Some thoughts on MySQL and Oracle [LWN.net]

By Jonathan Corbet
December 15, 2009

Your editor wishes to take no position on whether Oracle's acquisition of Sun Microsystems should be allowed to proceed by the European Union. Such a decision certainly involves a number of antitrust considerations which go beyond the free software community. That said, some of the positions being taken around this acquisition shine an interesting light on how parts of our community work.

Fear #1 is that Oracle will kill MySQL, which Oracle is said to see as a threat to its cash-cow relational database management system. One might respond that similar fears were expressed after Oracle's acquisitions of Innobase and Sleepycat Software, but that things have not turned out that way so far. One might say (as Eben Moglen has) that keeping MySQL healthy is in Oracle's economic interest. One might also respond that Oracle could arguably do more damage to MySQL by breaking off the acquisition and allowing Sun to simply die. But what is most interesting about this particular concern is the lack of faith it shows in our community's ability to cope with such an outcome.

MySQL is licensed under GPLv2; it is free software. It can always be forked; indeed, some groups have already done so. There is nothing Oracle could do about that. Oracle could stop developing the free version of MySQL; it could even release future improvements which are available only on proprietary terms. But all it can take from us is the stream of future development which (we assume) we would have otherwise had from Sun. We might wish we had some of those enhancements, but it is another thing altogether to say that we are entitled to them. Free software generally does not come with a promise of future enhancements; what it does come with is the freedom to make those enhancements ourselves.

To say that Oracle would kill MySQL is to say that our community is not strong enough to continue its development outside of Oracle. That suggests that MySQL never really was an independent free software project. MySQL users who believe that should be clear about the position they think they have put themselves in: in this view, they are users of a proprietary product which happens to put out its code under the GPL. If this code has no future without its supporting company, the fact that it is freely-licensed has relatively little value. But such a view essentially writes off the community that has built the amazing collection of free software that we use every day. We are stronger than that.

Another interesting claim is that MySQL's license is the problem. Richard Stallman signed his name to a letter which expresses this worry:

Many other FLOSS software projects are expected to move to GPLv3, often automatically due to the common use of the "any later version" clause. Because the current MySQL license lacks that clause, it will remain GPLv2 only and it will not be possible to combine its code with the code of many GPLv3-covered projects in the future. Given that forking of the MySQL code base will be particularly dependent on FLOSS community contributions - more so than on in-company development - the lack of a more flexible license for MySQL will present considerable barriers to a new forked development path for MySQL.

The "more flexible license" in this case would be to add the "or any later version" language to MySQL's GPLv2 license. This statement looks like an attempt to push a license change onto MySQL, based on the assertion that GPLv2 somehow inhibits community contributions. Your editor is unaware of any study showing that developers are less willing to contribute to GPLv2-licensed projects; if such a study exists, it could certainly benefit from wider exposure.

That is not the only attempt to use this situation to bring out regime change on the licensing front, though. Consider Monty Widenius's "Help saving MySQL" post from December 12. He is asking readers to send messages to the European Commission; suggested text is helpfully provided. It includes:

That MySQL should be released under a more permissive license to ensure that forks can truly compete with Oracle if Oracle is not a good steward after all.

Back in the days of MySQL AB, Monty and others were happy to put the GPL onto the MySQL code. It allowed them to release the code freely while building a business around selling proprietary licenses to companies which did not want to be bound by the GPL's terms. But the right to engage in this kind of business was sold to Sun with the company. Now Monty would like to get it back so that he, too, can sell proprietary versions of the software. This certainly looks like a bit of a request to have his cake and eat it too; it is not surprising that some observers have not been entirely impressed.

What we are really seeing here is the logical outcome of the corporate-controlled open source project model. Such projects may well create an external development community, but that community tends to be weak compared to well-established, independent projects. Additionally, the use of copyright assignments - common with company-owned projects - puts control of the entire code base into a single company's hands. As Eben Moglen noted in his submitted opinion on the acquisition, the single ownership of the MySQL code is part of the problem:

The crucial issue is not the license under which MySQL is distributed, although GPLv3 might be preferable to GPLv2 if one were writing on a clean slate. Rather, the central issue is an increase in the copyright diversity of the project, in which multiple parties have significant code in the main line. This would be sufficient to prevent anyone having an exclusive right to make proprietary enhancements or to undertake distribution under non-free licenses.

