×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Mickos Urges EU To Approve Oracle's MySQL Takeover

Soulskill posted more than 4 years ago | from the it'll-be-cool-trust-me dept.

News 67

mjasay writes "Former MySQL CEO Marten Mickos has written to EU Commissioner of Competition Neelie Kroes to urge speedy approval of Oracle's proposed purchase of Sun, including the open-source MySQL database. The EU has been worried that Oracle's acquisition of Sun could end up hurting competition by dampening or killing MySQL's momentum. But in his letter, Mickos separates MySQL-the-community from MySQL-the-company, arguing that Oracle's takeover cannot hurt the MySQL community: 'Those two meanings of the term "MySQL" stand in a close, mutually beneficial interaction with each other. But, most importantly, this interaction is voluntary and cannot be directly controlled by the vendor.' In a follow-up interview with CNET, Mickos indicated that he has no financial interest in the matter, but instead argues he 'couldn't live with the fact that [he's] not taking action,' and is 'motivated now by trying to help the employees still at MySQL and Sun, and by an urge to bring rational discussion to the matter.'"

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

67 comments

Good! (4, Insightful)

Futurepower(R) (558542) | more than 4 years ago | (#29704347)

I'm happy he is taking action.

Too often, technically-knowledgeable people don't recognize or accept the need for them to be social leaders.

Re:Good! (1)

davecb (6526) | more than 4 years ago | (#29711311)

And especially to offset FUD from people who don't want the Oracle/Sun merger to happen.

--dave (who wants them to finish merging so he can get more consulting gigs) c-b

Why don't they want Sunoracle to happen? (1)

Futurepower(R) (558542) | more than 4 years ago | (#29711595)

"... people who don't want the Oracle/Sun merger to happen."

Who are they, and why?

Re:Why don't they want Sunoracle to happen? (1)

rtb61 (674572) | more than 4 years ago | (#29715643)

I see the Sun Oracle merger producing and very interesting mix of closed source and open source software 'systems'. Proprietary and open applications on top of an open operating system. They are certainly large enough and have sufficient commercial activity to do very interesting things not only with MySQl but also OpenOffice and even Linux. So potentially an enormous and probably lethal threat to the single monopolist incumbent. Of course the convicted monopolist can't really say all that much so it is stuck with behind the scenes shenanigans and operating through puppets 'er' trusted partners.

Re:Why don't they want Sunoracle to happen? (1)

davecb (6526) | more than 4 years ago | (#29718299)

Any competitor in either software or hardware, who sees Oracle + Sun as the kind of full-line business that IBM used to be in the mainframe business.

Right now, Sun sales are down as people wait to see if the deal will go through, and the competition is going door to door saying "buy from us, Sun is going out of business".

--dave

Re:Good! (0)

Anonymous Coward | more than 4 years ago | (#29713391)

The bigger reason not to approve the acquisition is to of sun by oracle and oracle by Microsoft is the open office packeage is better than the microsoft office package. It is available free at sun.com or openoffice.org .

Alternatives (3, Interesting)

Anonymous Coward | more than 4 years ago | (#29704411)

if they mess MySql up(even more that is), people can just move to Postgre, Firebird, Couchdb, Drizzle, etc.. There's anything but a shortage of open-source databases.

Re:Alternatives (4, Insightful)

jadavis (473492) | more than 4 years ago | (#29704551)

Moving from one database system to another is no trivial matter, even if that other system is a fork.

I agree with Marten Mickos here. There's no benefit to dragging this process out. If Oracle owning MySQL would be a problem, the time to fix it was months ago (I realize that may have been impossible); leaving it in extended purgatory is worse. MySQL has some degree of protection by using the GPL license, anyway.

Disclaimer: I'm a PostgreSQL user, and I haven't used MySQL for a while.

Re:Alternatives (1)

Korin43 (881732) | more than 4 years ago | (#29705997)

If it's GPL, couldn't they just fork it now? And by that I mean:
  • Oracle buys Sun
  • Oracle kills MySQL
  • Someone else forks MySQL (gives it a new name, makes it run on the same port)
  • From the perspective of the programs using it, nothing has changed

Re:Alternatives (1)

FallLine (12211) | more than 4 years ago | (#29706141)

From the perspective of the programs using it, nothing has changed

Except MySQL's users generally are not just installing a static set of code that will never need changing. They are implicitly wedding themselves to the product and depending on its backers to supply patches as-needed and on the promise of future enhancement to support new hardware, better design, etc. In other words, the quality of the group behind the fork and their ability to attract funding to continue its development is a major question for any serious user.

Even if all the original MySQL developers were to actually join with some forker's company (providing they do not have non-competes in place), they would need to abandon the dual-license model that provided MySQL with a lot of revenue because Oracle would retain copyright ownership. That would call into question the ability for the company to survive and fund continued development efforts.

I, for one, would place low odds on that organization's ability to find a viable business model or find enough charitable contributions to operate as a non-profit. Even if it _could_, mere uncertainty alone can prevent it from acquiring the critical mass it needs.

