Wikipedia Moved To MariaDB 5.5

Peetke writes "As we all know Oracle is not the biggest friend to the Open Source Community. Long standing OSS supporter Wikipedia has now moved from an optimized fork of MySQL 5.1 to MariaDB 5.5, for both its English and German sites. Wikipedia expects all other languages to follow within a month. Performance-wise, this move has no big implications, but it will ensure our biggest community database will live long and prosper."

Comments

seriously?

[Anonymous Coward]

Soo they don't like Oracle too?

Re:seriously?

[Ultra64]

Right, they couldn't possibly have a good reason for doing it.

Re:seriously?

[jellomizer]

If they have a good technical reason, or support reason, sure... But just because they are like Ohhh Oracle is Bad, I better switch, is just stupid.

Re:seriously?

[Anonymous Coward]

It's a waste of money when there database is free.

Re:seriously?

[rvw]

So an organization who asks for donations, waste their money changing Database systems for the sole purpose that they didn't like the company that bought the old one, although they didn't show any signs that they are going to damage the product or make it worse for them in any ways? Sounds like a wast of donated money to me.

So you didn't RTFA???

For our most common query type, 95th percentile times over an 8-hour period dropped from 56ms to 43ms and the average from 15.4ms to 12.7ms. 50th percentile times remained a bit better with the 5.1-facebook build over the sample period, 0.185ms vs. 0.194ms. Many query types were 4-15% faster with MariaDB 5.5.30 under production load, a few were 5% slower, and nothing appeared aberrant beyond those bounds.

Better performance on such a heavy traffic site is neither a waste of time nor money! ;-)

Re:seriously?

[Anonymous Coward]

Since Oracle took over MySQL they've shown they routinely delay releasing patches for CVE security flaws for months until they can all be released together without documenting what fixes what issue. Several times updates have either ignore issues, removed fixes to earlier ones or in at least on case I remember applied a fix for an issue which failed to fix it and actually introduced a new one. This despite multiple FOSS projects (Debian,RHEL,MariaDB,Percona) having developed working patches separately which Oracle chose not to use.

They also don't disclose the details of many security vulnerabilities. That sounds reasonable on first glance, but it makes it a nightmare for sysadmins to assess whether it is worth system downtime to apply a patch, especially when that means upgrading to a newer DB version not tested against the application and which might break the application in several cases (for example due to the newer reserved keywords lists). A firewall or other measures may be sufficient to mitigate the threat, but that can't be assessed without seeing the details.

Re:seriously?

[jellomizer]

Yes, because there man hours needed to be put into this. Even if they are volunteers they could be working on doing something else that may have saved them money.

Re:seriously?

[Dancindan84]

Ummmm... that's not what happened. They weren't using a stock release of MySQL. They were using an old 5.1 fork that Facebook created and has been maintaining. They decided they wanted the enhancements that the newer releases offered, and had a choice of migrating to a newer release of MySQL, or migrating to a newer release of MariaDB. Either way, they were migrating and had to put forth the effort to do so.

Re:seriously?

[backslashdot]

As a WIkipedia donor .. I support this move. I got nothing against Oracle, but why would they use a database that is not the flagship database of the company? It's bound to cost more and lack features. MariaDB is owned by a foundation with goals more in line with Wikipedia.

Re:seriously?

[Anonymous Coward]

They aren't changing database systems. They are upgrading to the latest mainline version of the database they were already using. Don't be confused by the name change: MariaDB is a recent fork of MySQL where most (all?) new open development occurs. See MariaDB for the relevant history. Basically, "switching" from MySQL to MariaDB is like "switching" from OpenOffice.org to LibreOffice or from XFree86 to Xorg. MySQL got taken over by Oracle so the real development was forked with the new name of MariaDB.

Re:seriously?

[ShanghaiBill]

Even if they are volunteers they could be working on doing something else that may have saved them money.

Have you ever managed volunteers? They work on what they want to work on. You can't just reassign them to a task they don't like, or they will walk away and donate their time to some other organization. Most likely they had some MariaDB fanboi (or fangoil) who was willing to do this, but was not willing to upgrade MySQL instead.

Re:seriously?

[Mordok-DestroyerOfWo]

Not only this (but please mod up anyway!), but as far as I know MariaDB is compatible with plugins designed for a comparable version of MySQL. At least for my Django and PHP work this holds true. I mean, isn't this the reason most developers abstract the database library anyway?

Re:seriously?

[Anonymous Coward]

Clearly you've never done business with an Oracle rep....

Re:seriously?

[gmuslera]

If well MariaDB is backward compatible with MySQL, have some advantages on its own, like more choices for storage engines (i.e. Aria as a better myisam than myisam, xtradb instead of innodb, and others), and should have better performance in general than Mysql for the same equivalent version in the same hardware.

That Oracle is being bad right now with their concept of "open" (like suing Google for using Java) is an extra motivation.

Re:seriously?

[Synerg1y]

Welcome to the OSI community, we do things cause we think its cool and then wonder why we fail. We need wikipedia though, it serves a purpose beyond its maintainers, so either join the community and protest, fork over some more money, or... do nothing and hope it stays up.

Re:seriously?

[drakaan]

Most likely they had some MariaDB fanboi (or fangoil) who was willing to do this, but was not willing to upgrade MySQL instead.

Doubtful. More likely they wanted to be able to get decent community support for the forseeable future. Something that's not a given for a previously community-based software product that got gobbled up by a succession of commercial entities.

Oracle has gone to great lengths to make MySQL a second-class citizen to its own database in terms of support, and worse, they're not really getting the whole community part of why people used MySQL in the first place...or maybe they *do* get it and just want MySQL to go away so they can sell Oracle DB.