Anybody who has dealt with corporations for any period of time has probably learned one fundamental lesson: the company that one deals with today may differ significantly with the company one encounters tomorrow. Even in the absence of acquisitions, corporations tend to be just one bad quarter away from a total change of attitude. Being acquired will almost certainly change a company's approach to a project it owns - especially if that company is the sole copyright owner for the code in question.

Developers who contribute to a corporate project should be aware that they are signing their code over to an entity which may take a distinctly unpleasant turn tomorrow, regardless of how friendly it seems today. Users of this type of software should be aware that they cannot count on any promises which do not exist in a signed agreement with the owning company. The only exception is the license that the existing code is released under: that will not be going away. For a lot of MySQL users, the GPLv2 license is a more than sufficient promise for the future. Companies which have based products on the availability of affordable "GPL exception" licenses will be on less certain ground - though it is worth noting that Oracle has promised to extend those licenses for at least another five years.

Users of PostgreSQL (for example) need never worry about a takeover by Oracle or any other company; it is an independent project which will never be controlled by a single organization. Users of MySQL probably need not worry either; it is a well-established project which should survive a shift to a more community-oriented mode of development, should such a shift prove necessary. But the worries about this acquisition - at least, those which are not motivated by personal agendas - shine a light on what can happen with software which is controlled by a single organization. Being used as a political football in a regulatory fight, with all the associated uncertainties, is just one of the risks involved.

to post comments

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 15:21 UTC (Tue) by hummassa (guest, #307) [Link]

People who are afraid of this:

1. Stakeholders that have a MySQL-based house and use a lot of MySQL support
(which is bound to get more expensive) -- they'll have costs associated with
the migration to another database (even if it's only MariaDB or other MySQL
fork)

2. Developers that make proprietary MySQL-only-based software -- they cannot
use any MySQL fork, so they'll have to bear the costs of the migration

A bit of history

Posted Dec 15, 2009 15:55 UTC (Tue) by prl (guest, #44893) [Link] (3 responses)

Those with long memories will remember mSQL. It was started by an Australian guy called (Robert?) Hughes. When he realised that his dbms was becoming popular he decided to remove the GPL from future versions, and make his fortune by commercially licensing it.

Bad move. mSQL was out there and GPL'd and with no more than an extra letter in the name, MySQL simply forked away out of his control. Mr Hughes's company wasn't bought by Sun and I don't think many people remember him. Mr Widenius's company was, of course - and the only reason Mr Widenius made his money is because he was able to take over GPL'd code and had the wit not to repeat Hughes's mistake. He's welcome to dual license anything he produces, but I'm sure he knows the true power of the GPL; that's why he wants the EU commission to give him a get of jail free card.

You'll notice that nothing has happened at Hughes Technologies for 2 years at least.

A bit of history

Posted Dec 15, 2009 16:11 UTC (Tue) by jimparis (guest, #38647) [Link]

> Bad move. mSQL was out there and GPL'd and with no more than an extra letter in the name, MySQL simply forked away out of his control. Mr Hughes's company wasn't bought by Sun and I don't think many people remember him. Mr Widenius's company was, of course - and the only reason Mr Widenius made his money is because he was able to take over GPL'd code and had the wit not to repeat Hughes's mistake. He's welcome to dual license anything he produces

If MySQL was forked from GPL code, how could he offer to dual-license the resulting product?

According to everything I've found, your story is not correct: e.g. http://dev.mysql.com/doc/refman/5.0/en/history.html says "However, after some testing, we came to the conclusion that mSQL was not fast enough or flexible enough for our needs."

Is this true?

Posted Dec 15, 2009 16:24 UTC (Tue) by gbutler69 (guest, #54063) [Link]

Something doesn't sound right. There is NO WAY MySQL would be legal to be distributed under a
dual-license if it was a fork of what was originally GPL-Only. Could someone who knows please
clarify. There is definitely something not correct in the parent's history of mSQL vs. MySQL. Was
mSQL originally BSD or somesuch perhaps? Was it originally dual-license with others having equal
copyright ownership and ability to choose their license differently?

A bit of history

Posted Dec 15, 2009 16:56 UTC (Tue) by epa (subscriber, #39769) [Link]

AFAICR, early MySQL versions were mostly compatible with mSQL, but AFAIK not based on it. See the book MySQL and mSQL.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 16:26 UTC (Tue) by nim-nim (subscriber, #34454) [Link] (7 responses)

While the GPL protects the code, it does not protect the MySQL trademark. And MySQL's main value today is a positive brand recognized by many decision makers all around the world.