Re:Alternatives (1)

jjohnson (62583) | more than 4 years ago | (#29704767)

The problem with moving is that MySQL (like PHP) has a huge install base in ISPs. If you're using shared hosting, you've got MySQL available, guaranteed, and that tends to determine what people use and know.

I develop with both PHP and Python/Django, and every time I try to push for the latter, I have to overcome the perception that PHP is king and should be used, even in environments that don't require it.

However... (1)

IANAAC (692242) | more than 4 years ago | (#29704883)

The problem with moving is that MySQL (like PHP) has a huge install base in ISPs. If you're using shared hosting, you've got MySQL available, guaranteed, and that tends to determine what people use and know.

I'm seeing more and more ISPs offer PostreSQL in addition to MySQL. They don't really support it like they do MySQL, but it is being offered more.

Re:However... (1)

jjohnson (62583) | more than 4 years ago | (#29704923)

I've been paying attention to this as well, and invariably the version of PostgreSQL is older, usually 7.x when 8.4 is the latest production-ready release. I would love to see more current versions of PostgreSQL.

Re:However... (1)

IANAAC (692242) | more than 4 years ago | (#29705321)

I just checked my hosting... they offer 8.3.3. But anything I've done should so far with PostgreSQL should be doable with the 7.x series (I don't do anything difficult, just dictionaries and glossaries.

Re:However... (1)

jjohnson (62583) | more than 4 years ago | (#29705451)

Who's your host? My issue with 7.x is that a large part of the work on 8 was improving performance, something they were largely successful with.

Re:However... (1)

IANAAC (692242) | more than 4 years ago | (#29706031)

Siteground.

I had heard some bad things about them, but they've never given me any trouble. I can pretty much run whatever I want.

Re:However... (1)

Antique Geekmeister (740220) | more than 4 years ago | (#29707129)

The problem isn't the database itself. For a lot of use, there's not much difference. It's the tools to write to, and access, that database. If I've got to replace all the CPAN published, developer abusing Perl that my MySQL users have become accustomed to, I want really compelling benefits from the switchover.

Re:Alternatives (0)

Anonymous Coward | more than 4 years ago | (#29709399)

They have already killed mysql 6.0 "MySQL 6.0 was not developed beyond Alpha status and new releases have not been made for some time, so the manual has been withdrawn as well."

http://dev.mysql.com/doc/refman/6.0/en/index.html

so what is next ? 5.x on the death row

maybe is time to move to postgresql , firebird ...

Re:Alternatives (2, Informative)

Zontar The Mindless (9002) | more than 4 years ago | (#29709563)

They have already killed mysql 6.0 "MySQL 6.0 was not developed beyond Alpha status and new releases have not been made for some time, so the manual has been withdrawn as well."

http://dev.mysql.com/doc/refman/6.0/en/index.html

Wrong. We have NOT 'killed' 6.0, but it not going to be the focus of our development for a while, and we won't do any more official 6.0 releases for some time to come. But we HAVE moved it to the back burner, so to speak.

If you really want 6.0, you can get it right here:

https://code.launchpad.net/~mysql/mysql-server/mysql-6.0-codebase [launchpad.net]

No guarantees right now as to how well it'll build and run on any given day; I think I last built it last Tuesday or Wednesday, and it seemed to do okay, but of course YMMV. That being said, go get the code and knock yourself out.

In the meantime, we're backporting what we think are the best bits of 6.0 to 5.X.

so what is next ?

You'll find some answers to that question here...

http://dev.mysql.com/doc/refman/5.1/en/mysql-next-series-plans.html [mysql.com]

And here...

http://dev.mysql.com/doc/refman/5.4/en/ [mysql.com]

And what's this? The latest release, from less than a week ago:

http://dev.mysql.com/doc/refman/5.4/en/news-5-4-3.html [mysql.com]

That's the clone-off date, BTW. Binaries should be available here

http://dev.mysql.com/downloads/mysql/5.4.html [mysql.com]

in a few days.

5.x on the death row

I think that's a bit of a stretch. Why don't you see what we say about it on the site?

http://www.mysql.com/about/legal/lifecycle/ [mysql.com]

NB: We have a choice between (a) honouring the policies on this page and (b) breaking contacts with paying customers.

Can you guess which of these we're more likely to do?

maybe is time to move to postgresql , firebird ...

That's one of the reasons why it's called Free Software -- you're absolutely free to move to something else if you like, and we wish you every success with it if you choose to do so.

Fork it (2, Insightful)

Ghubi (1102775) | more than 4 years ago | (#29704433)

It's not that easy to kill an open source project.

Re:Fork it (0)

Anonymous Coward | more than 4 years ago | (#29704457)

Didn't that already occur when Sun bought MySQL (the company)? For the exact same reasons that seem to be the direct concern of the EU?

