Go down Moses....
...let MySQL go...
I just wanted to say that first.
Richard Stallman demanded the EU cut MySQL loose from Oracle yesterday in an open letter to Brussels' competition supremo Neelie Kroes. The self-described software freedom activist's intervention came just a day after MySQL founder Monty Widenius made a similar call, saying that Oracle could offset the EU's go-slow examination …
...let MySQL go...
I just wanted to say that first.
You don't explain how Oracle would prevent a fork.
Nice FAQ here:
Stallman has been a force of good for software and most of the time knows what he's talking about. I think he's right on this one, because even if MySQL is forked, the 'brand' is now synonymous with free databases, so a lot of people will continue seeking it, despite whatever the new fork aims to achieve (a bit like gNewSense really). Not to mention vested interests.
The Commission's remit is to investigate anti-competitive behaviour. It's about choice not "freedom". In fact Stallman's line is often horribly anti-competitive itself.
Erm lets get this straight, one tech company buys another, Richard Stallman doesn't like it so he moans to the EU.
What gives him the right to complain about this, the MySQL code base as it stands is GPL and has to remain GPL unless it's totally re-written.
As the very nature of the GPL encourages sharing, there is nothing anyone can do to prevent a fork unless they started using trademarked names.
Just like Oracle killed off InnoDB ... oh, wait...
Even if Oracle DID "kill off" MySQL - there are branches such as MariaDB or you could just say "stuff it" and use PostGRES. I'm not really sure how MySQL got to be the almost defacto OSS DB server anyway - PostGRES has always been a bit better but it doesn't have a cute dolphin I guess.
Stallman has been supported by the US taxpayer for a good chunk of his adult life, especially during his formative years working at MIT for DARPA and it's offshoots. His whole viewpoint on "free software" is distorted by the fact that he has never had to really work and make money - there were always US government grants to MIT labs to keep him employed, and he used that as the basis to declare that all software should be free (and later, all intellectual property).
Stallman stands as the antithesis of actually getting paid for your work. His single handed efforts to undermine Symbolics (a commercial offshoot from MIT) destroyed the LISP-machine, as he single-handedly reverse engineered their code and gave it to competitors for free - he hated the fact that Symbolics was set up to actually make a profit (and thus employ people and help them pay their mortgages, pay for food, etc. - all things HE didn't have to worry about sitting in MIT labs on the US government payroll!).
There certainly IS a niche for "free software" - however, MySQL hasn't been "free" for a long time. The single best indicator is that Sun aquired MySQL by purchasing THEIR STOCK. Now, you don't need to stock and to be incorporated unless you are in this business to make money. So MySQL was all about making money since before the aquisition, and now they will simply be making money for Oracle rather than themselves...and Stallman apparently can't stomach that.
Hey Stallman, there is always Postgres isn't there? You know, the ACID compliant SQL database that IBM donated to the freeware community a decade ago, one that scales perhaps even better than MySQL? Or does the fact that IBM donated it make it tainted?
Can't see how trademarks get around GPL?
You can't fork MySQL and *call* it MySQL. There is absolutely no barrier to creating a new GPL fork providing it doesn't infringe Sunacle's trademark. The other barrier to forking is those who, for commercial reasons, would want to change the licence as they will never own the copyright on the code and have no authority to change the terms under which the code is distributed. In other words, whatever your fork would be called would have to be pure GPL and any code changes must also be GPL as that is the only authorisation you have for distributing the original source. Stallman should be pleased since he keeps harping on about one licence to rule them all. See the Monty Program fork MariaDB for niggly little details.
What I suspect Stallman wants is what he can never have. He can have the entire codebase, mess with it, release half-arsed versions of it and add code from other sources then release the results under the existing terms of the GPL, a licence he himself considers the epitome of freedom. He just can't have the MySQL name. It's exactly the same situation as Mozilla's browser source: Change with the default build (even a custom mozconfig triggers this) and you cannot brand it "Firefox" as it's a trademark. Stallman should recognise this. One word: Iceweasel.
This is yet another "look at me! I'm St. Ignucius, saviour of free code!" moment and his hypocrisy is showing in the way he's not satisfied with the application of his own damned licence. You wrote it, Richard. If it's deficient in some way, you only have yourself to blame. Expect a GPLv3.x with anti-Tivo style trademark clauses shooting itself in the foot again. There is *no way* Sunacle will surrender that trademark, just as there is no legitimate way the EU can take it from them. In fact, that trademark is probably worth more to Sunacle than the entire codebase.