Forking MySQL if Oracle mismanages it is easy. Building another brand with the level of acceptability MySQL has today is a multi-year risky project.

Oracle could effectively kill the competition MySQL offers it in the conservative wealthy Enterprise market by restricting use of the MySQL brand. And this is where MySQL has been hurting Oracle most (cash cow of corporate customers that use oracle everywhere from big DBs to small because that's the only solution that got approved at all levels. That is, until the MySQL name managed to carve a place there too)

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 16:48 UTC (Tue) by iabervon (subscriber, #722) [Link] (6 responses)

I think the user base would be perfectly able to understand a dispute between the creative talent in the project and the owner of the name, and (particularly if Sun simply stopped doing anything with it) would be perfectly willing to use "the database formerly known as MySQL" and would recognize it as the spiritual continuation of the project.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 16:52 UTC (Tue) by bangert (subscriber, #28342) [Link]

As I try to argue below, Widenius' actions make me believe this is what he
is preparing for.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 18:19 UTC (Tue) by nim-nim (subscriber, #34454) [Link] (3 responses)

The user base would just stop using MySQL and wait a few years for the dust to settle and see if a fork with a reliable management and reliable support structure emerges. It's as simple as that, the savings in using a free database are not worth the risk of betting on the wrong horse.

That's what the MySQL name means today: reliable project. No-risk choice. And not reliable because the code is good, reliable because of market acceptance, and number of concurring reports it works in the field.

Seeing the number of vultures circling MySQL now that Oracle is likely to get its hands on it, the chances for a single fork to emerge quickly are very low. Everyone is going to launch its own fork, and fight for the free database crown with Oracle laughing on the sidelines. (see how long it took for Centos to emerge as the "free" whitebox RHEL clone, how many promising projects went nowhere, and they were not aggressively fighting each other for mindshare).

Corporate market hates choices. Winner takes all. Oracle, RHEL, Ubuntu, are all example where a single actor cashed on customer unwillingness to choose between multiple similar choices. Brand is a very strong factor, more than code quality.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 19:52 UTC (Tue) by iabervon (subscriber, #722) [Link] (2 responses)

I think the main concern really is the "reliable support structure" aspect; I don't think being named "MySQL" is actually at all relevant to that assessment. What matters is that the software under active development is supported by established organizations providing support. This is actually probably the biggest concern, because the most obvious organization providing support is the one Oracle is buying (although other companies provide commercial support for MySQL). I think the fact that this organization is also the one that holds the trademarks isn't a major factor.

As far as development is concerned, I think that, if Oracle were to become too hard to work with, the community would take it over with approximately the efficiency that X.org took over from XFree86. If you've got a group of people who know each other as the development community, they're likely to keep working together. And, if Oracle owns the bulk of the copyrights and is unresponsive, there is no crown (in terms of a codebase that some branch can offer non-GPL licenses to) to fight over. It's less likely that everybody would try to take over the project from themselves than that everybody would try to sucker somebody else into doing the project administration.

Of course, there's still the question of whether third-party corporate support organizations would declare the Oracle branch dead and support the community branch instead. But I could easily see OpenLogic telling their customers, "Starting with the next release, MySQL as we support it has been renamed to OurSQL. It is the same software and the same contracts apply to it." And the corporate market would say, "Whatever. We still think our cell phones are Cingular. As long as our business relationships stay valid, it doesn't matter." And, of course, there's the issue that, if Oracle stops their MySQL development, the version that their support organization supports presumably also stops, and that support organization is probably the most significant one currently.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 20:21 UTC (Tue) by nim-nim (subscriber, #34454) [Link]

XFree86 was never sold as a distinct product. Entities that used it got it as part of a larger whole, which is why the transition was graceful. Only technical people had to know about it.

MySQL's case is a lot more complex

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 4:18 UTC (Thu) by nowster (subscriber, #67) [Link]

Debian would probably call it IceSQuirreL. :-)

Some thoughts on MySQL and Oracle

Posted Dec 18, 2009 10:33 UTC (Fri) by jschrod (subscriber, #1646) [Link]

The problem are not the user base, but their management who have to sign off usage of a potential MySQL successor in enterprise environments.