Re:Fork it (5, Informative)

Z00L00K (682162) | more than 4 years ago | (#29704483)

There is already a fork: MariaDB [askmonty.org] by Monty, one of the MySQL founders.

Re:Fork it (1)

Tablizer (95088) | more than 4 years ago | (#29704779)

I'd prefer it named "MontyDB" over "MariaDB". The second sounds Catholic.

Re:Fork it (5, Informative)

Anonymous Coward | more than 4 years ago | (#29704955)

"The name MySQL (just like the MyISAM storage engine) comes from Monty's first daughter "My". MariaDB continues this tradition by being named after his younger daughter. "

Re:Fork it (0)

Anonymous Coward | more than 4 years ago | (#29705185)

Where is the +1 chokawaii when I need it?

Re:Fork it (0)

Anonymous Coward | more than 4 years ago | (#29710593)

So how did the first daughter loose favor so quickly?

Sorry My, even though you made me a lot of money, you're no longer the database of choice.

Re:Fork it (1)

munwin99 (667576) | more than 4 years ago | (#29727843)

Don't forget OurDelta - http://ourdelta.org/ [ourdelta.org] - [about] "OurDelta produces enhanced builds for MySQL 5.0 and builds for MariaDB 5.1, on common production platforms. James Purser of Open Source On The Air describes OurDelta as “a new distribution for MySQL”."

And then there's the under devellopment Drizzle - http://drizzle.org/ [drizzle.org] - MySQL rewritten from scratch by (amongst others) Brian Aker.

Re:Fork it (4, Interesting)

jadavis (473492) | more than 4 years ago | (#29704783)

It's not that easy to kill an open source project.

It takes a long time to put together a real community; it doesn't happen overnight. However, dismantling one can happen overnight, and that may be what has taken place already.

There is a promising amount of development, excitement, and support behind some of the forks like Drizzle and MariaDB. But losing a development team and then trying to reassemble it somewhere else is going to be a serious setback, and it will fracture the community.

When the forks start to diverge there will be even more problems. Application developers will pick fork X, and then people will start asking in the mailing lists "I am having a problem and I am using fork Y". Whether or not the difference between X and Y is causing the problem is not important -- what's important is that it will take effort for the application developers to figure it out.

The "open source can't be killed" idea is great in theory, but in practice it takes more than a license. It takes a real community effort, and requires real leadership, full-time people, a consistent message, and they have to be able to deliver a product, not a stream of patches. These challenges are all magnified for a database system, where it's hard to find those critical few developers that you can rely on, and the need for quantized releases is greater (to avoid the pain of data migration).

All that being said: I think MySQL can pull itself together. But it takes a lot of work, and the thinking that "open source can't be killed" is a sure-fire way to make sure the necessary work is not done, and that will lead to the death of the project.

Let them (2, Informative)

Norsefire (1494323) | more than 4 years ago | (#29704463)

I use MySQL exclusively and it would nice if Oracle were given a shot at supporting MySQL. Even if they do try and kill it to gain leverage for their own database, there's always MariaDB [askmonty.org] (a MySQL fork by Monty Widenius, the original creator of MySQL).

Re:Let them (3, Insightful)

lordandmaker (960504) | more than 4 years ago | (#29704973)

The problem is that there's also all sorts of other forks, each with their own claim to being the first heir to MySQL.

Were Oracle to kill of MySQL and there to be One True Fork that everyone switches to, that'd be fine. But there isn't. There're several, and they seem to be multiplying.

Re:Let them (0)

Anonymous Coward | more than 4 years ago | (#29709347)

Monty these days is in bed with Microsoft maybe he should call it MicrosoftDB

http://monty-says.blogspot.com/2009/09/codeplex-foundation-why-is-microsoft.html

Good news for PostgreSQL (3, Interesting)

ZipK (1051658) | more than 4 years ago | (#29704473)

Even the threat of Oracle owning MySQL is motivating commercial users to look more closely at the BSD-licensed PostgreSQL. If the sale goes forward, it may the biggest boost yet to the PostgreSQL community.

Re:Good news for PostgreSQL (1)

shentino (1139071) | more than 4 years ago | (#29704553)

It's like everyone jumping overboard before they even know if the boat's sinking.

Then, the bankrupt captain, having lost all his passengers, has to scuttle the boat.

Re:Good news for PostgreSQL (1)

jjohnson (62583) | more than 4 years ago | (#29704749)

The threat of Oracle leaving MySQL to whither on the vine is part of it, but it's also the actions of the MYSQL AB founders forking their product, and several more forks in response. No one is quite sure which is the 'correct' or standard MySQL anymore.

MySQL's install base in ISPs is huge, though, and will determine a lot about usage going forward. There'll really only be a seismic shift from MySQL to PostgreSQL if the ISPs make the move.

Re:Good news for PostgreSQL (0)

Anonymous Coward | more than 4 years ago | (#29709313)

No one is quite sure which is the 'correct' or standard MySQL anymore.

Why? MySQL.com [mysql.com] hasn't gone anywhere, and they seem to have plenty of downloads...

