A commenter says that no fork will be able to offer a dual license, and that a fork is only for the GPL portion of the license. Even so, what's the big deal there? Any sizable company can offer to sell a support contract for an open source license of the DB. What am I missing here?
I'm firmly disagree with Monty's duplicity, but I'll try to shed some light on it.
The "big deal" is that only two things provide a revenue stream around MySQL sufficient to support the heavy R&D costs associated with building a database product:
- The name ("MySQL")
- The copyright.
If you own the name, you can sell branded binaries and services. If you own the copyright, you can sell non-GPL licenses to commercial interests. Without these, creating a viable revenue stream is incredibly difficult.
The only thing we can speculate about is what we would have done in his shoes, unless he provides a convincing insight. If it was me, and I cared about a project of mine for the sake of the project and thought it was a vital project (as opposed to one of my tinker toys), I would have open sources the whole thing, no dual license. That way nobody can "steal" it from the community, by buying some company. I would also have tried to give it to someone like the Apache Foundation, so that when I die, my project does not.
On the other hand, if I thought my project was going to be a cash cow, I would have done exactly what Monty did, except the whining after the fact. I think if I buy a car, then sell it to someone and that person sells it later to my arch nemesis for a discount and he plans on wrecking the heck out of it, my screaming about how this is not right is not going to attract many supporters, is it?
When you say "I would have open sourced the whole thing", what you really mean is "I would have BSD-licensed the whole thing". There is no difference between dual-licensed GPL code and straight-up GPL code. In both cases, your rights as an end-user are identical. In neither case can you pick up the codebase after it's sold to Sun and start a new commercial endeavor on it that isn't GPL'd.
What happened to MySQL isn't a consequence of the GPL. It's a consequence of MySQL selling out to Sun. About the worst you can say about the GPL in this situation is that the GPL made MySQL much more attractive to Sun.
because companies weren't just paying for support, some were paying for non-gpl'd versions of the code to integrate/resell in their products. only mysql ab (now sun, soon oracle) was able to sell those licenses.
this is the same situation with asterisk. digium forces all users that contribute patches to hand over copyright so they can sell a private version of their open-source product to companies.
whether this is fair is up for debate (and is partly why many users hate gpl software, preferring more "open" licenses like bsd/isc). a company open sources an initial version of their product with the gpl, giving everyone access to it, its users contribute many bug fixes, new features, and free support (through forums, mailing lists, bug trackers, etc.) but are unable to sell their own versions of that software. the company that retains copyright and distribution rights is of course able to bundle up everything contributed and make a profit directly from it without having to give anything back to those users.
"some were paying for non-gpl'd versions of the code to integrate/resell in their products"
If that's what they need, perhaps to migrate their products to a BSD-licensed RDBMS like PostgreSQL. Or to a public-domain one like SQLite. Or a fully proprietary one. It's their software and, since, they have already licensed it as proprietary, their users were already robbed the freedoms GPL'ed MySQL would provide.
As it is now, those companies have licenses for specific versions. If they didn't get permanent unrevokable licenses and were happy with the cheaper ones they got, well... Phineas Barnum would be happy to use them as examples.
"whether this is fair is up for debate"
If handing over copyright is too expensive, companies are always free not to contribute or not to use Asterisk at all. They will, however, have to compete with other companies who do. Economics will dictate who wins.
"(and is partly why many users hate gpl software, preferring more "open" licenses like bsd/isc)"
The only freedom GPL denies a developer is to close up software that was once free. The only freedom the allegedly "more open" licenses give developers is to deny freedoms to their users. For the users (and software freedom is about them), those licenses give nothing.
But we are drifting from the original topic.
The topic is that MySQL is Oracle-proof and doesn't need saving the same way other GPL'ed products don't need saving. They are either good - and thus worthy of surviving - or bad - and thus will die natural deaths.
Many people have emotional attachment to their software. I must confess I am thinking and rethinking moving from Linux to OpenSolaris for much too long to explain it in strictly rational terms. I will miss it when Linux dies, but, by then, I expect GNU/HURD to run nicely on my 64-core ARM-based netbook. Or, at least, ZFS to be under GPLv3.
The only freedom GPL denies a developer is to close up software that was once free.
You can't "close up" software that was free. The software remains free.
The only freedom the allegedly "more open" licenses give developers is to deny freedoms to their users.
It gives developers the freedom to create an aggregate product without also releasing their own code under the GPL, allowing them to pursue a business model that they feel will fund the ongoing R&D necessary to produce that product.
For the users (and software freedom is about them), those licenses give nothing.
If users do not feel that a product does not provide a fair trade, they are under no obligation to purchase or use it, and the original code remains to do with what they will. However, as indicated by vast market success, users do feel that these licenses and the products produced using them do provide something of value.
No, you can't take a BSD-licensed program and make it non-free. What you can do is use it as rev 1.0 of a commercial closed-source project and dump resources into improving the closed fork and capturing share from the original code.
What GPL prevents that BSD doesn't is closed forks. Obviously some people don't care about closed forks --- some people may even like them on principle --- or nobody would BSD-license anything.
You're arguing that users don't believe the current situation is a fair trade, and would rather enter into a GPL quodque pro quo agreement if they were not restricted by the purchasing department?
Well -- I for one am happy with the licensing status of Mac OS X, and think I've received a perfectly fair trade.
Many users would be happier with technologies different from the ones endorsed by their managers (frequently after vendor-paid trips to vendor-sponsored events) and anointed as corporate standards that one should not break if one really likes his/her job.