I see this problem right currently at one of my customers, where I wanted to migrate their old TWiki to FOSWiki. That name change makes the managers suspicious.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 16:50 UTC (Tue) by bangert (subscriber, #28342) [Link] (3 responses)

To me Widenius' request for help is not directly related to the FLOSS
nature of MySQL. He is concerned about the commercial support options for
the MySQL products, which BTW is the only thing that the EU EC is
interested in (market forces and all).

Interestingly MySQL being FLOSS also presents a solution to same problem:
namely a different entity being able to provide commercial support to MySQL
customers. And who would be better positioned to provided this than
Widenius?

I believe, Widenius is basically using this opporturnity to market himself
as an alternative provider of commercial MySQL support. MySQL mostly being
copyrighted by SUN (or Oracle) this would actually be a good thing, since
this time Widenius does not have the possibility to give out commercial
licenses but in all cases is bound by the GPL.

IOW from a FLOSS perspective, i don't necessarily approve of Widenius
actions, but their effect could lead to a much better position for FLOSS
MySQL.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 19:31 UTC (Tue) by rahvin (guest, #16953) [Link] (2 responses)

If what you say were true Widenius and Mueller wouldn't be petitioning the EU to force Oracle to relicense the MySQL code to the Apache or BSD license. They want the company back without paying back the money they were paid.

Besides, when Monty lied (about requesting the license change) and Groklaw exposed it he lost all credibility on the matter IMO. His current MariaDB AB company is reselling MySQL currently, he just wants the ability back to sell commercial non-GPL licensees because it was more lucrative than trying to compete by supporting GPL code.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 19:52 UTC (Tue) by bangert (subscriber, #28342) [Link] (1 responses)

That doesnt make any sense. If MySQL were to be relicensed as Apache or BSD that would be
even better for FLOSS and anybody would be able to issue proprietary licenses...

I fail to see, how you that can be a bad thing for FLOSS.

Some thoughts on MySQL and Oracle

Posted Dec 22, 2009 2:55 UTC (Tue) by foom (subscriber, #14868) [Link]

The scenario envisioned is that Monty Program AB wants MySQL force-relicensed under the BSD
license, so that they can take that code, add their own changes (MariaDB), and release that
combination under a proprietary or proprietary/GPL license (like MySQL was before).

If that's what they want, Monty has essentially stolen $13Million from Sun/Oracle, as he sold them
his company, including the copyright to MySQL, and this move would basically be giving it back to
him, for free.

It's not bad for FLOSS, but it's not particularly good either.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 17:45 UTC (Tue) by ballombe (subscriber, #9523) [Link] (11 responses)

> Your editor is unaware of any study showing that developers are less willing to contribute to GPLv2-licensed projects;

As I see it, the issue is that complex userspace applications like MySQL or even a larger software that would embed part of MySQL, are usually linked to a bunch of (L)GPL libraries that will eventually move (L)GPLv3. Due to the incompatibility between LGPLv3 and GPLv2, any GPLv2+ software linked to such (L)GPLv3 became GPLv3+ for all practical purpose. Example of such LGPLv3 libraries today include GNU readline and GNU MP.

As a result, GPL2-only complex userspace applications get less attractive because they are restricted in which libraries they can link to.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 18:23 UTC (Tue) by sfeam (subscriber, #2841) [Link] (10 responses)

"any GPLv2+ software linked to such (L)GPLv3 became GPLv3+ for all practical purpose. Example of such LGPLv3 libraries today include GNU readline and GNU MP"

This is not correct. The problem with linking to GNU libreadline arises exactly because it is not licensed under the LGPL. It is straight GPL. Linking to LGPL libraries of any version should not cause a problem.

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 18:45 UTC (Tue) by ballombe (subscriber, #9523) [Link] (1 responses)

You are indeed correct that GNU readline is GPL, but LGPLv3 is in fact incompatible with GPLv2-only, see <http://gplv3.fsf.org/dd3-faq>.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 5:55 UTC (Thu) by bronson (subscriber, #4806) [Link]

Wow. What a nightmare the GPLv3 is turning out to be!

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 20:03 UTC (Tue) by rahulsundaram (subscriber, #21946) [Link] (7 responses)

Actually GPLv3 is incompatible with LGPLv2 only codebases.

http://fedoraproject.org/wiki/Licensing#GPL_Compatibility...

Some thoughts on MySQL and Oracle

Posted Dec 15, 2009 20:05 UTC (Tue) by rahulsundaram (subscriber, #21946) [Link] (6 responses)

Ah, I meant LGPLv3 and GPLv2.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 17:39 UTC (Thu) by RobSeace (subscriber, #4435) [Link] (5 responses)

So, wait... LGPLv3 allows a totally closed-source proprietary app to link against it, but yet somehow DOESN'T allow a GPLv2-only app to link against it?? How does THAT work, exactly?!