Either way, the folks at Wikipedia must have seen value in moving to a compatible, open-source, community-based database...just like the one they started with.

Re:seriously?

[Tarlus]

The effort put into this transition is probably not as drastic as you make it sound.

Re:seriously?

[greg1104]

MySQL got taken over by Oracle so the real development was forked with the new name of MariaDB.

No it didn't. MySQL was sold in a way that allowed Oracle to then purchase it. There's no reason to believe MariaDB will be any different. The fact that contributors to MariaDB have to assign their copyright to Monty's company says he's at least thinking about doing the same trick again.

Re:seriously?

[Anonymous Coward]

In the future, read the article before piping up with a stupid comment.

Re:seriously?

[hairyfeet]

Nice to see I'm not the only one to think that, because 1.-They weren't using Oracle in the first place, they were using their own fork of MySQL, and 2.- They didn't have any reason to give a crap since again, they had their own fork which its not like Oracle can send them a C&D for forking a FOSS program.

So if they would have said "Look we are switching because the features we made the fork for are being included in MariaDB thus saving us money" or something like that? Cool, makes sense, smart move. But doing this because of political bullshit? Let old Jimmy pay for that shit, because that is just stupid.

Re:seriously?

[hairyfeet]

Sadly that is true and why FOSS will ALWAYS suck for anything bigger than a project that can be done by a handful in a garage, its a problem I noted years ago and gave the name "Busted shitter problem".

You see if I ask for somebody to paint me a picture or sculpt me a bust or write a song for free? I will get several offers, some of which might even be really good. If I ask someone to come fix that overflowing shitter for free? Well I better get used to pissing in the sink.

What does that have to do with FOSS? Its actually quite simple in that for every decent programming job you have a dozen shitty jobs which is why companies like Apple, MSFT, and even Red Hat have to offer competitive salaries, because nobody wants to do a shitty job for free. But sadly when you are talking about a project run SOLELY with volunteers you just won't see those jobs get done, nobody wants to fix the busted shitters. Don't take my word for it, go look at the bug tracker of any major distro...do you have bugs that are TWO years old? THREE years old? Do I hear FOUR? The bug tracker will have bugs as old as the distro itself because those bugs are shitty and would take a hell of a lot of work to fix and thus don't get done.

In a way its a lot like communism in on the surface communism sounds great, I mean everybody working together to make their their lives and the world a better place? How could that be bad? Well what you end up with is a billion artists and nobody doing the job cleaning up the puke at the Chuck E Cheese. It got so bad that the USSR had to actually assign soldiers to "potato duty" which was all the shitty jobs they couldn't get the people to do, but large FOSS projects don't have that luxury.

I'm sure every programmer here will instantly mod me down because they don't like to think that their job is shitty, but c'mon dudes, it really is. Bug hunts, regression testing, writing docs, dealing with those little weird errors that just seem to pop up in corner cases and are maddening to track down...those are shitty jobs. This is why you see so many new releases in FOSS when the previous one hasn't even had half the bugs fixed so they are adding new bugs onto old bugs because writing new programs? That is enjoyable, making something with your own two hands is something we humans have enjoyed and took pleasure in for thousands of years, being the guy that has to widen the ditch because the sewage is backing up? Not so much.

Re:seriously?

[hairyfeet]

And what is wrong with that? Give old Monty some credit, the man managed to sell the product and keep it at the same time so at the end of the day all the corp ended up with was a now worthless name! That is fucking brilliant! How he managed to get the corp to agree to buy without a non compete clause i don't know, probably by being just that fucking slick, but its a trick worthy of playing the WB "sucker!" music at Oracle, damned smart if you ask me.

Re:seriously?

[tibit]

You need to be particularly thick, or just trolling, to think that upgrading a database system must only be done because of politics. Protip: just because the successor (MariaDB) is a fork of MySQL doesn't mean the political reason was very important. Not at all, it was merely a nice windfall. They needed to upgrade anyway.

Re:seriously?

[tibit]

Sorry hairyfeet, this rant has nothing to do with Wikipedia's choice to upgrade. You've been taken for a ride by a troll, hook line and sinker. They didn't upgrade because it was someone's pet project or because it was "cool". They upgraded so that they'll maintain a viable upgrade path for the future. For all practical purposes, MySQL's release model right now is broken, so if you're running a serious project, you can't just hope Oracle won't break it for you. Your rant is simply wholly inapplicable to the situation at hand, even though you may otherwise have a point.

Re:seriously?

[tibit]

This!

Re:seriously?

[hairyfeet]

I don't see why the FOSSie faction has a shitfit when i point that out, how often do we see a little guy manage to totally fuck over a big corp just by having the big brass balls to pull it off?

That is why i say give the man some credit folks, he managed to pull a scam worthy of "The Sting" on a huge corp and got a big fat check AND got to keep the thing he was selling! That is fucking brilliant! Give old Monty credit, his balls are big and plentiful to pull something like that off and walk away scott free with a big fat bank account, big and plentiful indeed.

Re:seriously?

[Anonymous Coward]

Volunteers? What volunteers? Most open source projects are *PAID*. You were a fool when you thought that 1) people were doing this just for fun and 2) people don't make money from it. Example: Google uses the Linux operating system all the time for their main search product, and have used it to the exclusion of all else since day 1. When people talk about Google, lack of funds is not anything anyone says. Last I checked, they were making bucketloads of cash on a daily basis. If I had their income for 1 day, my great grandkids wouldn't have to lift a finger for their entire lives. Do you really believe they are letting the Linux OS slide? Why do you think they re-brand Linux and call it Android for smart phones and tablets? Why do you think they re-brand Linux for their Chromebooks and call it ChromeOS? Do you really think they don't have at least 2 dozen paid people to work on the Linux kernel full time? Likewise RedHat. Likewise Oracle. Likewise Intel. Likewise IBM. Likewise Cray. Likewise at least a dozen other companies. There *are* thousands of volunteers, make no mistake. But there *are* thousands of full time paid engineers too. And that's just the kernel. Pick any other money making "GNU Free/OSS" project. Firefox has had mostly paid people for years. GCC too (and they are usually PhD candidates and post docs). Remember, the guy who started MySQL sold it to SUN Microsystems for $1Billion U$. He is the guy behind MariaDB, and can pay a dozen skilled engineers exemplary wages --out of pocket without any return even though he is making money-- till he is a grey haired little old man, and his grand kids won't have to lift a finger their entire lives. I laugh at your stupidity: "managed volunteers". Ha!