Re:Good news for PostgreSQL (1)

ZipK (1051658) | more than 4 years ago | (#29712043)

It's like everyone jumping overboard before they even know if the boat's sinking.

It's more like everyone rushing to the lifeboats as they see who the new captain is going to be. Commercial users, who currently license MySQL from Sun, will have to think long and hard about those contracts coming under the control (and renewal parameters) of Oracle.

Oracle already owns an open source database (1)

Antique Geekmeister (740220) | more than 4 years ago | (#29704539)

Oracle owns Berkeley DB, from when they bought Sleepy Cat Software. Has anyone heard of _any_ useful progress in Berkeley DB, which used to rule Linux for lightweight, small databases? I thought not: they supported it a little bit, and it's been profoundly ignored for years now, by both its owners and the open source community at large.

I'm afraid that MySQL is fated for the same end: Oracle has little incentive to support it properly or to expand its role when it competes directly with their core products. It can be forked, but how much will be left of the core development team that really understood its features and trade-offs after 3 years working at Oracle?

Re:Oracle already owns an open source database (2, Informative)

nxtw (866177) | more than 4 years ago | (#29704609)

Oracle owns Berkeley DB, from when they bought Sleepy Cat Software. Has anyone heard of _any_ useful progress in Berkeley DB, which used to rule Linux for lightweight, small databases?

Berkeley DB is still being developed [oracle.com] with new features - such as those in version 4.8, released less than a month ago [oracle.com].

Anyway, Berkeley DB is a different kind of database than MySQL or Oracle Database.

I thought not: they supported it a little bit, and it's been profoundly ignored for years now, by both its owners and the open source community at large.

Surely the rise of SQLite [wikipedia.org] has something to do with what you perceive to be Berkeley DB's decline?

Re:Oracle already owns an open source database (1, Informative)

Anonymous Coward | more than 4 years ago | (#29704953)

Surely the rise of SQLite [wikipedia.org] has something to do with what you perceive to be Berkeley DB's decline?

Don't tell that to the OpenLDAP people. [openldap.org]

Re:Oracle already owns an open source database (1)

Antique Geekmeister (740220) | more than 4 years ago | (#29707301)

I'm looking at the Berkeley DB 4.8 release notes: thank you for the pointer. The claim in your first URL that Berkeley DB "requires zero administration" has me laughing so hard, I almost sprayed something nasty on my keyboard. And this release is roughly 18 months lsince the last minor version release: please exuse my surprise. Now, reviewing the release notes, is there anything that we should care about? Not much. db_sql might be handy, if anyone used Berkeley DB anymore. The only reason to have it is to _extract_ data from Berkeley DB and transfer it to a better database.

You have a point about SQLite. The rather thorough discarding of Berkeley DB in Fedora and RHEL seems to have been in favor of SQLite: the only thing left in Fedora 11 that really requires it seems to be RPM itself, which strikes me as foolish.

Re:Oracle already owns an open source database (1)

jjohnson (62583) | more than 4 years ago | (#29704647)

MySQL has been forked several times since Sun bought them. There's now a confusing welter of forks of MySQL, and no one is sure which is the 'real' one anymore since the original owners of MySQL AB are responsible for one of the forks.

MySQL has an ace in the hole, however: a HUGE install base in ISPs. MySQL is THE default database you're going to be exposed to for web hosting, and a perception that it's being allowed to whither on the vine will kick up a backlash against Oracle (just as it did with Sun).

Re:Oracle already owns an open source database (1)

physburn (1095481) | more than 4 years ago | (#29707481)

I use both Berkeley DB and MySQL. Berkeley is still open source, still regularly updated, and works well enough. Its more like a disk backed hashtable than a full database though. MySQL is an eccential part of the current popular open source platform. Its the M in LAMP. So its very important it stays safe. Fortantuately becuase Open Source is Open Source, if Oracle break MySQL, another term could developed a new fork of MySQL.

---

Databases Feeds [feeddistiller.com] @ Feed Distiller [feeddistiller.com]

what competition? (2, Informative)

mcover (1653873) | more than 4 years ago | (#29704541)

I don't see why the EU is worried in the first place. First of all MySQL could never compete with Oracle's DB. They will never compete and never have. Completely different use-cases. Apart from that, I'm still using PostgreSQL and if i had an app specifically designed for MySQL, I'd go with Drizzle(fork).

Re:what competition? (2, Insightful)

jadavis (473492) | more than 4 years ago | (#29705205)

Completely different use-cases.

There's some truth to that, but that's hyperbole. The use cases are not disjoint.

Additionally, MySQL may represent a general shift away from the traditional SQL architecture towards things like MySQL and non-SQL database systems. In some sense, Oracle is not just fighting to keep its customers on Oracle, it's trying to keep customers using traditional SQL systems.

I happen to think that the traditional SQL architecture is a pretty good one, and much better for general purpose development than the alternatives. But some people disagree, and Oracle doesn't want to lose part of the market, even if that part of the market is misguided.