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 17:46 UTC (Thu) by ballombe (subscriber, #9523) [Link] (4 responses)

You frame it the wrong way. LGPLv3 certainly allow GPLv2 apps to link against it. However, GPLv2 does not allow to link against LGPLv3 libraries, because the LGPLv3 contains additional restriction that are not in GPLv2.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 18:42 UTC (Thu) by RobSeace (subscriber, #4435) [Link] (3 responses)

Wait, if that were the case, then GPL2 apps would be unable to link against proprietary libcs and such on other systems, which is clearly allowed... You just aren't allowed to distribute the two together as a unified whole... So, what may be forbidden is distributing a combination of an LGPL3 library along with a GPL2-only app that links to it... I can see that as possibly problematic, but how could it be problematic to write and release a stand-alone GPL2-only app which happens to link against an LGPL3 library? If one can link against closed-source system libs without poluting the GPL2 code, surely LGPL3 is no worse/dirty??

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 18:58 UTC (Thu) by mjg59 (subscriber, #23239) [Link] (2 responses)

The FSF's stated position is that you can only link against closed libraries because of this
section of the gpl:

"However, as a special exception, the source code distributed need not include anything that is
normally distributed (in either source or binary form) with the major components (compiler,
kernel, and so on) of the operating system on which the executable runs, unless that component
itself accompanies the executable."

The FSF's belief is that without this clause, a GPLed application can only be distributed if the
libraries it links against are also available under terms compatible with the GPL. The
GPLv2/LGPLv3 incompatibility follows from this - LGPLv2 also includes restrictions that are
incompatible with the GPLv2, but LGPLv2 explicitly allows you to use the code under GPLv2
instead. LGPLv3 only allows you to fall back to GPLv3, which itself is incompatible with GPLv2.

Now, the fact that companies like Sun and Apple distribute GPLed code linked against their non-
GPL compatible system libraries as part of their default install (and so would seem to trip over
the "unless" component of the above) indicates that they appear to have a different belief about
what a derivative work is. I don't know of any cases which have actually tested this, and I
suspect it's in the FSF's interests to leave it as a grey area for the moment.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 20:33 UTC (Thu) by RobSeace (subscriber, #4435) [Link] (1 responses)

But, given the hypothetical situation of a distro shipping some LGPL3 library as standard, then why couldn't a GPL2 app (not distributed as part of the distro) make use of that same exception you quote in order to use the LGPL3 library without changing its own license? (And, following from that, one could easily produce a distro that distributes the library as standard, if one chose to, which means that effectively ANY LGPL3 library fits into this exception... And, of course, you couldn't stop end-users from using your app on a different distro that didn't ship the lib as stock, and what they do with combining your app with non-compliant libs on their own systems is out of your hands and out of the scope of the GPL...)

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 21:53 UTC (Thu) by mjg59 (subscriber, #23239) [Link]

You can certainly do that, but the majority of Linux distributions consider themselves to distribute
all their packages alongside all their other packages - that is, you effectively rule yourself out of
being included by default in any mainstream Linux distribution. That doesn't matter to some people,
but it does to others.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 14:03 UTC (Thu) by hingo (guest, #14792) [Link] (14 responses)

I work for Monty and have been heavily involved with the EU process since August.

Jonathan, I just wanted to say, having followed various pundits for months now, making all kinds of bizarre claims, I was glad to see that at least LWN is still my island of sanity. This is why I subscribe to these news. Thank you. (And it's not like the article is pro or against anything, I'm just glad to see there are still people who understand how GPL and some open source business models work.)

We, at Monty Program, have not made much efforts to correct the many things that have been said about this, since that would have been an effort to drink the ocean. But since this article is more or less perfect, let me comment on the one thing that I can give more details on:

About "have the cake and eat it too": We always consistently said that divesting the MySQL business to a 3rd party was the desirable outcome. This way MySQL could continue with its old business models. Monty has no interest and certainly not enough money to be a buyer, it would of course had had to be one of the big players of the open source world. In short, there was no benefit to ourselves in this scenario, which was the primary alternative.