Re:seriously?

[hairyfeet]

Sorry tbit that you don't like the fact that like communism the entire premise of FOSS is built on a flaw, don't blame me that the emperor's balls swing in your face bothers you.

Do I ultimately give a shit WHY Wikipedia changed? Not really, anybody who has had a run in with the "Wikitards" and the deletionists really doesn't give a shit if Jimmy's toy goes down in flames.

What I DO care about is when politics and religion are allowed to overrule common sense, we saw that with communism then, see it now with libertarians not seeing how the end result will ALWAYS be feudalism, and we see it badly with FOSS because as one wise sage put it "Linux is always 3 to 5 years from being ready for the masses...only the 3 to 5 years never changes" and like it or not that is a fact and it ALL comes down to the busted shitter problem. That is why you see systems that are finally getting stable thrown out for alpha quality shit like Pulse and tossing KDE 3 for 4, why amateur hour bugs never seem to go away, why new versions just add new bugs to the old bugs, and why you will find placeholders where important docs should be...it ALL comes down to shitty jobs and the fact that humans just won't take truly shitty jobs for free.

Re:seriously?

[David Gerard]

No, MariaDB tested as fast or faster, or they wouldn't have made the move.

Re:seriously?

[David Gerard]

No, MariaDB tested as fast or faster, or they wouldn't have made the move.

Re:seriously?

[VortexCortex]

Sadly that is true and why FOSS will ALWAYS suck for anything bigger than a project that can be done by a handful in a garage, its a problem I noted years ago and gave the name "Busted shitter problem".

You see if I ask for somebody to paint me a picture or sculpt me a bust or write a song for free? I will get several offers, some of which might even be really good. If I ask someone to come fix that overflowing shitter for free? Well I better get used to pissing in the sink.

Which is why your "busted shitter problem" works the opposite way in FLOSS. Because when a FLOSS shitter breaks, it's not just you, it's a whole mess of people. Some of them are willing to pay to have the problem fixed, and some might even be capable of cleaning the shit off themselves. When ONE of them does fix the issue, then then everyone's busted shitter is fixed all at once. Compare this to a proprietary shitter than no one is allowed to fix but the shitter manufacturer: You have to wait on a specialist to come out with a fix, if they find it in THEIR interest to fix it... So, that's why Linux is better and faster than Microsoft is at patching OS vulnerabilities -- Linux, a successful project that runs damn near every web server on the planet, and powers the most smartphones as well, I might add. The many successful FLOSS projects that are bigger than a handful of devs does not completely obliterate your points, but makes you look pretty damn foolish, IMO.

Don't get me wrong, I agree that a core team of maintainers should be small. When starting out these maintainers are also developers. However, when the project gets bigger it's restructured so that devs get to keep developing and maintainers just merge and test and verify, etc. Lather rinse and repeat. Linux is successful because the dev became a maintainer quickly and let others do the dirty work. Protip: Linus doesn't write much code these days, but every kernel patch still crosses his desk. Ballmer and the late Jobs could only dream of such levels of control... Aside: What happens when Linus dies or quits? He's already set up the system of trust so that anyone can now replace him immediately.

This flexibility and scalability in structure is something that all companies should take a look into. Many are doing so. Some, companies are letting users fix their broken shitters for free to the benefit of all. Others claim control over all shiter functions, and thus become synonymous with their broken shit.

Re:seriously?

[JasterBobaMereel]

Unlike Communism FOSS projects and distros can make money, (See e.g. Red Hat, Canonical etc ..) and so can pay people to do the shitty jobs that otherwise would not get done

The USSR had for a large part of it's history a large army standing around with nothing to do, so yes they got them to do jobs that others did not want to do....see also the US army when they are not busy ....

Re:seriously?

[Man Eating Duck]

<Something something> communism <something something>FOSS

<more communism>

It seems like you haven't used proprietary software at all... I've seen a lot of QA issues like those mentioned in your rants in proprietary software, as well as OSS. On the other hand, I regularly use two very slick OSS projects both privately and at work: calibre and Sigil. Both are hands down the best option available in their category, proprietary or not. Nothing else even comes close. Both are maintained by extremely competent devs, have quick issue turnaround, and are relatively simple to run from source, as I have done to make (and contribute) a couple of fixes and improvements myself. In the case of calibre, millions of non-tech users are happily using it to catalogue their ebooks.

In your case, as it seems like OSS ate your dog, feel *very* free to look elsewhere. I've done so as well when I can't find anything that suits my requirements. There have been a few of your kind visiting the forums of those two projects. These people make incoherent, irrational demands, rant, won't listen to reason, and even refuse to explain what they mean so that people can help them. None of this is constructive for anyone. Although they're generally treated politely, we're frankly better off without them. Then you have people who bring rationally presented and relevant complaints to the table while behaving themselves, they usually walk away with a fixed issue, a workaround they're happy with, or a good explanation why a solution is not forthcoming (and yes, this can be "I'm not personally interested in implementing this feature, patches are welcome"). The project benefits from these people as well. Of course there are also bad and irrational maintainers out there, as well as projects so bad they're worthless, the barrier of entry isn't exactly high.

