Nobody is complaining that Redis ia using the same 'OSS core, proprietary add ons' business model that NGINX has been using for decades.
They're complaining about the weird/impossible 'clause on top of open source' and 'existing OSD' (as if there's another new one) language in Redis' announcement and the 'Commons Clause' website
To be fair most of the complains were about the model. And such complains where pretty pointless since no resource to write Redis core is take in order to write Redis modules. Fortunately (haha) Redis OSS was always just me and sometimes (like now that there is Fabio part time just to OSS side) some other collaborator, very few people allocated there, plus the very important contributions from Redis Labs devs, Alibaba devs, and so forth. I agree that criticizing the Commons Clause makes more sense, I did not follow closely how the license was created, but the process was to let N vendors to use the same clause so that the basic license was still valid but with the added restriction. It was AFAIK a non evil process, but actually now that I read the complains about Commons Clause I can see more how that could be better with new names for each license that each vendor used. Maybe you just cut&paste the clause into the OSS license you are using, but then you call it <company> Commercial License.
> To be fair most of the complains were about the model.
To be fair, if they'd just said hey we're selling these proprietary licensed modules then there wouldn't be much of debate. As others have pointed out, many other OSS projects do just that.
Instead mass confusion has erupted and literally every word debated ad nauseam -- including, now, the model.
This is not about 'the open source community' this is just about Redis. The open source community is doing just fine (and Redis is doing just fine), no need for overbroad or inaccurate claims.
Besides that the obvious point is that every open source producer is entirely free to release their product under whatever license they see fit and if you don't like it then simply don't use the product.
Of course Redis is doing just fine. But that depends on continued development (investment) and continued community involvement. To me it looked like the villagers have been gathering with pitchforks and torches for the last 24 hours and I think much of it is unfair to Redis.
There's Redis (OSS), Antirez (OSS developer), and RedisLabs (commercial company and employer of Antirez).
RedisLabs made some extra software available (nice) under a noncommercial-only license (well duh but ok), pretending some material relation between the noncommercial-use-only license and the Apache License (not nice).
The pitchforks and torches are (to 98%) about that last bit. Everyone is happy that Redis still exists and that Antirez and RedisLabs have a sustainable business around Redis. No one - and potentially especially not the Apache foundation - enjoys the bait-and-switch of using the Apache name for a proprietary shared source license.
I’d like to make all of the software I write freely BSD licensed.
If I do that, it seems with very high probability that it might turn out to be a time sink and a money sink with no benefit other than +reputation. Which is worthwhile, but perhaps not as worthwhile as those fat stacks.
I don’t know. We’re heading for a future where the only people that can afford to write OSS projects that actually influence the world are those who work for FAANG or in academia.
Redis Labs is doing fine but probably despite the license of Redis + the Cloud setup we have currently, just because Redis OSS is massively popular. I think that a less popular project that could survive with AGPL may be completely non viable with BSD. However note that AGPL is a weak protection. It takes just the Amazon legal department to say "ok, let's do it" because it is perfectly legal to just sell as a service AGPL licensed stuff. Let many popular OSS projects switch to AGPL and you'll see Amazon switching mentality about AGPL, IMHO.
AWS is probably unstoppable via licensing, they can do a clean room implementation from the wire protocol - if they really want to. (Though with the Oracle v Google API copyrightability lawsuit ongoing, this might change, but that seems like a small win for a big loss.)
Could go one step further (out of the Oracle playbook) and include a copyrighted poem in the wire protocol itself so that any clean room build automatically violates that copyright too.
GPLv3/AGPLv3 would still force Amazon and the like to share all the updates they make to the software itself, which is good for the software - forcing companies to be part of the software ecosystem - but not necessarily helping the (main or otherwise) developer.
The elephant in the room here is that the usual OSS-as-an-offering model works from the fact that intimate knowledge of the software and visibility creates a strong market position for selling services (SaaS or consulting) that others cannot match. Whereas if "others" includes Amazon, GCP etc., their economies of scale and market visibility tilt the playing field in the other direction, making it harder for the OSS-as-offering model.
AGPL would do nothing to stop cloud providers having hosted versions.
What it would do is mean that a lot less people would be able to use the software as it's a license that is not allowed at most companies. You can argue that this is for the wrong reasons, or it's due to fearmongering, but that doesn't change the fact.
I think you have it backwards: a less popular project could be viable with BSD and wouldn't survive as AGPL.
Yeah, because the 'sell support/hosting' model has been such a failure /s You don't have to be in academia to afford to write F/OSS nor is it financially draining to do so.
It's not that hard to make a living, if you write custom software instead of selling products. The niche I'm aware is development of modules for Odoo (a kind of SAP alternative); I've worked for such a small bootstrapped company for years, and 90% of what we wrote was A/LGPL licensed. Most clients don't care if you release useful generic parts of the code as OSS modules, nor if you use such modules from other developers.
I know companies following this model all over the world: Spain, Brazil, India, Argentina, Venezuela, etc.
> I’d like to make all of the software I write freely BSD licensed.
Why? If you want to make money off of it, there is nothing wrong with selling proprietary software. I'll gladly open source lots of stuff but I'm not just going to work for other people for nothing -- that's just madness. And that isn't really what free software / open source is supposed to be. It's not a platform for martyrdom.
> We’re heading for a future where the only people that can afford to write OSS projects that actually influence the world are those who work for FAANG or in academia.
I think the strength of an OSS project (or really a single individual at that stage) can be to innovate - make something that in a corporate setting, FAANG or not, can't be pursued because it hasn't a chance to become projected since business value isn't tangible, and anyway needs meandering around to find solutions in its problem space. Once something demoable is created, it might attract corporate backing. But corporate development is actually very bad at developing the kernel of something not seen yet. Maybe there should be a market place for initial designs/concepts to grab beyond VC funding?
For most of my time as a software developer and engineer I released code under GPLv2 and BSD licenses. It paid the bills because I knew how to sell what I was doing. I charged enough to cover the time sink. Now I am a world-class expert in some common BSD-licensed software.
There still isn't a simple, sustainable and uncomplicated model for developers to earn a full-time living selling their open source product.
The Patreon approach is not sustainable in the long-term, neither is the company sponsorship route.
Charging for support as the principal source of your income is probably one of the least appealing options for most developers.
Open-core is one option - GitLab wouldn't be a financial success without it. Dual-license is another option.
What does the FSF (Free Software Foundation) have to say on the matter? Charge for distribution [1]. Yet, no-one can realistically charge for distribution today - not least when so many projects are hosted for free on GitHub.
If you want to make source code available and still be paid for your product - consider something like the Fair Source License. It's neither closed source or open source. It will appear blasphemous to some open source advocates, but it feels like a reasonable compromise between closed source and permissive open source licences: https://fair.io/
This software is licensed under a mixed OSS/commercial license, the Apache 2 OSS license with the commercial Common Clause restricting certain freedoms.
The problem here is that the end result is not the Apache 2.0 license anymore. It’s something else. Using the Apache name in this context raises some serious eyebrows.
By all means keep the common core concept for Redis but I urge you to get as far away from this toxic “Common Clause” pseudo-license as possible.
That would be actively misleading - "This software is licensed under a mixed OSS/commercial license" implies the traditional dual-licenced model where you offer both an open source licence (e.g. GPL) and a commercial licence. Those who want an open source licence can choose that one, and those who want some extra options can choose the commercial licence. That's not the case here, there's no choice to get an OSS licence.
The difference between this and the restrictions in other OSS licences is that the "common clause" restricts core freedoms - if you attach it to something, that something ceases to be an open source licence, this clause transforms into a commercial licence that prohibits free use of the software. There's nothing mixed about it, it's a "not-OSS" licence.
Just because you disagree with the terms of a license doesn't make it "proprietary". Is the AGPL also "proprietary"?
It would be far more productive to state your particular disagreement(s) than to inaccurately apply labels you think will tarnish the target of your ire.
I will remove "proprietary" from my post. The gist of my objection is not the content of the license, but the implied endorsement of the Apache Software Foundation.
I think that Common Clause is proprietary while AGPL is open source. This should not be a matter of opinion. AGPL asks you to redistribute the work back, however if you complain with the share back clauses, you can do whatever you want with the code. This is not the case of Common Clause. However if we don't focus too much on the Open Source license yes/no, we can see how there is a spectrum of licenses... and Common Clause allows many freedoms to actual users compared to other non-OSS licenses, and in certain respects even VS AGPL.
Thanks for the thoughtful reply. I now understand and agree with this position, and observe that it was the "spectrum" aspect that caught me out: I'm used to thinking of a "proprietary" license as being far more restrictive than Common Clause. It's possible see both the OSS and proprietary aspects in this license.
Any license that includes the ""Commons Clause"" is by definition proprietary, it precludes incorporating the code into anything commercial and even prohibiting anyone from providing paid support for it (which seems even more extreme).
It's a set of anti-commercial license terms and is definitely proprietary by Free Software and Open Source definitions.
That is exactly why it should be renamed. It seems to be designed to be confusing. No one will care as much, if they just made it clear that they are transitioning parts of it to a commercial license that is not confusingly named. They are trying to make it sound like Apache is endorsing this license.
> Open core is a fairly honest way to go about developing open source software.
Only if you're equally honest in admitting it's the most risky approach wrt sustainability and therefore the one most subject to strange capitalization mechanisms (and endless blog post justifications for them :-)). They go out of business all the time, leaving stale software in their wake that might have otherwise had steady maintenance if independence and growth were less paramount (like before commercialization where, in many cases, the burden was just as large). Either way, it's definitely the hardest model and if you have a piece of open source software that must have growth or you are currently no longer able to sustain it, I would try to find any other way possible to finance it...but I agree sometimes it's just not there so you build a business around it.
And I'd change "developing" to "financing" since there are many more honest ways, just not at scale.
I find this article's angle to be somewhat refreshing, simply because there's a segment of the OSS world that is allergic to open discussion of funding. Some of the anti-commercial arguments I've seen are absolutely fantastical, amounting to demanding OSS to be a massive entitlement to the free time of others.
This also gets to the biggest criticism I have of the FSF: the only measure of success is a narrow definition of "freedom" -- with little to no discussion of what social systems, economic structures, and advocacy might be useful and/or necessary to support those aims.
I've also seen that many people FOSS users mainly because it's typically free of charge. It really should become more normalized to 'pay' for Free Software and give money regularly to projects the user supports and finds useful.
People underestimate how difficult it is to create a popular open source project. It's very competitive. You're competing not just with other open source projects but also with SaaS solutions from well-funded companies; and you have to compete with a $0 marketing budget.
Also, if the project does become popular eventually (which is very unlikely), maintenance is a lot of work. To make matters worse, there is a general tendency that if things go wrong, the community will blame you but if things go well, you get almost no credit.
Even if your project has several thousands of stars on GitHub, it means nothing financially. Most of these OSS authors can't afford to quit their day job even though their project may be generating millions of dollars of value for other people.
Money is an important topic to open source authors because after a decade of working for free during nights and weekends, most of them would love to be able to afford to quit their day jobs to focus exclusively on their projects but they can't do that; they're tired but they have to keep soldiering on with the 80+ hour work week.
Being an open source author is not a wise life decision right now. I think that any business innovation which helps keep open source projects going should be welcome.
>after a decade of working for free during nights and weekends
Think I spotted the problem.
This whole thing feels like someone martyring themselves for a cause and being recognized as a hero, but then complaining that martyrs are heroes and we should be _supporting_ our heroes and not forcing them to be martyrs. But that was the whole point to begin with and why there was any opportunity for recognition at all.
It's not that simple, what would be this industry without Linux, without the FSF, without the BSD network stack (that is in all the macs...)
The growth of our industry, and probably of a good chunk of the whole economy for the last 10 years has a lot to do with those people working during nights and week-ends imo.
No they’re not, because it’s increasingly expected of developers to have a track record of open-source contributions if they want to move up into the higher tiers of their profession. That amplifies inequalities in the field between:
1) the top 1% employees in the field who have the luxury to invest unpaid time, and the social clout to make their projects more successful, or to join more successful projects. (example: how much more likely are you to be a key contributor to Kubernetes if you are, or were at some point, employed by Google?);
2) the other 99% who cannot afford to invest the spare time, and do not benefit from the social proximity to successful open-source contributors (and success in open-source is as much about social proximity today than technical skills, hopefully we can change that in the future);
3) The corporations who benefit financially from open-source, sometimes enormously, with very little pressure to contribute back because of the aforementioned social pressure on individuals to contribute on their free time. The expectations are so low that even modest corporate contributions (joining or starting a foundation, hiring a few engineers, paying for a conference) will be hailed as generous and benevolent, because the bar is so incredibly low.
>it’s increasingly expected of developers to have a track record of open-source contributions if they want to move up into the higher tiers of their profession.
Is it really? Of the people I know who have had excellent career paths in software/computing none have a track record of open source contributions.
> Of the people I know who have had excellent career paths in software/computing none have a track record of open source contributions.
Right, that’s how they built their career in the past. All things being equal, it will become gradually harder to build a career that way in the future. I already know several major employers who look at a candidate’s github activity as a factor in the standard hiring process. And that trend is only accelerating.
If employers score higher the candidates with a high Github activity, then eventually the employers get workers who continue to spend a lot of their time in Github.
If these people spend a lot of their free time, and perhaps even some of their professional time, in Github contributing to various public projects, then they might not necessarily have 100% focus on the employer's tasks. I am not sure if this is what the employers expect or even want to happen.