Expecting that the Commission indeed would be strong armed by Oracle into some kind of compromise, we also explained that it would still be helpful then if MySQL was available under a license that gives MySQL users (in particular the proprietary SW vendors) a choice other than Oracle. Like you say, neither the proprietary applications or even GPLv3 software is today compatible with MySQL GPLv2 or a fork of that (like MariaDB). Stallman, Greg Stein and some others understood this too, surprisingly many (including lawyers and your favorite paralegal) quite didn't, so the debate has been... "interesting".

Whatever you may think of the above, Monty Program itself is committed to producing open source software, so our objective definitively was not to start doing proprietary software out of MySQL. (And if MySQL ceased to be GPL licensed, nobody could really do dual licensing anyway.)

PS: Exercise to the reader: Which prominent kernel hacker contributed to the paper co-signed by Stallman? A little web research should reveal it.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 14:17 UTC (Thu) by rahulsundaram (subscriber, #21946) [Link] (11 responses)

Monty was fully aware that dual licensing model leads to vendor lock-in for the customers relying on the proprietary version of MySQL. It also mandated copyright assignments which directly resulted in MySQL never becoming a community oriented project. Crying foul about it now is certainly self serving. I don't think I have much sympathy for that. I am not buying the claim that MySQL cannot be self sustaining or profitable based on support/services instead of dual licensing either. I would also note that Monty's business now around MariaDB is not based on dual licensing.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 14:30 UTC (Thu) by hingo (guest, #14792) [Link] (10 responses)

Hi

Everything you say is correct, except I don't understand the logic of you then saying what we do is self serving?

As you say, part of the MySQL user base is locked in and (for all practical purposes) not helped by the GPL. This was by design, and Monty is to blame, yes. So what we have done is to explain to the EU and in public that MySQL being under the GPL is not a solution to all current MySQL users, they are locked in. So it is the regulators responsibility to do something about it. This is no different than if Oracle buys a proprietary SW company.

Personally I have no doubt that MariaDB or some other descendant of MySQL will be stay alive. (self sustaining can mean many things...) But it cannot save the entire MySQL universe.

So my point is, there is nothing self serving in asking the EU to protect the MySQL users which we cannot help. (This is the whole point of having anti trust regulation.) Monty or our company doesn't get anything out of it (quite the contrary, Oracle getting their hands on MySQL will probably be good for our business).

Btw, it may be healthy to remember that a large part of the EU investigation had to do with market shares, "constraining Oracle" and other things, and the open source aspect of MySQL was just an interesting side track. But it is quite natural that in the community that is the side tract that most are interested in.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 15:32 UTC (Thu) by farnz (subscriber, #17727) [Link] (9 responses)

The reason you are being accused of being self-serving is simple: Monty (as owner of MySQL Inc) created this situation deliberately, because it made him more money. He then sold the company to Sun, without putting restrictions on the sale that would have prevented Oracle owning MySQL, because that made him more money.

Now that the sale to Oracle is in progress, you're complaining (effectively) that Monty didn't mean to sell the company without those restrictions, because Monty's that sort of nice guy, but he just forgot, y'know? You don't seem to be talking about offering the $1bn that was paid for MySQL Inc to Oracle to buy it back from them, or even a significant fraction of that; instead, you appear to be complaining that your reputation and future business model will be damaged if MySQL isn't treated specially.

People other than Monty and his employees complaining is fair enough - they could not have taken action at the time MySQL was sold to avoid the problems you're now complaining about. However, Monty chose not to put these restrictions on the purchaser of MySQL; it's now rather self-serving to ask competition regulators to (effectively) fix his past mistakes that might doom his new venture, especially since I suspect he would have received rather less money for MySQL had he put these restrictions on from day 1.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 16:28 UTC (Thu) by hingo (guest, #14792) [Link] (8 responses)

So half of what you say is right, the other half less so. (But it is probably right that this is why people accuse Monty, even if it is baseless, so I'm with you there.)

It is absolutely right that MySQL business model was designed the way it was on purpose, deliberately, even to make money. And such a business can be sold to another business, etc... And from an open source point of view you might think this was a bad model and the reason we are now in this mess (debatable, but valid point).

I guess it is good to point out yet again, that Monty was not in a position to decide anything about selling MySQL to Sun, and he did not get more than a few percent of the billion dollars. But that's not the point, there was nothing wrong in selling MySQL to Sun, whoever is responsible for it.

