Sun Announces New MySQL, Michael Widenius Forks

viktor.91 writes "Sun Microsystems announced three new MySQL products: MySQL 5.4, MySQL Cluster 7.0 and MySQL Enterprise Partner Program for 'Remote DBA' service providers." which showed up in the firehose today next to Glyn Moody's submission where he writes "Michael Widenius, founder and original developer of MySQL, says that most of the leading coders for that project have either left Sun or will be leaving in the wake of Oracle's takeover. To ensure MySQL's survival, he wants to fork from the official version — using his company Monty Program Ab to create what he calls a MySQL "Fedora" project. This raises the larger question of who really owns a commercial open software application: the corporate copyright holders, or the community?"

Comments

It depends

[raffe]

It depends on the license of the software. Always.

Re:It depends

[tolan-b]

Well, with FOSS it doesn't. It depends on whether the maintainers require copyright assignment.

In any case, the authors own the code (unless they reassign the copyright) and everyone else can do whatever they want with it provided they comply with the license.

The question in the summary is a bit stupid IMHO.

Re:It depends

[digitalunity]

Well, even if the maintainers have the copyrights, that only means future versions can be closed source. They can't terminate the already-outstanding licenses without a breach of terms. They also own the trademarks to the MySQL also.

IMO, Sun lost the hearts and minds of the developers which is where the real value was. The trademarks and copyrights are worthless if the community views MySQL's direction is wrong and moves entirely to a fork.

Re:It depends

[0xB00F]

It depends on the license of the software. Always.

It also depends on where you live. In Soviet Russia, software owns you.

Re:It depends

[Anonymous Coward]

In the case of a corporate OSS project, no, it doesn't. The developers don't own copyright, if the work in question was a work for hire.

In the case of a work work hire, the company owns the copyrights, if MySQL AB sold the copyrights to Sun, and there's no reason to think they didn't, Sun owns the copyrights. Oracle bought Sun and all of it's assests, ergo, the copyrights, and thus the code, belong to Oracle.

GPL is a copyright license, governed by copyright law. The owner of the copyrights, also owns the right to relicense, and litigate against anyone they deem is violating their copyright, and, in theory, that includes forks, which are derivative works. The copyright holder, also, in theory, reserves the right to revoke any licenses that were given out.

Individual, "community" developers are treated as freelancers and retain copyright (unless they've reassigned it) on their contributions, but any work done by MySQL AB, most likely now belongs to Oracle (unless they donated the copyrights to the FSF, which is how it's "supposed" to be done).

People erroneously like to think that OSS somehow is immune to normal copyright law. It isn't. Normal rules still apply.

The question in the summary is indeed stupid, though.

Re:It depends

[thrillseeker]

The question in the summary is a bit stupid IMHO.

You must be new here.

Re:It depends

[AKAImBatman]

They can't terminate the already-outstanding licenses without a breach of terms.

On the flip side, the forking company can't use the same business model as MySQL AB. Since MySQL owned the copyrights, they could see non-GPLed versions of the software under terms that were more palatable to corporations. To a certain degree, it served their purposes to fuel GPL fears.

Now that the forking company is 100% bound by the GPL, they must attempt to undo any misplaced fears about the GPL and seek to convince companies that what they really want is a support licene, additional tools, or trained consultants.

Re:It depends

[jedidiah]

Most companies don't need a more "palatable" license for an RDBMS. They typically
use it as a product, not something to build a product from. This is the key area
where the GPL can be a problem for a corporate entity. Most of Oracle's database
(or apps) customers don't have any reason to be concerned about their RDBMS having
a copyleft license.

They want assurance that their data will be protected and their operations won't suffer outtages.

Re:It depends

[paulthomas]

Agreed. See x.org for how quickly a community can switch to a fork.

Re:It depends

[digitalunity]

I don't think anyone questions that MySQL AB sold the copyrights to Sun, or that they can't again be transferred to Oracle.

The copyright holder, also, in theory, reserves the right to revoke any licenses that were given out.
 
 

No, not in theory. Have you actually read the GPL? In GPLv3, read sections 8 and 10. In GPLv2, read section 4.

You obviously aren't aware of OSS projects where community contributions are only accepted with copyright assignment to the software maintainer.

You're correct that Sun, or Oracle(if they buy Sun) could litigate breach of license terms in MySQL, but so could any copyright holders that contribute to MySQL forks.

Did I get trolled?

Re:It depends

[TheLinuxSRC]

The copyright holder, also, in theory, reserves the right to revoke any licenses that were given out.

I am going to have to see a citation for that. Unless that is written into the original license, that is completely wrong. In the case of the GPL, that is most certainly *not* the case. I am calling FUD.

Re:It depends

[TheRaven64]

What do you think most companies do with an RDBMS if they are not building software on top of it? Most either build in-house software on top of it, or license third-party software to run on top of it. In both of these situations the license is important. There's a reason MySQL AB made so much money selling licenses that let people avoid the GPL.

Re:It depends

[AKAImBatman]

Sounds like you haven't been following MySQL AB very closely. Their interpretation of the license was that any time you paired a MySQL database with an application, you needed a MySQL commercial license. Only if the application supported but was independent of MySQL would you not need to follow the terms of the license.

MySQL even tried to reinforce the idea by purchasing all the third party drivers and changing the licenses to GPL instead of LGPL or otherwise.

While MySQL's licensing info has changed over the years (interestingly not archived by the WayBack Machine...) even their current page on licensing is designed to steer users toward purchasing a commercial license:

Q3: As a commercial OEM, ISV or VAR, when should I purchase a commercial license for MySQL software?

A: OEMs, ISVs and VARs that want the benefits of embedding commercial binaries of MySQL software in their commercial applications but do not want to be subject to the GPL and do not want to release the source code for their proprietary applications should purchase a commercial license from Sun. Purchasing a commercial license means that the GPL does not apply, and a commercial license includes the assurances that distributors typically find in commercial distribution agreements.

For quite a few legal departments I've worked with, "the GPL does not apply" is magic words to their ears. They will instruct the business to grab the commercial license to get around the restrictions. In addition, there is the MySQL libraries issue I referred to above:

Q4: What is Sun's dual license model for MySQL software?

A: Sun makes its MySQL database server and MySQL Client Libraries available under both the GPL and a commercial license. As a result, developers who use or distribute open source applications under the GPL can use the GPL-licensed MySQL software, and OEMs, ISVs and VARs that do not want to combine or distribute the MySQL software with their own commercial software under a GPL license can purchase a commercial license.

The MySQL forking company is going to have to undo all of the anti-GPL ideas they've been riding, and convince companies that they don't need a commercial license. (Since it's not in the forking company's power to provide one.)

Re:It depends

[Crudely_Indecent]

Perhaps Widenius and the developers will change its name to "OurSQL" (Our as opposed to My) or "MoonSQL" (Moon as opposed to Sun)

Either way, below there are mentions of the InnoDB functionality that is third party, but nobody mentioned the SAP contributions in MAXDB (a MySQL derivative), what will happen to that?

It's happened before with XFree86 (forked by developers and now maintained as Xorg), Mambo CMS (forked by developers and now maintained as Joomla! CMS).

Does anyone else have examples of software that was forked by the developers where the original version became obsolete (or something like it)?

Sun/Oracle had an excellent opportunity to embrace their community, but it seems they're not going to do that.

Re:It depends

[MarkvW]

If you'd add mindshare to that, I think it would answer the question perfectly.

Re:It depends

[Anonymous Coward]

How about picking an antonym for 'Oracle' as their new name, given who owns them now :D

Or maybe just call it: 'The Full Monty DB' :D

Re:It depends

[rackserverdeals]

What do you think most companies do with an RDBMS if they are not building software on top of it?

Most people don't build software on top of an rdbms, they build software that uses and rdbms as a backend data store.

Re:It depends

[MrMarket]

Makes you wonder if Widenius saw the Oracle/Sun acquisition coming. I can't see Ellison putting a lot of support behind a competitor to Oracle's DB. If I were the founder and saw that coming, I'd split off a fork ASAP before my baby died on the vine post-acquisition.

Re:It depends

[Seth Kriticos]

Exactly. Most OSS projects source files supply a header detailing the Copyright holders and the License it is distributed with. Mostly they grant you a non revocable right/requirement to use, extend and redistribute the code with your modifications added. If you add code, that peace of code is owned by you and is automatically under the parent license. It's the same with mySQL, although their License text is a bit murky, but the FOSS exception basically aims for GPL v2.

As for the trademark: it lies with Sun/Oracle and they can basically do whatever they want with it. The worst thing of which is that you are no longer allowed to call it mySQL. A simple rename to something else would solve that problem (you could for instance fork the code and call it ourSQL).

That's what most open source is about: play in one sandbox as long as you like, and fork once you have different non-compatible ideas.

It's all written in law and license texts.

Re:It depends

[C0vardeAn0nim0]

you completelly ignore how a client talks to a RDBMS. here's how it works:

it works just like a web server (*). no, really.

a client connects to a tcp/ip port, sends a bunch of SQL satements and reads the answer.

if you know the protocol, you can interface with oracle, mySQL, sybase, whatever without touching oracle's code, which means oracle's license is irrelevant if all you want is to build a client app.

* this is a gross oversimplification, i know. sorry. but it was better than use a car metaphor, right ?

Re:It depends

[tolan-b]

Yes but they still *own* it. Everyone else is using it under license.

Re:It depends

[Z00L00K]

And also consider that Oracle also has the Sleepycat Berkeley DB engine, which you really have to know where to find to get.

And they also have the old Digital database engine.

Their method is to acquire competitors and then slowly decrease development and avoid promoting the products. Then the products can silently die.

But I wonder if they haven't bitten into something a bit too hard to bite into this time...

Re:It depends

[marcosdumay]

That is not the case for the GPL family and BSD. The original author can't revoke at will any code licensed under those licenses.

Also, being irrevocable is a prerequisite for both OSI and FSS accepting a license, except when that revogation comes as a consequence of one act you practice, and is limited to the person practicing it.

Re:It depends

[MrKaos]

It depends on the license of The Source. Always.

Patents, licensing, DRM, the Dark Side of The Source are they. Once you start down the dark path, forever will it dominate your destiny.

Re:It depends

[vux984]

if you know the protocol, you can interface with oracle, mySQL, sybase, whatever without touching oracle's code, which means oracle's license is irrelevant if all you want is to build a client app.

Until you want to sell the client app.

Re:It depends

[Jurily]

Agreed. See x.org for how quickly a community can switch to a fork.

And EGCS, the fork of GCC now called GCC.

Re:It depends

[SerpentMage]

And here is the crux of the argument and why MySQL is doomed...

You see if you have a product that binds to MySQL you will have to GPL your product. Why? Simple...

1) All (most?) drivers are GPL'd.
2) The MySQL notation uses a specific parameter delimintator that is specific to MySQL. And a 4 year old court decision said that there is no binding between application and RDMS if the same code can be used on other databases. With the special notation, it is not possible and hence constitutes a GPL binding.

Personally I see MySQL falling off to the way side...

Re:It depends

[TheLinuxSRC]

You may want to re-read my comment... I quoted the GP and said exactly what you just reiterated.

Re:It depends

[SerpentMage]

>The MySQL forking company is going to have to undo all of the anti-GPL ideas they've been riding, and convince companies that they don't need a commercial license. (Since it's not in the forking company's power to provide one.)

Actually I am going to sit back and laugh at this one. Because now they are going to have to eat their own dogfood. No short cuts!

My bet is that this will not work, unless of course they rewrite...

I have nothing against the GPL, I avoid it, but those who want to use it should have the freedom to do so. Just like it is my choice to avoid it. But the MySQL forking...

Re:It depends

[Jane Q. Public]

No, the "authors" do not necessarily "own" the code, if some or many of the authors work for a commercial interest. If you are an employee, normally your work belongs to the company.

The other contributors, on the other hand, do not necessarily "own" the code, either. It depends on what kind of license they have created. It is certainly possible to waive copyright ownership in a license.

Re:It depends

[Anonymous Coward]

Interbase vs Firebird SQL all over again...

Re:It depends

[rackserverdeals]

Problem easy to solve. Don't distribute the MySQL JDBC drivers in the case of a Java app, for example.

Tell the users to download and install MySQL and the drivers themselves. Not a big deal and not uncommon that the DB is not bundled with the application.

Better yet, use the application online and provide it as a service. You're not distributing anything and you don't have to worry about the GPL.

Re:It depends

[rackserverdeals]

Or even better yet. Don't use MySQL.

Re:It depends

[petermgreen]

amusingly in that case the old upstream basically gauranteed the success of the fork by licensing new versions of certain key libraries in a way that was unacceptable to the distros. The distros considered it gpl incompatibile which was a problem because those were used in gpl apps*

*yes i'm aware of the "major component of the operating system" exception but many distros preffer not to rely on it because it is not clear exactly what counts as a "major component of the operating system" or what the case is when both the app and the library are distributed with the operating system.

Re:It depends

[CarpetShark]

Now that the forking company is 100% bound by the GPL, they must attempt to undo any misplaced fears about the GPL and seek to convince companies that what they really want is a support licene, additional tools, or trained consultants.

...or PostgreSQL ;)

Re:It depends

[CarpetShark]

Well, even if the maintainers have the copyrights, that only means future versions can be closed source.

Not unless they're prepared to rewrite or throw away any parts contributed by other authors under the GPL.

Re:It depends

[DragonWriter]

Well, even if the maintainers have the copyrights, that only means future versions can be closed source. They can't terminate the already-outstanding licenses without a breach of terms.

If the "already-outstanding" licenses are license contracts, that would be true, but that is only the case where there is mutual consideration (i.e., where the licensee has given something to the licensor in exchange for the permissions in the license.) Gratuitous licenses (which most FOSS licenses would seem to be) are revocable at will, though the effect of revocation may be limited where a court would find that it result in an injustice on an estoppel theory.

The kind of legal uncertainty a license revocation would create for any fork would probably kill any serious business adoption and paid support opportunities until and unless the cloud could be cleared up, though personal users might not care.

Re:It depends

[DragonWriter]

Until you want to sell the client app.

If you do it based on a clean-room reimplementation of the database's communication protocol rather than, e.g., relying on the vendor's client libraries, as GP's post ("as long as you know the protocol...") suggest, the main barrier to the selling the client app is the challenge it poses to your ability to assure recipients that it won't get broken by future server upgrades.

Re:It depends

[CarpetShark]

if you know the protocol, you can interface with oracle, mySQL, sybase, whatever without touching oracle's code, which means oracle's license is irrelevant if all you want is to build a client app.

If that were true, any database system could easily implement all features of the SQL standards. Knowing the language isn't enough; you have to know how to parse that language --- often without losing compatibility with your existing tweaks ---, how to get the required answer given how your database stores stuff, and how to do it all in a reasonably efficient way.

Re:It depends

[x2A]

"a 4 year old court decision said that there is no binding between application and RDMS if the same code can be used on other databases. With the special notation, it is not possible and hence constitutes a GPL binding"

If your wording of the court case is accurate and complete (enough) - no it's doesn't mean that. If the court case has ruled on a situation where there is 'no binding', then that court case will only apply to other cases where there is similarly 'no binding'. Reversing a condition doesn't necessarily reverse the outcome.

Re:It depends

[fast turtle]

Interestingly enough is the fact that under the GPL; the code can be forked and used in any other project. It is true that Oracle can change the license from GPL to what ever they want on future versions if and only if they have the copyright to MySQL as one of the assets they purchased.