The intention is probably to use Github activity as a pre-screening proxy metric for technical capability and skillset. But... why would someone who is contributing a lot to public projects cease to do so, once employed? Is there a risk of getting a partially distracted, though skilled person?
If the wish from the company then is to ensure they get the most performance for themselves, the logical step would be to cut down on the public contributions via e.g. legal and contractual means.
So... why would someone who enjoys contributing to public projects join a company that selects based on Github activity, if it is in the company's interests that further contributions should be reduced?
Those developers either did it for the passion, which was it's own reward. Or, much more often, they actually got paid to do it. Doing it for free and then expecting a payoff later is very different.
Just because the motivations and reward mechanisms here aren't adequate in your eyes in terms of fairness doesn't mean that this system is somehow "out of equilibrium" or whatever and is headed in a bad direction.
In "Bullshit Jobs: A Theory" the author has a really interesting theory of how open source and employment in software interact.
His model was, that programmers are expected to donate their time to OSS creating core components, in order to get jobs integrating those core components into commercial systems.
When you think about the issue like that, it makes a lot of sense, but its a terrible deal for the programmer.
A part of me thinks open source is this epic joke and conspiracy played on programmers. Some cabal of business owners somehow managed to convince us that it's admirable and ideal if we build all this software infrastructure for free so they could profit off of it. And that's pretty much what is happening and things like OpenSSL show how terrible it is.
You're not really wrong, there is an epic conspiracy. Ours is the only profession I can think of that's expected to work slave hours for no additional compensation. At least when a lawyer works 80 hour weeks, he might make partner someday.
Exactly. We'd all be better off if more people worked for free working as medics, nurses and doctors, saving lives, but nobody seriously expects them to do any of that work with zero compensation. But somehow it's okay for programmers to build the tech the entire world depends on and get nothing in return, and programmers defend it!
In a Star Trek utopia, that idealism would be great, but we live in a world where people are exploited wherever possible to ensure endless profit.
I think the main reason people do OSS is because they can work on a project they chose and that it's an interesting technical challenge for them. Giving the work for free and being recognized for it is what they're also after, but I don't think anyone would refuse to get paid doing the same thing.
Right all this talk of "what it takes to be a success in OSS" ... programma please!
"Success" in the OSS world is putting code out that others can use. Either they find it useful, or they don't. If they don't that's not failure. That's ... someone else put out something better, or that your user community is small or just not out there.
OSS is not a business model. It's a SHARING model.
God it's just like the music scene these days. Did you get into it for the Art? Cool you're gonna be ok. Did you get into it wanting to be a rock star? Might as well buy a lottery ticket. Your chances are better.
OSS success is providing something of use to a community that contributes back to it.
Yeah, maybe you can find a way to make a nickel off that, but that's not what it's FOR. It's not a marketplace, it's a public library.
This celebrity/glamor seeking is so apparent on github. Every single project's readme.md reads like marketing copy these days.
Just like the music scene again. Reminds me of the 90's when the record labels still had money and steamrolled the alternative scene looking for a cash grab.
Making software open source doesn't mean you can't make money from it. Depending on the details of what you build, who uses it, how you license it, etc., it may be harder to make money from it than if you made it proprietary. But that certainly isn't something you can just take as a given. You also can't take it as a given that the project would make money if it was closed source either. Maybe you made something nobody needs. Or they need it, but there are better alternatives out there (possibly some that are Open Source). Who's to say?
Somebody yesterday (maybe DannyBee, not sure) made a very insightful comment in saying that sometimes the value created by Open Source software is created exactly because it's Open Source.
Anyway, you can always charge for software even if the code is Open Source. You can stick it on the web and make it a SaaS app, or you can make the source available but charge to download binaries (ala JBoss back in the day), or you can make source and binaries freely available and sell subscriptions, etc. Aaah, wait, I know, you're asking "who would buy a subscription for something that's free to download?" Companies. Companies will, for things that are mission critical, because they typically need for their to be a contractual relationship (eg, somebody to sue) in place. Or you have a manager who needs to spend her budget or lose it, and she understands that you need to make a profit in order to stay alive and keep enhancing the software. Etc., etc, yadda, yadda, ad infinitum. My point is, it's not as simple as "Oh, I made this Open Source, so there's no way to profit from it".
The underlying issue they are trying to solve is that developers believe they should be able to extract some of the wealth they feel they created. In open source, that wealth is often only created because the software was free. Otherwise, people would have used something else that was free but worked well enough. Developers like to often argue this isn't true, but history shows it to be true basically always :) So saying you should be able to extract this wealth is probably wrong. Saying you should be able to get paid a reasonable amount of money is not.
That’s the problem: we all take this particular public library for granted, but there isn’t a sustainable model for paying the librarians. Then a librarian tries a creative way to continue serving us without starving, and we tell them “you’re just doing it for fun, if you wanted to make money you shouldn’t have been a librarian!”. This is an ignorant response, because it places on the librarian’s shoulders the responsibility for solving a problem that concerns all of us - because we all benefit from open-source more than we contribute to it. That’s not sustainable.
Nope. OSS was here long before redhat and everyone else, and it will be here long after.
Because, in your analogy, the librarians are all volunteers. That’s how we got here. If the pain of not having a problem solved is great enough the volunteers will fix it. Or they won’t. If you feel like the pain level is high enough. Hey maybe YOU should try volunteering to solve it.
This is what makes OSS powerful. It is a community service. If you aren’t happy with the service p become a member of the community and contribute.
All open-source contributors are certainly not volunteers! Open-source contributors come in all shapes and sizes, and how much they get paid for it has no incidence on the quality or authenticity of their work. It does, however, have a huge impact on who can contribute, and how much.
Since we as a society rely on open-source enormously, it seems to make sense to try and allow as many people as possible to contribute as much as possible. And that requires giving up on antiquated notions that “only unpaid open-source is real open-source!”. Wouldn’t you agree?
> Since we as a society rely on open-source enormously, it seems to make sense to try and allow as many people as possible to contribute as much as possible.
We already do, anyone can do it and we shouldn't restrict that; that's the point.
> And that requires giving up on antiquated notions that “only unpaid open-source is real open-source!”. Wouldn’t you agree?
Not GP, but I most certainly don't, nor is it antiquated. For that matter, calling volunteer work "antiquated" is... A bit cynical, to say the least.
I also don't see how this follows from what you said before, could you walk me through your reasoning?
> For that matter, calling volunteer work "antiquated" is... A bit cynical, to say the least.
I emphatically did not say that volunteer work is antiquated! I myself do plenty of it - but I also acknowledge that I’m privileged to be able to afford it. What is antiquated is the notion that only unpaid work is authentic open-source work. It’s important to realize that providing unpaid labor is a luxury. If you exclude or devalue paid work, you exclude and devalue the people who cannot afford to do open-source work for free. As it turnd out, that’s the majority of people.
"The top 10 organizations sponsoring Linux kernel development since the last report are Intel, Red Hat, Linaro, IBM, Samsung, SUSE, Google, AMD, Renesas, and Mellanox."
Hmm that report is confusing. I thought the fact it said sponsoring meant money donated to the linux foundation. But page 14 of that report seems to imply it’s changes contributed to the kernel. Not money.
There’s a sustainable model for librarians who don’t want to be paid.
OSS is diverse and there are many, obviously, who don’t mind not being paid directly.
But what’s confusing is that there are proprietary licenses. Anyone who doesn’t like volunteering time to OSS can write commercial software and charge for it. That’s the model that works for them.
But the talk of a 40 year old model not being sustainable is pretty funny. It’s more sustainable because it takes so much volunteer time.
> But the talk of a 40 year old model not being sustainable is pretty funny.
It’s not the open-source model I called unsustainable, but a particular response to new forms of open-source monetization.
At any rate, the current model - of open-source as a mainstream R&D model, and as the critical infrastructure for the largest businesses in the world, primarily funded by corporate sponsorship and venture capital... that model is definitely not 40 years old.
You could argue that the model originated at the Linux hype of 1999, when Red Hat was the hottest IPO and IBM was spray-painting penguins on the sidewalks of San Francisco... Or you could argue it really started in 2004 when the Google IPO showed how much more scalable and profitable a business can be when you don’t pay software licenses.
In any case, the current model for open-source is really not that old, and it’s too early to tell how sustainable it really is.
Or you could argue that it originated with Unix coming out of Bell Labs and being passed around universities on reels of tape, with new extensions contributed along the way.
When Stallman thought seriously about open source in the 1980s, it wasn't because he was the only one to ever imagine a culture of sharing around software and information. I think he was reacting to a sense that an existing culture was being erased by growing commercial interests.
One could also argue it originated with gentleman-scholars circulating letters about their new physical science discoveries during the renaissance and feeding into the subsequent industrial revolution. But, I expect it is an emergent property of humanity which is probably exhibited throughout history and pre-history.
It's essentially the same idea of commons and culture applied to yet another domain of knowledge and technology. Was it sustainable for the first farmer to teach his techniques to another? For one chef to teach another how to prepare his dishes? Open source software can be sustainable as long as there are needs and resources available to produce and execute software that does something different or better than could be done without the software (or at a lower cost than other available options). The subsequent sale of additional copies need not be a motivating factor at all.
> Or you could argue that it originated with Unix coming out of Bell Labs and being passed around universities on reels of tape, with new extensions contributed along the way.
That was not motivated at all by the open source movement. The only reason AT&T Bell Labs didn't charge a significant amount for UNIX is because they were under a consent decree which precluded them from entering other industries at the time, see https://arstechnica.com/tech-policy/2011/07/should-we-thank-...
It doesn't actually matter why Bell Labs allowed it to happen. The cultural evolution of Unix development involved many other participants who could not have cared less about Bell Labs' interests. The later developments reflected another phase of culture where certain people started to think about the implications and work the system to their own ends.
> That’s the problem: we all take this particular public library for granted, but there isn’t a sustainable model for paying the librarians.
I don't think the fact that RedisLabs doesn't have a sustainable revenue model indicates that open source infrastructure in general doesn't.
It's true that there are a number of relatively new venture-funded OSS infra firms that were built on a step 1, build software, step 2 ..., step 3 profit model. But I don't think that's an OSS problem so much as a startup culture problem, as the same problem is rampant in startups in other spaces.
> Then a librarian tries a creative way to continue serving us without starving
It's not particularly creative or novel; the basic outline has been tried since the 80s at least.
> This is an ignorant response, because it places on the librarian’s shoulders the responsibility for solving a problem that concerns all of us - because we all benefit from open-source more than we contribute to it. That’s not sustainable.
I don't understand this. It seems the argument is "I can't keep giving away software for free unless you pay for it". Then it isn't free software, right? Nothing wrong with that, most of us write proprietary software and sell it. If it's worth more to the people who buy it than what it cost, everyone still benefits.
We use open source software to do our work more effectively. There are all sorts of OSS projects, many entirely run by unpaid volunteers. Many people, myself included, sometimes contribute code to the OSS projects, to benefit myself and anyone else who happens to use it. I don't expect anything else.
But you can't feel entitled for people to pay you for work that we all agreed was volunteer. It's not worth it to do OSS anymore? It's ok to stop doing it. There are enough of us that are rewarded by just being able to contribute, we'll have our own projects. So, it's ok for Redislabs to make their code proprietary and make money off it, if they can. There's nothing wrong with that. But the rhetoric that boils down to "people have been stealing from us by using our code and not paying for it", "it's not fair that they are using our code and making money and not giving us any" is wrong. You gave the code away for free. That was the understanding before you wrote the code, and after it was written. That was one of the reasons people were even using the code. It's not an injustice.
When someone gives you a nice gift, they deserve your gratitude. But if they turn around and say "you never paid me for that", then it wasn't really a gift was it?
Redis remains open-source. RedisLabs is using a “hybrid” license for its enterprise add-one, with some properties of open-source but not all. So you can use the source, modify it for yourself or for others. But if you want to make money from it, you have to fork.
The goal for RedisLabs is to make more money from the enterprise add-ons they’re developing. A lot of that money (perhaps most of it) goes right back into maintaining Redis, which remains open-source.
So it absolutely is a matter of funding more open-source gifts, and not taking back gifts.
That makes sense, and like I said, I don't have a problem with what they are doing. It's what they said and how it was presented and how it has been defended. Like evil BigCorp has been stealing from them.
Just a "Hey what we are doing now isn't working, for our business model we need to make more money and we need to make some of our code around Redis proprietary so we can sell it. Don't worry, Redis itself will remain open source, free to use for everyone."
Instead it was "today’s cloud providers have repeatedly violated this ethos by taking advantage of successful open source projects and repackaging them into competitive, proprietary service offerings". It was open source, they could do that. If you want to make them pay in the future, that's fine. But they aren't the bad guys for using OSS.
Anyway, I hope it works out well for RedisLabs, but I also hope that the licensing of OSS software is simple and open.
> we all benefit from open-source more than we contribute to it. That’s not sustainable.
This way of thinking is, in my opinion, destructively pessimistic. Everything is not always a zero sum gain. We can all benefit more than we contribute and it can still be sustainable. For example, 10 people can spend 10% of their time on a project, and get something out of it that is worth 50% of their time. They took the equivalent of 1 persons work and made it worth 5 people's work. I think most popular OSS projects are orders of magnitude more effective than that at creating benefit. Thousands of people contributing code, a few times a year, makes a project that supports all their careers and provides software that benefits society as a whole.