Re:what competition? (0, Flamebait)

mogul (103400) | more than 4 years ago | (#29705833)

EU think they have to! They fried MS, and now they think they have to do over and over again with every big corp.

Not so fast (1)

FallLine (12211) | more than 4 years ago | (#29704777)

Oracle can effectively kill MySQL. Although anyone can fork MySQL and support it in theory, the reality is not this simple.

To continue to be a force in the marketplace (albeit at the low end), the maintainers will need real funding to drive and manage the development. The primary problem is that essentially the only business models that actually works for open source companies is the hybrid licensing model and this option would almost certainly be unavailable to them.

The support business model frankly isn't a great way to make money and have enough margins to pay for expensive and continued development efforts. No doubt someone could make money creating patches and offering support, but I do not believe that would cover continued improvement.

A company would also need to emerge as the dominant MySQL offering and convince would-be users that they are viable.

Perhaps IBM or some other large company would subsidize development, but I would not want to count on it.

If Oracle decides to bury MySQL, I would bet that it disappears from the marketplace after a few years for all intents and purposes. Sure, you could still find a handful of hackers playing with it, making interesting hacks, etc... but most users will simply move on to greener pastures.

That being said, I do not believe this is an antitrust issue and the EU should STFU. :-)

Re:Not so fast (1, Insightful)

ClosedSource (238333) | more than 4 years ago | (#29705039)

I got modded down to negative numbers last time I suggested this but perhaps the European Commission is looking for a new revenue stream since it looks like MS won't be supporting them anymore.

Re:Not so fast (1)

jadavis (473492) | more than 4 years ago | (#29705285)

Although anyone can fork MySQL and support it in theory, the reality is not this simple.

Exactly right.

The primary problem is that essentially the only business models that actually works for open source companies is the hybrid licensing model

That's just not true. PostgreSQL doesn't use hybrid licensing. Linux doesn't, either. Yet both projects seem to attract valuable people and turn them into great communities that accomplish a lot.

I think your argument is too focused on money being sent to some central authority, which distributes the money to various people to make a database system and provide support. However, that's not the only way that money can move, and money is not the only incentive for people to accomplish things.

Re:Not so fast (1)

FallLine (12211) | more than 4 years ago | (#29705837)

Linux doesn't, either.

Much of the Linux's kernel development is subsidized by several large companies with an interest in seeing it grow as platform and a handful of embedded-device type companies. Then you have distros like Redhat that make money by selling update services essentially. It's an unusual model to fund development and not one that is applicable to most other software markets. Its success as a product is almost entirely in the server-space and this in part an accident of history (in my opinion), thanks to the web and the emergence of commodity hardware.

It's fine for what it does (web servers, devices, non-mission critical servers, some scientific computing, etc). That said, I do not see it making any real in-roads on the desktop, replacing microsoft's whole network infrastructure (Active Directory, Exchange, etc), and so on and a big part of the reason for this is because the license in the entire linux eco-system precludes the ability to build a successful for-profit COMPANY to drive that kind of end-to-end development effort.

PostgreSQL doesn't use hybrid licensing.

PostgreSQL is not a company and they have had many years to acquire commercial supporters (like Redhat, EnterpriseDB, etc) and make their products more user-friendly. I actually like PostgreSQL these days and I'm fairly impressed with their tools.

I suspect that EnterpriseDB may have played a substantial role in their recent enhancements (they're certainly one of their biggest financial contributors) and they enjoy the ability to sell proprietary enhancements to PostgreSQL. Regardless, I do not see it as a model that can be reliably replicated to maintain MySQL (not to mention create whole other new products and services).

I think your argument is too focused on money being sent to some central authority, which distributes the money to various people to make a database system and provide support. However, that's not the only way that money can move, and money is not the only incentive for people to accomplish things.

Money is the primary reason substantial groups of people organize to work on projects with focus and direction and to a schedule that they generally would not otherwise follow. Some software projects just need a handful of devs working on it at their whim, but there is a whole world of software out there that requires so much more (more developer hours, more focus, more insight, more varieties of contributors, i.e., technical writers, etc)

Granted, there are some exceptions, but that's just what they are: EXCEPTIONS. I do not think you can reasonably extrapolate from a handful of barely comparable situations and say the mere possibility of support or no-business development model means there is no cause for concern.

I do not think it is an accident that those few open vibrant open source products have been developed by companies that made money almost entirely by selling proprietary rights (dual licensing) or have been subsidized heavily by some other business for other reasons.

At the very least, if Oracle drops MySQL it will impose a substantial delay and create great uncertainty until some kind of organization can coalesce, acquire support, convince users of its viability, attract developers, etc. The very fact that a couple hundred people can fork it can actually harm this effort.

In short, I do not see GPL3 is being a magic bullet for the survival of products just because anyone can fork.

Re:Not so fast (2, Informative)

jadavis (473492) | more than 4 years ago | (#29706551)

Granted, there are some exceptions, but that's just what they are: EXCEPTIONS.