In the case of Open Office, it was released under the LGPL instead of the GPL, which appears to have been one of the reasons that Sun had many problems getting outside developers involved in the community. From a personal standpoint, though I am not a GPL purist, I switched to using K-Office some time back as it's almost impossible to get Open Office to build reliably under any version of GCC later then 3.4 (Yes I'm a Gentoo User but not a Gentoo Ricer).

Re:It depends

[DragonWriter]

Not unless they're prepared to rewrite or throw away any parts contributed by other authors under the GPL.

In order to offer both GPL and commercial licenses, MySQL already had to demand copyright assignment for any code they included; it wouldn't have been legal for them to use the dual-licensing model they did if they accepted contributions under the GPL. So no worries on that front.

Re:It depends

[SerpentMage]

The case related to a company who used ODBC and whether or not they binded to MySQL. It was not the NuSphere case, but one that used ODBC and MySQL.

The question was if your application used ODBC and MySQL was it binding in the GPL sense?

The answer was in the fact whether or not the application could function with another database. At the time the result was that MySQL lost the case since the application could function with another database.

It was around that time MySQL GPL'd all drivers, and changed their syntax so that it would only work on their servers. That way it is a GPL binding as per the court case.

Re:It depends

[Belial6]

So, someone makes the absolute bare minimum database that is nothing more than a glorified text file, but uses MySQL syntax. Tada! The application can function with another database.

Re:It depends

[vux984]

If you do it based on a clean-room reimplementation of the database's communication protocol rather than, e.g., relying on the vendor's client libraries

Who wants to do THAT.

as GP's post ("as long as you know the protocol...") suggest,

No. His comment was trying say that the client app is separate from the server app. That you communicate with the server, but that your client app ultimately stands separate. So you can sell your client app without worrying about the license on the server. Which is true. In precisely the same way that I can write and sell a browser ('client app') that talks to IIS7 without worrying about the license for IIS7. Because I'm not selling / distributing IIS7.

However, in reality, a RDBMS client app is worthless without the RDBMS it talks to; and in the majority of cases, when you sell an rdbms client app, the customer requires their own server. And they generally expect you provide them with it... and at that point the rdbms license becomes a big issue.

I have to either license the RDBMS for resale so I can bundle / integrate it with my software, or instruct customers that they have to provide an rdbms server that meets my specifications.

For REALLY BIG enterprisey stuff, people expect the latter.

But for smaller stuff... like a point of sale system for a small/med business, or the software to run your medical practice, or your brake and muffler shop... or handle the accounting for your 22 location chain of restaurants... people want to just 'buy a system' and install it. And there are a lot of systems like this.

With a GPL rdbms, I can't distribute it -with- my proprietary application, which means the former option is closed to me, and I have to distribute mysql separately from my app or have them obtain it themselves... either is a hassle.

The main reason people bought mysql licenses was so that they could sell complete turnkey solutions based on mysql and distrubte mysql WITH their solution, install it all from one CD, one wizard, etc.

Re:It depends

[Anonymous Coward]

When Oracle bought innodb, and then swiftly followed with sleepycat, it was "checkmate in X moves" for MySQL.

Yes that was 3 years ago. But 3 years isn't a long time if you are going to pick a DB for your company. You have to look at the long term. Just from a strategic POV you would have avoided MySQL or start to migrate off it if you were already using it.

Whether it was Oracle buying Sun or Oracle strangling MySQL, it was still "eventual checkmate unless Oracle makes a really stupid move".

Re:It depends

[digitalunity]

Servers aren't upgraded in a vacuum. Proper QA should include thorough testing to ensure the server, client libraries and application are all compatible.

Sure, a clean room implementation of client libraries is possible but it is pretty unlikely to be able to keep up with the DB maintainers. That's yet another reason(other than the long list I already have) to always use PostgreSQL. It's BSD, meaning you can do whatever you want, even roll it up and smoke it if you so choose.

Re:It depends

[Crudely_Indecent]

'The Full Monty DB'

I LIKE IT!

I've always been partial to writing code alfresco, although it takes more preparation. I must use a laptop cooler (upside down) and put a towel on the leather couch.

Re:It depends

[mr_mischief]

Repeat this sentence three times: "Bundling is not binding." There, feel better?

The GNU General Public License v2 - An Overview (subsection on bundling and aggregation).

You can also read the GPL, v2 and the GPL, v3 for more information rather than spreading disinformation.

Re:It depends

[mr_mischief]

The company that writes some application that talks to a GPLed MySQL will have to follow the GPL in distributing MySQL. They will not have to license their own code as GPL. What's the problem with giving someone the source to something they could have gone to the website and downloaded themselves anyway?

Re:It depends

[mr_mischief]

Actually, there's nothing about InnoDB or SAP that are third-party to Oracle, since it already acquired both of those.

Re:It depends

[mr_mischief]

Crap, they didn't buy SAP. They bought PeopleSoft. Sorry.

Re:It depends

[(H)elix1]

And also consider that Oracle also has the Sleepycat Berkeley DB engine, which you really have to know where to find to get.

Hidden prominently on the main database page (http://www.oracle.com/database), is a link to Berkeley DB, as well as why it might be the right tool for the job.

Re:It depends

[Anonymous Coward]

It also depends on where you live. In Soviet Russia, software owns you.

And in Redmond, WA.

Re:It depends

[Hatta]

What's the difference? Most companies add another layer to their RBDMS. Arguing whether the RBDMS is beneath or behind the custom software is simply stretching a spatial metaphor too far.

Re:It depends

[Crudely_Indecent]

Oracle acquired SAP? I believe you are mistaken.

MAXDB is a closed source (formerly GPL'd) and is owned and maintained by SAP AG.

http://en.wikipedia.org/wiki/MaxDB

From the MAXDB Wiki:

Our customers were able to buy MaxDB directly from SAP for running SAP applications. From October 2003 to August 2007 they were also able to buy the database from MySQL for running non-SAP applications. SAP owned MaxDB development and IP. SAP provided the development support for users of MaxDB under both SAP applications and non-SAP applications where as MySQL provided primary support for non-SAP applications.

Re:It depends

[Elwood P Dowd]

Once x.org started, the switch was light speed, but xfree86 was godawful for a long, long time before that.

Re:It depends

[mr_mischief]

Oh, sorry to reply yet again. Yes, I can name a number of project forks in which the original fell by the wayside as a fork took off. Not all the originals are dead, mind you, but the forks are much more popular. A few of these are situations in which the original is still viable (Debian, for example), but in which the fork has a huge number advantage or a lot of momentum.

GCC -> EGCS -> GCC
Mosaic -> Netscape
Netscape -> Mozilla
Mozilla (Seamonkey) -> Mozilla Firefox
KHTML -> WebKit
Debian -> Ubuntu
XFree -> X.org
StarOffice -> OpenOffice
SSH -> OpenSSH
Hack -> NetHack
osCommerce -> ZenCart
AT&T Unix -> BSD

Re:It depends

[eliot1785]

Well, even if the maintainers have the copyrights, that only means future versions can be closed source. They can't terminate the already-outstanding licenses without a breach of terms.

Oracle doesn't need to take MySQL closed-source or even make any money from it directly. By buying Sun and hence MySQL, they'll disrupt MySQL's development, giving Oracle DBMS a jump in development time, and presumably in features/quality, market share, etc.

It takes time to set up a real fork, and Mr. Widenius may not have the resources to hire as many employees as MySQL AB did without the commercial dual licensing arrangement. As for the development within the MySQL subsidiary, Oracle can slow that down simply by rewriting the release schedule.

In sum, if the goal of this acquisition was to address the threat of MySQL to Oracle DBMS (which I think it may have been, in part), then that task is accomplished not by making money from MySQL but by slowing down its development. Even a difference of 6 or 12 months would be a huge advantage for Oracle DBMS.

That's why I felt that antitrust laws should be considered here, because it's an anticompetitive strategy, but it would be hard to prove because the effect is only indirect.

Also, a less-informed regulator or judge would look at industry figures that show the database market to be highly competitive and not at risk of a monopoly. But industry figures are based on revenue, not adoption, which masks MySQL's true penetration because most MySQL installations are of the free version.

Re:It depends

[rackserverdeals]

What's the difference?

When you build on top of something, you are dependent on that as your foundation. When you use it on the backend, it is an external piece of the application that can be replaced with relative ease.

Re:It depends

[Anonymous Coward]

You seem to have missed the part about MySQL drivers being GPL.

Re:It depends

[vux984]

I interpret 'mere aggregation' as permission to put a copy of mysql and its installer on a CD of software, including some which might be non-GPL.

But as a key part of a complete system? Does that still qualify for 'mere aggregation'?

Re:It depends

[Anonymous Coward]

And also consider that Oracle also has the Sleepycat Berkeley DB engine, which you really have to know where to find to get.

And they also have the old Digital database engine.

Their method is to acquire competitors and then slowly decrease development and avoid promoting the products. Then the products can silently die.

You are so full of it! And here is why you're so full of it:

http://www.oracle.com/database/berkeley-db/index.html

Re:It depends

[jaseuk]

Wrong - most people USE software, that USES an RDBMS. The majority of businesses and individuals don't build any software at all. Of course for those providing the software, the client being GPL might cause a few problems, but then there is always postgres.

Jason.

Re:It depends

[JAlexoi]

Ahh.... You clearly have not heard of www.tia.dk :)

Re:It depends

[Estanislao Mart�nez]

The company that writes some application that talks to a GPLed MySQL will have to follow the GPL in distributing MySQL. They will not have to license their own code as GPL.

From GP's comment, it sounded to me like MySQL AB has argued that any application that absolutely needs MySQL to function is a derivative work of MySQL, and thus, cannot be distributed without license from MySQL AB; and that therefore, to distribute such an application, one must either license it under the GPL, or obtain a commercial license from them. The argument seems to hold that whether the application links to a GPL-licensed MySQL client library or not is irrelevant; what would matter is whether the application can be functionally severed from MySQL.

Re:It depends

[rackserverdeals]

Did you notice that this story was posted in the "Developers" section?

Re:It depends

[Anonymous Coward]

a client connects to a tcp/ip port, sends a bunch of SQL satements and reads the answer.

if you know the protocol, you can interface with oracle, mySQL, sybase, whatever without touching oracle's code, which means oracle's license is irrelevant if all you want is to build a client app.

While there are sql standards, there are differences. Very simple example: get the first 10 records from a table.

MS SQL Server:
SELECT TOP 10 product, descr, email
FROM products

ORACLE:
SELECT product, descr, email
FROM products
WHERE ROWNUM = 10

MySQL:
SELECT product, descr, email
FROM products
LIMIT 10

If your client app will work with multiple databases, you need to know these differences.

Re:It depends

[mishehu]

I guess you never read the EXCEPTIONS-CLIENT file which lists other licenses that you can link your software to if you happen to use one of the listed licenses.

Re:It depends

[ais523]

Debian -> Ubuntu

I don't think this counts as a fork at all. Ubuntu is still very much reliant on upstream changes, whereas a true fork wouldn't be.

Re:It depends

[CarpetShark]

MySQL already...demand copyright assignment for any code they included

Ahh, I didn't know that. This explains why it's been largely the MySQL team's own code that went into it too.

Re:It depends

[multipartmixed]

> Until you want to sell the client app.

Last I checked, you could legally dynamically link libclntsh.so off a client's oracle installation to make OCI calls as a third-party software developer.

You can't do that with mysql if it is GPL-only. It would GPL-encumber your whole G-D code base.

Re:It depends

[Anonymous Coward]

From GP's comment, it sounded to me like MySQL AB has argued that any application that absolutely needs MySQL to function is a derivative work of MySQL

That will teach 'em to write their code to not be database specific.

Seriously. I've been looking at buying a music streaming device, and Logitech makes one. Except their server software requires MySQL to be installed. For a freaking music player.

The worst part? They used to use SQLite, the embedded database system meant for exactly this kind of thing. But some of their users wanted MySQL, so they changed it to work with MySQL *ONLY*.

Re:It depends

[Crudely_Indecent]

Ubuntu would be considered a derivative of Debian. A symbiotic relationship.

Re:It depends

[ToasterMonkey]

It was around that time MySQL GPL'd all drivers, and changed their syntax so that it would only work on their servers. That way it is a GPL binding as per the court case.

So is the question of whether the GPL is viral or not even up in the air anymore? I've never heard this before, but it's f'ing nuts. More power to MySQL devs, leave, fork, whatever, and take your damned OSS purism ideals with you, it wont be missed.

Re:It depends

[Anonymous Coward]

Your argument is wrong. You don't have to open source anything by using any GPL product. If you publish your app. and it uses Free software, then it must be GPL because GPL is in there. If it uses GPL software 'on the side' as a library or computing system, or RDBMS, and all of the application is proprietary, then it can stay that way, be published (for money) and no harm done.

Re:It depends

[Hognoxious]

Most people don't build software on top of an rdbms, they build software that uses and rdbms as a backend data store.

Sorry, I don't see the difference.

SAP uses Oracle as a backend data store. SAP runs on top of Oracle. Those are equivalent to me.

Since when did pedantic (and wrong) asshole = insightful?

Re:It depends

[rackserverdeals]

Sorry, I don't see the difference.

I accept your apology.

Right

[ShieldW0lf]

Did anyone else notice that his little toy database is practically useless without InnoDB, which was written by a third party and is owned by Oracle?

What's he going to do, make another toy and cross his fingers that someone will come along a second time and redeem his craptastic creation?

What a joke...

Re:Right

[tolan-b]

Why would the fork have to stop supporting InnoDB?

Re:Right

[yakatz]

Did anyone else notice that his little toy database is practically useless without InnoDB, which was written by a third party and is owned by Oracle?

If you mean for transactions.
If you want a really fast free database that supports fulltext indexing, and you don't need transactions, MyISAM in the engine to use.

Re:Right

[tolan-b]

It's not necessarily true that MyISAM is faster actually.

http://www.mysqlperformanceblog.com/2007/01/08/innodb-vs-myisam-vs-falcon-benchmarks-part-1/

Re:Right

[Haiyadragon]

Yeah, you're right. Foreign keys, who needs 'em?

Seriously though, this could be good news for PostgreSQL. Fingers Crossed.

Re:Right

[Daniel_Staal]

Or you can use SQLite, get more speed, and still have transactions. (Although fulltext indexing does require a loadable extension.)

Re:Right

[Anonymous Coward]

As long as you don't need concurrent users. As soon as someone tries to update anything in the database, SQLite will lock the entire database until that update is completed.

I always consider SQLite to be a replacement for ad-hoc file formats, client-side storage, or anywhere else where you've got some data and something like SQL would be handy to manipulate it. Not so great as a replacement for database servers, unless everything's read-only.

Re:Right

[rvw]

Seriously though, this could be good news for PostgreSQL. Fingers Crossed.

AFAIK more and more people are using PostgreSQL. More and more providers are supporting it. Five years from now, it could be a whole different landscape...

Re:Right

[K. S. Kyosuke]

Or you can embed a Firebird engine and have a really well thought out transaction model...

Re:Right

[TheRaven64]

SQlite has supported per-table locking for a while, and I believe it supports per-row locking in some situations. It is not designed for concurrent writes, but it can be great for anything read-heavy workloads. It's certainly not suited for situations where you have a lot of concurrent writes, but for a CMS it can be a very good fit.

