, such that you might more accurately say, “the network plus the database is the computer.” The acquisition is also a great fit because Sun has staked its future on open source, releasing its formerly proprietary crown jewels, including Solaris, Java, and the Ultra-Sparc processor design. But even beyond those relatively recent moves, Sun was arguably the first great open source success story, co-founded by Bill Joy, who not only led the Berkeley Unix project but wrote the open source TCP/IP stack on which so much of the internet was built. And even leaving out other open source projects at the company such as openoffice.org and netbeans, Sun has long been the single largest corporate contributor to the open source ecosystem. (For further support for that claim, see page 51 in last year’s EU study on open source software [pdf].) This has been a bit of a lightning courtship, and I haven’t had a chance to discuss yet with Sun CEO Jonathan Schwartz just how he plans to take advantage of MySQL’s leadership position in the open source and internet-connected database market, but I do think that there is great potential for both companies. With one bold stroke, Sun has reshaped both the database and open source landscape. We’re all going to be chewing on the implications for some time. Update: Jonathan’s blog includes details on Sun’s plans for MySQL. Zack Urlocker, MySQL’s Executive VP of Products, has a blog post on Infoworld explaining the vision from the MySQL point of view. tags: • It makes me really wonder how they want to monetize their investment. One billion is a lot and OSS is usually not very profitable. All my CMS run on LAMP and i was just starting to use MySQL from C# (not an easy task!), so i’m really worried. • Congratulations to M√•rten, David, Monty and crew! Sun CEO Jonathan Schwartz has good commentary on the what the acquisition will mean for Sun over on his blog. • ale Great news. • mattiasw What does this mean for PostgreSQL? Sun promoted PostgreSQL a lot for a while, why did they change their mind? Is PostgreSQL crap? • I share Michael’s worries. • Douglas Great news.Congratulation! Chinese version of the post: http://www.oreilly.com.cn/dokuwiki/doku.php?id=blog:sun_to_acquire_mysql • Foobar And does it mean “Bye bye MySQL for Iranian”? As Sun has always restricted Iranian from downloading stuff from it’s website. • Michael — Neither Sun nor MySQL are disclosing MySQL’s sales, but all you have to look at is Red Hat to see a company directly monetizing open source at reasonable scale. It’s not Microsoft, but few companies have ever been that profitable. Just because open source offers a better deal to customers doesn’t mean that it can’t be a very profitable enterprise. I think it’s OK for me to say that the deal is not out of line with industry norms (for either acquisitions or public companies) when it comes to multiples given on revenue and growth. And it seems pretty clear that Sun is in a good position to accelerate both. This deal is a real opportunity for both companies. • Michael R. Bernstein Hmm. I wonder what this means for Sun’s relationship with Oracle… • Ed Johnston Lets see how long it takes Sun to screw MYSQL up. Sun has a great tradition of self destruction. The great thing about open source software, there will always be another group to fill the void. • nobody @Michael: My SQL also has commercial clients as it is dual licensed… A lot of people think that only one option is available for software developers, often it is possible after careful consideration to have several licenses attached to a certain piece of software. • Fabio Gurgone What do you think about Oracle InnoDB (and BDB) acquisition ? Now Sun is owning the company but Oracle has the most complete table managers. • ted I agree w/ Ed that Sun really used to screw things up. I will take a wait and see approach though and remind everyone that those bad times were under the Scott McNealy era(he did a lot of good things too, just couldn’t keep it up). Let’s see what Jonathan Schwartz can do. All in all I have to say that he has done a good job so far putting Sun back into the spotlight…and it seems like exciting times ahead • Cement Head The claim of Sun being the largest contributor to Open Source software that is referenced is, in fact, based largely upon Open Office, so it can’t be discounted. And later in the document, they describe why that isn’t really accurate, either: “Sun alone, in particular, is credited with 30% of the total code contribution in our sample, which highlights one of the flaws inherent in the technique used for identifying company code contribution, which is based on copyright credits. In the case of Sun, most of its contribution is accounted for by OpenOffice, for which Sun holds the copyright. The entire codebase of OpenOffice is not, in fact, Sun’s sole creation, but contributors – individuals and other firms, small and big – sign an agreement assigning Sun joint copyright of their contributions, in order to simplify licensing and liability management.” (p. 57 of Tim’s referenced PDF document) I think Sun makes a great OS and has made good contributions to the standards community, but they aren’t really all that great (at least not yet, despite trying pretty hard) at the “Open Source thing”, and saying they are and backing it with self-proclaimed inaccurate studies don’t make it so. • Robert I don’t have any worries at all. I think Sun is pretty much going to let it run as is. • In my opinion,$1 billion is a steal, considering all the companies built on mySQL technology (e.g. Facebook, valued at something like \$15 billion now?) Also, consider the number of terabytes of data out there that’s stored on mySQL databases? And how fast is this data growing? I don’t have the numbers, but I image they’re pretty big.

It’s nice to see that there’s a company out there that sees the big picture and the true potential of Web 2.0. On the other hand, the low purchase price indicates how the market undervalues this concept on the long-term scale.