The point is: No, OSS devs aren't your employees. Neither are you their paying customer, and you have no right to make demands. No, not even if you donate $3. Take what they offer, or not. Nonetheless, if you can't see the indescribably huge value in a plethora of OSS projects, including Wikipedia, I feel sorry for you. There are millions of people with better people skills and/or technical knowledge than you who actually make OSS work for themselves, every day.

Re:seriously?

[rbrander]

Everybody regards a different kind of problem as your "busted shitter" though. Proprietary software companies regard problems that won't affect sales by much to be such. There are bugs in Excel VBA 2010 that were identified in Excel 97. Microsoft just can't be bothered; nobody is going to stop the whole corporation automatically buying MS-Office because a few VBA macro fans in accounting and engineering have to do awkward workarounds. So despite having all those piles of cash to pay people do dig in to that hard, dirty thicket of code, they don't.

There's some truth in your thesis, but mostly it is disproved by the amount of large, complex, and good FLOSS projects out there. Since they are large and complex, it is utterly certain that they contained a number of your "busted shitters" on the long road from Line 1 to being PostGIS or GIMP or Drupal. And here they are. Even that fact that some or all examples have *current* "busted shitters" proves nothing as long as there are some fixed ones in the project history.

Oh, and the old FLOSS/communism association - actually FLOSS is *worse* than communism from the point of view of somebody who imagines work-organization systems to be proof of moral value or failure. It's not "from each according to his abilities, to each according to his needs", it's worse. From each according to his ability to write code that others agree to commit, to everybody, everything. We'll just dump mountains of free stuff on you that you didn't even ask for and *don't* need. Talk about moral hazard that disinclines you to the virtue of doing your own work! Truly, we are horrible people.

Re:seriously?

[MikeBabcock]

What FUD are you trying to spread exactly?

Re:seriously?

[MikeBabcock]

What on earth do you think you're talking about? And how do you think it made sense?

Re:seriously?

[MikeBabcock]

Obviously you're not a programmer. If you are, you're not a very good one.

Using proprietary software and hoping someone else fixes your problems is like playing Russian roulette. It might seem like a good idea at the time, but it ends badly.

Re:seriously?

[MikeBabcock]

Although I might add, we programmers love it when requests come attached to donations rather than solo.

Bribery really does work.

Re:seriously?

[MikeBabcock]

You do realize MariaDB *is* MySQL for all intents and purposes, right?

This is a lot like upgrading from Debian to Ubuntu (except that Debian is not controlled by a multi-national corp).

Re:seriously?

[Synerg1y]

That OSI projects often don't make good business decisions. And before anybody says anything, they do have limited resources and they can be using them to improve their "product" instead.

Re:seriously?

[MikeBabcock]

And yet I don't see this failure everywhere that you imply in your original comment.

Given that many OSI projects aren't failing and are in fact very successful and well-regarded in general, I'd say it has nothing to do with the model at all.

That is to say, unless you can show that OSI projects fail more frequently than commercial software projects, you have no point at all. Bear in mind that this must include *all* commercial software projects, which is very hard because unlike FOSS, those aren't necessarily public.

Re:seriously?

[Synerg1y]

I like how you selectively read my post and are completely oblivious to the word "often" : http://www.merriam-webster.com/dictionary/often

Comparing OSI failure rates to commercial failures rates is very difficult as per THE TOPIC of TFA because there is no set standard for OSI to be deemed a failure outside of subjective analysis. In which case, there's tons and tons of failed projects, but a few major successful ones.

Information

[neokushan]

For more information, Wikipedia has a statement regarding MariaDB: http://en.wikipedia.org/wiki/MariaDB

Re:Information

[kthreadd]

What statement? It looks like an ordinary article to me.

Re:Information

[Anonymous Coward]

Whoosh!

Re:Information

[theVarangian]

What statement? It looks like an ordinary article to me.

As a fork of a leading open source software system, it is notable for being led by its original developers and triggered by concerns over [the] direction [taken] by an acquiring commercial company Oracle.

It may only be an article but it says all that needs to be said. Oracle bought up MySQL and immediately dropped support for a range of systems that had previously been supported, probably in the hope it would drive scores of customers into the open arms of the Oracle sales team and their extortionate license prices. I now have to migrate several MySQL databases that previously lived a happy life on AIX 5 to something else and MariaDB is a welcome alternative because I'm sure as [Expletive deleted] not going to shell out thousands of $ for overpriced Oracle DBs with a pile of features that I don't need.

Re:Information

[Anonymous Coward]

AIX? Here's a nickle son, go get yourself a real operating system!

Re:Information

[Errol backfiring]

Correct.At the last line on the page, there is a link to the statement: https://blog.wikimedia.org/2013/04/22/wikipedia-adopts-mariadb/

Re:Information

[theVarangian]

AIX? Here's a nickle son, go get yourself a real operating system!

Young man, I'll have you know that I was using UNIX long before Linux was a 115 kB compressed tarball on the funet.fi FTP server.

Re:Information

[Gothmolly]

-1, KarmaWhore

Re:Information

[Anonymous Coward]

AIX? Here's a nickle son, go get yourself a real operating system!

Young man, I'll have you know that I was using UNIX long before Linux was a 115 kB compressed tarball on the funet.fi FTP server.

If you're that old, a nickle should seem like a lot of money!

Re:Information

[sydneyfong]

I think he was aiming for +5 Funny.

Stupid mods...

Re:Information

[theVarangian]

AIX? Here's a nickle son, go get yourself a real operating system!

Young man, I'll have you know that I was using UNIX long before Linux was a 115 kB compressed tarball on the funet.fi FTP server.

If you're that old, a nickle should seem like a lot of money!