If you want full-text indexing, transactions, and lots of concurrent users, PostgreSQL is generally a better bet. MySQL is being squeezed at the bottom by SQLite and at the top by PostgreSQL, and both have less restrictive licenses (public domain and BSD, respectively). I'm amazed that it's survived this long.

Re:Right

[rackserverdeals]

Yeah, you're right. Foreign keys, who needs 'em

They should even get rid of SQL support altogether and just call it My. Who needs all that complicated elitist crap anyway?

Re:Right

[Anonymous Coward]

Did anyone else notice that his little toy database is practically useless without InnoDB, which was written by a third party and is owned by Oracle?

If you mean for transactions.
 

How about data integrity? Is that desirably in a database?

Re:Right

[rackserverdeals]

Yeah right, if postgres was popular you'd be pimping mysql, just to be a trouble maker. In any case postgres is dangerously small and uns

Awww.. Look he tried to make a funny. See.. he was going to write unstable, but he stopped in the middle because his postgresql backed keyboard locked up.

What a darling to try and be clever.

People like Postgresql not because it's not popular like mysql. People like it because it's not crippled like mysql.

My experience, in general, has been that people moving from big commercial databases like postgresql. Those that that are new to rdbms's like mysql.

Re:Right

[TheRaven64]

Depends on the application. It's always desirable, but it may not be the most important feature. If something like Wikipedia, for example, loses a day's worth of edits, do you think many people would notice or care?

Re:Right

[Anonymous Coward]

I think the point was that it support FullText.
Try that with InnoDB.

Re:Right

[ShieldW0lf]

Only two-bit operations that are run out of their mothers basement use PostgreSQL.

Operations like Skype

Re:Right

[tolan-b]

Sure, but there was an implication that MyISAM is faster too.

Personally I use a mix of the two. InnoDB mainly for tables that kept crashing under MyISAM ;/

Re:Right

[Z00L00K]

That's why they started work on the Falcon engine.

Re:Right

[marcosdumay]

You know, I never understood the point of InnoDB. One may want a complete, fully functional DBMS, in that case, there is PostgreSQL, or one may want a lightning fast data indexing/accessing machine, and for that case there is MySQL. InnoDB brings something that is slower than Postgres and still isn't a complete DBMS by any point of view.

Re:Right

[MrKaos]

If you want a really fast free database that supports fulltext indexing, and you don't need transactions, MyISAM in the engine to use.

I can't really get by without being able to lock records in InnoDB, as opposed to the whole table in MyISAM. But I think that the Falcon engine is the long term way to go.

Maybe if MySQL was GPL3 this mess might look a lot different, then again so would MySQL.

Re:Right

[amicusNYCL]

If you want a really fast free database that supports fulltext indexing, and you don't need transactions, MyISAM in the engine to use.

That's exactly what I thought for a while. Then I built a system where many, many people were using it all at once and figured out that MyISAM has another "feature" that InnoDB doesn't: full table locking. It locks an entire table when running queries on it. When I've got a table that holds 400k records for a total of 90k users, and many of them are trying to access or update at the same time, MyISAM's table locking is a deal-breaker. InnoDB implements row-level locking instead. MyISAM is great for smaller hobbyist things, but it's terrible for concurrent access on large data sets. I would open the process list in phpMyAdmin and see 30 or 40 queries in the queue waiting for one to finish. So if your data set is small enough that MyISAM can handle it, then it's also going to be small enough for InnoDB to operate on it quickly, at least as quickly as MyISAM. I don't see any reason to use MyISAM when InnoDB is an option. I'm not trying to get into the politics of which engine is better to use, but InnoDB walks all over MyISAM in the real-world performance area, unless you're building yourself a shopping list or something similar.

Re:Right

[alta]

I have some tables in my database where I need a fully function DBMS with transactions and such, where I use InnoDB in Mysql. And I have other tables where I need lightening fast data indexing/accessing, in which case I use MyISAM. Sometimes I have need both and I get to make the decision which to use. My app has financial info, user info, logging, products, all sorts of different needs. I think you'll find that most apps have situations where data integrity is the most important, while other times speed is important.

And every now and then I'll implement a heap, archive or merge table... It all depends on my needs.

So for this reason, I choose MySQL. On the same server, i get to have my cake and eat it too.

Re:Right

[rackserverdeals]

Why would the fork have to stop supporting InnoDB?

If it's forking because Oracle would own MySQL, then it implies you can't trust Oracle. If you can't trust Oracle, you shouldn't use InnoDB.

Re:Right

[DragonWriter]

You know, I never understood the point of InnoDB. One may want a complete, fully functional DBMS, in that case, there is PostgreSQL, or one may want a lightning fast data indexing/accessing machine, and for that case there is MySQL.

The point of MySQL isn't a "lightning fast data indexing/accessing machine". The point of MySQL is the modular backends which enable it to serve as a common gateway to tables that each use the storage engine most appropriate to the way the table is used. (some of which may require a lightning fast data indexing/acessing engine and accept some risks to get it, some tables may not.)

The point of InnoDB (and, presumably, Falcon) is to support the kind of usage scenarios for which traditional RDBMS are designed, while the point of certain other MySQL table drivers is to support other types of loads.

Re:Right

[Anonymous Coward]

I always find it funny when people point out how inferior MySQL is on a somewhat sizable site run by - MySQL.

Re:Right

[Glendale2x]

I started with MySQL because it had replication. Yeah, I know Postgres has slony or whatever it's called, but I cringe at the though of maintaining disparate pieces of software (that sometimes don't get along) on something as critical as a database. Maybe it's changed, maybe it hasn't, but that's how it was when I chose to use MySQL.

Re:Right

[x2A]

"My experience, in general, has been that people moving from..."

Shows what you know. I use postgrexperience, cuz myexperience is somewhat limited...

Re:Right

[Anonymous Coward]

I really hope you use the Sequoia replication for MySQL if you think your data is important. The default replication in MySQL is as safe as walking next to lions.

Re:Right

[Anonymous Coward]

Too bad SQLite doesn't support any concurrency in writes :(

If there would at least minimal support for that I would be tempted to use it for something. But currently it isn't worth the trouble to write heavy error checking for failures because of random concurrent write errors. (or write some application level locking to prevent concurrent writes)

Re:Right

[Dhalka226]

You're assuming that people have both choice and knowledge. I'd venture to guess that most people using MySQL are doing so with things like phpBB on shared hosting, and that a significant portion of the remaining MySQL users are ones who began their RDBMS career working with MySQL on shared hosting. Of course we all know stories like Google using MySQL and that's great--I don't have nearly the problems with MySQL that some people here do--but I don't think it's the norm by any stretch of the imagination.

So what's the point of InnoDB? If you want transactions (primarily) but aren't in a position to simply use a different database server. Maybe you don't have physical access to the server (shared hosting, etc); maybe your applications are coded in such a way that they don't support it. Or, people simply don't know about Postgres and think the next leap up from MySQL is forking out cash for Oracle. (And please: I'm not trying to make any comment that if people simply knew what they were doing they'd use Postgres instead of MySQL. I'm simply offering rationale to "why would somebody use InnoDB on MySQL?")

InnoDB is your classic jack-of-all-trades, master-of-none: Not as fast as MyISAM, not as full-featured as something like Postgres. Sometimes, though, that's what the situation calls for.

Re:Right

[Anonymous Coward]

Seriously though, this could be good news for PostgreSQL. Fingers Crossed.

AFAIK more and more people are using PostgreSQL. More and more providers are supporting it. Five years from now, it could be a whole different landscape...

I sure hope so!

PostgreSQL is an excellent database, and with a little more integration work in the field of multimaster replication, it will be ready to compete with Oracle.

PostgreSQL deserves to be the dominant free open source database.

Re:Right

[Anonymous Coward]

Problems with data integrity always point to errors in memory handling logic. Those are exploitable bugs.

Re:Right

[maraist]

We fought back and forth with the idea of Sequoia (CJDBC), and it never really made sense.

A) no cross DB/schema talk
B) Requires a configuration for each DB separately (we have 50 independent apps that share hardware)
C) uses a database to store journal data (thus still a single point of failure)
D) SLOW for write-mostly transactional apps (we're 20/80 read/write)
E) Crippled command-line client.
F) Requires slightly extra hardware, especially w/ redundant front-ends.

Granted w/ Master/slave, you can't use cross DB updates. And certainly, w/ MySQL master/slave you're at the mercy of bugs.

Re:Right

[JAlexoi]

If you want a really fast free database that supports fulltext indexing, and you don't need transactions, MyISAM in the engine to use.

That is a load of bullshit. There are a number of cases where MyISAM will be actually slower. It depends not on the engine, but on the usage patterns of the database and database structure. And, interestingly, it varies from application to application.

Re:Right

[Anonymous Coward]

That is a load of bullshit. There are a number of cases where MyISAM will be actually slower. It depends not on the engine, but on the usage patterns of the database and database structure. And, interestingly, it varies from application to application.

It seems that everyone is missing the point. The OP said MyISAM supports Fulltext and not transactions. who said one is faster than the other?

Re:Right

[alanium]

Near as I can tell, the trends are downward for both.

Re:Right

[ShieldW0lf]

I always find it funny when people point out how inferior MySQL is on a somewhat sizable site run by - MySQL.

Is there a less trivial site?

Re:Right

[Glendale2x]

I can't say I've hit a bug in MySQL replication in years. It's not like I'm still trying to use 3.23. I'm not saying there aren't any bugs, just none that I've come across personally.

Who Owns Open Source Applications?

[Rary]

This raises the larger question of who really owns a commercial open software application: the corporate copyright holders, or the community?

No one. Or, perhaps, everyone. That's kind of the point, isn't it? It isn't locked into anyone's individual grip.

Re:Who Owns Open Source Applications?

[disi]

Agree, this reminds me of XFree86 :) Maybe the forked project will become a successor like Xorg?

Re:Who Owns Open Source Applications?

[anss123]

No one. Or, perhaps, everyone. That's kind of the point, isn't it? It isn't locked into anyone's individual grip.

In case of mysql I think they made a living on selling versions without the gpl license. That business model will not work for a gpl fork.

Re:Who Owns Open Source Applications?

[Anonymous Coward]

Sun owns MySQL, they can do whatever they like with its source code, because MySQL the company requested copyright to be assigned to them to all contributors, now Oracle can even use that code inside their proprietary databases and even create a "MySQL mode" for Oracle DB like FreeBSD have a linux mode for its kernel, but using the real MySQL sources to implement it.

What can Monty Program Ab? do improve (deteriorate is an option too) the code, distribute it following the GPL, but creating a dual licenced version like they did previously, I do not think so, unless they request copyright again to all the people that worked building MySQL other than the developers at Monty Program Ab

I Forked a Couple Nights Ago

[Anonymous Coward]

I forked a couple nights ago. I'm not trying to brag - I'm just a geek that's happy to be forking.

My girlfriend even looks good and is skinny!!

All I did was move to the (somewhat - I don't want to live on the drunks' 2am driving route) cool part of town, and go out as often as possible.

Geek in the pink ftw.

Re:I Forked a Couple Nights Ago

[bhtooefr]

The problem with forking is all the child processes, though.

Re:I Forked a Couple Nights Ago

[Anonymous Coward]

Heh... Sometimes the child processes are worth it.

Re:I Forked a Couple Nights Ago

[TooMuchToDo]

kill -9?

Re:I Forked a Couple Nights Ago

[Anonymous Coward]

Not always. Ever tried to fork a child process in a sandbox?

Re:I Forked a Couple Nights Ago

[multipartmixed]

setrlimit(3)

"r" in this case being "rubber"

It's GPL.

[AltGrendel]

This raises the larger question of who really owns a commercial open software application: the corporate copyright holders, or the community?"

So the answer is yes.

Re:It's GPL.

[amicusNYCL]

It's not a question of who owns it, it's obvious that Sun owns it. They bought it, they own it. That's not the issue. The issue is that the license that it's distributed under allows anyone to use and modify it, it doesn't matter who owns it. Sun owns it, everyone else can use and modify it, what's the question?

Repercusions for FOSS licenses

[syousef]

It's not the first time. I've seen supposedly open source die a cruel death at the hands of its creators. Anyone remember the Free Internet Chess Servers? The FICS code is still on dark corners of the net, but you'll have a fight on your hands if you want to try to use it, and I believe the guy who claims to own it because he contributed to it used it as the base of the current incarnation of FICS which is actually a paid service. You can't get the source to the server from there anymore.

So if Oracle are able to somehow prevent the use of this code, either due to terms of employment of the pricinple devs or by claiming ownership of the code and rescinding the free license, it'll make all these licenses worthless. Oracle has deep pockets. Individual developers don't.

Hell even if they can't impose their will legally but still manage to get their way due to fragmentation of the group, it's a black day for FOSS.

I really REALLY hope the devs are able to fork and move on.

Re:Repercusions for FOSS licenses

[Burkin]

So if Oracle are able to somehow prevent the use of this code, either due to terms of employment of the pricinple devs or by claiming ownership of the code and rescinding the free license, it'll make all these licenses worthless. Oracle has deep pockets. Individual developers don't.

Unless those principle devs are still working at Oracle they can't do the former, and the latter is only possible on future versions of MySQL so one can fork the last free version of the software and Oracle can't do a damn thing about it.

Re:Repercusions for FOSS licenses

[Andy Dodd]

But what license was the FICS code under? Was it really "open source"?

There are plenty of licenses that provide for distribution of source but are so restrictive that no one considers them to be "open source".

Re:Repercusions for FOSS licenses

[Anonymous Coward]

What surprises me is that Sun didn't ask the main devs to sign a non-compete contract as part of the original deal.

Re:Repercusions for FOSS licenses

[mrcparker]

FICS has been replaced by chessd: http://chessd.sourceforge.net/index-en.html

Re:Repercusions for FOSS licenses

[backwardMechanic]

Isn't this the whole point of a real open source license, rather than just publishing stuff and saying 'go play'? Either you're spouting misinformation, or the GPL et al are completely worthless...

Re:Repercusions for FOSS licenses

[binarylarry]

I don't believe Non Compete agreements are valid in California.

Re:Repercusions for FOSS licenses

[JasterBobaMereel]

Free internet chess server - source GPL'd

"http://sourceforge.net/project/showfiles.php?group_id=86389"

You didn't look very hard ...?

Once it's GPL it's free forever you cannot close source it ... you can only close source newer versions...

Re:Repercusions for FOSS licenses

[Burkin]

What surprises me is that Sun didn't ask the main devs to sign a non-compete contract as part of the original deal.

Yes, I too am surprised that Sun didn't ask the main devs to sign an illegal non-compete agreement.

http://lawzilla.com/content/noncompete.shtml

Re:Repercusions for FOSS licenses

[rackserverdeals]

Isn't this the whole point of a real open source license, rather than just publishing stuff and saying 'go play'? Either you're spouting misinformation, or the GPL et al are completely worthless...

This is one of the weaknesses of the GPL that doesn't work well with internet services.

The GPL only requires you to provide the source, and modifications, if you distrubute a product.

Installing the software on a server, and letting others access the software, doesn't count as distributing the software.

Re:Repercusions for FOSS licenses

[rackserverdeals]

Unless those principle devs are still working at Oracle they can't do the former, and the latter is only possible on future versions of MySQL so one can fork the last free version of the software and Oracle can't do a damn thing about it.

But what would the business model be? Any MySQL fork no longer has the ability to dual license the software since the copyrights have been sold. That's how MySQL AB made money.