I’d hazard a guess and say that this is just the first step for Sun. To quote Jonathan Schwartz’s blog: “Starting with the letter M.” Can’t wait to see what letter we end with. Can anyone spell “L A M P”?

By the same measure, you have to look at any metric that counts the percentage of Linux, for example, that was contributed by the Free Software Foundation. They too require copyright assignment. Not only that, a chunk of the FSF contribution was re-implementation of stuff from BSD, so that it could be put under GPL. And of course, some of that code was created by Bill Joy, who went on to found Sun….

So yes, this is a flawed methodology. But it’s definitely not “self proclaimed.” This study was done by the EU.

• Good post by Matt Asay from Alfresco: http://blogs.cnet.com/8301-13505_1-9851655-16.html

Just to be clear, by self-proclaimed, I meant to indicate that the study itself proclaimed that it was inaccurate, not that Tim had created the study. Sorry for the lack of clarity, and apologies for any offense.

I do think, though, that selling Sun as the greatest corporate contributor to open source (particularly if open office is excluded, as Tim indicated could be in his original post) is WAY overstating what they’ve done in that arena, and using this study to prove it is misleading.

So tell me again why Open Office doesn’t count?

But in any event, I think Sun got a bad rap because of the ambivalent way they handled Java and open source. But I would still assert that Sun has been a major friend to open source.

It’s always bemused me how much credit IBM got for supporting open source, when IBM is far more fiercely proprietary than Sun in other parts of their business, while Sun somehow has never been able to satisfy the open source community.

I do think that Sun’s been working hard to get better at this, and hopefully, with MySQL, we’ll see a turnaround. Personally, I’m hopeful about this acquisition, and I’d hope that the community would be supportive rather than cynical, and would be excited to help make it a success.

I was just reflecting what you said about Open Office not counting in your original post – “And even leaving out other open source projects at the company such as openoffice.org and netbeans…”

I think Sun made their enemies by cheesing off the Linux community from time to time, and they were viewed as trying to jump on the open source bandwagon late in the game with the opening of the Solaris 10 source code – I certainly agree, like I had said, that Sun has been one of the best friends of the open _standards_ community, but they seem to be occasionally bipolar in their support of the open _source_ community. I do, however, also believe that they are headed now in the right direction – I just hope that they stay headed that way this time (which, given the way the environment is going, is more of a business necessity for them now then ever)

• Bill

Unfortunately PostgreSQL starts with P instead of M. It’s all about marketing and money, the quality of the product and integrity of data is second. I do hope Sun still continue the support for the elephant.

• I think Sun is the right partner for MySQL. Can’t imagine a better one. I’m excited to see how things will improve, there are so many things Sun can do at MySQL.

• I’m eagerly awaiting Oracle’s reactions (if any) – now that MySQL has aquired the prominent status of a household name to CIOs through the backdoor, this will change Orcale’s perception in the market enormously.

• Perhaps the residual resentment of Sun in the open source community is left over from the very early days when GNU/Linux was touted as a “Sun killer”?

They were an “enemy” to some of the early firms because those firms were competing for some of the same customers. “Beating Sun” was one way that volunteers were encouraged to measure their success, even if it was someone else collecting the money from beating Sun.

(And that was quite a shift. GNU had previously aimed to create a “complete system,” available to all, so that everyone could use computers in freedom. With the rise of the first GNU/Linux distribution and support firms, the aim shifted towards helping those firms succeed.)

I’m skeptical of claims coming from the open open source community that Sun has been or is a bad player. The problem is that the social norms of the open source community — the way so-called merit is judged — are untrustworthy.

In its own mythology, the open source community is officially defined in legal terms, by dedication to certain licensing practices.

In its practice, though, the open source community creates itself out of polarizing judgements of good guys and bad guys, and the more and less meriting. The polarizations are formed around some strange metrics — people are judged by (and compete over) things like perceived project activity levels, personal name recognition, sizes of project membership lists, sizes of volunteer-supported user communities, trade-press mentions, book sales, numbers of patches per contributor, and leadership skills about collecting volunteer labor to raise all of the above statistics.

So, far from being simply a group that agrees about licensing, the open source community has gotten into a cycle in which it creates a perceived need for people to compete on those polarizing metrics. And when people compete effectively, one side effect is that they continue the cycle by reproducing the perceived need for further competition along those lines.

You can call that situation whatever you like as long as you don’t call it “business” because it isn’t based on trade. (And that’s a bit odd because it is based on labor and labor pricing.) One can hardly fault a company for not rushing to substitute the social games of the open source community for actual commercial development.

Meanwhile, with Java especially, Sun has been a leader in using inter-firm cooperative development in an orderly process to develop novel software of a quantity and (more conventionally measured) quality that finds few equals in the open source community.

Part of my hope, as Sun opens up more and more code, is that their influence will begin to break the hype cycle of the open source community and introduce a new norm of what counts as “professionalism”.

-t

• one more perspective :