My wife had to build an intranet on top of Sharepoint because a PHB said so. I have to read my e-mails off an Exchange server because corporate IT finds it nice. Far too many servers are Red Hat while I would prefer Debian (I am a fan of APT). Perhaps you have to make your presentations on PowerPoint for Mac, even knowing how much better they would look with Keynote. Life's not perfect.
And, while I love the NeXT side of OSX, I find the Unix side far too early 90's for my taste ;-)
Perl programmer Tim Bunce (DBI/DBD) has been sending mass emails to support this cause, and I didn't understand it either. There's some mumbling about how you can't really fork mysql but no clear explanation as to why. GPL is GPL, and the whole codebase is under GPL. You can do what you want. If they need to come up with a new name, because Oracle starts asserting a trademark issue, so be it. Mandriva survives, doesn't it? I had to think for a second to remember that it used to be called Mandrake.
MySQL does the job pretty well for applications in which stored procedures and other advanced features are not needed. It is really fast and very effective at doing what is absolutely needed: INSERTS, UPDATES, and SELECTS. I don't think it is "awful" or ready to die.
Quality rarely drives market share in software. In this case I blame nearly all of the 5.2 million PHP and MySQL tutorials out there for exemplifying shoddy practices that inevitably lead to quietly truncated fields, phantom reads, and SQL injection attacks.
An open-source project is more than the code. It also includes the community and the resources for continuing development. Although the code is under the GPL, Sun (and now possibly Oracle) has control of the brand and many of the potential revenue streams. A fork would lack the benefit of a clear, centralized community and many resources to fund development. Both of these things threaten the future of MySQL as an open-source project.
Forking + rebranding is perfectly viable option. A similar thing happened to the popular Mambo CMS, which had a big community and user base. Over time, everyone switched to Joomla, the rebranded free version - I don't see Mambo mentioned anywhere anymore.
If the same zeal and effort being spent to oppose Oracle were being spent to create a community around a fork, helped by the controversy it'd be a widely known thing already. Couple that with OS vendors choosing the fork over the "mainline", and in a few years the fork would be "the" MySQL, albeit with a different name.
I can guarantee that for the $1Bn Sun's shareholder's paid him for it - 14% of the total price Oracle is paying for all of Sun - Monty could buy MySQL back from Oracle and do with it as he pleased. So why doesn't he?
You have no way of knowing that. Ellison could view even $2 billion as a small price to pay for owning MySQL. MySQL has been a thorn in his side for years, and now he owns it.
Which means nothing really. If Monty announces that MarisDB is "the real MySQL" everyone who's ever heard of MySQL will go with him. Everyone else probably doesn't even know what database is under their company's data.
All Sun got for their $1Bn was the domain name mysql.com...
I really don't think that's true. The "average" MySQL user has never heard of Monty -- I used MySQL for more than a decade before I heard the name of its inventor.
MySQL's battle for the last decade has mostly been getting enough brand recognition that corporate environments consider it a serious, practical alternative. The brand IS the value of MySQL: it is a promise that the technology has been worked on by serious people, and will continue to be patched and improved for the forseeable future (however realistic that promise turns out to be).
Starting again with a new brand is possible, but by definition it would be a new company which would have to prove its chops and its longevity all over again.
Losing the MySQL brand would be a serious setback to MySQL adoption in corporate environments, and thereby a setback to the amount of money that gets invested in improving it. It is in the interests of all users of MySQL that the brand remain trusted.
Monty did not 'invent' mysql, it is a partial implementation of a standard.
Losing the mysql brand would mean less than nothing, in the corporate world a brand has value in the open source world the name changes and everybody moves on.
See joomla vs mambo and a ton of other examples.
The only person that really seems to have a problem with this is Monty himself since he can't let go of his baby even after selling it down the river. Which plenty of people spoke up about when he sold it to SUN in the first place.
Do you think he'd hate them jumping ship to SQLite?
Supposing he actually is being Machiavellian, he wants MySQL not to be so strong a competitor for the middle and high end of commercial installations. He can afford to concede the very low end - Oracle doesn't play there, except perhaps with BDB (a minor sideline).
Really? Java (which is most of the value of Sun) and MySQL are the only Sun products in mainstream use. Solaris is in the same state as HPUX was 10 years ago, VirtualBox and Sun's Xen Product don't have significant market share. Why would Oracle want to sell MySQL back?
Because they bought Sun for the hardware-business and Java. They sort-a already have a database, and if selling MySQL to Monty can save them some trouble with a bored EU regulator that gets its kicks from telling US business how to operate, I'm sure they'll be happy to. It'll have to be at market value, though, and it seems a bit like Monty wants a solution where he stays very rich.
So they didn't buy Sun to get the largest OSS database in the world?
Sun's storage products are top notch but I doubt Oracle specifically wanted to move into the field. Selling replacement SPARCs to the banks that still have Solaris won't last long, Oracle know that.
Sun has a very good range of hardware that goes far beyond SPARC, and Oracle is marketing it aggressively[1]
They specifically cited Java and hardware in the acquisition announcement, and I have a really hard time seeing the strategic value of the MySQL asset. Oracle has much more to gain from a "come to us when you're ready to sit at the grown-ups table" position.
Ah, I thought you were using to site that 'Sun has a very good range of hardware that goes far beyond SPARC'. Yes, Oracle are marketing SPARC, because they're going to own it. It doesn't mean that it has a future and people will start running new projects on SPARC again.
I don't really understand how that's the case. In 15 years as a professional DBA working with both Oracle and MySQL I have never found myself in the position of thinking, hmm, Oracle or MySQL for this project? It just doesn't really happen (no matter how much Monty would like to believe it does). I recently decided between MySQL and SQLite on a smaller project (SQLite won that one). I can see myself deciding between Oracle and Postgres one day (never happened in practice).