**Sigh** I'd explain the concept of inflation to you but I don't have the time. I'm busy loading shotgun shells with rock salt so I'll be prepared for the next time Larry Ellison makes the mistake of stepping onto my lawn.

Re:Information

[lgw]

As someone who was using mainframes back when: AIX? here's a nickel son, go get yourself a real IBM box.

Two steps forward, one step back.

[Anonymous Coward]

Too bad they're not on PostgreSQL instead. So many retardisms in MySQL.

Re:no way!

[Anonymous Coward]

Correct. There is no way you did get first post.

slash next?

[axehind]

Is slashdot next?!?!?!

Re:slash next?

[Anonymous Coward]

Not until MariaDB is completely broken. Right now, it's much too stable for Slashdot.

Re:slash next?

[rvw]

Is slashdot next?!?!?!

Slashdot needs MariaMagdalenaDB for all those wankers here!

But... But... Why?

[CRC'99]

So reading the links etc, there is no real reason that stands out apart from "Oracle may screw MySQL".

Is there a reason for this other than ifs, buts and maybes?

Re:But... But... Why?

[h4rr4r]

What Oracle has done in the past and how they operate their business in general I think.

Not moving now, is ensuring Oracle will screw you. Why wait for it to happen?

Re:But... But... Why?

[Annirak]

RTFA.

For the last several years, we’ve been operating the Facebook fork of MySQL 5.1 with most of our production environment running a build of r3753. We’ve been pleased with its performance; Facebook’s MySQL team contains some of the finest database engineers in the industry and they’ve done much to advance the open source MySQL ecosystem.
That said, MariaDB’s optimizer enhancements, the feature set of Percona’s XtraDB (many overlap with the Facebook patch, but I particularly like add-ons such as the ability to save the buffer pool LRU list, avoiding costly warmups on new servers), and of Oracle’s MySQL 5.5 provide compelling reasons to consider upgrading. Equally important, as supporters of the free culture movement, the Wikimedia Foundation strongly prefers free software projects; that includes a preference for projects without bifurcated code bases between differently licensed free and enterprise editions. We welcome and support the MariaDB Foundation as a not-for-profit steward of the free and open MySQL related database community.

It's part performance and part philosophical. Given that wikipedia is a strongly philosophical enterprise, this seems reasonable.

Re:But... But... Why?

[serviscope_minor]

"Oracle may screw MySQL".

Is there a reason for this other than ifs, buts and maybes?

One definition of madness is to try the same thing again and again and keep expecting different results. It's Oracle. You will get screwed.

Re:But... But... Why?

[Anonymous Coward]

MariaDB has quite a few improvements over MySQL. More information here: https://kb.askmonty.org/en/mariadb-versus-mysql-features/

Note that this does not address the specific storage backend features, which are quite attractive on their own. There are even plans to revive a key-value store backend at some point.

Said improvements may or may not be a factor in their decision to move, but it's almost a completely drop-in replacement. So the real question would by why not. Simply having it running on my servers gives me that clean and tidy feeling. You know the one I mean.

Re:But... But... Why?

[mexsudo]

The philosophy is sound. Open Source has a future beyond the fall of any individual enterprise

Re:But... But... Why?

[Alastair Gilfillan]

Is there a reason for this other than ifs, buts and maybes?

The performance increases were negligible (with some decreases, as well) so: No reason to switch other than a symbolic statement and to avoid any potential future licencing issues or litigation. The MariaDB project will mirror the free-as-in-beer (and maybe paid?) features of future MySQL versions while aiming to be a "drop-in" replacement. On the other hand, it's better to do it now in case the projects do diverge and the MySQL upgrade path becomes problematic or expensive.

Re:But... But... Why?

[theVarangian]

"Oracle may screw MySQL".

Is there a reason for this other than ifs, buts and maybes?

One definition of madness is to try the same thing again and again and keep expecting different results. It's Oracle. You will get screwed.

Amen!

Re:But... But... Why?

[CRC'99]

It's part performance and part philosophical. Given that wikipedia is a strongly philosophical enterprise, this seems reasonable.

Well, the performance difference didn't seem to be huge - in fact, some stats were slower.... I don't buy for a second that it was for performance reasons.

Philosophy - maybe - however Oracle contribute quite a bit to OSS - more than a lot of companies - See: http://www.oracle.com/technetwork/server-storage/linux/technical-contributions-1689636.html

In a nutshell, they are working on NFS over IPv6, data integrity checks for ext3, they maintain libstdc++, they worked hard on BTRFS, If anything, they have helped open source much more than most other companies.

Again, I don't see the philosophical reasons other than 'because we can'.

Re:But... But... Why?

[Dishevel]

If you arfe a business looking forward to the long term those ifs, buts and maybes are important.
You serve yourself well by not waiting till the last minute.

Re:But... But... Why?

[Anonymous Coward]

however Oracle contribute quite a bit to OSS

Looking like you support OSS is not a bad business move as even Microsoft has learned. It also makes underhanded sabotage of OSS much easier because they can "We support OSS and aren't greedy scum" FUD most people.

Re:But... But... Why?

[Anonymous Coward]

Oracle has a few employees that are solid OSS contributors, and apparently they have some management support. That's been true for years (e.g. their OCFS filesystem...). However, they're only an OSS contributor in a tactical sense. Many years ago (and much earlier than one would've expected!) they came to the realization that Linux was the future (or at least, a large chunk of the future) in the server space, and they made the very smart tactical decision that they didn't want to be relegated to a dusty corner where their products only ran (well) on legacy Sparc/Solaris, HP/UX, IBM AIX, etc environments. So they made their core Oracle products work on Linux, and as a part of doing that job fully and trying to make their stuff really shine on Linux, they necessarily had to get involved in the OSS community.

