Bruce Perens, one of the founders of the Open Source movement, is ready for what comes next: the Post-Open Source movement.
First of all, our licenses aren’t working anymore," he said. "We’ve had enough time that businesses have found all of the loopholes and thus we need to do something new. The GPL is not acting the way the GPL should have done when one-third of all paid-for Linux systems are sold with a GPL circumvention. That’s RHEL.
Thanks for the share.
Obviously Perens is one of the FOSS OG figures and he makes a lot of good points. Lately the RHEL/IBM situation has shown a mere license text file isn’t going to keep megacorps from finding ways to circumvent the ideology and the purpose behind it. They have simply too many resources both in development and in legal departments and too many ways to work around the legalese of its intended purpose .Also there’s been an increasing trend where products (Elastic etc) start off with FOSS license and as soon as they gain critical mass, they split their product and switch to their own FOSS-light license and gimped “community edition” downloads. Again, all still legally above the board, but at the same time completely ignoring the intended purpose of the license in the first place.
I think what Perens is proposing is too complicated. I understand that “contract” has far more binding legal fire power compared to a “license”, but as he also points out in the article, it complicates things to the point where it’s hard to adopt. The problem is of course far deeper than just licensing and has its roots deep somewhere in late-stage capitalism and deregulation of corporate entities and those are of course not problems that Perens or the free software community can easily solve. Unfortunately.
It’s clear that something new is needed and I appreciate the work he is doing. I’m not sure it’s the right direction to take, but can’t say I have any rabbits I can pull out of my hat either, so I’ll follow this with interest.
Thanks for sharing!
It’s not the most well-written article, but it does convey what Perens is concerned about - particularly the fundamental flaws we are all accustomed to - usability issues of OSS from a UX perspective and lack of time/interest to improve, the flaws that lead to closed source conversions and clauses. We really do need a new path forward. Even what’s proposed feels akin to the sponsorship model, that I wonder is just a dream more than reality.
Maybe we do need some sort of regulation in the software market to truly incentivize or keep the FLOSS wheel going…
I felt like I was going crazy sometimes with how often people in the FOSS community insist that nothing is wrong when large companies are massively profiting off of unpaid labor that is meant to help people, by turning it into part of their closed-source product, so it’s nice to see that well-known figures in the community are starting to wake up to this being a problem.
I think that non-commercial-use clauses are a good way forward for certain projects, and commercial licenses for others. I wish that the upstream contrib requirements had taken off, but clearly Capitalism and the FOSS mindset aren’t compatible, and capitalism is more widespread.
If you let corporations have something for free, they’ll find some way to ruin it.
I felt like I was going crazy sometimes with how often people in the FOSS community insist that nothing is wrong when large companies are massively profiting off of unpaid labor that is meant to help people
Starting a project as just source-available or with noncommercial clause is just fine. But the definition of free software and open source are pretty unambiguous - software with noncommercial clause can’t be either. The problem really happens when certain companies/projects want the advantages of the FOSS label, but don’t want to make the compromises associated. Any FOSS project has certain advantages that comes with the label - promotion by individuals and industries, widespread training and external contributions. Some companies/projects start off as FOSS (almost always with CLA) and take advantage of all those. Then when they’re popular, they switch to non-commercial, citing competition. Hashicorp and Redis are examples of these. When they cite unfair competition, they’re outright denying the contributions of external players like contributors and industry that popularized it. It’s basically a rug pull.
Another form of this is a recent trend of people claiming that non-commercial clauses count as FOSS. I’ve heard weird claims like the FSF and OSI don’t a monopoly on the definition of what’s FS or OSS. Yet others simply ignore these definitions. Any project that wants to be source-available should compete on their own merit, rather than riding and exploiting the world’s preference for FOSS.
I think that non-commercial-use clauses are a good way forward for certain projects, and commercial licenses for others.
Just want to reiterate - it’s ok as long as it starts as such, instead of doing a bait and switch. But another method is to use AGPL or similar license for all your code. The corporations that exploit FOSS code hate this license. And that’s why they widely promote the idea that copyleft licenses are less-free compared to permissive licenses (less free for them to exploit, perhaps). Unfortunately, many FOSS developers have bought this BS.
by turning it into part of their closed-source product
Corporations convert a lot of FOSS code into part of their closed source products. Using copyleft instead of permissive license is a good way to prevent that. But that aside, there is one class of software whose exploitation can’t be solved with copyleft licenses - cloud software. Many companies that switched licenses were offering cloud services. And then AWS of GCP comes and offers their cloud version, forcing the smaller companies to go source-available. The main problem I see is, why are they cloud software? The main goal of free software is freedom - especially the freedom to privacy, to own the data and to decide on the computation. That’s much better served on local machines than on the ‘cloud’. That’s much easier today with machines that are magnitudes of order more advanced than a decade old ones in terms of computational power and storage. Yet, we see companies wanting to turn all that computing power and storage into mere thin clients with everything from note-taking tool to entire operating systems offered as SaaS. This entire problem happened because the ‘freedom’ part of ‘FOSS’ got de-emphasized in-lieu of profiteering.
