Slashdot Mirror

← Back to Stories (view on slashdot.org)

Redis Changes Its Open Source License -- Again (zdnet.com)

Posted by EditorDavid on from the taking-license dept.

"Redis Labs is dropping its Commons Clause license in favor of its new 'available-source' license: Redis Source Available License (RSAL)," reports ZDNet -- adding "This is not an open-source license." Redis Labs had used Commons Clause on top of the open-source Apache License to protect its rights to modules added to its 3-Clause-BSD-licensed Redis, the popular open-source in-memory data structure store. But, as Manish Gupta, Redis Labs' CMO, explained, "It didn't work. Confusion reigned over whether or not the modules were open source. They're not open-source." So, although it hadn't wanted to create a new license, that's what Redis Labs ended up doing....

The RSAL grants, Gupta said, equivalent rights to permissive open-source licenses for the vast majority of users. With the RSAL, developers can: Use the software; modify the source code; integrate it with an application; and use, distribute, support, or sell their application. But -- and this is big -- the RSAL forbids you from using any application built with these modules in a database, a caching engine, a stream processing engine, a search engine, an indexing engine, or a machine learning/artificial intelligence serving engine. In short, all the ways that Redis Labs makes money from Redis. Gupta wants to make it perfectly clear: "We're not calling it open source. It's not."
Earlier this month the Open Source Initiative had reaffirmed its commitment to open source's original definition, adding "There is no trust in a world where anyone can invent their own definition for open source, and without trust there is no community, no collaboration, and no innovation."

And earlier this week on Twitter a Red Hat open-source evangelist said they wondered whether Redis was just "clueless. There are a lot of folks entering #opensource today who are unwilling to do the research and reading, and assume that these are all new problems."

11 of 68 comments (clear)

  1. Bye, Redis by null+etc. · · Score: 2

    This will be the thing that causes Redis to lose whatever prominent marketshare it currently has.

    1. Re:Bye, Redis by Entrope · · Score: 2

      This.

      The headline is wrong. This is not "Redis changes its open source license" -- as the company acknowledges, this is "Redis goes non-open source". In-house developers are not going to want to build on Redis if its license forbids them from using modified versions for the things that Redis is designed to do.

    2. Re:Bye, Redis by amp001 · · Score: 4, Informative

      You're right that the headline is wrong, but not entirely in the way you assume. The license for Redis itself (the core database, including failover and clustering/distribution, etc.) isn't changing. It's still BSD licensed. What they changed was the license for the extensions they've written (RedisSearch, RedisGraph, RedisJSON, RedisML, and RedisBloom). Now, I personally think that's a mistake (and I suspect they'll end up realizing that at some point), but it also doesn't affect how I use Redis today, nor does it affect how the cloud vendors use it in the services they provide.

  2. Redis alternative in 3,2,1 ... by Qbertino · · Score: 3, Interesting

    Does anyone know any neat Redis alternatives/forks?

    --
    We suffer more in our imagination than in reality. - Seneca
    1. Re:Redis alternative in 3,2,1 ... by Anonymous Coward · · Score: 2, Interesting

      One issue you’re ignoring is that a lot of us open source developers would love to spend more time on open source, but have signed "inventions" contracts which force us to turn over any "inventions" we make, in our spare time or not, over to our employer (California has a "on your own time, with your own computer, as long as it is not related to what your employer makes" exception, but some companies argue they make everything, so the third clause means they own everything a developer can possibly do in their free time).

      I know a lot of companies use Redis, so hopefully there are enough with managers with some clue who are willing to allow some of their developers to take the time to maintain an open source fork of it. Or, maybe the US can be like Germany and mandate the employees must have six weeks of paid time off every year; that would allow developers to contribute to open source during those six weeks.

      It's a question of money, and whether we want to benefit the 1% (more proprietary software) or the 99% (more open source software libre) when all is said and done.

  3. Fork it :) by stooo · · Score: 2

    Fork it hard and right before the license change. :)

    --
    aaaaaaa
  4. What a load of bollocks by drinkypoo · · Score: 3, Interesting

    Earlier this month the Open Source Initiative had reaffirmed its commitment to open source's original definition, adding "There is no trust in a world where anyone can invent their own definition for open source

    Uh, bullshit. Open Source means you can see the source. That's all it means. That's why we have all these various Open Source licenses, and also why Free Software is different from Open Source. When you don't invent the term, which was provably in use before the leading lights of the OSI claimed to have coined it, you don't get to define it.