Later came the MySQL acquisition, which was another tactical decision along the lines of "Well, that worked great and we retained our corporate Oracle customers that wanted to move from Sparc/Solaris (etc) to x86/Linux, but... we can't get all these exiting Linux/OSS users to adopt Oracle because MySQL works well enough for them and its free, so lets take over MySQL too and own the Linux relational database space".

It's all tactical, and it's all designed to corner the market on relational databases (and various other bits that go on top of them) as hard as they can. Philosophically, as an organization, Oracle doesn't have any real interest in promoting Open Source or doing right by the community. Their vision isn't long-term enough for that. It was just barely medium-term enough to make the right calls to get involved in OSS at all. Their big-picture motivation isn't "Build awesome free software for the world to share", it's "Let's find a way to trap all these Linuxy people into paying us for something".

Re:But... But... Why?

[Anonymous Coward]

Well, the performance difference didn't seem to be huge - in fact, some stats were slower.... I don't buy for a second that it was for performance reasons.

??

For our most common query type, 95th percentile times over an 8-hour period dropped from 56ms to 43ms and the average from 15.4ms to 12.7ms.

Emphasis mine. This is a 18% drop in average query time. This is easily like getting an extra server for every 9 servers you already have. I don't know how many servers they have, but the transition represents thousands of dollars in savings if any new servers will need to be bought.

Re:But... But... Why?

[Viol8]

Does the reason matter so long as they do it? And I doubt anyone ever believed that Oracle did it out of the kindness of their heart. They're a commercial organisation, they have to make a profit - or at least break even. And they won't do that buy not selling anything. In fact more people buying Oracle keeps them in business and that'll keep them employing OSS coders.

There's no such thing as a free lunch - OSS coders either have to be paid for their work or they do it in the free time when they can around their real jobs. And while the latter approach is laudable it wouldn't alone have led to OSS as we know it today.

Re:But... But... Why?

[geek]

Does the reason matter so long as they do it?

Yes because it means their support is shakey and their motivations can swing them in another direction at any time. Oracle can sabotage any project they are a part of and knowing who their CEO is should make people wary of them.

I'm not a purist but I get where the purists come from when it comes to Oracle. I don't trust them and I don't give them my money.

Re:But... But... Why?

[greg1104]

I don't understand why people trust MariaDB either. The result the last time everyone jumped onto Monty's open-source ship? He cashed out and put all of his customers on a road that led to being screwed by Oracle. There's a bit of madness expecting different results adopting MariaDB I wouldn't buy into either.

Hint: when a contributor agreement like the MariaDB one says copyright must be assigned to "Monty Program AB", your contributors are usually being setup so that the owner of that copyright can then profit from the community's free work on the project, a decision that will be motivated by what's best for them. There are a few software projects that require copyright assignment that aren't doing that, like the gnu projects. Monty Program AB is not a non-profit with a decades long history of anti-commercialism like gnu though. It's a regular company run by someone who has screwed both his paying customers and his open-source user community exactly this way once. Why are people signing up such that he could do it again?

Re:But... But... Why?

[Anonymous Coward]

As someone who migrated to postgres for anything that wasn't tied to only mysql (through libraries or sql tables that were more of a hassle than they'd be worth to rewrite for importation into postgres) I agree with this question.

Moreover everybody seems to have forgotten when libmysql got pushed from lgpl to gpl'd because not enough people were licensing mysql since most people only needed their code linked to the client library, which didn't require source redistribution even for commercial applications and thus didn't require them to purchase a commercial license to sell a proprietary application using a mysql db as a backend.

And anyone fool enough to assign copyright to a third party deserves what they get. That's the one Linus and I would agree on: Keep your contributions under your copyright, that way if someone feels strongly enough about changing the license, or screwing you over, they'll have to put in the work to excise all your copyright claims themselves, which will often be more trouble than it's worth.

Re:But... But... Why?

[Annirak]

Does the reason matter so long as they do it?

Again, I said it was a philosophical motivation. Philosophy is the place where the reasons for things matter.

Re:But... But... Why?

[Anonymous Coward]

This is a very fair point. The Wikipedia article on MariaDB clearly states:

Its lead developer is Michael "Monty" Widenius, the founder of MySQL and Monty Program AB. He had previously sold his company, MySQL AB, to Sun Microsystems for 1 billion USD.

Clearly indicating that, if someone dangles a carrot big enough, he's quite likely to sell out again.

Re:no way!

[Anonymous Coward]

No way did I get first post!!!

Nope.

I Don't Like The Name

[Anonymous Coward]

MariaDB is a crappy name. I know crappy names are the usual preference for open source projects, but MySQL seemed to be the exception. It was a good name that clearly described what the project was. MariaDB is no good.

We need to DDoS Oracle until the give back MySQL. Who's with me?

Re:I Don't Like The Name

[Anonymous Coward]

Replacing one database named after one of the author's daughters, with a database named after another of the author's daughters. Seems pretty consistent to me.

Re:I Don't Like The Name

[hoggoth]

The first daughter's name is MySQL? I should introduce her to my nephew, little Tommy ;'Drop Tables.

Re:I Don't Like The Name

[kthreadd]

Her name is just My.

Re:I Don't Like The Name

[rvw]

Replacing one database named after one of the author's daughters, with a database named after another of the author's daughters. Seems pretty consistent to me.

Well I guess we are talking about the fifth generation of Maria's then. They certainly know how to procreate!

Re:I Don't Like The Name

[Gyske]

He also has a son named Max: http://en.wikipedia.org/wiki/MaxDB

Re:I Don't Like The Name

[hoggoth]

And a dog named "Mongo".

