Time To Move To PostgreSQL

“MariaDB Corp. has announced that release 2.0 of its MaxScale database proxy software is henceforth no longer open source. The organization has made it source-available under a proprietary license that promises each release will eventually become open source once it’s out of date.
 
MaxScale is at the pinnacle of MariaDB Corp.’s monetization strategy — it’s the key to deploying MariaDB databases at scale. The thinking seems to be that making it mandatory to pay for a license will extract top dollar from deep-pocketed corporations that might otherwise try to use it free of charge. This seems odd for a company built on MariaDB, which was originally created to liberate MySQL from the clutches of Oracle.”
 
See Uproar: MariaDB Corp. veers away from open source
Sigh… I understand that businesses need to make money but proper businesses don’t jerk their customers around in the process. That drives them away.

Large businesses that use MySQL/MariaDB depend on the MaxScale component and changing the licence for that jerks them around. In the process, MariaDB is preventing a larger community from sharing in the development, a major plus of FLOSS. So, this is essentially kicking a large segment of the market for SQL databases to a non-Free solution. It really is time to go to PostgreSQL, a truly Free/Libre Open Source database from top to bottom.

I’m a small user and don’t need MaxScale but I don’t want to wait for the next shoe to drop. Oracle and MariaDB both seem reluctant FLOSSies. I will move everything I can to PostgreSQL as rapidly as I can. Certainly all my new projects will skip MariaDB.

UPDATE Widenius responds: “the chief technology officer of MariaDB corporation called his firm’s embrace of a commercial licence for part of MariaDB “critical” to delivering new revenue and for the continued development of open-source software.
 
Widenius told The Register in an interview that he believes criticism of MariaDB’s commercial licence for its new database proxy sever, MaxScale 2.0, is motivated by a “religious” belief in free and open source software.”

For me, FLOSS is not a religion but a logical point of view. FLOSS is the right way to develop software. No small collection of individuals can do better than the collected intelligence of a large group of users/developers. e.g. Cities are not best which exclude diverse kinds of people. Houses are not best built by a single tradesman. Automobiles use JIT to bring together the best products and ideas from a wide variety of specialized organizations. Software can and is made by the world not just a few friends of Widenius. Small groups may or may not produce good ideas and implement them. Large diverse groups certainly will do the job.

Widenius should have invested more time in considering how MariaDB could take its place in the world of FLOSS rather than trying to limit MariaDB to his way of thinking. If he was short of funds there must have been 100 better ways to go than the one he’s chosen. How is MariaDB better off with fewer users paying more? How is FLOSS better off with that?

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in technology and tagged , , , , . Bookmark the permalink.

10 Responses to Time To Move To PostgreSQL

  1. Ivan wrote, “it’s about doing the job cheaper. This is why Takata airbags are killing people. Some jackass had a quota to meet and he did so. People died.”

    Bad things do happen. Meanwhile, the wheels are round, engines are lighter and more efficient, and everything fits together better than ever. I never claimed auto makers were perfect but they are pretty damned good, better than M$, I’d say. M$ ships products with thousands of defects in every copy. Users can meet a few every month of use. Meanwhile owners of automobiles may go many years between problems. I remember selling a car to pay for my wedding… I bought it new and owned it 4 years and never had a single problem with it. That was in the 1980s… It was a Toyota… My father bought USAian products and was always having problems. One time he bought a pickup truck. It had unimaginably high fuel consumption and the cab had not been fully bolted to the frame and one wheel bearing never saw grease…

  2. oiaohm says:

    Ivan in fact is a mixture of both cheaper and attempting to be safer is the Takata case.
    Bullshit, it’s about doing the job cheaper. This is why Takata airbags are killing people. Some jackass had a quota to meet and he did so. People died.
    To be correct this is a very big question. Takata has been making air-bags for a very long time. From 1988 to now.

    –“We just said, ‘No, we can’t do it. We’re not going to use it,’” said Robert Taylor, Autoliv’s head chemist until 2010.–
    Over ammonium nitrate. Autoliv is competitor to Takata. Autoliv used explosives were it was not possible to stuff up in production but from a toxic gas production not as safe.

    Before Takata used ammonium nitrate everyone fairly much used sodium azide that heats up to 300C and produces a few toxic things and light up fuel fumes(100 percent not wanted).

    There is another issue 5-Aminotetrazole that is one of the other airbag explosives is suspected to be long term toxic that Takata attempted to avoid by using ammonium nitrate.

    So cheaper product was not Takata only consideration. Less toxic was one of Takata consideration so less staff sick days making them this is also cost. Also Takata was attempting to avoid being on the receiving of future toxic exposer cases from their air bags. Yes out of all the explosives tested so far for airbags ammonium nitrate is still the one that produces the least amount of toxic chemicals. Yes problem with ammonium nitrate is that it blows the casing it bits and those fragments kill .

    If you think you can design the casing to take ammonium nitrate force without turning into fragments using ammonium nitrate looks like a super good idea.

    Add in ammonium nitrate exposed to moisture becomes a more powerful explosive this was not allowed for in the Takata design. So even a perfect Takata product was not safe.