I get that the problem is the undue burden on the relatively few people who organize this work for really big projects. I've noticed that some of these people are employed by large corporations who encourage their OSS work, because the result, and the thousands of contributions from others they harness are very valuable to these organizations.
I wouldn't be surprised if the large corporations that were listed as the ones using Redis software without paying for it would be willing to sponsor a few employees to represent their interests in the OSS projects. That works for the individual engineers, but that doesn't exactly work for RedisLabs to make $$. So I still don't see RediLabs as a beleaguered crusader trying to save OSS.
I think everything you said is also true of musicians, artists, actors, writers, athletes, game developers, youtubers, and twitch streamers, just to name a few.
The problem for all of these categories is that lots of people want to do the job. There is basically an unlimited supply of people that like to write novels, make programming languages, play soccer, or make video games. That makes it a buyers market. "What? This iPhone game is $0.99? No way. I'll play one of the million free iPhone games instead."
For every singer, basketball player, or instagram personality that hits the big time, there are thousands that never made a dollar doing it.
For better or worse, authoring an open source library is one of those things that there is no shortage of people willing to do, so I am not optimistic about use of "business innovations" to solve this problem. If those innovations are in any way non-consumer friendly, consumers are just going to choose a different open source tool/project.
Sure, I agree that consumers use video games, for example, for personal enjoyment, whereas cloud providers use OSS to make millions.
But that difference does not change the market dynamics in any way. In both cases, the effects of a near unlimited supply of creators is having the same effect, regardless of how their creations get used.
Perhaps the difference changes the ethics of the situation, I don't know, but I remain equally skeptical of "business innovations" being able to help the OSS creators as I am "business innovations" being able to help aspiring rock stars and shortstops.
Except games either (1) are pay-per-purchase, or (2) are free-to-play and subsidized by whales. In both cases, high usage games tend to make enough to justify the economics of continued support and development.
That's almost exactly the opposite of very popular OSS software, which generates an endless wave of free users. And generally no whales. And high demands for bug fixes. And serious temper tantrums (this has happened to me too) when you explain to users of your free software that you owe them exactly fuck. and. all. and won't be hopping-to in order to fix something for them.
> In both cases, high usage games tend to make enough to justify the economics of continued support and development.
Sure there are some runaway indie hits that make money to cover the development. And Linus Torvalds gets paid $10 million per year to work on open source. But nearly all indie game developers never make anything from their games, same as OSS authors. (Indie games are an the appropriate comparison since we are talking about independent OSS authors, not Facebook developers working on React).
In any case, I am not arguing that all of these endeavors are equivalent in all ways. My only point is that all of the fields have an oversupply of creators. The result is that creators don't have leverage over consumers. To really spell it out: if OSS library creators try sticking the commons clause or other "innovations" on their projects, most businesses are just going to go find another library that does the same thing, without the restriction.
Perhaps a good open source analogy might be libcurl. And just take the many ruby bindings. Since about 2007 or so I’ve maintained a libcurl ruby binding. Other than perhaps using it in some of my own commercial works - that project has netted me $0 and yet has plenty of alternative open source libraries that compete with it for the mind share of the ruby community.... I want great open source projects to prosper so that we can use them to build cool things solve real problems and make real money
The big difference is that those Indie games make money for no one. It isn't like there is some giant cloud company hosting your indie game and selling it to consumers and cutting you out of the equation. There is no demand for that game. On the OSS side there is often huge demand with no compensation to the developer. I'm not making a judgement on whether that is good or bad, I assume most OSS developers know that is the way it works, but I am saying that it isn't comparable to a bunch of games that have no demand and people are making them for their own satisfaction.
I have an OSS project used by 1000s of people and many huge companies. I don't get compensated for that directly and I don't expect to. I'm just happy to be able to use it on my own projects no matter what company I am working with.
Thats the deal with open source. Don't like it commercialize and see if you can make it your day job. You'll have to pay programmers too though, no more free work from the community.
Because none of those cases have people giving permission for their work to be used by others to make millions of dollars. If someone gave their work away into the public domain, and then someone used that work to make millions, their shouldn't be an expectation that the other person somehow owes the other money.
If you think your code is worth money, sell it.
LGPL/GPL/AGPL exist for a reason. Not my fault you used BSD/MIT and didn't consider what you were explicitly allowing and encouraging people to do.
> their shouldn't be an expectation that the other person somehow owes the other money.
There btw.
Also, it would be common decency but we live in an era where if something cannot be enforced in court then you should not be surprised somebody is [not] doing it. Oh well.
That's the whole point, isn't it? That's a good thing, right? Of course society at large should incentivize and facilitate the creation of open source software, but it seems the alternative you're suggesting (that the people making millions must financially support the original creators) would render the software proprietary.
As a matter of fact yes, if I were an OSS creator and people made millions from my work, some basic appreciation would be quite nice actually. If companies X, Y and Z make $10 million a month and this is partially enabled by my software, giving systemic donations in the orders of $10k - $20k a month would only be common sense in my mind.
(And I know I would have a "donation budget" if I was a businessman because I am not stupid and I know the work of which people outside my company facilitates my business further.)
But appreciation is not what businessmen do. 99% are cut-throats in all areas of their lives.
When the concept of open source software was first introduced, it was seen as a desperate way to get users but it was an effective strategy at the time.
Today, giving software for free is not enough - I think that the next trend will be to pay people to use your OSS software. Probably with cryptocurrency.
>When the concept of open source software was first introduced, it was seen as a desperate way to get users but it was an effective strategy at the time
This was not the motive for free software's beginning, nor does it describe the beginning of the term "open source." The concept rose out of early computing allowing and sometimes encouraging users to share and improve software, and the true start of the movement envisioned a model similar to Red Hat, where the software is sold though sharing is not restricted.
Forcing user payment is effectively impossible for open source software, someone can just buy a copy and redistribute it without the required payment, so your prediction is a movement away from open source software.
Forcing is not possible indeed. But I think this is comparable to the tax system, not everyone contributes equally or as they should but most public utilities still work from those people's contributions who are honest. It's kind-of depressing to think about people trying their best to freeload instead of paying a (symbolic) price, don't you think?
>It's kind-of depressing to think about people trying their best to freeload instead of paying a (symbolic) price, don't you think?
Not particularly. It's depressing to see businesses profitably exploit that work, and irritating to see people develop a sense of entitlement without ever contributing back.
Though I wish more general users donated to projects, people using the software without paying seems better than people not using or paying for the software, and far better than paying and being unable to use the software.
Depressing it is but I found this way of thinking self-destructive. We cannot change the fact that most people view themselves as the best hagglers ever. It is probably the only area in their lives where they have actual influence. Maybe if the system in general was not so cut-throat there would be less hagglers (although I kind of doubt it; for many people it's their air and water).
In the end it's about what choice you make. If you choose to help the community then you must be prepared that it will be a thankless and hard job. If you cannot bear it, don't help the community.
Trouble with open source is, many people don't realize this phenomena until they are burned out.
There is basically an unlimited supply of people that like to write novels, make programming languages, play soccer, or make video games.
There is an unlimited supply of people who say they would like to write novels, make programming languages, play soccer [professionally], or make video games. The supply of people who have the drive and patience to complete a novel or programming language or video game for free or on spec is arguably much smaller. The same is also true of open source libraries. There may be an effectively unlimited supply of GitHub repositories containing new JavaScript frameworks and template languages, but the majority of them haven't gotten very far past the README file. And, we haven't even addressed the question of whether they have the skill and talent to make a good novel or programming language or video game or open source library.
Having said all that, I think the original article is on point in at least one crucial respect: all of these cited kinds of work require a huge investment of time, and it's very hard to sustain that investment without money. This is why I always get very frustrated with the particular kind of pseudo-libertarian who blithely asserts that artists and writers don't need to be fairly compensated for their work, because clearly they'd keep doing it solely out of the love of creation. In reality, artists and writers--and open source programmers--aren't likely to put in "second job" level hours without compensation indefinitely. They're going to have a family, or develop more relaxing hobbies, or possibly just get frustrated by the ongoing effort.
I don't know if "business innovations" can solve this problem, but I think it's worth looking for more ways for developers (and writers and artists and...) to bring in some kind of stable recurring revenue.
I agree that some kind of disruption of this space would be great, to reward creatives.
...but practically how do you do it?
The parent comment is quite right, going from 'pay x' to 'pay y' is one thing... but from 'pay nothing' to 'pay anything' is a massive and difficult step.
You can't make it mandatory; theres a glut of stuff out there, and people will just move on to the next free (perhaps crappier, but free!) version.
'business innovations may help' is just a nice way of saying 'well darn, nothing we've tried in the last 10 years to fix this has worked... hm. Don't really have a solution for this, but it'd be great if someone figured one out!'
...which I totally agree with, sure, but its not really helpful or practical.
Seriously, nothing people have tried (tips, patreon, foundations) has worked very well.
No one is going to solve this problem any time soon.
Of you're an open source library author/maintainer, you've got to acknowledge you're not just magically get rewarded for your efforts; you have to work for it, and (probably, for most people) your attempts to monetize it will fail.
If I had an answer to that, I'd be a full-time fiction writer. :)
Seriously, I suspect most people aren't going to ever make a lot from creative efforts, which we'll lump open source libraries into; the vast majority of us are going to be much farther out along the low long tail of the compensation curve.
I suspect the only thing that will help -- not to the degree that everyone will be able to make a living wage doing whatever they want, but to the degree that more people will feel like they're getting something back for the time they're putting in beyond the Warm Fuzzy Feeling of Sharing -- is a cultural shift of sorts. We need to uncouple the concept of "zero marginal cost" from the concept of "should be free." (I'm aware that leads down a whole rabbit hole about IP and "piracy," but I'm leaving it aside.) In the open source world, we may all say we're here for the free-as-in-freedom part rather than the free-as-in-beer part...but we're sure disinclined to pay for those beers.
> This is why I always get very frustrated with the particular kind of pseudo-libertarian who blithely asserts that artists and writers don't need to be fairly compensated for their work
Who gets to determine what is “fair compensation” and to whom?
Who gets to determine whether you're being paid enough for whatever it is you do? Who gets to determine whether college tuition is reasonable? Who gets to determine whether you paid a fair price for that burger? Who gets to determine if "who gets to determine..." is actually a good faith question?
Universal Basic Income, adjusted per country, is a very good start. Probably assign varying coefficients to it depending on demand and supposed rarity of the job. There are not that many building architects for example so probably multiplying UBI times 2.5 is "fair".
> Universal Basic Income, adjusted per country, is a very good start. Probably assign varying coefficients to it depending on demand and supposed rarity of the job.
There's nothing wrong with open source as long as you are honest to yourself regarding the intentions.
If you go into open source expecting to get paid, then you're gonna have a bad time.
If you go into open source for the accolades, then you're gonna have a bad time.
If you go into open source with a true desire to improve the state of the art, then you might be able to succeed.
We (https://sheetjs.com, our most popular library https://github.com/SheetJS/js-xlsx has over 11K GitHub stars) combat that by intentionally removing our names and egos from the equation. We are happy to see others use our open source code in their applications, and are constantly surprised when high-profile projects like React end up using our code (https://github.com/facebook/react/blob/v16.0.0/src/renderers...). Our intentions are clear: the open source offerings give us a test surface far larger than any proprietary solution, and for that we are extremely satisfied with the results.
If you want to tell everyone else that they should be in OSS just for advancing the state of the art, you might want to reconcile that with SheetJS earning money from non-OSS extensions and consulting services, because that discrepancy certainly doesn't look very honest of you.
Your interpretation is far from the original point "There's nothing wrong with open source as long as you are honest to yourself regarding the intentions." If you're entering OSS to make money, it's incumbent upon you to find a business model.
Many open source developers start with intentions like "advancing the state of the art" and over time become disappointed because others build businesses and profit off of their creations. The intention at the onset wasn't to make money, and continued investment of time and money into OSS should be weighed against the original intention.
We are elated when we find others using open source SheetJS libraries and code in their profitable business applications. Others may regret and wish they didn't open source, but you don't know how things would have played out if you didn't open source. The yardstick we apply to our open source contributions is whether we're advancing the state of the art, since that's why we entered the space in the first place.
> There's nothing wrong with open source as long as you are honest to yourself regarding the intentions.
My takeaway from your bullet points is more specific – that you shouldn't bother with open source unless you want to advance the state of the art while being perpetually unpaid and unrecognized, which I can not agree with in any way.
> Many open source developers start with intentions like "advancing the state of the art" and over time become disappointed because others build businesses and profit off of their creations.
I just don't buy this distinction of starving artists vs commercial enterprises. Especially as the project grows in scope, complexity, and popularity you might need to spend more time on it to keep it in good shape, and time does not come for free.
While you might have been willing to spend 10 hours a week on it for free, the now-grown project would really benefit from you working on it full time, but that's not feasible to you, so you turn to monetizing it. There is nothing wrong with wanting to both advance the state of the art and to be paid for it.
It's not about being disappointed with others making money off of your project, it's about ensuring sustainable development of your project.
If OSS projects weren't funded by a chance of money or fame, we'd have much less such projects available to us, especially significant ones, because not everyone who currently contributes to OSS can afford to or even wants to work full time for no pay. I don't know why that would be a desirable outcome to anyone.