Re:I Don't Like The Name

[Gyske]

I kid you not: http://en.wikipedia.org/wiki/Michael_Widenius :)

Re:I Don't Like The Name

[Anonymous Coward]

Funny name. Is her brother named Your?

That's simple

[Anonymous Coward]

As we all know Oracle is not the biggest friend to the Open Source Community

That's because the "communitah!" is a bunch of sniveling brats. I wouldn't be their friends either. Either way Oracle has made and contributed to plenty open source software products over the years.

Re:That's simple

[Pi1grim]

Please point out the projects it contributed to?
OpenOffice is dead and shipped off to apache, MySQL is stagnating, Oracle linux is nothing but a Redhat clone with no changes but copyright names change. Oracle is not a friend to open source. Never been either.

Re:That's simple

[kthreadd]

Btrfs, VirtualBox, Java. Just to mention a few.

Re:That's simple

[Anonymous Coward]

https://oss.oracle.com/projects/

Re:That's simple

[Anonymous Coward]

I'll raised you a dead opensolaris, complete with now closed source ZFS and dtrace.

Re:That's simple

[Lunix Nutcase]

Also, if Oracle is not friends with open source why have they sponsored numerous Linux and FOSS conferences?

Re:That's simple

[M. Baranczak]

Java was already open source when Oracle bought Sun. And since then, Oracle has been trying to close it back again with bullshit patent claims.

Re:That's simple

[kthreadd]

Java was already open source when Oracle bought Sun. And since then, Oracle has been trying to close it back again with bullshit patent claims.

As far as I know they actively contribute to the open source implementation, which was what the question was all about.

Re:That's simple

[serviscope_minor]

That's because the "communitah!" is a bunch of sniveling brats.

Screw what the open source community community has to say about it.

I've been in two private organisations that have been utterly fucked over with overpriced underperforming shit spewed forth by Oracle unthe the guise of some "solution" for something or other. Speaking to collegues at other organisations, my experience is not unique. Many of them have been oracled. The experience is never pleasant.

There is no chance in hell I would trust them with anything that they are very much in charge of (there are enough checks and balances in the Linux and GCC community I feel to curb their worst instincts).

And OSS has nothing to do with it.

Oracle being a bunch of bastards is the reason.

Re:That's simple

[Anonymous Coward]

I didn't say Oracle wasn't a dickish company. But to try to act like they've made or contributed nothing to FOSS because they aren't "friends with the communitah" (whatever that is supposed to mean) is flat out wrong. If they would stop acting like spoiled brats maybe Oracle might be less of a dick.

Re:That's simple

[serviscope_minor]

If they would stop acting like spoiled brats maybe Oracle might be less of a dick.

Why on earth would you think that? Oracle is far more of a dick to its paying customers than to the OSS community, probably because the OSS community can act like dicks and make a fuss and fork and/or not accept their changes.

The best way of getting fucked by oracle is to pay them money. But honestly, there are more pleasant ways of getting fucked for money and it will cost less than $10,000,000 a pop.

Re:That's simple

[Nerdfest]

Advertising?

Re:That's simple

[gorzek]

A concise summary, if worded provocatively.

In other words, the real problem is people expecting "open source" to mean the same as "free software." They don't. They are two different things. "Open source" is a pragmatic approach to making life easier for developers without overly inconveniencing profit-driven institutions; "free software" is a philosophical movement designed to ensure user freedom, even at the expense of corporate interests.

Of course, confusing the two concepts has been one of the key goals of the open source movement, in order to extinguish "free software."

Re:That's simple

[Dan667]

to upgrade you can't let your package manager just grab the latest java runtime. You have to login to oracle and click "accept" before you can download it. I am switching to icetea.

Re:That's simple

[lgw]

Oh, yeah, Java: I'm sure they'll have Lamba ... any year now. For years C# and Java were leapfrogging one another in functionality - each would add everything the other had, plus a little, with each release. But that stopped when Oracle bought Sun, and Java fell far behind even their own plans.

Map, reduce, and similar list-processing operations have been one-line-with-lamba for many years now in C#, while Java has pushed lamba back yet another year, and I'm not sure it will even have inline list processing to go along with lamba when Java 8 finally does come out. Oracle has ruined it.

As we all know??

[Crimplene Prakman]

"As we all know Oracle is not the biggest friend to the Open Source Community." This is a bit weasely. We all don't know any such thing. For example, Oracle was in the top ten of organisations that contributed code to Linux last year: http://www.linuxfoundation.org/news-media/announcements/2012/04/linux-foundation-releases-annual-linux-development-report Since then it has been very public with Oracle Linux, and made several large contributions from that front. Shucks, it's even got its own OSS portal: https://oss.oracle.com/ I'm happy to agree it's a big bad corporate beast and does a lot of wrong in the world, but if you're going to criticize it, at least be factual.

Re:As we all know??

[SIGBUS]

True, but then again, Oracle has closed off access to MySQL test cases, and let's not forget what they did to OpenSolaris.

Re:As we all know??

[Anonymous Coward]

As much as I hate Oracle, Guy Steele works there, so they aren't all bad.

Re:As we all know??

[kthreadd]

As far as I know they just stopped contributing to OpenSolaris. But they never attempted to stop anyone from picking it up and keep running with it, which led to Illumos and a few other projects taking its place. It's unfortuneate, but not evil.

MySQL is a bit of a different story.

Re:As we all know??

[Anonymous Coward]

Oracle only supports free software inasmuch as they feel that it will help them drive sales of their not-free-at-all software and support. Oracle is not your friend! If their contributions to the community did not somehow equate to new yachts for Larry, they wouldn't contribute anything.