If you want to get clever with your fork's name, UnnameDB or SQLEWNN seems logical in that it conveys the loss of the right to use the MySQL name whilst keeping the codebase and upgrade path intact.
Beer. That's not free either.
That FAQ boils down to "you CAN fork MySQL, but you wont be able to call it MySQL". So Stallman should STFU. As usual.
There are very few precedents for projects whose code has been GPLed, but where the copyrights to the codebase have always been wholely owned by one commercial entity (this will, in fact, be one of those precedents - so the outcome could have far wider repercussions than just this).
MySQL was licensed to commercial software developers for integration with proprietary software products. This licensing right is what Oracle will, essentially, be acquiring. These protective agreements avoided the need for proprietary projects to assume any sort of GPL status, themselves, simply because they used a GPLed product.
A fork project would not have the commercial rights to do that sort of licensing. Only Oracle would.
Personally, I think Oracle will behave, since - in a world where every electronic device is gradually turning into a computer - a growing number of devices will need fast, light weight databases. The licensing opportunities could be gigantic, once you stop thinking about a 'database' as being something that runs the company payroll. The problem, really, is that that's exactly what Oracle have traditionally regarded a 'database' as being.
If the GPL works, then Stallman is wrong.
If the GPL doesn't work, then Stallman is wrong.
In any case, MySQL can fork and has forked.
Move on, it's just a database. I thought that was pretty much a solved problem for 99% of use cases. The real fight is in the clouds.
Spot on about the Symbolics shenanigans. However, IBM didn't "donate" the Postgres source code, it originated at University of California, Berkeley.
Monty knows very well that Oracle will not stifle MySQL - in fact, he wants it cut out of the acquisition to *benefit* his MariaDB fork because he knows the best nudge for recognition and adoption is for MySQL to become an uncontrolled orphan.
Just ask him - he's probably not shy to admit it.
(posted as a/c to avoid getting a slap)
If Stallman is so against property rights, he can hand me his wallet.
Oh, wait, on second thought, there's nothing in it and I don't want to deal with the toxic waste. Nevermind.
Mine's the one with Hayek's "Road to Serfdom" in the pocket.
"If Stallman is so against property rights, he can hand me his wallet."
Wasn't handing you gcc on a plate to make a fortune with enough for you? Greedy bastard.
Well, just switch to PostgreSQL then. It's all MySQL has always wanted to be, but has only recently approached to.
I've always been amazed by the way MySQL suddenly took over as lead "free DB" when Postgres had the leading edge in most things, including those basic RDBMS things like "Transactions". Which, by the way, MyISAM tables can't handle thanks to Monty. Sheesh.
Oracle didn't kill InnoDB, I doubt it'll kill MySQL. After all, killing InnoDB would've taken out the transactional capabilities from MySQL (except for the BDB engine hack, but I haven't seen that in years) and no sane DBA would use a transaction-less DB.
you are all thinking and caring for proprietary developers and software vendors. How about the end user ?
Please stand-up each of you in turn and explain (preferably using solid arguments) why should we have to be locked-in, WGA'd, audited, enslaved by the software we're buying ? Come on, speak up loud ! All you mention is the word free but you're all carefully avoiding to use the word freedom.
FOSS is an escape hatch, some sort of protection circuit that allow end users to escape from greedy vendor lock-in and Stallman fights to make sure those greedy bastards can't take it away from you and from me. And please don't come up and tell us the world would be over and the software will perish without innovative proprietary software vendors.
You're not forced to accept my view, just read a bunch of relevant EULAs and TOSes
Robert Hill, taking a leaf out of the tabloid scandal journalist's manual, writes...
"His whole viewpoint on "free software" is distorted by the fact that he has never had to really work and make money - there were always US government grants to MIT labs to keep him employed, and he used that as the basis to declare that all software should be free (and later, all intellectual property)."
Nonsense! Stallman's recent statements on copyright in a more general sense - not limited to software - admit the need for certain restrictions of various kinds of work, such as those which could be altered, republished, and then used to misrepresent the content of the original work and its authors. Although it's fashionable for various people (including a bunch of people who regularly comment on Register articles) to cherry-pick stuff out of scientific papers to support their fantasy position on something or other, Stallman is clearly at odds with anyone advocating a complete free-for-all on content.
And "intellectual property" is your term, not Stallman's. I imagine that while he may be against many forms of patent, he is quite likely to recognise trademarks as a valid tool of doing business.