In the case of Oracle, it is quite obvious (well, to me) that Oracle has a completely different interest in acquiring MySQL. The anti-trust regulator in Europe is rightfully concerned about that, it is their job to stop such mergers if proven to be harmful. So it is quite normal for anyone, whether Monty or a MySQL customer, to expect that MySQL can indeed be sold to anyone, but not to a competitor that just wants to reduce competition. (I realize especially in the US there is a common political opinion that you should be allowed to sell anything to anyone and the government shouldn't interfere, but that is a political view, and as it is now, we do have regulation.) So there is nothing wrong in helping the EU to do their job, which they are supposed to do, and there is no benefit to Monty (perhaps emotional, that's all) to be doing this.

Anyway, I get what you are saying, that this is why people are blaming Monty, because many of them probably think along the lines you just said.

Some thoughts on MySQL and Oracle

Posted Dec 18, 2009 10:42 UTC (Fri) by rahulsundaram (subscriber, #21946) [Link] (7 responses)

"It is absolutely right that MySQL business model was designed the way it was
on purpose, deliberately, even to make money. And such a business can be sold
to another business, etc... And from an open source point of view you might
think this was a bad model and the reason we are now in this mess (debatable,
but valid point)"

I don't think there is much room for debate on that. Monty in his discourses
isn't even pointing to this fact which is one of the reasons for me to
believe that this is a self serving position. If I create a mess and wanted
someone else to clean it up, the least I should be doing is to acknowledge if
not apologize for the fact that I created the mess in the first place.

Some thoughts on MySQL and Oracle

Posted Dec 18, 2009 12:19 UTC (Fri) by hingo (guest, #14792) [Link] (6 responses)

Perhaps so :-)

Since personally I'm not the one who invented dual licensing, so have nothing to take credit nor apologize for, I can say that my personal opinion is that it remains to be seen if it is a sustainable model or not. It is a good way to produce great software which is GPL licensed. In some cases it may even turn out to have a happy ending, like is the case with Qt now. On the other hand, this risk was always there.

Due to MySQL's and the model's financial success, it is becoming an increasingly popular model and more software is being created under a similar setting. It may be good for us to also be aware of potential pitfalls.

Some thoughts on MySQL and Oracle

Posted Dec 18, 2009 13:09 UTC (Fri) by rahulsundaram (subscriber, #21946) [Link] (5 responses)

"It is a good way to produce great software which is GPL licensed. In some
cases it may even turn out to have a happy ending, like is the case with
Qt now. On the other hand, this risk was always there."

Heh. Qt's happy ending turned out to be leaving behind dual licensing by
releasing it under LGPL and striving to become a real community project
instead of centralized ownership. That's because Nokia realized it made
better business sense to strive for ubiquity. I agree, that is the way to
go forward and Oracle might still do that since they don't need rely on
dual licensing. MySQL's business according to then CEO was increasingly
moving towards support and services as well. Anyway, you wanted to know why
I consider it self serving and I hope I have explained that clearly to you.

Some thoughts on MySQL and Oracle

Posted Dec 18, 2009 13:18 UTC (Fri) by hingo (guest, #14792) [Link] (4 responses)

Yes, thank you.

And: My point exactly with Qt, less optimistic with Oracle but time will tell. If they wanted to LGPL MySQL, they should have saved us these months of agony and done it long ago...

Some thoughts on MySQL and Oracle

Posted Dec 18, 2009 14:47 UTC (Fri) by rahulsundaram (subscriber, #21946) [Link] (3 responses)

Sun is too busy keeping its head out of the water to think about such
matters. Oracle cannot do it before acquiring them. I won't even say that
it is likely but it does make sense for Oracle to consider it at some
point after the acquisition.

Some thoughts on MySQL and Oracle

Posted Dec 21, 2009 8:17 UTC (Mon) by hingo (guest, #14792) [Link] (2 responses)

Actually, AFAIK, Sun has discussed it with Oracle and Oracle has rejected it.

On the other hand, we shouldn't read too much into it. Majority of Oracle people haven't yet understood what significance it would have anyway (for instance, in conversations they would confuse MySQL commercial licensing (non-GPL) and subscriptions (GPL) and just in general when you can use the GPL version and when not). So if they one day turn around and do LGPL anyway, it can be explained with pure cluelessness.

That, and having seen how Oracle plays this game, they could just out of principle refuse to do any compromises with the EU, then do the same thing anyway later. Just to make it clear they don't negotiate with anyone about these things.

But as I said, assuming rational and informed behavior from Oracle, it would seem they are against LGPL.