Case in point, I had to renew maintenance for Primavera (Oracle scheduling software) a month ago at a cost of $2.6K for one year - my biggest complaint is that it (frequently) corrupts its own database, and the "patch" that oracle issued several months ago consists of about eight JAR files that must be copied to each client, by hand, into the install directory. They can't even be bothered to package it up into an installer, or release an updated installer for everything - you have to deploy 8.2 and then manually upgrade it to 8.2.1. Oh, and it force-installs Java 6u28, even if you have an updated JRE, and even though it then bugs you to upgrade to Java 7 latest! All that nonsense aside, it still has security holes big enough to drive a Mack truck through - any user with a mind to do so could take over the admin account, or trash the DB, trivially. No, it is not my config, it's the way the thing was designed. If any other big mean corporation (Microsoft, Apple, ) put out a product like this, they'd be drawn, quartered, and the pieces hung on pikes all down Wall Street. QED.

Re:As we all know??

[Anonymous Coward]

We do know it's "not the biggest", surely, even by your own metrics. That's all Peetke asserted. Seems pretty solidly factual. And that's before we get into the business of the amici briefs in Oracle vs. Google that are arguing for an interpretation of the law that would _permanently_ _fuck_ open source and the mere interoperability of closed source packages.

Re:As we all know??

[lowlands]

"Since then it has been very public with Oracle Linux"

Ah you're maybe referring to the CentOS-like clone they created from Red Hat's source packages. Once they spun their isos and slapped together a 3 page website they went after Red Hat's customers saying "their" so called unbreakable Linux is better than Red Hat's. That same unbreakable Linux which is based *entirely* on Red Hat's source packages. That's pretty evil in my book. Add how they are shielding off MySQL bugs and development and what they did to OpenSolaris and you get a pretty clear picture of a dinosaur-going-the-way-of-the-dodo who's trashing anything Open while giving in to its insatiable hunger for ginormous license fees and sending ginormous invoices to misguided customers for their $1000-suit "consultants".

It makes total sense to migrate to MariaDB, Percona, EnterpriseDB or PostgreSQL as fast as you can. While you are at it replace BerkeleyDB with LMDB. Last time I looked OpenLDAP, Postfix, OpenDKIM and other projects already support it. And it's faster and more reliable than BerkeleyDB too.

The only way Oracle will (hopefully) make some attempt to become a proper member of our Community is when they feel it where it hurts most: revenue streams. So slam the door in the face of that pompous Oracle rep, don't renew the license(s), drink the F/OSS Cool Aid and enjoy the view.

Re:As we all know??

[Raenex]

As much as I hate Oracle, Guy Steele works there, so they aren't all bad.

As much as I'd like to like Guy Steele, he's just another corporate whore filing software patents for his paymasters.

Me too!

[alonso]

Me too, in all my production server!

Maria

[Anonymous Coward]

The most beautiful sound I ever heard:
Maria, Maria, Maria, Maria . . .
All the beautiful sounds of the world in a single word . .
Maria, Maria, Maria, Maria . . .
Maria!
I've just met a database named Maria,
And suddenly that name
Will never be the same
To me.
Maria!
I've just kissed a database named Maria,
And suddenly I've found
How wonderful a sound
Can be!
Maria!
Say it loud and there's music playing,
Say it soft and it's almost like praying.

Maria,
I'll never stop saying Maria!

The most beautiful sound I ever heard.
Maria.

Web scale database

[hoggoth]

But is MariaDB web scale?

Re:Oracle killed OpenSolaris.

[Anonymous Coward]

andnothingofvaluewaslost.png

Now

[TheSkepticalOptimist]

If they can only move to Web 2.0, they might bring the website into the "early" 21st century.

Re:Now

[Tarlus]

they might bring the website into the "early" 21st century

I don't understand what it means when people say that. It's a website with organized, searchable content that can deliver varying forms of multimedia. What is not "21st century" about it?

But, did you donate already?

[Herve5]

Otherwise, otherwise... in my country (France) there is a nice saying, 'Les conseilleurs ne sont pas les payeurs', that basically describes you.

(attempt at translating: 'Most prominent advisors are generally not the investors to begin with')

Re:But, did you donate already?

[MoxFulder]

Herve, this is one of my favorite French expressions. We don't have an exact equivalent in English, but there are a couple of similar ones:

"Talk is cheap."
"Those who know don't talk. Those who talk don't know."

but Amazon doesn't offer Maria

[cellurl]

What can others (non self hosters) do? Will Amazon offer MariaDB? Is there a reason they will never do such?

Help eliminate stupid speeding tickets.

SluttySQL

[Anonymous Coward]

SQL for the masses!

From non-Oracle to non-Oracle; why mention Oracle?

[DragonWriter]

Wikipedia was using a non-Oracle fork of MySQL (a Facebook maintained fork of MySQL 5.1) and moved to a different non-Oracle fork (MariaDB). The comment about Oracle not being a friend of OSS seems to be a non-sequitur.

Maria DB

[CanEHdian]

"MariaDB"? "Ishmael view" Jeebus, what's going on here? And now http://mariadb.org/ has been slashdotted^H^H^H DDoS'ed to death by the criminal organisation known as Slashdot!

Maybe we'll see some coverage in the TechEye Bible, wonderful proze!

Oh well..

[h8sg8s]

Larry kills another one. Are there any "open" OSS projects left at Oracle?

You and the 200-something other residents...

[Anonymous Coward]

of 'his' island, eh?

Oracle contributions

[martin_dk]

Ever since Oracle decided to bundle the Ask Toolbar with Java they showed the world what they are really made of.

Make them stop!

A company their size doing such shady business is unbelievable to me. No matter their contributions in other areas they will stand out as real bad dishonest guys. Add to this the Oracle interpretation of openness regarding MySql development.

Postgres

[Blackknight]

Should have switched to PostgreSQL instead. But that would require cleaning up shitty SQL statements that only work with MySQL/MariaDB.