[More tittle-tattle deleted]
"There certainly IS a niche for "free software" - however, MySQL hasn't been "free" for a long time. The single best indicator is that Sun aquired MySQL by purchasing THEIR STOCK. Now, you don't need to stock and to be incorporated unless you are in this business to make money. So MySQL was all about making money since before the aquisition, and now they will simply be making money for Oracle rather than themselves...and Stallman apparently can't stomach that."
MySQL is "free" in the Free Software sense. However, Stallman is suspicious about the likelihood of further development and whether it will remain competitive. I don't think that there's much danger, and that the only people who will lose are those who make proprietary software - they'll be nagged by Oracle endlessly now to upgrade to vanilla Oracle products.
And a "niche" for Free Software? Google runs on the stuff: MySQL, Linux, with and without the GNU part of GNU/Linux (Android is without). You might want to clue up on the whole free as in freedom thing. (Beer icon in place to remind you that it isn't about free beer - sheesh!)
"Hey Stallman, there is always Postgres isn't there? You know, the ACID compliant SQL database that IBM donated to the freeware community a decade ago, one that scales perhaps even better than MySQL? Or does the fact that IBM donated it make it tainted?"
As someone else pointed out, PostgreSQL didn't originate from IBM.
You know, a lot of the concern is about the competitive landscape once Oracle starts pulling strings and presumably limiting the OEM deals that might well be in place around MySQL, and anyone - even Stallman - has a right to point this out to the competition commissioner for what good it may or may not do. You may doubt the value of doing so, but by maintaining a gossip magazine tone (and failing to exercise familiarity with what Stallman actually does stand for), we get to doubt the value of your critique.
Personally, I'm fairly fed up with the primary school level of debate whenever people have stuff to say about Stallman. It's yet another sign of the dumbing down of society in general as people gawp at their celebrity-drenched television coverage of non-news and vote the same cretins into government again and again.
... it is, but it's arguable that therein lies the road to riches.
JBoss didn't have clear ownership of its code but that didn't stop Red Hat paying top dollar for the trademark and all that went with it. By all accounts they are doing better now out of the acquisition than many originally anticipated. Most people agree that "licensing" of supported configurations is the way to go (it hasn't hurt Red Hat) and you don't need to own the copyright on GPL source to do that.
Of course "licensing" is the sort of thing that makes RMS' head explode.
Yes Stallman is a bit on the fringe but he does serve to balance out pure greedy commerical interests with the good of the public. Seems to me many on here are open hostile to the public owning anything. Tragedy of the commons applies rather poorly to bits that potentially last forever with very low maintence. Socialism across a whole economy may be a mess but in this small realm it has done well so far. Besides I find it ironic that biggest defenders of an economic status quo where %1 of the people own 99.9% of resources are not the %1 but their salaried middle manager lackeys. Our economic system is currently not sustainable just like it wasn't in the early 20th century when the robber barons owned everything. Eventually another Teddy R will come along and go after the trusts or else the whole house of cards will collapse again like it almost did a year ago.
First of all, I have a certain suspicion that your handle is false advertising.
Anyway, NO, the helpfully provided FAQ does not just boil down to "fork but don't use the name". I too thought that was the case, before reading it. Go and read again, since you clearly did not understand. Pay attention this time, specially to the point about exclusive copyright and the ability of the forkers to use the code in a commercial manner with non-GPL code.
So, no, they can't kill the code. But they can kill the commercial product.
Not that I care, I use MySQL in an academic setting and it works great for the things we do.
I think this text is quite analytical about the whole question.
It's called EU-sql. When queried it performs no transactions at all...
(Actually since English overloads that word I'd say "Libre", but that only adds precision to the description).
And some of you are very unpleasant.
Clever. Of course open source serfs will be less likely to trust the sell out artist this time.
MySQL lost what oss heart it had when sold to Sun. I guestimate that postgres has many more shrewd brains behind its development now.
I hear there is a new fork of mysql called EUsql.
When queried, it performs no transactions at all.
How easy it is to punch others on the head when you hide behind a mighty anonymous nick. People should stop behaving like children who just lost their launch box filled with candy. RMS has done a wonderful job for all of us using FOSS-systems in our everyday computing, GNU/Linux. The very same system that many MS fan boys have written off for dead a gazillion times. It's as if they can't touch the system they go for the people that uses it. Go shils, go. MySQL stays GLP'ed software. No matter how much shils curses about RMS.
fscked by SHA-1 collision? Not so fast, says Linus Torvalds