There's a better way to frame this: Do you think an open source project creates a perpetual obligation? Even if many other developers depend on your project, are you forced to continue or can you walk away? If you assume you can walk away, you can then ask whether it makes sense to stop working on a project. And that answer requires a bit of introspection, being "honest to yourself regarding the intentions".
"advancing the state of the art" is fundamentally different from "money" and "fame" insofar as the latter involve social effects that you have less direct control over. To make money, someone else has to pay you for something.
Interestingly, one line from the original article jumped out to me:
"Lastly, we have consulting and support, which I think is a dead end. It certainly works on a smaller scale with individual developers, but it falls apart when trying to build a company. "
That seems like a very odd claim to make in support of Redis Labs making this controversial change restricting other people's rights to charge for consulting and support.
Me neither, but telling others that they will have a bad time trying to do the same is either confusing or disingenuous depending on how charitable you want to be. His second comment seems to indicate having regrets in going that way presumably because this strategy turned out to not be very profitable, and that was the reconciliation I was looking for, sort of. I still disagree with the original bullet points.
> they will have a bad time trying to do the same is either confusing or disingenuous depending on how charitable you want to be.
How? For me the interpretation is straightforward: if you create OSS in the hopes of making money from it, you might succeed but the odds are overwhelmingly against you.
Ultimately, the only success guaranteed when you open source a piece of code is the fact that you... Open sourced a piece of code. There's now a new library/tool/application out in the world for everyone to see.
That's how I interpreted the other comment, and I wasn't even too charitable.
Even if you launch a startup you have the odds heavily stacked against you making money. That's just the nature of such variable return endeavours and has nothing to do with OSS.
In my sibling comments I explained more about disagreeing with the starving artist vs business enterprise false dichotomy.
Sure, but if you want to make money from your work, it's a bit of a requirement. Gargon, the author of Mastodon, is making good money from his work on Mastodon. I'll bet folks would make money off Redis if they build a LibrePay or Patreon for it too. Now if you're a random small OSS author trying to become an OSS rockstar, then nobody would contribute.
Well, the question was what's wrong with these platforms with regards to OSS.
You can't just provide software that 1000s of people use. You now have to be creating some story around it on Patreon or whatever, have a way to get people to patreon or your website in the first place, which will only take away time from your side project.
Numbers just don't add up. You can perhaps expect $50/month with those number of users if you really, really tried and spent countless hours on propagation. That's like 1-2 hours of work in a real job.
For a cli tool distributed in major linux distros you don't even have a way to get people to your website, because mostly just packagers will go there and some people with exotic preferences who build from source, which is a tiny minority.
I think the biggest chunk of people directly benefiting from OSS are companies, for whom Patreon might be an inefficient intermediary. Personally, I'd prefer my employer funnel a few thousand dollars to the the FSF, Python Software Foundation, etc. than setup a Patreon donation for each project we use. (Maybe some big ones like Redis and nginx and BackboneJS should get direct donations, but now many npm packages do we use? Oh god, so many.)
That said, if you maintain an OSS tool, please do set up an easy way to donate. When it's easy to donate a dollar or two, I do (well...occasionally). I've found very few developers have Patreon's setup. Some will have a PayPal link, which is especially nice for random online tools that I use once or twice. Most don't provide an easy opportunity to contribute financially. Though I guess I should keep more of an eye out for them.
> People underestimate how difficult it is to create a popular open source project. It's very competitive. You're competing not just with other open source projects but also with SaaS solutions from well-funded companies; and you have to compete with a $0 marketing budget.
So? What makes your "popular open source project" worth my time? Too many people are using open source as a way to try to turn it into a business, and have a problem turning it into a business. Being an open source project doesn't mean you deserve special treatment.
I think the problem is the mentality you present in your first question: "People underestimate how difficult it is to create a popular open source project."
I imagine it's very hard to create a popular open source project, when in reality, all the popular open source projects I know of started out by being useful open source projects that just happened to catch on.
I don't think it's responsible to write about "open source" software while ignoring the legacy of "free software" which kicked off the entire movement in the first place. The pioneers of "open source" believed in utterly free software and (at least RMS) believe furthermore that cloud solutions (wherein someone else does your computing for you) are anti-free.
We do have a license that deals with anti-free cloud solutions but people feel "it's too restrictive" but it's actually only restrictive to developers who'd want to infringe users' rights. Not understanding the point behind GPL causes developers to go with MIT and BSD and then complain about other people using their software for profit and giving nothing back... that's a pity.
>Lastly, we have consulting and support, which I think is a dead end
Given that the only large company that is actually successful with open source as a business model long term (Redhat) follows this model, that shallow statement seems pretty tone-deaf.
As other's have pointed out, but it bears repeating, this is not a thing.
There is actually nothing particularly shocking about what Redis is doing here; it's been done by many open source companies in the past and present and nobody cares. Yes, it can be considered a somewhat less tasteful way to make money from an open source platform but not enough to get a big thread on HN.
What they did to stir this debate is confusingly and unnecessarily tie this move to open source licensing and "commons" naming. This is what brought all this attention to something that is otherwise not very controversial.
I hope the Apache Foundation speaks out against this horrible potential for confusion. Could this possibly be turning into a case of trademark dilution (of the "Apache" trademark)?
It almost certainly is, and I sure hope the Apache Foundation starts to enforce their trademark. Otherwise, they might lose it.
They have a trademark on "Apache" and even a lot of software under the "Apache Commons" brand. This is extremely confusing and clearly dilutes their branding.
I have no problem with Redis Labs prioritizing (although I think an AGPL dual license would be more appropriate). I have a serious, serious problem with the intentional confusion they're spreading.
On the whole, I think this will backfire. Redis Labs undercut their credibility. I'm unlikely to do business with a company I don't trust. The main value-add of being the official developer of an open source product is that trust. A lot of that is likely to spill over on Redis itself.
Seriously, there is nothing controversial about layering proprietary offerings on top of your own Open Source software. Tons of companies do this. They've done it for ages.
It's only controversial because it's been phrased as an evolution of the Open Source model. We have articles like this getting thrown around about how the Open Source community needs to grow up and accept Open Core, right next to articles about how the Common license is going to be the death of Open Source.
This is only a debate because Redis doesn't know how to write a decent press release.
But in reality, most people already accept Open Core. We're not getting a ton of articles on Hacker News about how Gitlab is evil. Just don't act like it's something new or innovative. It's not some brand new problem for us to solve, or a sign of the Open Source community's immaturity.
We really don't need a new license for Open Core, and even if we did need some kind of shared source available/proprietary license, it probably wouldn't look like the Commons license.
> It's only controversial because it's been phrased as an evolution of the Open Source model.
It would have been controversial anyway because they've taken existing software and put a more restrictive license on it; this will always generate headlines. The error was trying to avoid the inevitable backlash by spinning it like it was an original improvement on the first principles of licensing. Now they have a double backlash: against the new restriction, and against the concept of this being the New Frontier of OSS licensing. By trying to be clever, they've shot themselves in the foot.
I updated the language in the post for clarity. However, I wasn't confused by the Redis post and I didn't for a second think this was an official Apache Foundation thing. I seriously doubt that others thought that either. I suspect that even if they had called it RedisLabs' Open License, they still would have been met with the same level of vitriol.
The confusion isn't whether it's an official Apache thing. The confusion comes from trying to a tie an non-open source license to an actual open source license and confuse matters further with this "common" naming scheme. It's purposefully obfuscating and maybe even teeters into purposely dishonest.
I stand by my original disagreement. Removing the open source license from some of their own software would have been met with some discussion but not anywhere near "the same level vitriol" that you seem to believe. The proprietary module business model is used by plenty of open source companies including Nginx. So why would you think Redis would be exceptional here?
I still don't see how they were obfuscating anything. They stated very clearly that those parts of code had restrictions and the purpose of those restrictions were to enable them to profit from providing them as part of their offering. There was no subterfuge here. You're taking the extremely uncharitable view on the language because you don't like the message.
You have it completely backwards; I have absolutely no problem with them taking the software that they've written and making it proprietary. No one should be forced to open source anything and we should all be thankful those modules were open sourced before now. If they own all the copyright (which they must to change the license) they are free (and welcome) to do what they want.
But this whole Commons Clause licensing thing is absolutely bizarre. The naming itself is obfuscating the fact that this actually removes this code from the commons which is the exact opposite of the similarly named Creative Commons licenses. Secondly, it's highly questionable whether or not you can simply attach an external rider like this to existing open source license; Apache License + something else is no longer the Apache license.
I don't understand why any company would go through this awful mess except obfuscation. I don't think they have any intention to defraud; I think they just wanted to make their proprietary plan sound better. And I think that is a mess but their actual business model is both sound and reasonable.
You wrote an article about how open source must "get real" and accept these business models and I'm sorry that the entire premise is undermined by the hundreds of other open source projects that have a variety of business models either closely or exactly matching what Redis is trying to do here. The only thing Redis did poorly was try and hide it under flowery language and poor legal advice.
re: "Developers free time is where it often starts..."
Words matter. We need to stop calling it free time. It's not free. Nothing is free, and certainly not a limited resource such a time.
Personally I'd like to see one of the major OSS code hosts (e.g., GitHub, GitLab, etc.) incorporate payments (read: donations) into their system. If I could load say $25.00 in credits to my account I could make a donation with a click. The receiver - or anyone for that matter - could keep the "cash" or put a bounty on a bug or new feature.
Patreon currently exists, and a lot of game emulators make use of it. There are a variety of one-off tip mechanisms too (kofi is a popular one). And feature bounties also exist (elementary, for example, uses Bountysource for a lot of their development).
I think all of these things work for certain projects. They work less for mission-critical but un-flashy components (i.e. the "how do we adequately fund OpenSSL?" problem, which was solved by having big money corporates endow funding for it). They work more for user-facing stuff. So they're a good option.
But I don't think they're materially superior to the other models for sustainable OSS (maintainers working on OSS projects as 20% type tasks at their day jobs; companies selling service agreements and hosted versions of OSS components; very large companies simply being public good providers). They're just another part of the puzzle. I do agree it might be slightly more convenient to offer them as part of the public-facing code repo stack.
In this case, it's not clear to me Patreon style development funding would have worked near as well as Redis Labs being a corporate sponsor.
One emulator earning 20k (!) a month is an outlier - it's also not open source.
There's still the marketing issue. If I was to work on open source full time I'd need to get about $2kAUD a month. That kind of income doesn't just happen - you've got to have an extremely popular project first.
Also remember that Patreon's gotta start making money for the VCs soon so there will be user hostile changes eventually.
The real game-changer would be to link downloads/checkouts to user accounts. Like, you can download this repo anonymously once or twice a day; more than that, and you have to pay -- on a sliding scale, so that companies with build systems and so on are forced to pay more.
If Github and Gitlab (and maybe Bitbucket) had the balls to build something like that, I think things would improve. Users would be impacted, so it would be essential to get some help on the PR side from established projects and bodies like FSF, Python Foundation, and so on. Getting the big companies onboard should be easy enough - just give them a flat-fee yearly subscription option that they can spin in a nice Bloomberg release on how they support innovation and blablabla.
You would likely still see cheapskates setting up their internal mirrors and so on, but with the right price structure, I bet most medium-to-big shops would accept it and start generating substantial amounts for OSS projects.
So a paywall similar to newspaper sites? And make it per project per organization? You work at ACME Inc. and you already used your 2 free downloads of Redis for today?
One problem I see is that most open source doesn't get consumed as source code but via language package servers (Rubygems/NPM) or OS distributions (Debian packaging Redis).
Language-specific packagers that source directly from VCS services, could easily accommodate the need to specify credentials for source repositories -- in the same way you need cloud-provider credentials to run cloud-provider command-line utilities. So that's not a problem.
For other packagers and OSes: packages don't happen out of thin air: someone has to write recipes and scripts, and test them. So that's already some revenue right there, especially from the big players. Companies like RedHat and Ubuntu could just cut a flat $10m check every year, distributed to projects proportionally in accordance with OS-provided stats like popularity-contest. That's the easy option. Ideally, you would also have some buy-in from the major OS distributions that could somehow "trickle down" the model to their own packages (say, linking a VCS repo to an OS package and reporting activity accordingly: "You asked to apt-get install Redis, but I've checked with a webservice and you have already downloaded it 2 times from Github.com today: time to pony up!". In most cases, this sort of link is already documented formally, somewhere in the package definition.)
Like newspaper paywalls, it doesn't need to be perfect; it just has to be enough hassle that most people who can afford to make the pain go away, will just do that.
I think also, it would help diminish forking. Mind you, forking is important to OOS. I get that. But if, especially early on in a project's birth, (e.g.) two groups focused on one product (e.g., library) instead of each trying to do it "alone" I think on average we'd have better OSS instead of more of it.
Maybe it's also time to realise for OSS developers that they cannot win with the cloud in the long term, so maybe it would be better to work with it.
The reality in a few years time will be a multi-cloud environment where it's the cloud itself that will be commoditized. So let's make great open source projects work well with that:
1. Provide X options for deployment on many clouds/orchestrators - k8s, terraform, serverless, etc.
2. Work on top of clouds and easily integrate with them, e.g. use existing permission systems of AWS, GCloud, Azure, service discovery systems, secret management systems, parameter systems, log aggregators and many others. So in this approach, pricing would still be controlled by the OSS company, by paying for machine time or per-request.
> which will prohibit them from being used freely by hosting and cloud providers and other software makers that profit directly from them
All the pro-"Common Clause" reporting I've seen has focused exclusively on cloud providers, omitting the prohibition on even "consulting/ support services related to the Software". This reaches further than most license agreements for fully-proprietary software, to the point where I'm stunned that it's even legal.