I’ve heard weird claims like the FSF and OSI don’t [have] a monopoly on the definition of what’s FS or OSS.
I’m part of that group. If OSI and FSF want to control the definition of something, they should make new and unique terms, not just attempt to take over a concept that predates both of them. (Interestingly, OSI’s website claims no one used “open source” to talk about software before 1998, and that’s patently not true; I remember seeing people use that in IRC channels back in the early 90s). If I came along tomorrow and said, “my org now controls the definition of ‘downloadable software’,” people would tell me to sod off. Even worse “Open Source” and “Free” are both terms with plain-English meanings (which most people naturally assume to be what people are calling “source available”, in OSS’s case). Trying to impose centralized control over a simple phrase isn’t really in line with the collaborative, community-led spirit of the FOSS community, imho
Call it OSI-Approved Zero-Restriction Licensing or something.
Any project… should compete on their own merit, rather than riding and exploiting the world’s preference for FOSS.
Funny, that’s how I feel about OSI stepping in to claim control of that term.
Just want to reiterate - it’s ok as long as it starts as such, instead of doing a bait and switch.
I agree with this for existing projects, absolutely.
I’m part of that group. If OSI and FSF want to control the definition of something, they should make new and unique terms, not just attempt to take over a concept that predates both of them.
Call it OSI-Approved Zero-Restriction Licensing or something.
Whether that term existed before it or not, that’s what people understand now. When talking about FOSS software, those definitions are what people expect - by a humongous wide margin. Calling those terms ‘generic’ is the weakest argument I have heard to dismiss the rigorous meaning people attach to it. Standards are centralized for a reason - so that everyone is on the same page. There’s nothing wrong with it. Claiming otherwise isn’t anarchy - it’s an intent to cause confusion*.
Trying to subvert those definitions and trying to pass of non-commercial as either Free software or Open source software are in my opinion, rather malevolent distortion of an existing paradigm meant to help only the companies that I mentioned before - those that want to exploit the FOSS ecosystem, but without making the necessary compromises. It’s an attempt to exploit a widely-held belief based on a rather vague and frankly misguided technicality.
Non-commercial sources already have an appropriate term - ‘source available’. It’s another generic term with a well-defined meaning. Hijacking the meaning of ‘open source’ and ‘free software’, instead of using this one is the perfect indication of the misleading nature of the hijack. And looking at the prevalence of this, I’m starting to suspect a widespread astroturfing/misinformation campaign.
Funny, that’s how I feel about OSI stepping in to claim control of that term.
Funny, they just used a generic term to mean something, while the exploiters use the term that means something to hide their true intentions and profiteer.
Which software was ruined by companies?
I didn’t read anything but thanks Bruce ha
🤖 I’m a bot that provides automatic summaries for articles:
Click here to see the summary
RHEL stands for Red Hat Enterprise Linux, which in June, under IBM’s ownership, stopped making its source code available as required under the GPL.
Pointing to popular applications from Apple, Google, and Microsoft, Perens says: "A lot of the software is oriented toward the customer being the product – they’re certainly surveilled a great deal, and in some cases are actually abused.
The reason that doesn’t often happen today, says Perens, is that open source developers tend to write code for themselves and those who are similarly adept with technology.
Perens acknowledges that a lot of stumbling blocks need to be overcome, like finding an acceptable entity to handle the measurements and distribution of funds.
Asked whether the adoption of non-Open Source licenses, by the likes of HashiCorp, Elastic, Neo4j, and MongoDB, represent a viable way forward, Perens says new thinking is needed.
Perens doesn’t think the AGPL or various non-Open Source licenses focus on the right issue in the context of cloud companies.
Saved 91% of original text.
I’m obviously a nobody in the field, but (since the mid-'90s) I’ve always seen two issues, here. First, there’s the general problem of not thinking of a public license as a kind of contract, where you get, because you give. If the community started thinking about copyleft or similar ideas as payment or contractual obligations for the use of the software, rather than “restrictions” or “a virus,” we’d probably see a rapid change in behavior, in enforcement or license-writing, if not compliance. …Except that, second, there’s the specific problem that the FSF has always ignored warnings until it was too late and a company went out of their way to offend them. For as long as I’ve watched the Free Software community, people asked what would happen if you embedded gcc on an appliance or served it up on the web, and those people essentially got a response of “the big guy only cares about code that he runs locally, so it’s fine.” And then a company did the thing, forcing them to gasp at the entirely-foreseen problem and issue a new license a decade too late to help. Which I guess is all to say that Perens is probably on the right track, and I wish that we had more people looking at these problems more frequently. I still vaguely remember the Copyleft Next project, but that (clearly) went nowhere.