Some thoughts on MySQL and Oracle

Posted Dec 21, 2009 12:06 UTC (Mon) by rahulsundaram (subscriber, #21946) [Link] (1 responses)

"Actually, AFAIK, Sun has discussed it with Oracle and Oracle has rejected it."

Reference?

"Majority of Oracle people haven't yet understood what significance it would have anyway (for instance, in conversations they would confuse MySQL commercial licensing (non-GPL) and subscriptions (GPL) and just in general when you can use the GPL version and when not)"

This was a common accusation against MySQL Ab and then later Sun as well. So not specific to Oracle at all.

Some thoughts on MySQL and Oracle

Posted Dec 21, 2009 13:00 UTC (Mon) by hingo (guest, #14792) [Link]

Reference?

Those discussions did of course not happen in public. (And no, I don't have first hand information, I could be misled myself.)

This was a common accusation against MySQL Ab and then later Sun as well. So not specific to Oracle at all.

I give you the Sun part, it was a big company and certainly most Sun employees had no clue about MySQL's business model either. But my point was exactly that it may be premature for me to say anything about Oracle's plans with MySQL licensing based on their current actions, since currently they don't even understand the basics of what the current situation is. So I tried to say that from all I can see I don't expect them to go LGPL, but if it happens, then this could explain why they didn't do it much earlier.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 20:29 UTC (Thu) by vblum (guest, #1151) [Link] (1 responses)

I also liked the article and its conclusions, but I still have not understood one fundamental point: What is wrong with GPLv2?

The license will work as a license, going forward, certainly equivalent to or better than other open source type licenses. For example, the Linux kernel works very well under GPLv2, and with a sufficient community out there, does not seem likely to die soon ....

The worst that _could_ happen is that Oracle could buy the entire "developer community" and pay them to shut down the GPLv2 development process. Short of that, the product is out there, GPLv2, and can be used and developed under that license. I fail to see how a change to an Apache style license or any other one would change this. Certainly I fail to see why the license would impact the antitrust aspect of the case.

So any good explanation of that would be helpful. How exactly would a change from GPLv2 to anything else alleviate the fact that Oracle could simply pay everyone under their control to not develop anymore?

Some thoughts on MySQL and Oracle

Posted Dec 18, 2009 7:53 UTC (Fri) by hingo (guest, #14792) [Link]

The significant difference is that the license used for Linux does not affect the software run on top of Linux (a.k.a user space). So you can run GPLv2, GPLv3, LGPL, BSD, proprietary software on a Linux system. This of course makes sense.

For the GPLv2 version of MySQL this is not possible, and as discussed above this is by design, since the business model was to require proprietary SW makers to pay money for the privilege of using MySQL. (Imagine if glibc and other libraries used on Linux were also GPL all the way.) MySQL does however provide for a FOSS Exception, so that also other open source SW than GPLv2 can use it freely, with one exception...

So if MySQL were controlled by a "hostile" party, proprietary software, and curiously GPLv3 software (which is missing from the list of FOSS exception licenses), could not use MySQL or a fork of it anymore.

Some thoughts on MySQL and Oracle

Posted Dec 17, 2009 20:13 UTC (Thu) by sbishop (guest, #33061) [Link]

It looks like this deal is going through.

What about the MySQL docs? They make forking hard.

Posted Dec 30, 2009 3:16 UTC (Wed) by kop (guest, #37748) [Link]

While the MySQL code is GPL(v2)ed the MySQL documentation is proprietary. There may be many forks of MySQL, but I have been unable to find any with useful documentation. Who wants to use a database that has no documentation? IANAL but it seems that re-creating the reference manual for a database, what should be the specifications for the code, is more than a huge task it is also fraught with legal risk should reference to the actual MySQL docs be part of the re-creation process. (There may be legal risk in any case. If nothing else in the threat of a lawsuit.)

To my mind the proprietary nature of the MySQL docs is a large dis-incentive to fork, one strongly reminiscent of traditional proprietary vendor lock-in. It puts MySQL in an entirely different category from the usual FOSS project when it comes to forking.

I would be entirely comfortable with any sale of the MySQL codebase -- if the documentation had a free license. As things stand the owner of the MySQL documentation has control over MySQL in a manner that's not comparable to other projects -- there are no forks of a similarly licensed code/documentation pair of which I am aware. This makes me very skeptical of arguments claiming that MySQL can be forked. I have seen no reasonable arguments that consider what it would take to fork the documentation along with the code.