Thought experiment: is a paid employee of a company working with one of these Redis modules considered to be providing support "for a fee or other consideration"---and thus in violation of its license? Is it illegal now to teach a class on how to use RediSearch?
This is my problem as well. Freelancers and consultants would be unable to teach, install, configure, or manage anything licensed under Common Clause. It's a terrible license.
Maybe, but that's still a very harsh restriction - in that regard pretty much every commercial product of random anti-open-source company is more free, their licences almost never require something remotely comparable to that.
If you haven't seen it yet, I highly recommend Henry Zhu's (Babel core team) talk from React Rally[1]. He talks candidly about the struggles of maintaining an extremely popular library and the demands and stress it puts on his life; especially now that he quit his job to work on Babel full time.
If your company uses Babel (which it probably does), please donate to Henry[2] or at the least, send him a thank you for his work.
Henry is a wonderful person, and this talk was excellent. I second the recommendation.
I posted my own thoughts about over-committing and potential burnout a few months back: https://blog.isquaredsoftware.com/2018/06/redux-writing-resp... . Since then, I've stepped back from trying to keep up with recurring treadmill tasks like cataloging articles and links, and am focusing my maintainer time on some specific Redux/React-Redux issues. Feeling a lot better as a result.
If your opensource project can't live without "life support," it's a failed project. It means nobody wants to keep it alive.
No such prerequisite is for 9 out of 10 software packages you see in any serious software repo.
I now run a graphical Linux desktop with fully functioning graphical UI, all done without a dime of sponsorship money, or corporate life support of any sort.
No "well funded corporation" in the world can pull out making an OS with graphical UI from scratch these days. That's simply beyond even heavyweights like google and apple with their countless battalions of coders.
Now say, who has more power there?
Why it is such a problem for guys making software for web ecosystem if open source desktop community can pull out doing things many many many times more hard and complex?
I myself draw a mental line in between a genuine open source projects, and those which are masquerading as such. In reality, those are failing commercial enterprises trying to earn "coolness cred"
This article seems to assume that the main problem with this "Commons" rider is a complaint about the "open core" business model. I think that assumption is false. Lots of organizations have "open core" business models, and do not get the rise that this announcement has received.
The problem is that this "Commons" rider is confusing and misleading. The term "Commons" makes it sound like this is a Creative Commons license - which it is not. The term "commons" is often used to refer to open source software licenses - which this is not. The coupling with Apache makes it sound like this licensing system is from or endorsed by the Apache Foundation - which it is not. (The article uses the term "Apache Common Clause license" - the fact that author thinks that's a reasonable name, even though Apache didn't create or endorse this license, shows how confusing the situation is.) The Commons FAQ admits that the rider isn't OSS, but then suggests that the OSI definition is some minor technicality - yet the OSI definition is what practically everyone uses, and the "technicality" is at the heart of the point (that everyone can use it for any purpose).
Yes, making money is hard. That problem is not limited to OSS either - most new restaurants fail within the first 2 years. I love to see new/better ways to fund OSS. Having a proprietary rider for some modules or having an open core model aren't new ideas, but if they want to do those things (and have the legal right to do), then let's see what happens!
But if you're going to create a rider, it's important to actively work to not be confusing. It's hard getting consensus on anything; I think the anger here is choosing names that appear designed to be deceptive. If these names and such aren't changed, I wouldn't be surprised if the Creative Commons Corporation or Apache Foundation start considering a trademark suit to prevent confusion in the marketplace. If they weren't intended to be deceptive - and I hope they weren't - then some renaming and clarification would go a long way.
This is exactly my perspective. An announcement post that Redis modules would now be distributed under a proprietary license with shared source would have caught nobodies attention. The problem for me is with trying to paint the Commons Clause as a legitimate foss license, thereby arguably attacking the very principles of foss.
I agree "Commons" might be unfortunate, but I understood it to stand for "license for common people" eg. those neither paying nor otherwise contributing. Maybe I'm wrong, English isn't my native language.
"Commons" means shared public resource that is free for use.
A phrase associated with it is "tragedy of the commons" which means everyone takes as much as they can get to enhance their own lives and doesn't give back. This is one of the things impacting Open Source and one if the things people are complaining about when they say companies use Open Source, make money off of it, but don't want to support it.
Historically, the commons was typically a field where animals grazed. Overgrazing by some could ruin it for all.
This is the best write up I have seen in a long time on this subject. Open source doesn’t exist in a vacuum without commercial influence. The best you can do is pick the influence.
FAANG sponsor open source either for strategic reasons (the Kubernetes example in the blog) or as a goodwill exercise. To the kinds of engineers Facebook would love to hire to help serve ads more devilishly, React goes a long way to making the company look developer friendly. These projects get a ton of marketing support (time to promote the project, brand name, etc) that an all-volunteer nights and weekends grassroots open source project can never complete with.
Then there’s open core, which seems to struggle from the difficulty that open source has a kind of altruistic halo around it, and any attempt to commercialise or restrict parts of it (especially for things that were once open) is seen as bait and switch. Yet to see an Open Core company that isn’t continuiously torn between what it takes to keep the business viable and what it takes to keep the community on their side.
Consulting and support has its own downsides too. It only scales linearly as you can hire people, and the margins are much lower so it takes a lot more people - much less fun of a company to run. And it influences the project roadmap too: “yes the setup process is broken and could do with a wizard, but the complex install process drives a lot of our consulting leads”.
It seems like the huge companies that were threatened by open source have been replaced by huge companies that know how to weaponise open source. Open source still wears a certain halo thanks the nights and weekends code hacker that does it for the love, but open source at any scale is as corrupt - maybe more corrupt - than any proprietary company.
(Disclaimer: my company sells proprietary software but gives customers source access/right to modify and open sources many of the building blocks. As someone who wrote code for the love of it, I often feel like there’s something evil in not making everything open source. But I’m yet to see a less evil open source model)
The most successful open source company of all time is RedHat. They built their core product 100% open source. I think there is a model out there where you can be open source and still provide a valuable service for customers.
Why do you think they're the only one, don't you think maybe they enjoyed a very special set of circumstances that will, more than likely, never happen again?
I think the only special set of circumstances is that they were able to go public and remain independent. At this point most companies get acquired as an exit, but thats not just Open Source companies, thats almost all companies.
The fact that they were the first to support an infant OS that will end up powering the majority of the server out there is a stronger reason imo, but you're right, it's definitely a reason. Another one is the internet bubble of course.
IIRC they weren't the first (Caldera...?), but they were among the few that managed to survive the dot-bust. One reason they survived was that they were (then) charging less aggressively than others, at a time when money dried up everywhere. When the bloodbath was over, only them and SuSE were left standing; and SuSE was eventually swallowed and mismanaged by Novell.
I always found the commercial open source community to be a bit too idealistic. On one end you have developers from very wealthy companies driving these hyped up projects.
Sometimes those same developers will leave that wealthy company to start the commercial vehicle for that project. They talk about the wonders of open source and free software.
Obviously their next move is to monetize the project with commercial add-ons and they end up charging 6 to 8 figure price tags for 'free' software. Their enterprise customers justify this by saying 'we have the source code for the core product...its open source so we won't get locked in'. But the likely hood that some overgrown Fortune 2000 IT team will use the forked source code and not the shipped enterprise product is basically zero.
At that point, the FOSS nature is just a farce. It might as well be proprietary software at that point. There's no way the same company will prioritize the community over the high paying customers. They take over the development of the core open source version. They end up hiring all the main contributors. They find new ways to monetize it. They build add-ons that are end up being essential to using the same project in production. Its no longer a community-driven engineering meritocracy, but simply an enterprise development team with a github page.
> Despite that confusion, I don’t think the community’s outrage is driven largely by that. RedisLabs’ post wasn’t ambiguous about the licensing of Redis itself.
Yes it was. It has since been edited to be clearer.
> Within InfluxDB, our dividing line is that anything related to high availability or scale out clustering is kept as a closed source commercial product
No, we've discussed this time and time again:
Open Core = crippleware
Articles like this occasionally come out and promote propaganda that masks "crippleware" with nice sounding words.
I maintain a tremendously popular Open Source project with 8.5K+ stars, and it is extraordinarily hard. (Hint: Imagine Firebase + IPFS had a lovechild, that is what it is. See my HN profile if you are curious.)
I should be the first person agreeing with this - last year was one of the hardest times of my life because of doing Open Source. But this is the moral, harsh, but important:
Financial hardship sucks but does not define the values of Open Source.
The values of Open Source has removed financial hardship for many, and made lives better for the masses.
Just like you should not gamble on crypto what you cannot afford to lose... you should not Open Source what you cannot risk.
Open Core = crippleware: Faux companies publicly wearing sheepskin to look like they're developer friendly, but privately are forced to shame you into the fox's den.
Open Source = giving, truly. It has hurt me at times, but those values have brought more win+win to my life and others, far beyond we can count.
So thank you, to all real Open Source developers out there. You are heroes.
I don't agree with the tweet at all. I too thought MIT/BSD are good licenses, until I got burned by a piece of code majorly consisting of MIT code infringing my user rights - unwarranted tracking and a few bugs I couldn't fix. I now understand why (A)GPL exists and make sure my projects infringe on no users' rights like projects containing MIT/BSD can.
>However, there are frequently hidden motives for these projects. For example, Google puts significant investment into Kubernetes, but it’s also in their financial best interest to do so.
This, in a nutshell, is the driving force behind the EU's antitrust investigation of Android.
254 comments
[ 3.0 ms ] story [ 242 ms ] threadThey're complaining about the weird/impossible 'clause on top of open source' and 'existing OSD' (as if there's another new one) language in Redis' announcement and the 'Commons Clause' website
Yep, I think that would make it better too.
Redis: BSD (or whatever OSS licence)
Redis X module: Redis non commercial License
Redis Y module: Redis non commercial License
To be fair, if they'd just said hey we're selling these proprietary licensed modules then there wouldn't be much of debate. As others have pointed out, many other OSS projects do just that.
Instead mass confusion has erupted and literally every word debated ad nauseam -- including, now, the model.
Besides that the obvious point is that every open source producer is entirely free to release their product under whatever license they see fit and if you don't like it then simply don't use the product.
RedisLabs made some extra software available (nice) under a noncommercial-only license (well duh but ok), pretending some material relation between the noncommercial-use-only license and the Apache License (not nice).
The pitchforks and torches are (to 98%) about that last bit. Everyone is happy that Redis still exists and that Antirez and RedisLabs have a sustainable business around Redis. No one - and potentially especially not the Apache foundation - enjoys the bait-and-switch of using the Apache name for a proprietary shared source license.
I’d like to make all of the software I write freely BSD licensed.
If I do that, it seems with very high probability that it might turn out to be a time sink and a money sink with no benefit other than +reputation. Which is worthwhile, but perhaps not as worthwhile as those fat stacks.
I don’t know. We’re heading for a future where the only people that can afford to write OSS projects that actually influence the world are those who work for FAANG or in academia.
Maybe something like this would be more fitting:
The elephant in the room here is that the usual OSS-as-an-offering model works from the fact that intimate knowledge of the software and visibility creates a strong market position for selling services (SaaS or consulting) that others cannot match. Whereas if "others" includes Amazon, GCP etc., their economies of scale and market visibility tilt the playing field in the other direction, making it harder for the OSS-as-offering model.
What it would do is mean that a lot less people would be able to use the software as it's a license that is not allowed at most companies. You can argue that this is for the wrong reasons, or it's due to fearmongering, but that doesn't change the fact.
I think you have it backwards: a less popular project could be viable with BSD and wouldn't survive as AGPL.
That is not universal across all the developer community. Here, in Colombia, open source is something you do on the sidelines.
Pretend you can live of it? That is non-sense.
I know companies following this model all over the world: Spain, Brazil, India, Argentina, Venezuela, etc.
Why do you think that?
Why? If you want to make money off of it, there is nothing wrong with selling proprietary software. I'll gladly open source lots of stuff but I'm not just going to work for other people for nothing -- that's just madness. And that isn't really what free software / open source is supposed to be. It's not a platform for martyrdom.
I think the strength of an OSS project (or really a single individual at that stage) can be to innovate - make something that in a corporate setting, FAANG or not, can't be pursued because it hasn't a chance to become projected since business value isn't tangible, and anyway needs meandering around to find solutions in its problem space. Once something demoable is created, it might attract corporate backing. But corporate development is actually very bad at developing the kernel of something not seen yet. Maybe there should be a market place for initial designs/concepts to grab beyond VC funding?
The Patreon approach is not sustainable in the long-term, neither is the company sponsorship route.
Charging for support as the principal source of your income is probably one of the least appealing options for most developers.
Open-core is one option - GitLab wouldn't be a financial success without it. Dual-license is another option.
What does the FSF (Free Software Foundation) have to say on the matter? Charge for distribution [1]. Yet, no-one can realistically charge for distribution today - not least when so many projects are hosted for free on GitHub.
If you want to make source code available and still be paid for your product - consider something like the Fair Source License. It's neither closed source or open source. It will appear blasphemous to some open source advocates, but it feels like a reasonable compromise between closed source and permissive open source licences: https://fair.io/
[1] https://www.gnu.org/philosophy/selling.en.html
No. This is reprehensible garbage. This license is not endorsed by the Apache Software Foundation.