    Redis IS Open Source. It is NOT Free Software. Equivocating the two is corporate whoredom.

    --
    "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    1. Re:What a load of bollocks by weilawei · · Score: 2

      I agree with you here, but I don't have mod points.

      A specific person probably used the term "Open Source" first, and that person has given a definition and description of intent, and control has been placed in the hands of the OSI.

      As far as I can see, they've got a monopoly on the meaning of that term. I'm not really sure why drinkypoo tried to claim anyone was equating it with Free Software, a thing that the inventor of the term Open Source stated he was deliberately trying to be distinct from. Smells like burnt strawman.

    2. Re:What a load of bollocks by drinkypoo · · Score: 4, Interesting

      Stupid bleeping...I'm not sure how I wrote "probably". A specific person, ESR, definitely used it first.

      First, ESR doesn't actually claim (any more) to have invented it, he claims Christine Petersen invented it. He changed his story.

      Second:
      1993: Jerome Schneider
      1996: Caldera (Written by Lyle Ball, whom I queried on the subject)
      1998: Christine Peterson (Writing in 2006, mind, and providing zero citations)

      So you tell me, who you gonna believe? The citations which prove that its use predates OSI claims by five years, provided by a person (me) who has nothing to gain by continuing this argument except the credibility which naturally comes from supporting the facts, or someone with something to gain economically from making such claims, like Christine Peterson or Bruce Perens? The only dog I've got in this fight is the truth.

      Sustaining this argument over the years (literally over more than a decade) has cost me substantial credibility, but only among people who value prejudice over fact. I'm okay with that. Better to suffer for the truth than promote a pack of lies. Buying into bullshit is why we can't have nice things.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
  5. Re:Missing Information by phantomfive · · Score: 4, Informative

    Redis makes one of the most popular NoSQL databases (I believe it's basically just a key-value store, and people use it a lot for holding high-volume reads that don't change very often, like cookies).

    Amazon wants to take their database, modify it so it can be used as a hosted AWS service, and not contribute back their code changes.

    If that bothers Redis, they should use the AGPL, which specifically prevents that kind of thing. The BSD licenses specifically allow it. Instead, Redis made their own licence, which annoyed RedHat enough to threaten removing Redis from their distro.

    --
    "First they came for the slanderers and i said nothing."
  6. Thanks much. You misrepresent Bruce's words by raymorris · · Score: 2

    Thanks for the link and your research.

    You have, however, misrepresented the statement Bruce Perens made. This unfortunate fact turns your work from useful to bullshit. I'm not sure why you would put in all that time, then produce a lie. You could have lied without doing ANY research.

    Bruce's statement is consistent with everyone else's. Bruce, ESR, Christine Peterson, et all discussed it and agreed upon that term. *Bruce then took the responsibility for drafting and maintaining a definition based upon the Debian document.* Perens never claimed to have originally coined the term, and your claim that he did is crap. Not only is it crap, but it's OBVIOUS bullshit, to anyone and everyone who has a basic grasp of the English language. What a waste of time and valuable research.

    Your attempt to present it as of there is disagreement between the people involved in similarly bull. Neither Perens, nor ESR, not Stallman, nor any other person claims to have originated the term, only to have discussed ans agreed to it at the meeting with Christine Peterson, where she says she brought it up. Nobody disagrees with that. Trying to portray it as an argument between the people who were present is disingenuous.

    You did find a single instance of someone talking about source code that is open; a subject line using the words "open" and "source code" together. That is interesting. It would be more interesting if it weren't buried in bullshit. It would have been more interesting if they said "open source software", but they didn't.

      It would also be more interesting had they used the combined term "open source" as a noun. They used the noun phrase "source code" with the adjective "open", which may seem like a subtle difference. There is a world of difference between mentioning your tube and naming something You Tube, though.

    You've misrepresented what Bruce said. He never claimed to have coined the term. He stated, correctly, that he took responsibility for putting together a definition document for the new term based on the Debian document. That is an undisputable fact. You've falsely misrepresented the different aspects of the story related by the people present as if they were fighting about it. That isn't the case. What a waste of your time and ours.