They aren't just "exceptions", you have hand-waved away a majority of the free software people use.

I don't know of any free software compiler that uses that approach (not gcc, ghc, python, ruby, perl, etc.); nor any OS (GNU+linux, freebsd, opensolaris); nor any desktop environment or GUI tools; nor any browsers or email clients; nor text editors/IDEs. For database systems, MySQL and BerkeleyDB do, but postgresql, firebirdsql, and sqlite do not. Let's say that OO.o does, as well.

When I think about the volume of quality software that I use, the part that uses a dual licensing model is there, but it's not the predominant portion. For one thing, any dual licensed software project requires that you sign over the rights if you are an outside contributor. Not many open source projects do that, because it generally eliminates outside development except from some special cases.

the mere possibility of support or no-business development model means there is no cause for concern

There's plenty of cause for concern, as with any project using any model. Ultimately, a lot of things need to line up for a project to really take off and sustain itself. The "hybrid licensing model", however, is not the only way to do it.

or have been subsidized heavily by some other business for other reasons

That is very common, and it's a very different model than the "single company plus dual licensing" model. I would also add that it's often many businesses. It's probably a lot better in many cases -- PostgreSQL and Linux are both backed by various companies. I don't think you should marginalize this as a model for a successful project.

Google contributes heavily to MySQL, yet they are a second-class citizen in mainline MySQL, because they have to sign over rights to get improvements accepted. I don't think this state of affairs will last very long -- they will throw their weight behind one of the forks, and become a first-class contributor to the project, among others.

I do not think it is an accident that those few open vibrant open source products

Which few? I use a lot of free software from vibrant projects, and a lot of it quite simply does not follow your "hybrid licensing" explanation at all.

Re:Not so fast (1)

FallLine (12211) | more than 4 years ago | (#29710877)

They aren't just "exceptions", you have hand-waved away a majority of the free software people use.

I was talking about those really vibrant projects with large and sustained development efforts. You know, as in, those that have a bunch of full time developers, QA teams, technical writers, and so on. I admit there are a large number of smaller projects and some of those have significant and unique value in this world. However, there is a big difference between, say, DRH working part-time on sqlite (never mind that it's a BSD-style license) and TrollTech (now Nokia) paying a bunch of developers to make sure that not only does their core code work, but that it actually works in all the key platforms, that it's well documented, that it can be easily installed and plugged into Visual Studio and other environments, and so on. I see a difference in those sorts of projects. I believe that MySQL, as a critical application that is widely used and used by more average users, needs to fall more into the latter than the former. In other words, it's not enough to say that a handful of hackers will make cool changes on their own time. Someone needs to ensure that it fits together, that it's safe, that it can be readily installed, that it's clearly documented, and so on. Almost as importantly, users need to be convinced of this for mainstream acceptance.

The way I see it, there are really just a handful of large vibrant projects like this in the open source world and they are almost entirely either funded by the dual-licensing model (for-profit) or as a non-profit organization that receives large and sustained donations from companies that have a vested interest in seeing the project succeed.

To be fair, I should explicitly mention a third model which is basically a company like RedHat that primarily makes their business by providing a certified packaged product and updates (which, I think, is very different than the "support services" envisioned by idealists) for businesses that value reliability over the bleeding edge. This model fundamentally depends on the users needing some entity to organize and bless the release because there is substantial complexity and uncertainty in the open source world. I do not think this model is that applicable to MySQL, given that they are essentially just dealing with one package and that it does not constantly need to be adapted to accommodate the latest and greatest hardware. It definitely limits what can be charged and is potentially marginalized by 3rd parties that perform the same function. That said, it can be a mission critical application for some businesses and some might be willing to pay for that sense of security. I would not want to build such a business though or rely on it for my livelihood.

For one thing, any dual licensed software project requires that you sign over the rights if you are an outside contributor. Not many open source projects do that, because it generally eliminates outside development except from some special cases.

You are confusing cause and effect. Those dual-licensing projects are dominated by the inside team because their scale and scope requires that kind of organization for most efforts and it's not worth a lot of time trying to involve the community in major efforts (which is not so different than the subsidized non-profit model), not so much because people are philosophically opposed to making contributions that might get used in a closed-source project. If you contribute code to a dual-licensed project your code will still be available under the GPL to everyone thats want to do an open source project. It simply gives a handful of closed-source developers an opportunity to build a downstream product with it (almost never any kind of competing product), but their financial contribution strengthens the open source efforts that you care about.

I don't think you should marginalize this as a model for a successful project.

Linux is a special case because it is a true platform for hardware vendors and device makers to sell their own products. Even there, there are limits to how much they can justify contributing to extending the platform to modern general purpose desktops (e.g., Windows and Mac OS), business networks (Active Directory, File sharing, and so on). I know you can and probably will point to ldap, samba, whatever the latest and greatest window manager is, and so on, but I see a huge difference between them all. This is basically the 80/20 rule at work, i.e., ~20% of the effort has been put in with maybe ~80% of the results. Unfortunately, that's not enough to compete against mature products when the overwhelming majority of users that have better things to do with their time then work through all the issues (presuming they even have the requisite skills).