EDIT: removed "proprietary"
This software is licensed under a mixed OSS/commercial license, the Apache 2 OSS license with the commercial Common Clause restricting certain freedoms.
Using the text of the Apache License 2.0 as a point of departure is fine. But please make it clear what you are doing.
By all means keep the common core concept for Redis but I urge you to get as far away from this toxic “Common Clause” pseudo-license as possible.
The difference between this and the restrictions in other OSS licences is that the "common clause" restricts core freedoms - if you attach it to something, that something ceases to be an open source licence, this clause transforms into a commercial licence that prohibits free use of the software. There's nothing mixed about it, it's a "not-OSS" licence.
It would be far more productive to state your particular disagreement(s) than to inaccurately apply labels you think will tarnish the target of your ire.
It's a set of anti-commercial license terms and is definitely proprietary by Free Software and Open Source definitions.
Only if you're equally honest in admitting it's the most risky approach wrt sustainability and therefore the one most subject to strange capitalization mechanisms (and endless blog post justifications for them :-)). They go out of business all the time, leaving stale software in their wake that might have otherwise had steady maintenance if independence and growth were less paramount (like before commercialization where, in many cases, the burden was just as large). Either way, it's definitely the hardest model and if you have a piece of open source software that must have growth or you are currently no longer able to sustain it, I would try to find any other way possible to finance it...but I agree sometimes it's just not there so you build a business around it.
And I'd change "developing" to "financing" since there are many more honest ways, just not at scale.
This also gets to the biggest criticism I have of the FSF: the only measure of success is a narrow definition of "freedom" -- with little to no discussion of what social systems, economic structures, and advocacy might be useful and/or necessary to support those aims.
Also, if the project does become popular eventually (which is very unlikely), maintenance is a lot of work. To make matters worse, there is a general tendency that if things go wrong, the community will blame you but if things go well, you get almost no credit.
Even if your project has several thousands of stars on GitHub, it means nothing financially. Most of these OSS authors can't afford to quit their day job even though their project may be generating millions of dollars of value for other people.
Money is an important topic to open source authors because after a decade of working for free during nights and weekends, most of them would love to be able to afford to quit their day jobs to focus exclusively on their projects but they can't do that; they're tired but they have to keep soldiering on with the 80+ hour work week.
Being an open source author is not a wise life decision right now. I think that any business innovation which helps keep open source projects going should be welcome.
Think I spotted the problem.
This whole thing feels like someone martyring themselves for a cause and being recognized as a hero, but then complaining that martyrs are heroes and we should be _supporting_ our heroes and not forcing them to be martyrs. But that was the whole point to begin with and why there was any opportunity for recognition at all.
You can't have your cake and eat it too.
It's not that simple, what would be this industry without Linux, without the FSF, without the BSD network stack (that is in all the macs...)
The growth of our industry, and probably of a good chunk of the whole economy for the last 10 years has a lot to do with those people working during nights and week-ends imo.
EDIT: people or corporations, same thing.
1) the top 1% employees in the field who have the luxury to invest unpaid time, and the social clout to make their projects more successful, or to join more successful projects. (example: how much more likely are you to be a key contributor to Kubernetes if you are, or were at some point, employed by Google?);
2) the other 99% who cannot afford to invest the spare time, and do not benefit from the social proximity to successful open-source contributors (and success in open-source is as much about social proximity today than technical skills, hopefully we can change that in the future);
3) The corporations who benefit financially from open-source, sometimes enormously, with very little pressure to contribute back because of the aforementioned social pressure on individuals to contribute on their free time. The expectations are so low that even modest corporate contributions (joining or starting a foundation, hiring a few engineers, paying for a conference) will be hailed as generous and benevolent, because the bar is so incredibly low.
Is it really? Of the people I know who have had excellent career paths in software/computing none have a track record of open source contributions.
Right, that’s how they built their career in the past. All things being equal, it will become gradually harder to build a career that way in the future. I already know several major employers who look at a candidate’s github activity as a factor in the standard hiring process. And that trend is only accelerating.
If these people spend a lot of their free time, and perhaps even some of their professional time, in Github contributing to various public projects, then they might not necessarily have 100% focus on the employer's tasks. I am not sure if this is what the employers expect or even want to happen.
The intention is probably to use Github activity as a pre-screening proxy metric for technical capability and skillset. But... why would someone who is contributing a lot to public projects cease to do so, once employed? Is there a risk of getting a partially distracted, though skilled person?
If the wish from the company then is to ensure they get the most performance for themselves, the logical step would be to cut down on the public contributions via e.g. legal and contractual means.
So... why would someone who enjoys contributing to public projects join a company that selects based on Github activity, if it is in the company's interests that further contributions should be reduced?
Ahh... Maybe I am just overanalyzing it.
His model was, that programmers are expected to donate their time to OSS creating core components, in order to get jobs integrating those core components into commercial systems.
When you think about the issue like that, it makes a lot of sense, but its a terrible deal for the programmer.
Nobody else does their job on weekends for free.
In a Star Trek utopia, that idealism would be great, but we live in a world where people are exploited wherever possible to ensure endless profit.
Then you remember that there was never any reward and you wonder why you did this to yourself. Lol.
"Success" in the OSS world is putting code out that others can use. Either they find it useful, or they don't. If they don't that's not failure. That's ... someone else put out something better, or that your user community is small or just not out there.
OSS is not a business model. It's a SHARING model.
God it's just like the music scene these days. Did you get into it for the Art? Cool you're gonna be ok. Did you get into it wanting to be a rock star? Might as well buy a lottery ticket. Your chances are better.
OSS success is providing something of use to a community that contributes back to it.
Yeah, maybe you can find a way to make a nickel off that, but that's not what it's FOR. It's not a marketplace, it's a public library.
This celebrity/glamor seeking is so apparent on github. Every single project's readme.md reads like marketing copy these days.
Just like the music scene again. Reminds me of the 90's when the record labels still had money and steamrolled the alternative scene looking for a cash grab.
Why is that nothing something we should reconcile? Why is that not "sharing"?
Somebody yesterday (maybe DannyBee, not sure) made a very insightful comment in saying that sometimes the value created by Open Source software is created exactly because it's Open Source.
Anyway, you can always charge for software even if the code is Open Source. You can stick it on the web and make it a SaaS app, or you can make the source available but charge to download binaries (ala JBoss back in the day), or you can make source and binaries freely available and sell subscriptions, etc. Aaah, wait, I know, you're asking "who would buy a subscription for something that's free to download?" Companies. Companies will, for things that are mission critical, because they typically need for their to be a contractual relationship (eg, somebody to sue) in place. Or you have a manager who needs to spend her budget or lose it, and she understands that you need to make a profit in order to stay alive and keep enhancing the software. Etc., etc, yadda, yadda, ad infinitum. My point is, it's not as simple as "Oh, I made this Open Source, so there's no way to profit from it".
https://news.ycombinator.com/item?id=17818407#17819123
Quote at the end:
The underlying issue they are trying to solve is that developers believe they should be able to extract some of the wealth they feel they created. In open source, that wealth is often only created because the software was free. Otherwise, people would have used something else that was free but worked well enough. Developers like to often argue this isn't true, but history shows it to be true basically always :) So saying you should be able to extract this wealth is probably wrong. Saying you should be able to get paid a reasonable amount of money is not.
That’s the problem: we all take this particular public library for granted, but there isn’t a sustainable model for paying the librarians. Then a librarian tries a creative way to continue serving us without starving, and we tell them “you’re just doing it for fun, if you wanted to make money you shouldn’t have been a librarian!”. This is an ignorant response, because it places on the librarian’s shoulders the responsibility for solving a problem that concerns all of us - because we all benefit from open-source more than we contribute to it. That’s not sustainable.
Because, in your analogy, the librarians are all volunteers. That’s how we got here. If the pain of not having a problem solved is great enough the volunteers will fix it. Or they won’t. If you feel like the pain level is high enough. Hey maybe YOU should try volunteering to solve it.
This is what makes OSS powerful. It is a community service. If you aren’t happy with the service p become a member of the community and contribute.
Since we as a society rely on open-source enormously, it seems to make sense to try and allow as many people as possible to contribute as much as possible. And that requires giving up on antiquated notions that “only unpaid open-source is real open-source!”. Wouldn’t you agree?
We already do, anyone can do it and we shouldn't restrict that; that's the point.
> And that requires giving up on antiquated notions that “only unpaid open-source is real open-source!”. Wouldn’t you agree?
Not GP, but I most certainly don't, nor is it antiquated. For that matter, calling volunteer work "antiquated" is... A bit cynical, to say the least.
I also don't see how this follows from what you said before, could you walk me through your reasoning?
I emphatically did not say that volunteer work is antiquated! I myself do plenty of it - but I also acknowledge that I’m privileged to be able to afford it. What is antiquated is the notion that only unpaid work is authentic open-source work. It’s important to realize that providing unpaid labor is a luxury. If you exclude or devalue paid work, you exclude and devalue the people who cannot afford to do open-source work for free. As it turnd out, that’s the majority of people.
https://www.linuxfoundation.org/blog/2017/10/2017-linux-kern...
No, it's not a community service, not really. It's a happy byproduct of some very powerful companies.
Instead of paying LF to hire devs, they hire devs who write code that is contributed to Linux.
OSS is diverse and there are many, obviously, who don’t mind not being paid directly.
But what’s confusing is that there are proprietary licenses. Anyone who doesn’t like volunteering time to OSS can write commercial software and charge for it. That’s the model that works for them.
But the talk of a 40 year old model not being sustainable is pretty funny. It’s more sustainable because it takes so much volunteer time.
It’s not the open-source model I called unsustainable, but a particular response to new forms of open-source monetization.
At any rate, the current model - of open-source as a mainstream R&D model, and as the critical infrastructure for the largest businesses in the world, primarily funded by corporate sponsorship and venture capital... that model is definitely not 40 years old.
You could argue that the model originated at the Linux hype of 1999, when Red Hat was the hottest IPO and IBM was spray-painting penguins on the sidewalks of San Francisco... Or you could argue it really started in 2004 when the Google IPO showed how much more scalable and profitable a business can be when you don’t pay software licenses.
In any case, the current model for open-source is really not that old, and it’s too early to tell how sustainable it really is.
When Stallman thought seriously about open source in the 1980s, it wasn't because he was the only one to ever imagine a culture of sharing around software and information. I think he was reacting to a sense that an existing culture was being erased by growing commercial interests.
One could also argue it originated with gentleman-scholars circulating letters about their new physical science discoveries during the renaissance and feeding into the subsequent industrial revolution. But, I expect it is an emergent property of humanity which is probably exhibited throughout history and pre-history.
It's essentially the same idea of commons and culture applied to yet another domain of knowledge and technology. Was it sustainable for the first farmer to teach his techniques to another? For one chef to teach another how to prepare his dishes? Open source software can be sustainable as long as there are needs and resources available to produce and execute software that does something different or better than could be done without the software (or at a lower cost than other available options). The subsequent sale of additional copies need not be a motivating factor at all.
That was not motivated at all by the open source movement. The only reason AT&T Bell Labs didn't charge a significant amount for UNIX is because they were under a consent decree which precluded them from entering other industries at the time, see https://arstechnica.com/tech-policy/2011/07/should-we-thank-...
I don't think the fact that RedisLabs doesn't have a sustainable revenue model indicates that open source infrastructure in general doesn't.
It's true that there are a number of relatively new venture-funded OSS infra firms that were built on a step 1, build software, step 2 ..., step 3 profit model. But I don't think that's an OSS problem so much as a startup culture problem, as the same problem is rampant in startups in other spaces.
> Then a librarian tries a creative way to continue serving us without starving
It's not particularly creative or novel; the basic outline has been tried since the 80s at least.
I don't understand this. It seems the argument is "I can't keep giving away software for free unless you pay for it". Then it isn't free software, right? Nothing wrong with that, most of us write proprietary software and sell it. If it's worth more to the people who buy it than what it cost, everyone still benefits.
We use open source software to do our work more effectively. There are all sorts of OSS projects, many entirely run by unpaid volunteers. Many people, myself included, sometimes contribute code to the OSS projects, to benefit myself and anyone else who happens to use it. I don't expect anything else.
But you can't feel entitled for people to pay you for work that we all agreed was volunteer. It's not worth it to do OSS anymore? It's ok to stop doing it. There are enough of us that are rewarded by just being able to contribute, we'll have our own projects. So, it's ok for Redislabs to make their code proprietary and make money off it, if they can. There's nothing wrong with that. But the rhetoric that boils down to "people have been stealing from us by using our code and not paying for it", "it's not fair that they are using our code and making money and not giving us any" is wrong. You gave the code away for free. That was the understanding before you wrote the code, and after it was written. That was one of the reasons people were even using the code. It's not an injustice.
When someone gives you a nice gift, they deserve your gratitude. But if they turn around and say "you never paid me for that", then it wasn't really a gift was it?
The goal for RedisLabs is to make more money from the enterprise add-ons they’re developing. A lot of that money (perhaps most of it) goes right back into maintaining Redis, which remains open-source.
So it absolutely is a matter of funding more open-source gifts, and not taking back gifts.
Just a "Hey what we are doing now isn't working, for our business model we need to make more money and we need to make some of our code around Redis proprietary so we can sell it. Don't worry, Redis itself will remain open source, free to use for everyone."