Developers, and their families, can't eat freedom and self righteousness.

Re:Repercusions for FOSS licenses

[Burkin]

But what would the business model be? Any MySQL fork no longer has the ability to dual license the software since the copyrights have been sold. That's how MySQL AB made money.

Ask Monty what that would be. He's the one making a new business off of the fork. I'm sure he has a plan otherwise I doubt he'd be sinking his money into it.

Developers, and their families, can't eat freedom and self righteousness.

I'm pretty sure Monty is paying the people to work on his fork.

Re:Repercusions for FOSS licenses

[Burkin]

Welcome to 7 years ago. http://en.wikipedia.org/wiki/Affero_General_Public_License

Re:Repercusions for FOSS licenses

[rackserverdeals]

Welcome to 7 years ago. http://en.wikipedia.org/wiki/Affero_General_Public_License

You do realize that the AGPL is not the GPL right? And that not many projects use the AGPL? Only GPLv3 has an Affero type clause in it and many GPLv2 licensed projects haven't moved to GPLv3.

Re:Repercusions for FOSS licenses

[base3]

Another example is the CD database (CDDB) that was locked down after accepting contributions from users. There is still an out-of-date dump of the database from a decade or so ago floating around, and a parallel project, FreeDB, that Gracenote goes out of its way to prevent anyone from actually using.

Re:Repercusions for FOSS licenses

[Burkin]

You do realize that the AGPL is not the GPL right?

Yes, but if you were worried about this you should have been using the AGPL since it came out in 2002.

And that not many projects use the AGPL?

Then they should have been using it had they cared enough.

Only GPLv3 has an Affero type clause in it and many GPLv2 licensed projects haven't moved to GPLv3.

Actually that's false. There is an Affero clause for GPLv2 which you would have noticed had you read my link.

The Affero General Public License, often abbreviated as Affero GPL and AGPL (and sometimes informally called the Affero license) refers to two distinct, though historically related, free software licenses:

* the Affero General Public License, version 1, published by Affero, Inc. in March 2002, and based closely on the GNU General Public License, version 2 (GPLv2),

Re:Repercusions for FOSS licenses

[rackserverdeals]

Actually that's false. There is an Affero clause for GPLv2 which you would have noticed had you read my link.

Oh yeah? Where?

Section 2(d) is not in GPLv2.

AGPLv1 is also incompatible with GPLv2 because of 2(d).

Maybe you should reread your link.

Re:Repercusions for FOSS licenses

[syousef]

But what license was the FICS code under? Was it really "open source"?

Directly from the source code:
"This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License"

It really PISSES ME OFF that uninformed crap comments like yours get rated insightful while my original gets overrated.

Re:Repercusions for FOSS licenses

[syousef]

FICS has been replaced by chessd

FICS was stable and usable. 10 years later chessd is in alpha. So it hasn't replaced a thing. Also, it doesn't look like it's based on the same code. So my original comment about seeing GPL code (and it was under the GPL that FICS was distributed) killed stands.

Re:Repercusions for FOSS licenses

[syousef]

It's not the same code base, and it's in alpha, you ignorant troll.

Re:Repercusions for FOSS licenses

[syousef]

Isn't this the whole point of a real open source license, rather than just publishing stuff and saying 'go play'? Either you're spouting misinformation, or the GPL et al are completely worthless...

Thanks for calling me a liar but the fact is I have the FICS code, which includes notices that it was distributed under the GPL and it is no longer publicly available on the web. So no I'm not spreading misinformation. No the licenses aren't worthless but the idea that you're secure once something is published as GPL has been proven wrong in some instances.

Re:Repercusions for FOSS licenses

[Anonymous Coward]

From the first paragraph of your link:

However, there are two exceptions in California. Non-compete agreements are enforceable for partnerships and when someone is selling their ownership interest in a company.

I'll give you some time to think about it.

Eh no. This raises no larger question

[GauteL]

.. about "who really owns a commercial open software application".

The copyright holders owns a commercial open software application. If all the copyright has been assigned to a corporation, then the corporation owns the copyright. This is fact.

Obviously, if the corporation has licensed other people to use and distribute it (i.e. with the GPL) then non-copyright holders may have some rights too.

This leads us to the only part of the GPL that I think is in any way legally questionable (IANAL). I'm not sure it is entirely legally clear if the copyright holder is allowed to revoke the GPL licensing terms or not, no matter what is said in the license. (i.e. They could argue that the license is not a binding contract).

In the opposite case of anyone breaking the GPL license by releasing binary only derivatives, the violator can not hide behind 'not a binding contract' because that would remove all of their rights to the software whatsoever (they have no rights other than what the GPL grants them).

If the copyright holder is allowed to revoke the license, they could close up any project that they own copyright to without allowing any forks. It would mean a loss of MySQL and OpenOffice.org as free software forever.

Re:Eh no. This raises no larger question

[Burkin]

This leads us to the only part of the GPL that I think is in any way legally questionable (IANAL). I'm not sure it is entirely legally clear if the copyright holder is allowed to revoke the GPL licensing terms or not, no matter what is said in the license.

No, they can't. The FSF has already stated that if the public has had the right to use the program under the GPL that it can't be revoked.

Can the developer of a program who distributed it under the GPL later license it to another party for exclusive use?
No, because the public already has the right to use the program under the GPL, and this right cannot be withdrawn.

http://www.fsf.org/licensing/licenses/gpl-faq.html#CanDeveloperThirdParty

Re:Eh no. This raises no larger question

[Nursie]

People have tried it, there have been a couple of cases, though I can't remember the names of the software off the top of my head. Generally goes like this:

write program,
releasing a few versions under GPL,
corporate interest/money arrive
I'm revoking the GPL, you all have to delete any copies of MY stuff you have

Now, I know that if a single entity holds the copyrights to the whole thing then they can release it under another license any time they like, and stop producing GPL'd versions, but I don't think they can tell their users that they no longer have rigths to stuff they've already got. It seems like that would go so far against common sense as to be ludicrous. But then we are taking about the law...

Re:Eh no. This raises no larger question

[russotto]

This leads us to the only part of the GPL that I think is in any way legally questionable (IANAL). I'm not sure it is entirely legally clear if the copyright holder is allowed to revoke the GPL licensing terms or not, no matter what is said in the license. (i.e. They could argue that the license is not a binding contract).

They could argue that, but even if that were to be found to be the case, for any program with significant distribution, I'd think the doctrine of "detrimental reliance" would apply.

Re:Eh no. This raises no larger question

[maxume]

The FSF doesn't actually make law (and I doubt it is a given that their intent would override the intent of some author that used the GPL).

I don't think it is particularly likely that anyone could successfully revoke the gpl on code that they had distributed, but you can't look to the FSF for that determination (only for their opinion).

Re:Eh no. This raises no larger question

[Burkin]

I don't think it is particularly likely that anyone could successfully revoke the gpl on code that they had distributed, but you can't look to the FSF for that determination (only for their opinion).

It's entirely unlikely because the license itself explicitly forbids it.

Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein.

Once the rights have been granted to copy, distribute and modify the program any attempt to revoke those rights is imposing further restrictions, which as the quoted section says is forbidden.

Re:Eh no. This raises no larger question

[squoozer]

IANAL either but I suspect as they are the copyright holder they can change the licence whenever they want to whatever they want. In fact I believe this happened with the X Server and is one of the things that spawned the X Org that we now use.

What I don't think that can though is change the licence retrospectively so you would just have to fork from the nightly before the licence change. If you could have retrospective licence changes how would you ever know if the software you were running was valid?

Re:Eh no. This raises no larger question

[mabhatter654]

they way MySQL and stuff from the GNU/FSF is set up, they require contributors to sign the work over to them. Then the body in charge has ALL the rights and can do what they will. Like when the FSF moved everything to GPL 3, they could do that unilaterally because they had assignment.

Mozilla also has assignment and releases just what you say under a tri-license, the same code base published 3 times. Two are open source (MIT & LGPL) but the main Firefox branded binary is actually NOT open source. Anybody can fork the MIT or LGPL versions but has to strip all branding and can't call it "firefox" or "mozilla".

They can't UNLICENSE things already in the wild though. But much like Red Hat/CentOS, they could beat you up over every little point of branding (because they own the name) and keep suing you for every little code comment if they were that petty, leaving mountains of work for somebody to get "every reference" to the old name/logos out before distribution. Of course a fork is only useful if enough people follow you, and that's where nearly all the projects break down.... only the new parent company is big enough to provide new features and timely support.

On the other hand Linux is pure GPL 2. Because Linus has no "foundation" when he wrote it, contributions are still owned by the individual coders... moving off of GPL 2 is nearly impossible because many early contributors no longer work on Linux or are deceased. The copyright sticks, so the only way to change the license would be to rewrite the modules entirely.

Re:Eh no. This raises no larger question

[Anonymous Coward]

for a specific version.

if i release v 1.2 under the GPL, that code is always GPLed.

now, as copyright holder i can then release version 1.3 under BSD or GPL3 or even make it closed source, because I own the copyright. The GPL CANNOT take that from me.

so i can make version 1.3 closed source. BUT version 1.2 will continue to be GPL. The code has been released as GPL.

also, you can dual or tri- or any type of liscense code you want. again if you are the copyright holder. if it is your code, you can do what you want with it.

Re:Eh no. This raises no larger question

[Burkin]

Of course you can relicense future versions under a new license and I never said otherwise (in fact in another post in this article I said just this). The person I responded to was talking about someone revoking the GPL for previous versions of the code which is not possible.

Re:Eh no. This raises no larger question

[Albanach]

If the copyright holder is allowed to revoke the license, they could close up any project that they own copyright to without allowing any forks. It would mean a loss of MySQL and OpenOffice.org as free software forever.

This is taken care of by section 6 of the GPL v2 (though it appears as section 7 in the MySQL documentation for version 5.0 at least.

• Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.

If I give you a copy of MySQL that I download today, you automatically get a license to modify and distribute from the copyright holder. Any copies you distribute will likewise have a perpetual license as long as you and the recipient obey the requirements of the GPL V2.

That's not something Sun or Oracle can take away from you. They can stop releasing new versions under the GPL as they own the code (anyone submitting patches must agree to the Sun Contributer Agreement). They cannot, however, unGPL the code that has already been released.

Re:Eh no. This raises no larger question

[maxume]

I'm not trying to deny the intent of the license, I'm pointing out that up until there is case law regarding specific clauses, there is some small possibility that a clause might not matter (the various GNU licenses are carefully written to avoid this, but that doesn't make them bulletproof, it just makes them reasonable to use).

Re:Eh no. This raises no larger question

[GauteL]

You're not really answering my post. The GPL is clear enough on this matter, but it isn't entirely legally clear if the GPL is to be considered a binding contract or not. Thus, my argument can't really be answered by referring to the GPL. The FSF has their opinion on the matter, but the FSF does not make laws.

This is not a problem at all in the opposite case where a recipient of the code breaks the license, because without the license, the recipient has no rights to the code.

However, without the license, the copyright holder has all the rights exclusively to the code.

I'm not saying there is a big chance of the copyright holder being legally able to revoke the licensing on GPL software, but I do think it is worth considering as a remote possibility. In any case I don't think it is a good idea to agree to reassign copyright on your contributions to SUN/Oracle.

Re:Eh no. This raises no larger question

[Estanislao Mart�nez]

This leads us to the only part of the GPL that I think is in any way legally questionable (IANAL). I'm not sure it is entirely legally clear if the copyright holder is allowed to revoke the GPL licensing terms or not, no matter what is said in the license. (i.e. They could argue that the license is not a binding contract).

From GPLv3: "All rights granted under this License are granted for the term of copyright on the Program, and are irrevocable provided the stated conditions are met. " GPLv2 doesn't have this text, though.

IANAL, but my understanding is that a copyright license is not a contract in the first place, so the example argument you cite just can't work. A copyright license is the permission (hence the word "license") granted by the copyright holder to another party, allowing that party to make copies of the work in question. Granting a copyright license to somebody doesn't create any obligations for either party; it just expands the receiver's privileges.

Re:Eh no. This raises no larger question

[Albanach]

You're not really answering my post. The GPL is clear enough on this matter, but it isn't entirely legally clear if the GPL is to be considered a binding contract or not.

I'm not sure what you're looking for then? Do you only trust every contract you enter into after it has been tested in court?

Either the GPL is binding, or you have no right to be using the software now, let alone after Oracle take over. If the GPL isn't binding we might as well all pack up and go home.

Certainly the GPL has never been found to be unenforceable - surely that carries some weight?

Re:Eh no. This raises no larger question

[arth1]

The copyright holders owns a commercial open software application.

The copyright holder is usually, but far from always the owner. The two can be separate.

Re:Eh no. This raises no larger question

[DragonWriter]

This leads us to the only part of the GPL that I think is in any way legally questionable (IANAL). I'm not sure it is entirely legally clear if the copyright holder is allowed to revoke the GPL licensing terms or not, no matter what is said in the license.

I think its pretty well settled case law, in the US at least, that a gratuitous license is revocable at will regardless of what the license says (of course, if you are also a GPL licensee, revoking the downstream license of your derivative of a GPL-licensed work breaches the license you have, with all the usual consequences of such a breach, but if you require copyright assignment for all code that you include, then that's not an issue.)

If revoked, the consequences of the revocation might be limited to some degree, to prevent injustice, on a promissory estoppel basis, but that's a case-by-case, minimum-to-avoid-injustice equitable remedy, not a clear-cut, absolute legal right.

Re:Eh no. This raises no larger question

[DragonWriter]

The FSF has already stated that if the public has had the right to use the program under the GPL that it can't be revoked.

The FSF can state whatever it wants, but its not going to change the law.
The reason that the GPL may be revocable, at least in the US, is that, as a matter of law, copyright licenses that do not meet the legal definition of contracts (among other things, requiring mutual consideration) are revocable at will by the licensor. Statements by the FSF, even where they are the copyright holder (which is often not the case on GPL software), have no bearing on that.

The FSF's statements may have bearing on whether or not promissory estoppel is invoked in particular cases by courts to limit the power of a copyright holder to seek a remedy for copyright violation after the GPL has been revoked, since they might have some bearing on what reasonable expectation the licensee had. But promissory estoppel is unlikely to be a complete and eternal bar to any action after a revocation; it is most likely to prevent things like the copyright holder revoking the license and then taking immediate action seeking damages for copyright infringement against someone distributing software consistent with the now-revoked license.

Re:Eh no. This raises no larger question

[butlerm]

The problem is that "gratuitous license" is a somewhat ambiguous term. If you have a citation for a case where a copyright license was held to be revocable contrary to its own text, please post.

Re:Eh no. This raises no larger question

[DigitalWallaby]

Once the rights have been granted to copy, distribute and modify the program any attempt to revoke those rights is imposing further restrictions, which as the quoted section says is forbidden.

Except that the licence is a contract from the copyright holder to the licencee. The copyright holder has the right to change the terms of the licence, or add new licences at any time. It may be possible for Oracle, as the new owner of the MySQL copyright, to revoke the GPL licence.

And who's going to stop them?

Since the FSF doesn't own the copyright, it might not be granted the right to sue Oracle for a breach of the licence. Besides who'd want to take a behemoth like Oracle on in court anyway. It would be like Lionel Hutz standing up against Burn's cadre of high priced on-retainer attorneys. At the very least it would drag on for years while Oracle systematically destroys MySQL. Even if the FSF won, there'd be nothing left.