We can argue about this back and forth till we both turn blue in the face, but I've said my piece. Thanks

Re:Not so fast (1)

jadavis (473492) | more than 4 years ago | (#29711597)

You make some interesting points, but I think we'll have to agree to disagree, mostly on matters of scale and importance.

One clarification:

not so much because people are philosophically opposed to making contributions that might get used in a closed-source project

That's not what I was trying to say. Any BSD software could potentially be used in a closed-source project, and lots of people contribute to BSD projects.

But I would not contribute to a project where I had to sign over my copyright unless I had a very clear reason for doing so.

Re:Not so fast (1)

FallLine (12211) | more than 4 years ago | (#29712117)

That's not what I was trying to say. Any BSD software could potentially be used in a closed-source project, and lots of people contribute to BSD projects.

But I would not contribute to a project where I had to sign over my copyright unless I had a very clear reason for doing so.

Why not? The reasons for the company wanting this are pretty clear. They need that to do any kind of transaction not compliant with the GPL license and their current-day business model depends on it. I suppose they might try to carve out a specific license for themselves from each contributor ahead of time, but with a large number of small contributors it would be hard to manage and presents a lot of potential legal issues for themselves and their clients (especially if they need to change their strategy in unforeseen ways).

Worst case, the company decides they want to become a proprietary software company at some point in the future and make a private fork from the GPLd codebase (or even offer it under BSD terms instead). That possibility seems generally unlikely to me and your code will still be available under the GPL terms (they cannot remove what's already out there). That aspect of it is not so different from contributing to a BSD-style licensed product, except that only THEY or someone whom they've assigned rights to can do that instead everyone (including yourself). If you're a GPL-leaning person, this seems a lesser "problem" than contributing to a BSD-style licensed project in terms of its practical effects.

Do you another specific reason or is it principle of it that bothers you? How is that any different in practical terms from contributing to a BSD-style project (and one that will in all probability behave as if it's a GPL or at least LGPL project)?

Re:Not so fast (1)

jadavis (473492) | more than 4 years ago | (#29720057)

Do you another specific reason or is it principle of it that bothers you?

Let's say I want to implement the same or a very similar patch for a different project. If I've seen the previous patch (because I wrote it) but I signed all of my rights away, maybe they can make a copyright claim over any similarities between the two implementations.

I should say that I don't really understand copyright well enough to know whether the above makes any sense legally.

It's mostly principle, but there are practical differences. I feel much more comfortable just licensing code as BSD or GPL than actively signing my own rights away. Of course, if I'm well paid, that is a different story.

mysql "community" == sun employees (1)

farble1670 (803356) | more than 4 years ago | (#29705323)

there are lots of kinds of open source project. some thrive on the mind share of individuals. on the other end of the spectrum are those that are backed 100% by employees that are paid to be involved in (to run) the community. both are perfectly valid and valuable.

sun-mysql is the latter. anything that affects the sun-mysql employees (greatly) affects the community. mysql couldn't exist without the involvement of those people, people that will be under the thumb of oracle.

also, you have to look at motivation. the mysql CEO is most certainly looking at a fat bonus as soon as the deal goes through. there's a pool in the bakyward of his mansion that needs to be funded.

Re:mysql "community" == sun employees (0)

Anonymous Coward | more than 4 years ago | (#29709657)

the mysql CEO is most certainly looking at a fat bonus as soon as the deal goes through

Do you have any evidence to back this claim up?

I worked for Mårten for 4+ years. I found him to be a very intelligent and capable person who genuinely seemed to care about FOSS, the community, and the people working for him.

(BTW, he's not been MySQL CEO since April 2008, when the position ceased to exist, and he's not worked for Sun since April 2009.)

During that time, I never found any reason not to take him at his word. If he says he has no significant financial interest in the outcome, I'm prepared to believe him until I'm shown hard evidence to the contrary. And the fact that you can type the word "certainly" certainly doesn't provide it.

EU has a point (0)

Anonymous Coward | more than 4 years ago | (#29705435)

~1 year ago, sun microsystems acquires mysql, the most popular and most used database for small to medium size deployments. mysql: a company with $100m in revenue, with a purchase price of $2b dollars. at the time, sun had it fingers in many other databases, including heavy involvement with postgres and derby, and others. by all accounts sun overpaid. the response to this is that mysql was acquired for it's open source value, and for it's community.

this year, oracle, the #1 database vendor, acquires sun microsystems, thereby acquiring mysql.

some conspiracy theorists might suggest that the only reason sun acquired mysql was to make itself more attractive as a takeover target. others might go as far to suggest that it was all a plot from day one between sun and oracle to squash mysql without drawing direct attention. would anyone have batted an eye if oracle had acquired mysql directly? maybe not, but it would surely be getting more attention then it is now.