Instead it was "today’s cloud providers have repeatedly violated this ethos by taking advantage of successful open source projects and repackaging them into competitive, proprietary service offerings". It was open source, they could do that. If you want to make them pay in the future, that's fine. But they aren't the bad guys for using OSS.
Anyway, I hope it works out well for RedisLabs, but I also hope that the licensing of OSS software is simple and open.
This way of thinking is, in my opinion, destructively pessimistic. Everything is not always a zero sum gain. We can all benefit more than we contribute and it can still be sustainable. For example, 10 people can spend 10% of their time on a project, and get something out of it that is worth 50% of their time. They took the equivalent of 1 persons work and made it worth 5 people's work. I think most popular OSS projects are orders of magnitude more effective than that at creating benefit. Thousands of people contributing code, a few times a year, makes a project that supports all their careers and provides software that benefits society as a whole.
I get that the problem is the undue burden on the relatively few people who organize this work for really big projects. I've noticed that some of these people are employed by large corporations who encourage their OSS work, because the result, and the thousands of contributions from others they harness are very valuable to these organizations.
I wouldn't be surprised if the large corporations that were listed as the ones using Redis software without paying for it would be willing to sponsor a few employees to represent their interests in the OSS projects. That works for the individual engineers, but that doesn't exactly work for RedisLabs to make $$. So I still don't see RediLabs as a beleaguered crusader trying to save OSS.
In the web space maybe. The desktop and CLI sector are much more relaxed.
The problem for all of these categories is that lots of people want to do the job. There is basically an unlimited supply of people that like to write novels, make programming languages, play soccer, or make video games. That makes it a buyers market. "What? This iPhone game is $0.99? No way. I'll play one of the million free iPhone games instead."
For every singer, basketball player, or instagram personality that hits the big time, there are thousands that never made a dollar doing it.
For better or worse, authoring an open source library is one of those things that there is no shortage of people willing to do, so I am not optimistic about use of "business innovations" to solve this problem. If those innovations are in any way non-consumer friendly, consumers are just going to choose a different open source tool/project.
But that difference does not change the market dynamics in any way. In both cases, the effects of a near unlimited supply of creators is having the same effect, regardless of how their creations get used.
Perhaps the difference changes the ethics of the situation, I don't know, but I remain equally skeptical of "business innovations" being able to help the OSS creators as I am "business innovations" being able to help aspiring rock stars and shortstops.
That's almost exactly the opposite of very popular OSS software, which generates an endless wave of free users. And generally no whales. And high demands for bug fixes. And serious temper tantrums (this has happened to me too) when you explain to users of your free software that you owe them exactly fuck. and. all. and won't be hopping-to in order to fix something for them.
Sure there are some runaway indie hits that make money to cover the development. And Linus Torvalds gets paid $10 million per year to work on open source. But nearly all indie game developers never make anything from their games, same as OSS authors. (Indie games are an the appropriate comparison since we are talking about independent OSS authors, not Facebook developers working on React).
In any case, I am not arguing that all of these endeavors are equivalent in all ways. My only point is that all of the fields have an oversupply of creators. The result is that creators don't have leverage over consumers. To really spell it out: if OSS library creators try sticking the commons clause or other "innovations" on their projects, most businesses are just going to go find another library that does the same thing, without the restriction.
I have an OSS project used by 1000s of people and many huge companies. I don't get compensated for that directly and I don't expect to. I'm just happy to be able to use it on my own projects no matter what company I am working with.
These are people who are providing a service, potentially because it’s a passion, but other entities soak up the profit.
You seem to be arguing that these people (in the grandparent comment and my own) do get paid, but my understanding is that they don’t, or barely do.
If you think your code is worth money, sell it.
LGPL/GPL/AGPL exist for a reason. Not my fault you used BSD/MIT and didn't consider what you were explicitly allowing and encouraging people to do.
There btw.
Also, it would be common decency but we live in an era where if something cannot be enforced in court then you should not be surprised somebody is [not] doing it. Oh well.
(And I know I would have a "donation budget" if I was a businessman because I am not stupid and I know the work of which people outside my company facilitates my business further.)
But appreciation is not what businessmen do. 99% are cut-throats in all areas of their lives.
When the concept of open source software was first introduced, it was seen as a desperate way to get users but it was an effective strategy at the time.
Today, giving software for free is not enough - I think that the next trend will be to pay people to use your OSS software. Probably with cryptocurrency.
This was not the motive for free software's beginning, nor does it describe the beginning of the term "open source." The concept rose out of early computing allowing and sometimes encouraging users to share and improve software, and the true start of the movement envisioned a model similar to Red Hat, where the software is sold though sharing is not restricted.
Forcing user payment is effectively impossible for open source software, someone can just buy a copy and redistribute it without the required payment, so your prediction is a movement away from open source software.
Not particularly. It's depressing to see businesses profitably exploit that work, and irritating to see people develop a sense of entitlement without ever contributing back.
Though I wish more general users donated to projects, people using the software without paying seems better than people not using or paying for the software, and far better than paying and being unable to use the software.
In the end it's about what choice you make. If you choose to help the community then you must be prepared that it will be a thankless and hard job. If you cannot bear it, don't help the community.
Trouble with open source is, many people don't realize this phenomena until they are burned out.
There is an unlimited supply of people who say they would like to write novels, make programming languages, play soccer [professionally], or make video games. The supply of people who have the drive and patience to complete a novel or programming language or video game for free or on spec is arguably much smaller. The same is also true of open source libraries. There may be an effectively unlimited supply of GitHub repositories containing new JavaScript frameworks and template languages, but the majority of them haven't gotten very far past the README file. And, we haven't even addressed the question of whether they have the skill and talent to make a good novel or programming language or video game or open source library.
Having said all that, I think the original article is on point in at least one crucial respect: all of these cited kinds of work require a huge investment of time, and it's very hard to sustain that investment without money. This is why I always get very frustrated with the particular kind of pseudo-libertarian who blithely asserts that artists and writers don't need to be fairly compensated for their work, because clearly they'd keep doing it solely out of the love of creation. In reality, artists and writers--and open source programmers--aren't likely to put in "second job" level hours without compensation indefinitely. They're going to have a family, or develop more relaxing hobbies, or possibly just get frustrated by the ongoing effort.
I don't know if "business innovations" can solve this problem, but I think it's worth looking for more ways for developers (and writers and artists and...) to bring in some kind of stable recurring revenue.
...but practically how do you do it?
The parent comment is quite right, going from 'pay x' to 'pay y' is one thing... but from 'pay nothing' to 'pay anything' is a massive and difficult step.
You can't make it mandatory; theres a glut of stuff out there, and people will just move on to the next free (perhaps crappier, but free!) version.
'business innovations may help' is just a nice way of saying 'well darn, nothing we've tried in the last 10 years to fix this has worked... hm. Don't really have a solution for this, but it'd be great if someone figured one out!'
...which I totally agree with, sure, but its not really helpful or practical.
Seriously, nothing people have tried (tips, patreon, foundations) has worked very well.
No one is going to solve this problem any time soon.
Of you're an open source library author/maintainer, you've got to acknowledge you're not just magically get rewarded for your efforts; you have to work for it, and (probably, for most people) your attempts to monetize it will fail.
The same is true for all creatives.
If I had an answer to that, I'd be a full-time fiction writer. :)
Seriously, I suspect most people aren't going to ever make a lot from creative efforts, which we'll lump open source libraries into; the vast majority of us are going to be much farther out along the low long tail of the compensation curve.
I suspect the only thing that will help -- not to the degree that everyone will be able to make a living wage doing whatever they want, but to the degree that more people will feel like they're getting something back for the time they're putting in beyond the Warm Fuzzy Feeling of Sharing -- is a cultural shift of sorts. We need to uncouple the concept of "zero marginal cost" from the concept of "should be free." (I'm aware that leads down a whole rabbit hole about IP and "piracy," but I'm leaving it aside.) In the open source world, we may all say we're here for the free-as-in-freedom part rather than the free-as-in-beer part...but we're sure disinclined to pay for those beers.
Who gets to determine what is “fair compensation” and to whom?
So based on supply and demand, i.e. prices?
In a pure supply/demand market races to the bottom can and do happen.
If you go into open source expecting to get paid, then you're gonna have a bad time.
If you go into open source for the accolades, then you're gonna have a bad time.
If you go into open source with a true desire to improve the state of the art, then you might be able to succeed.
We (https://sheetjs.com, our most popular library https://github.com/SheetJS/js-xlsx has over 11K GitHub stars) combat that by intentionally removing our names and egos from the equation. We are happy to see others use our open source code in their applications, and are constantly surprised when high-profile projects like React end up using our code (https://github.com/facebook/react/blob/v16.0.0/src/renderers...). Our intentions are clear: the open source offerings give us a test surface far larger than any proprietary solution, and for that we are extremely satisfied with the results.
Many open source developers start with intentions like "advancing the state of the art" and over time become disappointed because others build businesses and profit off of their creations. The intention at the onset wasn't to make money, and continued investment of time and money into OSS should be weighed against the original intention.
We are elated when we find others using open source SheetJS libraries and code in their profitable business applications. Others may regret and wish they didn't open source, but you don't know how things would have played out if you didn't open source. The yardstick we apply to our open source contributions is whether we're advancing the state of the art, since that's why we entered the space in the first place.
https://news.ycombinator.com/item?id=14886481 is a similar comment that probably better reflects the opinion expressed here.
(note: we ended up developing paid offerings after conversations with @patio11 and others. https://www.kalzumeus.com/2015/01/28/design-and-implementati... discusses these points)
My takeaway from your bullet points is more specific – that you shouldn't bother with open source unless you want to advance the state of the art while being perpetually unpaid and unrecognized, which I can not agree with in any way.
> Many open source developers start with intentions like "advancing the state of the art" and over time become disappointed because others build businesses and profit off of their creations.
I just don't buy this distinction of starving artists vs commercial enterprises. Especially as the project grows in scope, complexity, and popularity you might need to spend more time on it to keep it in good shape, and time does not come for free.
While you might have been willing to spend 10 hours a week on it for free, the now-grown project would really benefit from you working on it full time, but that's not feasible to you, so you turn to monetizing it. There is nothing wrong with wanting to both advance the state of the art and to be paid for it.
It's not about being disappointed with others making money off of your project, it's about ensuring sustainable development of your project.
If OSS projects weren't funded by a chance of money or fame, we'd have much less such projects available to us, especially significant ones, because not everyone who currently contributes to OSS can afford to or even wants to work full time for no pay. I don't know why that would be a desirable outcome to anyone.
"advancing the state of the art" is fundamentally different from "money" and "fame" insofar as the latter involve social effects that you have less direct control over. To make money, someone else has to pay you for something.
Why should someone be forced to walk away from a project that became popular if there is an option to monetize it to support its growth?
"Lastly, we have consulting and support, which I think is a dead end. It certainly works on a smaller scale with individual developers, but it falls apart when trying to build a company. "
That seems like a very odd claim to make in support of Redis Labs making this controversial change restricting other people's rights to charge for consulting and support.
How? For me the interpretation is straightforward: if you create OSS in the hopes of making money from it, you might succeed but the odds are overwhelmingly against you.
Ultimately, the only success guaranteed when you open source a piece of code is the fact that you... Open sourced a piece of code. There's now a new library/tool/application out in the world for everyone to see.
That's how I interpreted the other comment, and I wasn't even too charitable.
In my sibling comments I explained more about disagreeing with the starving artist vs business enterprise false dichotomy.
You can't just provide software that 1000s of people use. You now have to be creating some story around it on Patreon or whatever, have a way to get people to patreon or your website in the first place, which will only take away time from your side project.
Numbers just don't add up. You can perhaps expect $50/month with those number of users if you really, really tried and spent countless hours on propagation. That's like 1-2 hours of work in a real job.
For a cli tool distributed in major linux distros you don't even have a way to get people to your website, because mostly just packagers will go there and some people with exotic preferences who build from source, which is a tiny minority.
These websites solve none of this.
Patreon is owned by the VCs and will eventually make changes that negatively impact creators.
That said, if you maintain an OSS tool, please do set up an easy way to donate. When it's easy to donate a dollar or two, I do (well...occasionally). I've found very few developers have Patreon's setup. Some will have a PayPal link, which is especially nice for random online tools that I use once or twice. Most don't provide an easy opportunity to contribute financially. Though I guess I should keep more of an eye out for them.
(While Googling for more info, I did come across https://github.com/feross/thanks -- so that's neat.)
So? What makes your "popular open source project" worth my time? Too many people are using open source as a way to try to turn it into a business, and have a problem turning it into a business. Being an open source project doesn't mean you deserve special treatment.
I think the problem is the mentality you present in your first question: "People underestimate how difficult it is to create a popular open source project."
I imagine it's very hard to create a popular open source project, when in reality, all the popular open source projects I know of started out by being useful open source projects that just happened to catch on.
Given that the only large company that is actually successful with open source as a business model long term (Redhat) follows this model, that shallow statement seems pretty tone-deaf.
As other's have pointed out, but it bears repeating, this is not a thing.
There is actually nothing particularly shocking about what Redis is doing here; it's been done by many open source companies in the past and present and nobody cares. Yes, it can be considered a somewhat less tasteful way to make money from an open source platform but not enough to get a big thread on HN.
What they did to stir this debate is confusingly and unnecessarily tie this move to open source licensing and "commons" naming. This is what brought all this attention to something that is otherwise not very controversial.