Re:Eh no. This raises no larger question

[Anonymous Coward]

leaving mountains of work for somebody to get "every reference" to the old name/logos out before distribution.

find . -type f -exec grep -Eqi 'yourbrand|theirbrand' {} \; -print0 | xargs -0 perl -p -i -e 's/yourbrand|theirbrand/mybrand/gix'

That was tough.

You'd have to replace some graphics and filenames as well but it's not "mountains of work".

Re:Eh no. This raises no larger question

[DragonWriter]

The problem is that "gratuitous license" is a somewhat ambiguous term.

Actually, its not. Its a very clear term: its a license which is not given in exchange for consideration of value.

If you have a citation for a case where a copyright license was held to be revocable contrary to its own text, please post.

If you have a citation for a case, whether regarding copyright or not, where a license that was not for value was considered irrevocable because of its text (or, for that matter, for any other reason), contrary to the general law pertaining to gratuitous licenses, please post.

Re:Eh no. This raises no larger question

[GauteL]

"I'm not sure what you're looking for then? Do you only trust every contract you enter into after it has been tested in court?"

Certainly not. But in the case of a software license like the GPL, there is no explicit agreement between anyone. Nobody had to write their signature anywhere or even verbally say "yeah, I agree to these terms".

"Either the GPL is binding, or you have no right to be using the software now, let alone after Oracle take over. If the GPL isn't binding we might as well all pack up and go home."

You are again not reading my post before posting a reply. The copyright holder has given us a license, but isn't clear that the copyright holder has to abide by the terms in the license saying it is 'irrevocable'. Another poster talked about a 'gratuitous licenses' and I suggest you read that.

"Certainly the GPL has never been found to be unenforceable - surely that carries some weight?"

AFAIK no company has ever tried to revoke someone's GPL rights and very few companies are even in a position to attempt to do it. I.e. first of all, you will have to be the sole copyright holder, with nobody else having any claim to the code. Otherwise you'd be violating someone else's copyright by revoking the GPL.

SUN/Oracle is perhaps the only major player with significant Open Source projects where they own all the copyright due to having required copyright assignment for any patches.

Even so, I still think this is a fairly remote possibility, but I'd be wary of assigning your copyright to someone else when contributing.

Cases Like OpenOffice.org

[eldavojohn]

This raises the larger question of who really owns a commercial open software application: the corporate copyright holders, or the community?

No one. Or, perhaps, everyone. That's kind of the point, isn't it? It isn't locked into anyone's individual grip.

"Open source" is just too broad a term to address this way. You would have to look at individual licenses. On top of that, you have things like Open Office, which is "open source" but clearly controlled by Sun (or Oracle now I guess).

While you claim you can always fork an open source project, it's not always that simple. Especially in massive open source efforts (like Linux) where they have contacts and knowledge that are vital to the project. It isn't possession or control or fiscal ownership but instead a name you've made for yourself as the Father of some project that gives you "ownership" or "rights." And usually the market share of your user base reflects that.

You'd be surprised how many of your open source solutions are actually controlled and operated by a single entity. And this is great for those products because the entity is usually donating a lot of time and money to it. Should the entity ever drop out, that's when someone can pick up the cross and take it a new direction with everyone helping.

Re:Cases Like OpenOffice.org

[Anonymous Coward]

It isn't locked into anyone's individual grip.

"Open source" is just too broad a term to address this way.

There's a difference between "locked" and "dominated" by.

We'll Find Out For Sure...

[blcamp]

...after the coming litigation.

Licenses

[tsotha]

To ensure MySQL's survival, he wants to fork from the official version -- using his company Monty Program Ab to create what he calls a MySQL "Fedora" project. This raises the larger question of who really owns a commercial open software application: the corporate copyright holders, or the community?"

That's what all the lawyering over the license text is all about. This question is one of the more settled questions in the industry.

Get it here

[A beautiful mind]

Get the improved code here.

Re:Get it here

[AlterRNow]

The world's most advanced open source database

Shame they couldn't spend a little time and allow the website to fill the width of my browser. IMO: Fixed width site = lazy design = unimpressive.

Re:Get it here

[AbbyNormal]

I'm not too certain of that. I use the replication built into MySQL (with minimal glue) and I've found it to be extremely easy to setup with a few stock Ubuntu server images. Just doing a brief google search reveals a lot of postgres replication products (Slony/PGCluster/DBBalancer), but they did not seem to have any active development on them in a while. I am not knocking PostGres by any means as I find the different language options for Procedures/Functions and the admin tools to be a huge boon.

Re:Get it here

[carambola5]

Sometimes, a fixed width is more effective than full width.
Sometimes, a fixed width is cleaner than full width.
Sometimes, a fixed width is more graceful than full width.

There are very few hard-and-fast rules in web design. Always designing to full width is not one of them.

Shame you couldn't spell out "In my opinion." Abbreviations = lazy typing = unimpressive.

Re:Get it here

[TheRaven64]

Fills the full width of my browser, but then I actually know a bit about typography and set my browser width to a comfortable maximum line length for variable-width sites, rather than letting lines become very long and making them difficult to read.

That said, it doesn't reflow nicely if I reduce the width of the window to something that a typical mobile device might support. It fails, but for the opposite of the reason you suggest.

Re:Get it here

[drinkypoo]

I know a bit about typography as well, but I've been using computers for so long that line length doesn't make much difference to me. NoSquint made a much bigger difference in readability. Also I keep my all-in-one-sidebar continually visible, which narrows the view a bit. Many webpages are unusable if you don't widen the screen quite a bit.

Re:Get it here

[Anonymous Coward]

MySQL replication is easy to set up, and very very hard to get right.

At the risk of starting a flame war, it seems to me that the PostgreSQL project cares very deeply about data integrity, whereas the MySQL project thinks data integrity is a more or less worthwhile goal as long as it doesn't get in the way of any of their neat features.

Re:Get it here

[AlterRNow]

I've never seen a fixed width site that had any benefit from being fixed width. And besides being easier for the developers, I can't think of one.
Can you provide some examples?

Re:Get it here

[jjohnson]

Most of the designers I know prefer, these days, to create fixed width designs because most sites look like crap when the browser is maximized on a monitor with 1900x1280 resolution. The line lengths are far too long.

Re:Get it here

[Lord Ender]

Your web design philosophy died in the previous decade. It is damn hard to read extremely long lines. Good web design should set a fixed (perhaps using em) width for text lines. Try reading slashdot on a 26 inch display and see if you think having arbitrarily long lines is a feature or not (it is not).

What you want was a good idea when we were scraping together money to upgrade to 15 inch CRTs. That time is long dead.

Re:Get it here

[rackserverdeals]

Skype is one, if not the biggest users of PostgreSQL and has released some tools that they use to manage their PostgreSQL cluster that are probably worth looking into.

With postgresql, you also have the option of scaling vertically instead of just horizontally. It seems that Postgresql scales better than MySQL across multiple cpus/cores and handles heavy load better.

Another interesting benchmark shows that both Postgresql and MySQL handle load better on Solaris instead of Linux.

Re:Get it here

[raynet]

Webpage containing a very long table might be more readable in fixed width format.

Re:Get it here

[Anonymous Coward]

I'll admit to being a Postgresql biggot, but still, those performance notes are 2.5 years old.

Re:Get it here

[Glendale2x]

I see that as a core problem with postgres: too many replication tools that are potentially incompatible. The MySQL slaves I set up years ago have been going along just fine - even though upgrades - because it's part of the core. If I were to switch to postgres I'd have to bolt on some random tool to get basic functions back. What if the one I chose stopped updating out of nowhere and I'm stuck? It kills the postgres benefit of supposedly being more reliable than MySQL unless you're only using one database server. Does anyone know why the postgres team decided to omit replication?

Of course, at the end of the day, the customer doesn't care or even know what's going on behind the scenes as long as it keeps working. Just like how people sit on slashdot and claim MySQL could never operate something as big as that.

Re:Get it here

[tepples]

Fixed width site = lazy design = unimpressive.

Newspapers use five or six fixed-width columns instead of one page-wide column for a reason: text a column 20em to 40em wide is just more readable than text in a column 80em wide.

Re:Get it here

[Just Some Guy]

Most of the designers I know prefer, these days, to create fixed width designs because most sites look like crap when the browser is maximized on a monitor with 1900x1280 resolution. The line lengths are far too long.

...in their not-so-humble opinion. Screw 'em. Fixed widths look like crap when I'm standing a few feet away from my monitor with the font sized increased, unless
you think this
looks better than
letting the user
decide how long
theirs lines
should be.

Re:Get it here

[jjohnson]

For you, increasing font size so you can stand several feet away may be the best solution. But it's not for almost everyone else, and designers are all about designing for the largest number.

Besides, both Firefox and IE zoom the page, increasing layout as well as font size, rather than just jacking the font size. Doesn't that work for you?

Re:Get it here

[Just Some Guy]

It's tolerable, but I just hate the idea of "designers" trying to force the web to act exactly like print media. I can't stand fixed width pages because I don't want to have to find an acceptable zoom level, and I don't like when a webmonkey assumes they know more about what looks good on my screen than I do.

Yes, long lines are ugly. That's why I don't typically have browser windows maximized. Still, I want the text to flow to my preferences, which will always look better to me than what a fixed layout can manage.

Re:Get it here

[Anonymous Coward]

What, use a database that doesn't even support unsigned integer types? Or that *just* added support for enums? Or that doesn't have built-in replication capabilities (you expect me to choose between 3rd party replication suites?!).

Why would I switch to something that doesn't have ON DUPLICATE KEY UPDATE? Expecting me to write routines for something so simple is too much. The simple fact is that posgresql is still too immature, and they try to stay too pure to the SQL standards and thus don't provide enough useful extensions that make life working with a database simple. The fact is that currently, MyISAM is just easy and makes day-to-day operations a breeze.

Re:Get it here

[zerocool^]

Ok, that article on scaling is
1.) 3 years old
2.) Comparing some random postgres alpha to a mysql 5.1 beta
3.) Actually not comparing them at all, since there's all these pretty graphs of Postgres performance, and then it's like "lol this is better than mysql".

How about using a stable release? How about doing comparative testing?

As someone who's recently upgraded some mysql write masters to 8 core / 16 GB ram / 8x 300gb 2.5" SAS in Raid10 boxes, it scales vertically just fine.

Re:Get it here

[AlterRNow]

Apples and oranges in a way, I can't resize the "window" of a newspaper. Also, all the space on it is used whereas fixed width sites tend to waste a lot of space!
On the other hand, the line length on newspapers is ridiculously stupid at times. 5 words per line? Jeez.

Fixed width also means I can't balance the seeking time involved in finding the start of the next line and the read speed of a continuous line of text to my own personal taste..

Re:Get it here

[james_shoemaker]

What do you think about sites designed for a 1280 width, or even sillier, 1600 width. Several of the corporate sites where I work only work without horizontal scrolling if you have the browser maximized on a 1600 wide monitor. Unsurprisingly the original designer happens to use just that size screen. That's why fixed width is bad, it allows the designer to not think about how things look on any monitor other than their own.

Re:Should Read: Sun announces last MySQL products

[tolan-b]

I have to disagree, I think that would be immensely stupid of them. I think they'll just use it to try to funnel users butting up against its limits towards full Oracle. If they kill it they lose that potential sales channel.

I think it's time to switch...

[Khan]

...to PostgreSQL. Seriously, I already use it for GpsDrive. Now I just need to convince the Cacti devs to switch over.

Re:I think it's time to switch...

[shutdown -p now]

If there are multiple MySQL forks around, such a switch will follow for most projects that use MySQL exclusively today. The main reason why it was so popular was the install base on cheap hosters (where it usually goes hand in hand with PHP) - if you used MySQL as a backend, you targeted the largest market for a particular price range. But forks will inevitably become incompatible eventually, segmenting that market, and at that point both hosters and developers might as well move on to something better, like Postgres.

I'm so going to get flamed...

[squoozer]

As the owner of a software development company I think your would have to be stark raving nuts to open source your main product. It's not that the model can't work it just that if it becomes successful you are pretty much guaranteed to lose control of it at some point.

If we look at MySQL for example: here's a company that produces half way decent database engine that that make open source. They play the open source game "properly" producing code that a mortal can compile to get a working database. While the company is giving the community what they want everything is hunky dory and there is peace.

Enter Sun who buy MySQL and suddenly the community isn't happy and it's fork fork fork. Only one of those forks needs to be any good and all of a sudden Suns not bought very much at all. If a company plays nice with the open source community forks are fairly easy but rare. The problem is they hang like a knife (or maybe that should be fork) over the company and if they are unfortunate enough to annoy the community they could eaisly lose control of their product.

That said I think there are situations where companies can participate in open source. The Linux kernel and Plone being a couple of good examples. Both of those projects are structured very differently to the MySQL situation though as no one company is trying to make a living off the code.

Re:I'm so going to get flamed...

[Anonymous Coward]

If Sun was a good steward of MySQL then we would not be having this discussion. Apparently Sun has done something to piss of the dev community. I'm not sure what they did, but this is one of the reasons the GPL exists.

Re:I'm so going to get flamed...

[Abcd1234]

If we look at MySQL for example: here's a company that produces half way decent database engine that that make open source. They play the open source game "properly" producing code that a mortal can compile to get a working database. While the company is giving the community what they want everything is hunky dory and there is peace.

Enter Sun who buy MySQL and suddenly the community isn't happy and it's fork fork fork.

Congratulations! With your very example you actually managed to disprove your original assertion. See, your original claim was this:

"As the owner of a software development company I think your would have to be stark raving nuts to open source your main product."

But, the very first paragraph in that quoted text demonstrates that isn't actually the case. The community was very happy with and supportive of MySQL corporate.

The problem, as you pointed out, was the purchase by Sun. In that case, the customers didn't feel Sun would necessarily have their interests at heart, and so there was dissatisfaction. This is only increased by the fact that Sun has now been purchased by Oracle, a company that actually markets a product in the same general space (I would argue they aren't actually in the same market, and so MySQL has little to fear, but... people aren't exactly rational).

So the key to running a company on an open codebase seems simple: keep your customers happy, and don't give them the impression that their interested are being threatened. But, of course, that's a good general rule to follow regardless of the license your code falls under. The only real difference between open and closed source, in this case, is that if the source is closed, you may have achieved vendor lock-in, which gives you more freedom to buttfuck your customers, as they won't have a clear avenue for recourse... but if that's your strategy, well, frankly, fuck you.

Re:I'm so going to get flamed...

[tuffy]

By making MySQL open source, the founders benefit because they received a lot of money from Sun and don't lose their product's source code, even after leaving the company.

Sun, by buying the company, not only gain the product's source code, but also the MySQL name. And the name is the important thing to them, since Sun could've made their own fork from the source if they simply wanted a database.

So it seems like everyone got exactly what they wanted out of the deal.

Re:I'm so going to get flamed...

[windsurfer619]

I think you're forgetting the part where Sun can then get back the code from the fork. And Sun still has just as much control of their own version, as they have just as many devs working on it.

I think what you really mean is this: If you are open sourcing your main product in hopes of sacrificing direct purchase revenue for free developer time, you run the risk of alienating your community and losing developer time if they decide you aren't doing a good job.

Re:I'm so going to get flamed...

[Vellmont]

It's not that the model can't work it just that if it becomes successful you are pretty much guaranteed to lose control of it at some point.