Berkeley DB back to first tier please! (3, Interesting)

IGnatius T Foobar (4328) | more than 4 years ago | (#29705733)

Once upon a time, MySQL supported the use of Berkeley DB [sleepycat.com] as one of its back end storage engines. Then Oracle acquired Sleepycat Software, the makers of Berkeley DB (which was, and still is, open source). MySQL didn't like the idea of Oracle controlling their back end, so they phased out its support [linux.com].

Now it doesn't matter anymore. Oracle is going to own MySQL and Berkeley DB. In my opinion, Berkeley DB is the finest storage engine on the planet. Either with a relational/schema layer on top of it (like MySQL), or all by itself (in which case it's simple key/value pairs), it is insanely reliable and its performance is excellent. I can't say enough good things about it. So how about it, Oracle? Can we get these two great pieces of software together again?

Re:Berkeley DB back to first tier please! (1)

Enleth (947766) | more than 4 years ago | (#29706227)

Take a look at Tokyo Cabinet [1978th.net] then. Better in just about every technical aspect - it only lacks in corporate backing and marketing, probably because the author is a die-hard programmer, not very interested in those aspects of project management. Otherwise it's absolutely brilliant. Oh, and it comes with an optional, very simple and fully ACID server for network access and cuncurrency and a full-text search engine.

Re:Berkeley DB back to first tier please! (0)

Anonymous Coward | more than 4 years ago | (#29709353)

Once upon a time, MySQL supported the use of Berkeley DB [sleepycat.com] as one of its back end storage engines. Then Oracle acquired Sleepycat Software, the makers of Berkeley DB (which was, and still is, open source). MySQL didn't like the idea of Oracle controlling their back end, so they phased out its support [linux.com].

If that had actually been our motivation, we would have axed InnoDB as well.

As Brian Aker says in the linux.com article you cited (but apparently didn't actually read), we were *already* planning to drop BDB due to lack of interest; something like 0.03% of all MySQL users who responded to our polls said they were actually using BDB, and most of those indicated that it was not mission-critical.

Re:Berkeley DB back to first tier please! (1)

IGnatius T Foobar (4328) | more than 4 years ago | (#29733679)

Assuming that you are actually a Sun employee and not just some anonymous troll ... regardless of what the installed base was in the past, it should be implemented now. Berkeley DB is the finest storage engine in the world, and it belongs in MySQL. Oracle's imminent oversight of both code bases presents a good opportunity to make it happen.

It's a trap! (1)

CuteSteveJobs (1343851) | more than 4 years ago | (#29708185)

Having used MySQL in an enterprise computing environment, I reckon even Oracle's worst enemies would also urge the EU To Approve Oracle's MySQL Takeover. ;-)

This is why LAMP should be LAPP (1)

petrus4 (213815) | more than 4 years ago | (#29709949)

Linux, Apache, PostgreSQL, PHP.

Postgres is BSD licensed. Even if the parent company is the victim of a hostile takeover, that means you can fork the existing codebase, and keep developing and using it. It also means that it doesn't have the viral aspect of the GPL, either; so it's more business friendly as well.

Re:This is why LAMP should be LAPP (1)

Zontar The Mindless (9002) | more than 4 years ago | (#29710467)

Postgres is BSD licensed. Even if the parent company is the victim of a hostile takeover, that means you can fork the existing codebase, and keep developing and using it.

Which you can also do with something that's GPL. Such as MySQL... and the Linux kernel.

 

It also means that it doesn't have the viral aspect of the GPL, either; so it's more business friendly as well.

Only if by "business-friendly" you mean "easy for businesses to take while giving nothing in return".

Re:This is why LAMP should be LAPP (1)

petrus4 (213815) | more than 4 years ago | (#29711033)

Only if by "business-friendly" you mean "easy for businesses to take while giving nothing in return".

Paranoia about reciprocity, again. You know, it's interesting how few of the people I see that support this meme, actually write code.

Re:This is why LAMP should be LAPP (1)

DragonWriter (970822) | more than 4 years ago | (#29722781)

Only if by "business-friendly" you mean "easy for businesses to take while giving nothing in return".

PostgreSQL itself -- which is financially supported largely by businesses making money off it, particularly EnterpriseDB -- and SQLite -- likewise supported by a bunch of businesses using it to make money -- are both proof that, even if this might be true in theory from a certain viewpoint, its not necessarily how things work out in practice. The reason is pretty obvious, a licensing scheme (which a dedication to the public domain like SQLite's amounts to just as much as PostgreSQL's use of BSD-style license does) that enables businesses to use it in their own closed products also gives the same businesses an incentive to see the open source product thrive, remain as compatible as possible with the implementation in their closed products, and have an active user and development community, since all of that keeps the businesses costs down. Which gives the businesses a strong financial incentive to give something back -- which the businesses profiting from PostgreSQL and SQLite do, both financially and terms of code committed back to the open project.

Check for New Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...