They have a trademark on "Apache" and even a lot of software under the "Apache Commons" brand. This is extremely confusing and clearly dilutes their branding.
I have no problem with Redis Labs prioritizing (although I think an AGPL dual license would be more appropriate). I have a serious, serious problem with the intentional confusion they're spreading.
On the whole, I think this will backfire. Redis Labs undercut their credibility. I'm unlikely to do business with a company I don't trust. The main value-add of being the official developer of an open source product is that trust. A lot of that is likely to spill over on Redis itself.
It also does signal a financially weak position.
It's only controversial because it's been phrased as an evolution of the Open Source model. We have articles like this getting thrown around about how the Open Source community needs to grow up and accept Open Core, right next to articles about how the Common license is going to be the death of Open Source.
This is only a debate because Redis doesn't know how to write a decent press release.
But in reality, most people already accept Open Core. We're not getting a ton of articles on Hacker News about how Gitlab is evil. Just don't act like it's something new or innovative. It's not some brand new problem for us to solve, or a sign of the Open Source community's immaturity.
We really don't need a new license for Open Core, and even if we did need some kind of shared source available/proprietary license, it probably wouldn't look like the Commons license.
It ok if you want to say we have Apache open core and then non-free extensions, but you shouldn't associate the non-free extensions with Apache.
It would have been controversial anyway because they've taken existing software and put a more restrictive license on it; this will always generate headlines. The error was trying to avoid the inevitable backlash by spinning it like it was an original improvement on the first principles of licensing. Now they have a double backlash: against the new restriction, and against the concept of this being the New Frontier of OSS licensing. By trying to be clever, they've shot themselves in the foot.
I stand by my original disagreement. Removing the open source license from some of their own software would have been met with some discussion but not anywhere near "the same level vitriol" that you seem to believe. The proprietary module business model is used by plenty of open source companies including Nginx. So why would you think Redis would be exceptional here?
But this whole Commons Clause licensing thing is absolutely bizarre. The naming itself is obfuscating the fact that this actually removes this code from the commons which is the exact opposite of the similarly named Creative Commons licenses. Secondly, it's highly questionable whether or not you can simply attach an external rider like this to existing open source license; Apache License + something else is no longer the Apache license.
I don't understand why any company would go through this awful mess except obfuscation. I don't think they have any intention to defraud; I think they just wanted to make their proprietary plan sound better. And I think that is a mess but their actual business model is both sound and reasonable.
You wrote an article about how open source must "get real" and accept these business models and I'm sorry that the entire premise is undermined by the hundreds of other open source projects that have a variety of business models either closely or exactly matching what Redis is trying to do here. The only thing Redis did poorly was try and hide it under flowery language and poor legal advice.
Words matter. We need to stop calling it free time. It's not free. Nothing is free, and certainly not a limited resource such a time.
Personally I'd like to see one of the major OSS code hosts (e.g., GitHub, GitLab, etc.) incorporate payments (read: donations) into their system. If I could load say $25.00 in credits to my account I could make a donation with a click. The receiver - or anyone for that matter - could keep the "cash" or put a bounty on a bug or new feature.
I think all of these things work for certain projects. They work less for mission-critical but un-flashy components (i.e. the "how do we adequately fund OpenSSL?" problem, which was solved by having big money corporates endow funding for it). They work more for user-facing stuff. So they're a good option.
But I don't think they're materially superior to the other models for sustainable OSS (maintainers working on OSS projects as 20% type tasks at their day jobs; companies selling service agreements and hosted versions of OSS components; very large companies simply being public good providers). They're just another part of the puzzle. I do agree it might be slightly more convenient to offer them as part of the public-facing code repo stack.
In this case, it's not clear to me Patreon style development funding would have worked near as well as Redis Labs being a corporate sponsor.
There's still the marketing issue. If I was to work on open source full time I'd need to get about $2kAUD a month. That kind of income doesn't just happen - you've got to have an extremely popular project first.
Also remember that Patreon's gotta start making money for the VCs soon so there will be user hostile changes eventually.
Bring on universal basic income.
If Github and Gitlab (and maybe Bitbucket) had the balls to build something like that, I think things would improve. Users would be impacted, so it would be essential to get some help on the PR side from established projects and bodies like FSF, Python Foundation, and so on. Getting the big companies onboard should be easy enough - just give them a flat-fee yearly subscription option that they can spin in a nice Bloomberg release on how they support innovation and blablabla.
You would likely still see cheapskates setting up their internal mirrors and so on, but with the right price structure, I bet most medium-to-big shops would accept it and start generating substantial amounts for OSS projects.
One problem I see is that most open source doesn't get consumed as source code but via language package servers (Rubygems/NPM) or OS distributions (Debian packaging Redis).
For other packagers and OSes: packages don't happen out of thin air: someone has to write recipes and scripts, and test them. So that's already some revenue right there, especially from the big players. Companies like RedHat and Ubuntu could just cut a flat $10m check every year, distributed to projects proportionally in accordance with OS-provided stats like popularity-contest. That's the easy option. Ideally, you would also have some buy-in from the major OS distributions that could somehow "trickle down" the model to their own packages (say, linking a VCS repo to an OS package and reporting activity accordingly: "You asked to apt-get install Redis, but I've checked with a webservice and you have already downloaded it 2 times from Github.com today: time to pony up!". In most cases, this sort of link is already documented formally, somewhere in the package definition.)
Like newspaper paywalls, it doesn't need to be perfect; it just has to be enough hassle that most people who can afford to make the pain go away, will just do that.
The reality in a few years time will be a multi-cloud environment where it's the cloud itself that will be commoditized. So let's make great open source projects work well with that:
1. Provide X options for deployment on many clouds/orchestrators - k8s, terraform, serverless, etc.
2. Work on top of clouds and easily integrate with them, e.g. use existing permission systems of AWS, GCloud, Azure, service discovery systems, secret management systems, parameter systems, log aggregators and many others. So in this approach, pricing would still be controlled by the OSS company, by paying for machine time or per-request.
All the pro-"Common Clause" reporting I've seen has focused exclusively on cloud providers, omitting the prohibition on even "consulting/ support services related to the Software". This reaches further than most license agreements for fully-proprietary software, to the point where I'm stunned that it's even legal.
Thought experiment: is a paid employee of a company working with one of these Redis modules considered to be providing support "for a fee or other consideration"---and thus in violation of its license? Is it illegal now to teach a class on how to use RediSearch?
If your company uses Babel (which it probably does), please donate to Henry[2] or at the least, send him a thank you for his work.
1 - https://www.youtube.com/watch?v=h0sfFX7WH1c
2 - https://www.patreon.com/henryzhu
I posted my own thoughts about over-committing and potential burnout a few months back: https://blog.isquaredsoftware.com/2018/06/redux-writing-resp... . Since then, I've stepped back from trying to keep up with recurring treadmill tasks like cataloging articles and links, and am focusing my maintainer time on some specific Redux/React-Redux issues. Feeling a lot better as a result.
If your opensource project can't live without "life support," it's a failed project. It means nobody wants to keep it alive.
No such prerequisite is for 9 out of 10 software packages you see in any serious software repo.
I now run a graphical Linux desktop with fully functioning graphical UI, all done without a dime of sponsorship money, or corporate life support of any sort.
No "well funded corporation" in the world can pull out making an OS with graphical UI from scratch these days. That's simply beyond even heavyweights like google and apple with their countless battalions of coders.
Now say, who has more power there?
Why it is such a problem for guys making software for web ecosystem if open source desktop community can pull out doing things many many many times more hard and complex?
I myself draw a mental line in between a genuine open source projects, and those which are masquerading as such. In reality, those are failing commercial enterprises trying to earn "coolness cred"
The problem is that this "Commons" rider is confusing and misleading. The term "Commons" makes it sound like this is a Creative Commons license - which it is not. The term "commons" is often used to refer to open source software licenses - which this is not. The coupling with Apache makes it sound like this licensing system is from or endorsed by the Apache Foundation - which it is not. (The article uses the term "Apache Common Clause license" - the fact that author thinks that's a reasonable name, even though Apache didn't create or endorse this license, shows how confusing the situation is.) The Commons FAQ admits that the rider isn't OSS, but then suggests that the OSI definition is some minor technicality - yet the OSI definition is what practically everyone uses, and the "technicality" is at the heart of the point (that everyone can use it for any purpose).
Yes, making money is hard. That problem is not limited to OSS either - most new restaurants fail within the first 2 years. I love to see new/better ways to fund OSS. Having a proprietary rider for some modules or having an open core model aren't new ideas, but if they want to do those things (and have the legal right to do), then let's see what happens!
But if you're going to create a rider, it's important to actively work to not be confusing. It's hard getting consensus on anything; I think the anger here is choosing names that appear designed to be deceptive. If these names and such aren't changed, I wouldn't be surprised if the Creative Commons Corporation or Apache Foundation start considering a trademark suit to prevent confusion in the marketplace. If they weren't intended to be deceptive - and I hope they weren't - then some renaming and clarification would go a long way.
A phrase associated with it is "tragedy of the commons" which means everyone takes as much as they can get to enhance their own lives and doesn't give back. This is one of the things impacting Open Source and one if the things people are complaining about when they say companies use Open Source, make money off of it, but don't want to support it.
Historically, the commons was typically a field where animals grazed. Overgrazing by some could ruin it for all.
FAANG sponsor open source either for strategic reasons (the Kubernetes example in the blog) or as a goodwill exercise. To the kinds of engineers Facebook would love to hire to help serve ads more devilishly, React goes a long way to making the company look developer friendly. These projects get a ton of marketing support (time to promote the project, brand name, etc) that an all-volunteer nights and weekends grassroots open source project can never complete with.
Then there’s open core, which seems to struggle from the difficulty that open source has a kind of altruistic halo around it, and any attempt to commercialise or restrict parts of it (especially for things that were once open) is seen as bait and switch. Yet to see an Open Core company that isn’t continuiously torn between what it takes to keep the business viable and what it takes to keep the community on their side.
Consulting and support has its own downsides too. It only scales linearly as you can hire people, and the margins are much lower so it takes a lot more people - much less fun of a company to run. And it influences the project roadmap too: “yes the setup process is broken and could do with a wizard, but the complex install process drives a lot of our consulting leads”.
It seems like the huge companies that were threatened by open source have been replaced by huge companies that know how to weaponise open source. Open source still wears a certain halo thanks the nights and weekends code hacker that does it for the love, but open source at any scale is as corrupt - maybe more corrupt - than any proprietary company.
(Disclaimer: my company sells proprietary software but gives customers source access/right to modify and open sources many of the building blocks. As someone who wrote code for the love of it, I often feel like there’s something evil in not making everything open source. But I’m yet to see a less evil open source model)
Sometimes those same developers will leave that wealthy company to start the commercial vehicle for that project. They talk about the wonders of open source and free software.
Obviously their next move is to monetize the project with commercial add-ons and they end up charging 6 to 8 figure price tags for 'free' software. Their enterprise customers justify this by saying 'we have the source code for the core product...its open source so we won't get locked in'. But the likely hood that some overgrown Fortune 2000 IT team will use the forked source code and not the shipped enterprise product is basically zero.
At that point, the FOSS nature is just a farce. It might as well be proprietary software at that point. There's no way the same company will prioritize the community over the high paying customers. They take over the development of the core open source version. They end up hiring all the main contributors. They find new ways to monetize it. They build add-ons that are end up being essential to using the same project in production. Its no longer a community-driven engineering meritocracy, but simply an enterprise development team with a github page.
Yes it was. It has since been edited to be clearer.
> Redis has adopted a new license I've been working on w/ a talented team.
The Redis mentioned here. I believe, actually means Redis Labs, not Redis the project.
[0]: https://twitter.com/kevinverse/status/1032068995979956224
No, we've discussed this time and time again:
Open Core = crippleware
Articles like this occasionally come out and promote propaganda that masks "crippleware" with nice sounding words.
- Changelog has interviewed me about this https://changelog.com/podcast/236
- Hackernoon has published me on this https://hackernoon.com/the-implications-of-rethinkdb-and-par...
I maintain a tremendously popular Open Source project with 8.5K+ stars, and it is extraordinarily hard. (Hint: Imagine Firebase + IPFS had a lovechild, that is what it is. See my HN profile if you are curious.)
I should be the first person agreeing with this - last year was one of the hardest times of my life because of doing Open Source. But this is the moral, harsh, but important:
Financial hardship sucks but does not define the values of Open Source.
The values of Open Source has removed financial hardship for many, and made lives better for the masses.
https://twitter.com/marknadal/status/1008066378295750656
Just like you should not gamble on crypto what you cannot afford to lose... you should not Open Source what you cannot risk.
Open Core = crippleware: Faux companies publicly wearing sheepskin to look like they're developer friendly, but privately are forced to shame you into the fox's den.
Open Source = giving, truly. It has hurt me at times, but those values have brought more win+win to my life and others, far beyond we can count.
So thank you, to all real Open Source developers out there. You are heroes.
I don't agree with the tweet at all. I too thought MIT/BSD are good licenses, until I got burned by a piece of code majorly consisting of MIT code infringing my user rights - unwarranted tracking and a few bugs I couldn't fix. I now understand why (A)GPL exists and make sure my projects infringe on no users' rights like projects containing MIT/BSD can.
https://www.indiegogo.com/projects/prosemirror#/
I suppose this funding-model could work for maintenance work too.
This, in a nutshell, is the driving force behind the EU's antitrust investigation of Android.