The mistake you make is you keep trying to think of the model as "owning the software" rather than "employing smart people that make your product worth buying, and keeping those smart people happy". If you don't do both of those things, you fail.

The problem is they hang like a knife (or maybe that should be fork) over the company and if they are unfortunate enough to annoy the community they could eaisly lose control of their product.

Here's what I see from 10,000 feet after only knowing a little about MySQL and its history. (I don't use it, and if I want a free DB I use Postgresql).

I don't think they annoyed the community as you put it. I think what happened is the developer weren't kept happy, so they left. The project got derailed at that point and the product suffered. Closed source products have the same issues. They normally deal with this issue by trying to lock in users with proprietary interfaces, spreading FUD, etc. Basically stall until you can find people again to improve your product. That doesn't work as well with OSS.

As far as being crazy to open source your product.. well, it depends. If you're the top one or two guys in your market and aren't threatened by a Big Player.. you probably would be crazy to open source it. On the other hand if you're a small fish, or a Big Player is going to stomp you, it can make a lot of sense. In the end, it's not different than any other business decision. YMMV.

Re:I'm so going to get flamed...

[AbbyNormal]

But with the knife hanging over you, wouldn't you have more incentive to keep the imagination/heart of the community alive? It may be more difficult than simply answering to a board of directors/stockholders, but in the end you end up with a better quality product.

  Sun/Oracle/MySQL did not lose control of their product by any means, as they still own the name and brand. Most corporate customers most likely will stay with the main branch of the code and the supporting company rather than go with the "forked" branch. If one of the open source forks produces something that was missing in the original branch, then it will (hopefully) be merged back into the main branch.

  I agree there are risks to open sourcing your main product, but if you provide the better support and code base and keep an organized/energetic community, you'll garner a larger share of paying corporate customers.

Re:I'm so going to get flamed...

[kv9]

The Linux kernel and Plone being a couple of good examples. Both of those projects are structured very differently to the MySQL situation though as no one company is trying to make a living off the code.

RedHat and SUSE might disagree with you on that one.

Re:I'm so going to get flamed...

[zx75]

Why do you have to be nuts to open source it?

If you are planning to make most of your money by selling the software directly, then yes I agree that you would be nuts to open source it. But that is not how the game is played. You play the game by open sourcing your software in order to build recognition and acceptance, and you make money by selling support contracts to businesses who need up-time guarantees and someone to call when things go wrong.

Once you have business acceptance and money coming in from contracts you can mine the community for software improvements (as well as creating your own). Then if something happens (like a buy-out) and the community goes on a fork-frenzy, WHO CARES? This is NOT a bad situation to be in because this means that the community is fairly large and committed to your database. In fact this may simply increase your exposure and bring in even more customers, because remember that your customers are NOT the people who USE your software, they are the businesses that pay for SUPPORT.

That is how the game is played... or at least one way that it can be played.

Re:I'm so going to get flamed...

[Bob9113]

The problem is they hang like a knife (or maybe that should be fork) over the company and if they are unfortunate enough to annoy the community they could eaisly lose control of their product.

And that is one of the major reasons why professional software engineers love commercial Open Source software. The company is on no uncertain terms with the customer: Keep us happy and we keep giving you money for support contracts. Turn into a dick and we walk. It is a vastly healthier relationship for mission critical business information systems.

Before I go on, let me note that what I am about to say does not consider redistribution, which is its own thorny problem and is intrinsically linked to copyright.

On the matter of continued use and improvement, a critical matter to information systems, the fiat monopoly of copyright is extremely dangerous to the customer. If a commercial proprietary software vendor changes the terms of the bargain, the options are to use the existing software as it is, to migrate your information infrastructure, or to accept the new terms. The first, using the software as it stands, is usually impractical in this rapidly evolving industry. The second, migration, can be enormously expensive. The final option, paying the Dane Gelt, is often the least objectionable option at the time.

And so it is that while redistribution is an issue which copyright may handle well, continued use and improvement falls heavily in favor of Open Source or non-copyright. Because continued use and improvement is so critical to mission critical information systems, it is in our industry that the onus of copyright has become most visible, and in our industry that Open Source first became a significant market option.

Re:I'm so going to get flamed...

[Darth]

If we look at MySQL for example

MySQL would be a bad example with which to try to assert your point. MySQL AB did not lose control of MySQL because they open sourced their main product. They were a very profitable company that grew large enough to be sought after for acquisition by major international corporations because they open sourced their main product.

They lost control of it because they sold control to Sun. Sun is losing control not because people are unhappy that Sun owns MySQL now, but because Sun is not providing their customers with what they want.

Part of the problem is that MySQL is not the main product of Sun so it gets less attention and resources than its customers require.

So when it was the main product of the controlling company, it was profitable and successful. Now that it is just one element of a portfolio of products of a company, it is languishing and its customer base is talking about forking it. That doesn't support your premise at all.

Open source removes the ability for the company to use lock-in to control their customers. It forces the company to provide what their customers want and manage the perception that their involvement is valuable.

This is not intended as a flame. I just disagree with your assertion in general and your example specifically and am trying to articulate intelligently why i think you are incorrect.

Re:I'm so going to get flamed...

[Archangel Michael]

Enter Sun who buy MySQL and suddenly the community isn't happy and it's fork fork fork.

[voice = Swedish Chef]Skooer zee muusheee Fork Fork Fork [/ voice]

Re:I'm so going to get flamed...

[IcyHando'Death]

/*
  * Flame on
  */
You've got to be kidding me!

You would never have even HEARD of MySQL if it had not been an open source product. You think the closed source world had room for yet another database? I'd like to point out that even with just Oracle and IBM in there, that space was pretty crowded, and there were plenty of small players too, going for the crumbs.

MySQL, when it started gaining mind share and usership, was a buggy, feature poor product with nothing going for it EXCEPT its open source status and developer community. If it were closed source, you would never have heard of Michael Widenius nor MySQL. He would either have abandoned his project long ago or still be toiling in obscurity on this little hobby DB in his spare time while he paid the bills with a real job.

Open sourcing a product does mean losing some degree of control, but for all that, it's an excellent way to launch a company into the lime light. As we've seen over and over, there are plenty of ways to leverage that into wealth, with control or without.

Re:I'm so going to get flamed...

[owlstead]

Yes, they netted only 1 billion in total after becoming popular and selling off to Sun. Gosh, I wish I could make such a bad choice once in a while. And now they can do very interesting things with the money they've made. With 1 billion you can control an awful lot of applications.

Re:I'm so going to get flamed...

[Lord Ender]

If it weren't opensource, nobody ever would have heard of MySQL. It's creators made real money when they sold it. Sounds like they got things right and you got things wrong.

Re:I'm so going to get flamed...

[mkcmkc]

...suddenly the community isn't happy and it's fork fork fork

Okay, but unhappy customers are going to leave anyway. Is this really any worse than having them flee to your commercial competitor?

Also, with this "downside", there is a balancing upside, which is that having your source code be Open Source is a substantial benefit for your customers. When I'm choosing vendors, if there's a serious competitor offering an Open Source product, that all but rules out the closed alternatives.

Re:I'm so going to get flamed...

[_Sprocket_]

Enter Sun who buy MySQL and suddenly the community isn't happy and it's fork fork fork. Only one of those forks needs to be any good and all of a sudden Suns not bought very much at all. If a company plays nice with the open source community forks are fairly easy but rare. The problem is they hang like a knife (or maybe that should be fork) over the company and if they are unfortunate enough to annoy the community they could eaisly lose control of their product.

My current employer has heavily invested their infrastructure in Sun and Oracle. However, Oracle licensing represents an inordinately large cost to our budget. And frankly, there are a lot of projects that just don't need to use Oracle so the value of those licenses are getting a lot of scrutiny. There are discussions about moving some of those projects to MySQL which would do a lot to relieve our budget. And a part of those discussions involve taking the money saved on Oracle licensing and purchasing MySQL support from Sun. We know Sun. We have existing support contracts with them. And we're pretty happy with that arrangement. Adding in MySQL support is almost a given.

That's what Sun gets. As long as Sun doesn't abandon MySQL or otherwise break it, it is unlikely our folks will consider going with another fork and purchasing support from an unknown entity. Right now, Sun is considered the go-to folks for MySQL. And they're Sun. We're all but rushing to give them money for this.

It should also be noted that having options other than Sun for MySQL has it's own attraction. Today we're keen to get Sun support. But we know that tomorrow we might decide that Sun's MySQL support isn't a good value once we've tried it. Other options means that investing in Sun and MySQL today isn't a dead end; it isn't more of the same we get with Oracle. So we're much more open to making that investment today. And that means Sun gets sales even if it means they risk those sales tomorrow.

Re:I'm so going to get flamed...

[Anonymous Coward]

If we look at MySQL for example: here's a company that produces half way decent database engine that that make open source. They play the open source game "properly" producing code that a mortal can compile to get a working database. While the company is giving the community what they want everything is hunky dory and there is peace.

Enter Sun who buy MySQL and suddenly the community isn't happy and it's fork fork fork.

Congratulations! With your very example you actually managed to disprove your original assertion. See, your original claim was this:

"As the owner of a software development company I think your would have to be stark raving nuts to open source your main product."

But, the very first paragraph in that quoted text demonstrates that isn't actually the case. The community was very happy with and supportive of MySQL corporate.

The problem, as you pointed out, was the purchase by Sun. In that case, the customers didn't feel Sun would necessarily have their interests at heart, and so there was dissatisfaction. This is only increased by the fact that Sun has now been purchased by Oracle, a company that actually markets a product in the same general space (I would argue they aren't actually in the same market, and so MySQL has little to fear, but... people aren't exactly rational).

You miss the point -- most small software companies are hoping to be purchased at some point so the owners can cash out and retire. This clearly shows that a company with an open source product is a risky purchase, which means you won't get as much money for it.

Re:I'm so going to get flamed...

[OdinOdin_]

That depends if your main/only product is your life. Some people would be happy to take the money on offer at the time knowing that those who now owned the product would not be able to do as good a job as you with it, knowing that because you provided it as open source you could always take up the reigns again under a different name. You'd also have the insider knowledge about all the things you learned on your original journey to taking that big bankers cheque.

So it really is win win, sure you lost control of your brand name but you took a big chunk of cash to compensate. Sure you need to start all over again (building a company, employees, strategy, etc...) but wasn't that the original fun in the first place.

Those that matter in the community are wise enough to understand that if you start over and your product is good and starts to push the areas that matter again they will switch.

Re:I'm so going to get flamed...

[domatic]

Enter Sun who buy MySQL and suddenly the community isn't happy and it's fork fork fork. Only one of those forks needs to be any good and all of a sudden Suns not bought very much at all. If a company plays nice with the open source community forks are fairly easy but rare. The problem is they hang like a knife (or maybe that should be fork) over the company and if they are unfortunate enough to annoy the community they could eaisly lose control of their product.

From the point of view of a user, that isn't a bad thing. It is an assurance of good behavior. If you know something you are going to do will seriously annoy your customers then you should think three times at least about it. And it wasn't just that Sun bought MySQL and it was "fork, fork, fork". It was Sun bought MySQL and the core devs jumped ship. If the code was all that important than there should have been concern for those devs as they are the ones who know it best. And many projects that have "forked" have in reality had the core devs pushed out for one reason or another and the community simply follows the devs.

It seems to me that you want to relegate goodwill and trust to low or no importance. For users of FOSS code, it is almost everything.

Re:I'm so going to get flamed...

[Abcd1234]

You miss the point -- most small software companies are hoping to be purchased at some point so the owners can cash out and retire. This clearly shows that a company with an open source product is a risky purchase, which means you won't get as much money for it.

Agreed, there is an increased risk, there. But that just means the purchasing company needs to be careful to reassure the community that they are going to continue to support and develop the product.

I mean, most people don't *want* to fork. They'll only do it if they feel they're backed up against a wall. So as long as the purchasing company is reasonable, I really don't think the risk is that great.

Re:I'm so going to get flamed...

[Elwood P Dowd]

MySQL's only feature bullet point ever was that it could be had free as in speech and beer. If they hadn't open sourced it, they wouldn't have had a product.

If you've got a phenomenally successful software product that people are lining up to buy despite it being closed-source and expensive... well shit, I wouldn't open source that either. I hope your competition continues sucking.

Re:I'm so going to get flamed...

[Thantik]

Clearly RedHat and SUSE are two companies.
I know...But I HAD to.

Re:I'm so going to get flamed...

[El_Oscuro]

Or IBM or even.... Oracle

Re:I'm so going to get flamed...

[Anonymous Coward]

*yawn* this roughly translates to "I want you to give me something for free and if you don't then fuck you". Grow up, get out of your parents basement and hopefully one day you will learn the world does not owe you a living.

Re:I'm so going to get flamed...

[williamhb]

Congratulations! With your very example [the original poster] actually managed to disprove [his] original assertion. See, [his] original claim was this:

"As the owner of a software development company I think your would have to be stark raving nuts to open source your main product."

But, the very first paragraph in that quoted text demonstrates that isn't actually the case. The community was very happy with and supportive of MySQL corporate.

The problem, as [he] pointed out, was the purchase by Sun. In that case, the customers didn't feel Sun would necessarily have their interests at heart, and so there was dissatisfaction. This is only increased by the fact that Sun has now been purchased by Oracle, a company that actually markets a product in the same general space (I would argue they aren't actually in the same market, and so MySQL has little to fear, but... people aren't exactly rational).

Unfortunately, in a business sense that's not good enough. If the community is blissfully happy with SuperGPLSoftwareStartup but nervous enough to fork the product if they get bought out, then SuperGPLSoftwareStartup is screwed. Young companies have venture capital funders, who are interested in an exit strategy -- how they will get their money back. That isn't "earning profits over twenty years" but an exit through a sale to a larger company or (more rarely) through an IPO. Both of those involve changes in ownership (and likely management) of the company, raising those "will they still have our interests at heart" questions.

It's quite possible that what we are seeing is the death of the GPL/commercial dual license business model. If Sun do not get any value from their purchase (but the community quickly forks MySQL) then the lesson will be quickly learned (not just by Sun) that buying GPL software companies is probably not worthwhile. Very quickly, VC firms would then realise that their main exit strategy from GPL software firms probably won't ever work again. That would translate into a rule of thumb "if your business plan contains the letters GPL, it will instantly go in the bin and never be funded".

Re:I'm so going to get flamed...

[roman_mir]

It's quite possible that what we are seeing is the death of the GPL/commercial dual license business model.

- not until Netcraft confirms it we won't.

Re:I'm so going to get flamed...

[squoozer]

Both RedHat and Suse produce something that is hard for the community to copy, packaing up god knows how many pieces of software and managing them takes time and effort which is what RedHat, for example, sell along with a bunch of enterprise tools. It's hard to equate the main RedHat offering with MySQL I think but I take your point.

Re:I'm so going to get flamed...

[Anonymous Coward]

Sounds like the protestant reformation to me. And Christians have been forking ever since.

Re:I'm so going to get flamed...

[Anonymous Coward]

Agreed, there is an increased risk, there. But that just means the purchasing company needs to be careful to reassure the community that they are going to continue to support and develop the product.

Bzzzt, more risk == less money, that's all she wrote, that's the GP's point.

Re:Should Read: Sun announces last MySQL products

[RogerWilco]

I have to disagree, I think that would be immensely stupid of them. I think they'll just use it to try to funnel users butting up against its limits towards full Oracle. If they kill it they lose that potential sales channel.