    Some jackass had a quota to meet and he did so.
    Not the case in the Takata at first this is stage two of the Takata disaster making in Mexico with under-trained personal that bad they blow up the factory 2006 and caught reworking parts on the production like in 2011.

    So the Takata design was not solid because key effect on explosive over time had been got wrong. Add in sloppy not to spec construction design defect is going to show itself sooner. Yes a Takata defective air bag not made by a jackass attempting to meet quota made exactly to spec could attempt to kill you. Core of the Takata mess is a product design error made worse by poor construction. Perfect construction might have taken 30 years before the first kill.

    When playing with explosives get it wrong someone dies either from the blast or the toxic by products. Defective Takata airbags gets you with flying fragments and the other airbags give you minor toxic fume exposure. So this is not a case of win but what choosing lesser of the evils. You can say airbag makers are stuck between rock and hard place and Takata happened to choose hard place and got it wrong.

    The fact the air bag is an explosive device it would have helped if governments put more regulation in the form of mandatory quality control including certifying process for the designs.

  3. Ivan says:

    “So, why do dozens of companies contribute to modern automobiles? It’s not just about quotas but about who can do the best job.”

    Bullshit, it’s about doing the job cheaper. This is why Takata airbags are killing people. Some jackass had a quota to meet and he did so. People died.

    Stop drinking the buzzword kool-aid, old man.

  4. Dylan Hunt wrote, of PostgreSQL, “Not quite in the same league as the big three (Oracle, DB2 and MS SQL), but it’s the closest there is.”

    BASF thinks it’s in the same league. In 2002 they set up their global agricultural supply system on it. They saved 67% of the total cost using PostgreSQL. Fujitsu extends PostgreSQL for their customers and actively contributes to PostgreSQL. According to Fujitsu, PostgreSQL is the fourth most-used database with 2 million users.

  5. Dylan Hunt says:

    Postgres is by far the more robust, powerful and flexible solution, and it can do JSON/NoSQL, if that floats your boat. It also does much better vertical scaling than MySQL. Replication (slony, DRS) and binary storage (Bytea, Toast) are cumbersome at best, but aside from conservative defaults, there isn’t much else in terms of weaknesses.

    Not quite in the same league as the big three (Oracle, DB2 and MS SQL), but it’s the closest there is.

  6. Ivan wrote, “JIT is used so middle management can make their bullshit quotas set by some random jackass in a boardroom, it doesn’t result in a better product.”

    So, why do dozens of companies contribute to modern automobiles? It’s not just about quotas but about who can do the best job.

  7. Ivan says:

    “Automobiles use JIT to bring together the best products and ideas from a wide variety of specialized organizations.”

    JIT is used so middle management can make their bullshit quotas set by some random jackass in a boardroom, it doesn’t result in a better product.

  8. oiaohm wrote, “Maxscale feature set should have been going though a merge with mainline Mariadb to keep up with the competition. But no they are going the other way breaking Maxscale into a independent part.”

    Yes, at this late date, changing the rules of the road will create a lot of chaos for zero benefit. People hate chaos and want stuff that works for them. Given choice, people make choices. For people who’ve resisted doing things Oracle’s way, resisting doing things the MaxScale way makes sense. Further, with the way computing and storage and networking are going, MaxScale probably makes sense for a rather small group of large organizations who can easily contribute to PostgreSQL instead of MariaDB. These guys who have been confident of MySQL may well prefer the FLOSS DB, PostgreSQL, or they could fork MaxScale. Either way MariaDB is shooting themselves in one or both feet. Not good.

  9. oiaohm says:

    http://blog.2ndquadrant.com/postgresql-10-roadmap/
    http://blog.2ndquadrant.com/postgresql-solutions-roadmap/
    dougman exactly why fork MariaDB like Mysql was to make MariaDB.

    The idea of putting Maxscale under a commercial license would have been a sane idea a few years a go.

    Problem is
    https://2ndquadrant.com/en/resources/bdr/bdr-performance/

    BDR and XL from postgresql perform extremely well. londiste slony baucardo are all like the Maxscale idea of put a proxy in the middle.

    BDR currently goes 48 servers wide. So using postgresql you can go at least 48 server blocks before need to use a proxy. BDR and XL are a postgresql extensions that is planned to go mainline.

    Postgresql BDR and XL both proved something that most people could guess. The database proxy idea is slow.

    The problem is forking the MariaDB open code base will not give you the performance or the scale going across to Postgresql will give in a few years as mainline or what you can achieve currently with extensions.

    A few years ago when Postgresql was slow and was not adding mainline cluster replication including multi master support making MaxScale database proxy closed would have been a ok idea. Today making MaxScale database proxy closed is kinda pointless when you look at Postgresql and you see with XL and BDR that the non proxy built into main database engine is faster and superior in all metrics.

    Maxscale feature set should have been going though a merge with mainline Mariadb to keep up with the competition. But no they are going the other way breaking Maxscale into a independent part.

  10. dougman says:

    Future blog entry:

    “New DATABASE name” is a community-developed fork of the MariaDB relational database management system intended to remain free under the GNU GPL.

Leave a Reply