But that would give them an incentive to stop improving MySQL, as closing the gap with Oracle would make no business sense.

Let me make it easy

[Anonymous Coward]

If MySQL had a BSD license it would be owned by the community.
If MySQL had a "non-free" commecial license it would be owned by Oracle.
The mess MySQL, and you, find yourselves in is because of MySQL's stupid dual-level license bullshit. Nobody seems to be able to figure it out or agree on it and it has caused more column inches of claptrap on Slashdot than the MySQL/PostgreSQL threads themselves. MySQL's originator's wanted to have it both ways: Lots-O-corporate money AND GPL poster child. Well they got their money alright, but to get it they had to pray for a really wealthy, poorly managed corporation to come along and vet their convoluted business plan. That would be Sun.

Now, with a billion dollars spent to "buy" MySQL but a bunch of forks still out there, no company in their right mind is going to invest anything in MySQL because they'll be worried Widenius will just steal the improvements and fork it again. MySQL is pariah, it's poisoned.

If you're running any kind of data volume worth talking about you're better off with PostgreSQL. Not only is it faster with *real* queries and more robust, but now it's safer going forward.

Re:Let me make it easy

[jedidiah]

Except MySQL didn't want their code "owned by the community". That's why they dual licensed it to begin with.

If they wanted it to be "free and open like BSD" they could have simply made the relevant parts LGPL.

This notion of yours that FSF licenses are incompatable with business is just a fantasy.

Re:Let me make it easy

[TheRaven64]

If MySQL had a BSD license it would be owned by the community.

Interesting choice of words. I think you will find this makes more sense if you substitute 'controlled' for 'owned' and 'their customers' for 'the community'.

Re:Let me make it easy

[Eil]

MySQL's originator's wanted to have it both ways: Lots-O-corporate money AND GPL poster child. Well they got their money alright, but to get it they had to pray for a really wealthy, poorly managed corporation to come along and vet their convoluted business plan. That would be Sun.

Uh, no. MySQL, the company, was doing just fine all by themselves. They were making a tidy profit on value-added services, support contracts, and custom development. They were one of the shining examples of how to build a successful (read: profitable) business on top of an open source product. That's what made them an attractive purchase in the first place.

Now, with a billion dollars spent to "buy" MySQL but a bunch of forks still out there, no company in their right mind is going to invest anything in MySQL because they'll be worried Widenius will just steal the improvements and fork it again.

That just makes no sense. In the open source world, you're supposed to "steal" improvements. That's what open source is all about. And in any event, there have always been plenty of support/consulting options in the database world besides the original vendor. That has always been true no matter who owns the commercial version of MySQL or whether an open source fork becomes the de facto edition.

MySQL is pariah, it's poisoned.

Thanks for the FUD, but you can have it back. Because MySQL is open source, it is by design immune to such trivial things as corporate takeovers. MySQL is the most popular database engine on the planet and this is one of the big reasons why. Although the future of MySQL may be uncertain, anyone currently using MySQL can be secure in the knowledge that their copy won't magically shut itself off and that there isn't just one vendor capable of adequately supporting the current code base.

If you're running any kind of data volume worth talking about you're better off with PostgreSQL. Not only is it faster with *real* queries and more robust, but now it's safer going forward.

PostgreSQL is a fine database engine, but it has fuck all to do with this discussion.

Re:Let me make it easy

[ccady]

...Widenius will just steal the improvements and fork it again....

Steal? Do you understand with GPL means? It means that if anyone wishes to start with the original code, and make improvements and distribute them, then they are obligated to freely give those improvements to everyone. That is the purpose of the GPL. There is nothing to steal here!

You are caught in the trap of thinking that "MySQL" is a product. The trademarked name MySQL is what Oracle bought. The code is free, thanks to the freedom-loving license. Improvements to the "MySQL" code will be made by all sorts of people and companies with all sorts of agendas. When you say "MySQL is pariah, it's poisoned." I might agree, if what you are talking about is the trademarked name "MySQL". If you are talking about the very useful database server code which is currently being distributed by Oracle under the name "MySQL," but which is owned by the world, then you are blithering and spewing your own claptrap.

Alright. That was a bit venemous. I apologize. But I hope you get the point. The code and all distributed improvements has been rescued by the fact that it is under a free license. The only companies that would avoid contributing to that codebase are those whose purpose is to make money by selling the code itself. Obviously that was not the case when MySQL AB owned the name "MySQL"--lots of companies and people contributed to the code. If you are arguing that that will change now that Oracle owns that name or Monty is creating a fork, then you have not made your point.

Who is Spartacus?

[tsstahl]

Will the real MySQL please stand up.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Who controls?

[Todd Knarr]

If the code's GPL-licensed, the answer's simple: the user community. As long as the corporate owner continues to serve the users' needs, the users will stick with the "official" version. If the corporate owner (or any other fork maintainer for that matter) starts to send the code in directions that don't serve the users' needs, those users will tend to switch to a fork that does better serve their needs. If there isn't one now, there will be once some enterprising soul realizes there's money to be made giving users what the corporate owner won't. And the terms of the GPL mean the corporate owner can't legally stop this process. If it wasn't a copyright-assignment-required project the corporate owner can't even get out from under the GPL terms going forward.

If you think it won't happen, think back to the GCC/EGCS fork/reunification.

Re:Should Read: Sun announces last MySQL products

[oliderid]

Oracle will surely kill (or at least castrate) MySQL.

Well it would be stupid. MySQL has Open Source competitors, they can easily replace it. they would destroy their asset value and nothing else.

Secret to Success

[steltho]

1. Create a company around a popular open source database.
2. Sell company for $1 billion.
3. Profit
4. Fork it
5. ???
6. Profit again

Re:Secret to Success

[shutdown -p now]

1. Create a company around a popular open source database.
2. Sell company for $1 billion.
3. Profit
4. Fork it
5. ???
6. Profit again

I think it's actually:

    5. Goto #1.

and there's no #6.

Re:Should Read: Sun announces last MySQL products

[Abcd1234]

Who said anything about closing the gap? Continuing to develop and support MySQL doesn't mean turning it into a powerhouse database like Oracle.

The simple fact is, MySQL and Oracle do not, and have never, played in the same league, and I believe it would be a mistake to try and turn MySQL into a shitty Oracle. MySQL has a niche... keep it there.

Oracle wants more people writing SQL

[davecb]

So they improved InnoDB to make MySql more attractive to the small folks. If they become as big as eBay and PayPal, they probably will switch to Oracle (;-))

--dave

Re:Should Read: Sun announces last MySQL products

[rackserverdeals]

The gap between MySQL and Oracle is huge and not likely to be closed anytime soon.

Technology leaders in big companies aren't as into all the open source gossip as the slashdot crowd are and I wouldn't be surprised if many of them didn't even know there were MySQL forks or what that meant.

They would rather go with a MySQL that is named MySQL and has a big company like Sun or Oracle, the leading db vendor that also owns the only sane database engine for MySQL, than some noname fork. Even if it was started by the MySQL founders and all the developers went to it. If all the MySQL developers go to a fork, well then Oracle developers will take over.

What's more concerning is IBMs partnership with EnterpriseDB, which is based on PostgreSQL.

If you want an open source database that closes the gap with Oracle, use PostgreSQL.

Sun should have never bought MySQL. Instead they should have put more effort into PostgreSQL. Sun has had some big wins with Solaris and Postgresql in the past and offer support for it on Solaris.

Must be tough since Oracle is an important part of Sun's business but Oracle has done things that could be considered as stabbing Sun in the back too.

Proved me right!

[rgviza]

I told my team mate this would happen as soon as I heard the news about Oracle buying Sun on Monday. I had NO idea it would happen this fast.

Monty should call the new project OurSQL =D

-Viz

Re:Proved me right!

[Samah]

Monty should call the new project OurSQL =D

Let's hope they don't write it in Python.

Dead Horse Flogged - Again

[mdielmann]

I think we've already answered this a million times.
BSD - Do whatever you want, but give attribution. Fork you!
GPL - Do just about whatever you want, but you have to distribute source, and a few other conditions. Fork happens.
Standard Copyright - Don't even think about forking.

Since it's (or was) GPL, forks are allowed by the copyright. If the current copyright holders don't like it, they shouldn't have bought a GPL licensed product, or kept the community happy. They obviously failed to some degree on the second option, and so we have forks. And so long as the forks follow the requirements of the GPL, all the copyright holders can do is scream and shake their fists. Or straighten things out and hope the forks re-merge or die.

Re:Should Read: Sun announces last MySQL products

[ducomputergeek]

Most people I know that plan to start with a OSS database and move to Oracle start with PostgreSQL, since PostgreSQL mirrors the capabilities and features of Oracle pretty close, just it's not quite as fast. (But the PostgreSQL folks have been making progress).

A little OT but Monty will be talking in the US

[atari2600]

Bellingham, WA will be hosting their 10th LinuxFest Northwest this coming weekend (25th and 26th) and Monty Widenius will be a speaker on the 25th. Just a heads up for anyone in the area wanting to see and hear the man speak.

Nice --

[KingPin27]

So it appears that MySQL is going the way of the Microsoft model. whats next MySQL home premium platinum but somewhat not limited for use version or perhaps MySQL go jump off a bridge edition.

PostgreSQL FTW

A better question

[DaveV1.0]

Who really owns a commercial open software application: the corporate holders of the trademarked name of the project, or the community?

The copyright is pretty much irrelevant up to the point that the corporate holders change the copyright and licensing. But, if Oracle now owns the trademark "MySQL", then Monty Program Ab may not be able to use the name MySQL.

Re:A better question

[Delifisek]

MontySQL ?

It's obvious who owns it...

[argent]

The people who own an open source application are the people who are at any point in time putting in the effort of maintaining it.

Re:Should Read: Sun announces last MySQL products

[Bob9113]

I think they'll just use it to try to funnel users butting up against its limits towards full Oracle. If they kill it they lose that potential sales channel.

A rational assessment of a solid business model that simultaneously benefits both a proprietary and an Open Source product. It is clearly beneficial to Oracle and at worst is indifferent to the users of Oracle DBMS and MySQL. That is the sort of win/win thinking that benefits us all to consider. Thank you for your post.

GCC was forked off the FSF

[Mask]

The danger of forking is not reserved to commercial entities.
If the community is not happy with whoever controls the code then it's fork fork fork.

If the company that controlled the code plays this well then it may have a chance to merge the fork back.

Cygnus were fed up with FSF's attitude regarding GCC development, other developers were also fed up.
In fact most of the active community were fed up. So Cygnus forked GCC into EGCS which started to thrive.

FSF came to its senses and made an agreement with Cygnus and other developers to merge EGCS back to FSF.
In fact EGCS was renamed to GCC 2.8 or 2.95 (I don't remember).
The smartest thing that Cygnus and other developers did was to assign all copyright to FSF even during
the fork. This allowed the merge back.

If the major players play well then it is possible to merge any fork back.
Things should not be as bad as you say for creates an open source product. Things can be fixed if
the developing company is willing to avoid the arm wrestling game.

The question is, how interested are Sun, Oracle and the developers to avoid a fork.

There is no binding, only unbinding

[SuperKendall]

The GPL is clear enough on this matter, but it isn't entirely legally clear if the GPL is to be considered a binding contract or not.

That's irrelevant, because at the heart of things the GPL is not what about what is binding. It is about what is being unbound, specifically you are waiving your rights as copyright holder to some things. You don't really have a contract with anyone, you are stating to the world explicitly what rights you are relinquishing.

You cannot "unwaive" rights once revoked, just like you cannot put material in the public domain and then lock it down again.

Re:There is no binding, only unbinding

[DragonWriter]

That's irrelevant, because at the heart of things the GPL is not what about what is binding. It is about what is being unbound, specifically you are waiving your rights as copyright holder to some things. You don't really have a contract with anyone, you are stating to the world explicitly what rights you are relinquishing.

Yes, and under US law, such a statement can be revoked if it is not secured by a contract. It's called a "gratuitous license".

The FSF can say the GPL is irrevocable, the GPL can say the GPL is irrevocable, but as long as it is a license not secured by a contract (say, because it lacks mutual consideration), than barring a fundamental change in the law regarding licenses in the US, the GPL and other gratuitous licenses are and will remain revocable, at will, by the licensor.

Pwnership

[Anonymous Coward]

This raises the larger question of who really owns a commercial open software application: the corporate copyright holders, or the community?

Well we know what IBM and all the other mega-corps Lunix sold it's soul to will say. As a company, they consider their billion dollar investment in Teh Lunix to mean they now have defacto ownership of it.

Teh Lunix could have stayed pure, but in their desperation to chase Microsoft's tail lights, they were taking money from anyone who would offer it. Plus, Teh Lunis needs a new Ferrarri, and giving away free software doesn't put shoes on teh baby.

Now of course, Teh Lunix Community is going to make the case that all that money was a "donation", not an investment... because they want to have it both ways. I'm sure everyone would be happy to be given billions of dollars and still stay in control... but that doesn't happen in reality.

Re: FICS and others

[Lonewolf666]

Actually, there is a project on Sourceforge that appears to build on the old GPLed ICS code:
http://sourceforge.net/projects/chessd/

In general, it seems that forks only happen if the original group of developers really does a bad job. Because otherwise, it is simply easier to download the work that someone else does for you.
In the case of XFree86, it took dissatisfaction among developers and a license change that was seen as unacceptable by many Linux distributors. But once those happened, X.org was founded. By now it has mostly supplanted XFree.

So many things wrong with the summary.

[Bill, Shooter of Bul]

Monty had already forked Mysql, after he left Sun but before the Oracle announcement. Word to the wise, don't actually base any decisions off of a slashdot summary.

Re:So many things wrong with the summary.

[C_Kode]

He didn't fork MySQL before all the take over talks. He was taking Sun's MySQL code and adding his own stuff to it. When Sun would release new code, that code (or most of it) went into his code base. (at least thats how I understood it)

Re:So many things wrong with the summary.

[Bill, Shooter of Bul]

Well, the semantics between a branch and a fork are confusing, but his branch/fork was born before the take over talks .

Re:It depends, Well, change the fork from "My" to:

[davidsyes]

"OurSQL", or to "YourSQL"..., and tell SOracle: What's YOURS is MINE and what's MINE is.... MINE"

Dual licensing and forks

[GleeBot]

What I don't understand is... OK, the original MySQL developers only have the right to fork the GPL-licensed version of the code. Presumably, the ability to commercial license remains solely with Sun (and I guess, now Oracle).

Doesn't that sorta shoot their old business model in the foot, where they would charge for a commercial license for non-GPL clients? Sure, fork away, but haven't they still screwed themselves over by selling out to Sun to begin with? It's not an ideal situation for them, either.

The Party is Over

[Udigs]

I mean, who among us thinks this is actually going to turn out okay? You have the database manufacturer MOST hurt by MySQL now turning around and *BUYING* it. I give it 6-8 months before they come out with a "MySQL Lite" kinda like Oracle Express Edition, which is free, but only works on one CPU and supports like 10 connections at a time.

This is seriously bad news.

And don't tell me about the "fork." There's the thing about forks: a fork is like a militia. It sounds like a good idea, but in the end we all know it won't work. It's more of a last resort kinda thing. A fork only works in the end if one of the forks becomes the standard. If that doesn't happen, well, goodbye to supporting MySQL. Seriously, this is going to be a mess.

Re:The Party is Over

[shentino]

Just more "proof" that

Nice Guys Finish Last

Re:Should Read: Sun announces last MySQL products

[Anonymous Coward]

Why do you think Postgresql is not as fast as oracle? oracle's license forbids the publishing of benchmarks. Are you talking out of your ass or are you violating your oracle license?

Larry the Hutt wants to know.

I doubt it was Sun

[Jane Q. Public]

that "lost the hearts and minds", since the programmer exodus did not start until the Oracle acquisition. It is Oracle that people do not trust to carry MySQL forward, as it formerly existed.

Oracle to buy Sun...A tragedy for open source...

[giuffsalvo]

Seriously, given that Sun is still the biggest commercial contributor to OSS, and given the fact that I highly doubt Oracle will continue to feed money into many open-source project sponsored by Sun (like Netbeans, MySQL, ecc..), since this would 1 - Hurt them (MySQL) or 2 - Make them waste money into products not for their target markets I think that this merge is a tragedy for many open source projects, which will see a slowdown, or complete death. Not to mention the fact that the world is loosing one of the most open-minded, trasparent, and less "bastard" companies ever existed... Is there any chance that a like-minded company like Google, despite working in a completely different market (they provide services, Sun/Oracle provide the infrastructure), might try to save Sun and its legacy, for "historical" reasons? Or maybe take the financial burden of sponsoring Netbeans, MySQL, ecc...?

Re:Should Read: Sun announces last MySQL products

[jadavis]

What's more concerning is IBMs partnership with EnterpriseDB [cnet.com], which is based on PostgreSQL.

Can you elaborate? Why is that a big problem?

Re:It depends, Well, change the fork from "My" to:

[x2A]

Or go the Coke route, call it MysQlassic

Re:Should Read: Sun announces last MySQL products

[rackserverdeals]

PostgreSQL is the open source database that can comepete with Oracle in terms of features and reliability. EnterpriseDB aims to be a drop-in replacement for Oracle with a PostgreSQL backend. The partnership is going to create a path for EnterpriseDB to be a drop-in replacement for Oracle with a DB2 backend.

It creates a migration path from Oracle that IBM can take advantage of.

Re:It depends, Well, change the fork from "My" to:

[Anonymous Coward]

They should fork MySQL and call it LOL8BILLIONSUCKERS.

Re:Should Read: Sun announces last MySQL products

[ducomputergeek]

We run Teradata actually.

Appliances and isolated networks

[tepples]

Problem easy to solve. Don't distribute the MySQL JDBC drivers in the case of a Java app, for example.

If the application is going to be distributed as part of an appliance like Google-in-a-box, then it just wouldn't work.

Better yet, use the application online and provide it as a service.

Not practical for an otherwise Internet-disconnected network. And in many areas, this isn't even possible without an extra $720 per year or more to the cell phone company for a tetherable data plan, as dial-up just isn't good enough anymore.

SQLite

[tepples]

So, someone makes the absolute bare minimum database that is nothing more than a glorified text file, but uses MySQL syntax.

Or more practically, a patch against the SQL parser of (public domain) SQLite to improve its support for MySQL syntax. SQLite has already accepted such patches: see REPLACE, an alias for SQLite's own INSERT OR REPLACE.

Re:Should Read: Sun announces last MySQL products

[Anonymous Coward]

It's not the database size that matter, PostgreSQL performs almost as fast as Oracle with databases as large as a Petabyte.

However, Oracle has WAY better replication/distributed data storage technology.

or ... all that time spent forking could be ...

[Hohlraum]

used to enhance the usability of postgresql. Thats really the only reason people continue to use mysql. Its certainly the reason I continue to use it today. Postgresql just has too much setup/maintenance imo.

Does anyone else notice the stupidity in this?

[Anonymous Coward]

Sun buys MySQL AB. MySQL AB agrees to sell to Sun. MySQL employees then quit and programmers intend to fork now that they've sold out to the beast itself, Oracle.

But.. I have to ask.. if they were obviously not thrilled about being bought out by Sun, WHY SELL IN THE FIRST PLACE? This wouldn't be happening today if they remained independant, so it's kind of like saying "Yea, we sold out to Sun for $1 billion, then got pissed off, and forked the project anyway" - am I missing something or does this sound like a scam - Create open source product, create company around said product, ... profit!, fork product out of protest to what the corporate overlords have done... profit!

=/

Re:Does anyone else notice the stupidity in this?

[Anonymous Coward]

It looks like a scam to me if they are able to do that, but you'd think there would be some sort legal barrier to it though.

OurSQL

[halber_mensch]

Should definitely be the name of the forked project. East it, Oracle.

Re: FICS and others

[syousef]

Actually, there is a project on Sourceforge that appears to build on the old GPLed ICS code

It might be based on the same idea but it is not even written in the same LANGUAGE as the old FICS code. (This is C++, FICS was C).

I'm going to fucking tear my hair out. It seems what I've said is so unpopular that it's led to at least 4 people spouting the same uninformed and unresearched CRAP.

In the case of XFree86, it took dissatisfaction among developers and a license change that was seen as unacceptable by many Linux distributors. But once those happened, X.org was founded. By now it has mostly supplanted XFree.

That happened because X is fundamental to the operation of the entire Linux operating system. Hence there was a lot of incentive to do this and there was too much at stake. FICS was killed off quietly. You'll be lucky if you can find it on some dodgy Russian chess site that you won't feel safe downloading from.

Logical data integrity

[Estanislao Mart�nez]

Problems with data integrity always point to errors in memory handling logic.

You're thinking of what's called physical data integrity. Your statement is only right if we grant you that assumption.

However, relational database management systems are supposed to also guarantee logical data integrity. This is, in general, validity of constraints on the data (value range constraints, uniqueness constraints, foreign key constraints, all-or-nothing transactions, etc.). MySQL is correctly criticized for not enforcing this kind of data integrity.

Why the hurry?

[Ilgaz]

Can't they wait for the deal actually happen or demand official statement from Oracle regarding their policies for the future of mysql?

It is not like AOL acquired Sun for God's sake. As far as I know, Oracle does very high end database servers and I don't see why they would spend billions just to "kill" GPL'ed code. Don't you think Oracle as second largest software company on planet doesn't know what GPL is or the mysql developers/community?

I could understand if Sun was left with no buyers and in position of "rejected by IBM", the future of company would be in doubt really. It is not AOL acquired them nor MS (to kill). It seems immature to me.

Re:Should Read: Sun announces last MySQL products

[Anonymous Coward]

They would rather go with a MySQL that is named MySQL

You say that as if MySQL has a good reputation. If some random guy on the Internet thinks he can do more reliable design work than the core MySQL developers, well, judging by history he's probably right.

Re:YourSQL

[rdebath]

A name change is easy, do you even remember the old name of Firefox?

Not necessarily

[ILongForDarkness]

Say that your employer is paying you to work on the project and has a NDA/no compete in your contract. I don't think you could spin the product off then even if its open source.

Re:YourSQL

[DaveV1.0]

MySQL has what is known as "brand awareness" which is apart of "corporate goodwill". Changing the name means that all the effort towards making MySQL a known entity to non-technical people will be lost.

And, Firefox was originally Phoenix, but changed due to trademark issues with Phoenix Technologies. Then, it was Firebird, but that didn't go over well because of the Firebird DB.

If Firefox were to change it's name to SuperBrowser or RedFox, how many people would associate the two? By giving up the product's name, one gives up the positive feelings and respect associated with the old name. It is one of the reasons people defend their trademarks and copyrights.

Tabbed browsing likewise makes resizing harder

[tepples]

I can't resize the "window" of a newspaper.

Nor can you easily resize a web site if you are switching back and forth between web sites in tabs in one window.

Revocable licenses / promissory estoppel

[butlerm]

"A similar basis is to be found for the doctrine of an early Pennsylvania decision, which held that a license, in terms permanent, to divert a watercourse, could not be revoke after the licensee had made improvements and invested capital in consequence of it. This decision has been followed by other cases holding that a license cannot be revoked in violation of its terms after the licensee has seriously changed his position on the faith of it." (Samuel Willston, The Law of Contracts, 1920)

The Pennsylvania case is "Rerick v. Kern, 14 S & R 267".

Even when (as is usually the case), a license is revocable at common law contrary to its terms, the doctrine of promissory estoppel may allow the license to be enforce in equity.

i.e. "in the law of contracts, the doctrine that provides that if a party changes his or her position substantially either by acting or forbearing from acting in reliance upon a gratuitous promise, then that party can enforce the promise although the essential elements of a contract are not present." (West Encyclopedia of American Law, 2008)

Just because a license is considered technically "revocable" at law contrary to its own terms does not mean it won't be enforced. A copyright license all the more so.

Re:Revocable licenses / promissory estoppel

[DragonWriter]

Just because a license is considered technically "revocable" at law contrary to its own terms does not mean it won't be enforced. A copyright license all the more so.

Saying "all the moreso" doesn't make it so. Diverting a watercourse is a one-time event, the effect of such a license is essentially all or nothing, so if any effect is given by promissory estoppel, it is effectively permanent. (That is, the license effect is temporary -- only to divert the watercourse at one time -- but the change effected is permanent.) This makes it rather unlike the case with most copyright licenses (particularly, the GPL and related open-source licenses.)

A copyright license, OTOH, is more like a license to cross or use land on an ongoing basis, where estoppel (providing the terms of the license relied on didn't provide for immediate termination) would generally prevent from being revoked with immediate effect on a licensee in the course of use, but also would not generally make permanent in the effect of an attempted revocation.

Re:Revocable licenses / promissory estoppel

[butlerm]

If someone creates a significant business infrastructure around an open source software package that has no viable replacement, and in particular makes a significant investment in enhancing it, a revocation contrary to the license's own terms is a perfect setup for the doctrine of promissory estoppel to be applied.

"Certain elements must be established to invoke promissory estoppel. A promisor--one who makes a promise--makes a gratuitous promise that he should reasonably have expected to induce action or forbearance of a definite and substantial character on the part of the promisee--one to whom a promise has been made. The promisee justifiably relies on the promise. A substantial detriment--that is, an economic loss--ensues to the promisee from action or forbearance. Injustice can be avoided only by enforcing the promise.

A majority of courts apply the doctrine to any situation in which all of these elements are present. A minority, however, still restrict its applicability to one or more specific situations from which the doctrine emanated, such as when a donor promises to transfer real property as a gift and the donee spends money on the property in reliance on the promise." (West's Encyclopedia of American Law, 2008)

I say "all the more so" in the case of copyright licenses, because unlike a gratuitous license to cross one's real property, a gratuitous copyright license does not create an ongoing economic or physical burden on the part of the licensor. The very reason software is licensed under such terms in the first place (in most cases) is to get an economic benefit from the enhancements of others to the same software package.

Re:Revocable licenses / promissory estoppel

[DragonWriter]

BuIf someone creates a significant business infrastructure around an open source software package that has no viable replacement, and in particular makes a significant investment in enhancing it, a revocation contrary to the license's own terms is a perfect setup for the doctrine of promissory estoppel to be applied.

I agree; as an equitable doctrine, however, it usually will not be applied (I'm speaking generally, there is, AFAIK, very little if any gratuitous-copyright-license-specific case law invoking promissory estoppel) the same way a contract would; specifically, promissory estoppel usually won't create a permanent right, it will limit the effect of an exercise of the right to revoke a license. The exceptions I am familiar with are all cases (like the one cited earlier in this thread regarding a license to redirect a stream) where the circumstances make the right essentially all-or-nothing.

I say "all the more so" in the case of copyright licenses, because unlike a gratuitous license to cross one's real property, a gratuitous copyright license does not create an ongoing economic or physical burden on the part of the licensor.

IMO, this is simply not true: not allowing a copyright holder to revoke a gratuitous license creates exactly the same kind of ongoing economic burden on the licensor as does not allowing a real property owner to revoke a gratuitous license to another to make economically-significant use of his land. In either case, the economic utility of the property to the revoking licensor is limited by the courts refusal to give effect to the revocation. Now, there is a reason that estoppel will nonetheless limit the effect given to a revocation -- to avoid manifest injustice to the licensee who reasonably relied on the promise given with the license. But whereas in contract promises will be generally be strictly enforced, with promissory estoppel promises are not strictly enforced, but enforced to the extent of the reliance (or, worded another way, to the extent necessary to avoid substantial injustice given the reliance.)

Re:Revocable licenses / promissory estoppel

[butlerm]

I believe that a copyright case would likely be decided differently. Copyright licenses have radically different characteristics than licenses to use real property. Open source licenses, for example, are adopted with the intent that people create derivative works. If the license necessary to distribute those derivatives may become null and void at any time, no one would ever create them. Most courts tend to be rather pragmatic about that sort of thing. If a rational legal defense of the irrevocability of GPL type licenses can be made, I suspect most courts would be inclined to adopt it. On the other hand, if you are right, at some point the only viable open source "license" will be *public domain*.

WTF

[anomalous cohort]

The original post expresses what I believe to be some very real concerns about the future viability of MySql now that it will be in the hands of competitor Oracle. The tone of this discussion completely ignores this most salient point. Instead, we seemed to be more concerned about what constitutes the legal definition of binding from the context of the GPL or which DBMS is better MySql or PostGreSql.

The reality is that there are a lot of web sites out there using MySql. Is anyone here responsible for one of those sites? Do you have any concerns about this Oracle deal with Sun? What is your migration strategy were Oracle to poison or sunset MySql?

FLOSS communities can't be tied to one place

[bkuhn]

I think the fundamental problem is that FLOSS community needs diversity and cannot be tied to a single for-profit entity. For a project to succeed with a healthy community, it needs individual developers or a non-profit entity (run by developers) to control it. I've written a blog post about this specific issue in response to Monty's linked in the main article.

get your facts straight

[toby]

Er, no, chessd on sourceforge is entirely C. And even a brief perusal would show that it is based on the original [F]ICS C code. Did you even look at the source?

That codebase is pretty much dead; the project leader(s) have been working on a completely different solution that has been mentioned on the mailing list, is apparently in use in Brazilian schools (where chess is very popular) but unfortunately has not been published.

Re:get your facts straight

[syousef]

Er, no, chessd on sourceforge is entirely C. And even a brief perusal would show that it is based on the original [F]ICS C code. Did you even look at the source?

Are you lying, trolling, or just plain stupid?

Lets ignore hte filenames for a moment. Since when does C have constructors? From Player.cc //- Constructor
        Player::Player(int n, float wt) : name(n), floatValue(wt) { }

Or from the install file

--- Chessd ---

This is a implementation of a chess server, it was designed to replace the old
chessd. Hopefully the new one is more robust and more effcient. Note that is is
still in development and it is unstable, use it at your own risk.

Hell it even uses a Postgres SQL databae not a flat file system like the original FICS

I think you'll find even a brief perusal will show you it's NOT based on FICS code. Some of the same ideas may be in play, but that's it

That codebase is pretty much dead; the project leader(s) have been working on a completely different solution that has been mentioned on the mailing list, is apparently in use in Brazilian schools (where chess is very popular) but unfortunately has not been published.

So my point was the FICS code base has been killed so to refute you point me to a different codebase losely based on the same ideas, which is also dead??? Are you obtuse??

Another example

[SpammersAreScum]

Kowari -> Mulgara

Triggered by the purchase of Tucana (built on top of Kowari) by Northrop-Grumman and their subsequent alienation of the Kowari developers. As such, it seems like an example Oracle should keep in mind.