I think it’s not fair that we expect every developer tool to be OSS. Most things that are fully OSS like react, typescript, etc. and even Linux to a large degree are supported by humongous amounts of cash from large tech companies. When a smaller firm devoted to developer tools like datastar decides to monetize, moreover, they are doing so without leaning on ad monetization, which many big firms rely upon, or other means that are not particularly nice. I think it’s really important we cut folks working on paid dev tools a break. The alternative is that we end up in a world that is largely merely defined by huge tech companies.
The problem is the rug-pull itself, not that the rug isn't there.
I don't have an opinion on Datastar, as I'd never heard of it until this article, but over the past year or two there have been a _lot_ of open source projects that have been converted to proprietary licenses, very often after being invested by VCs or PEs. It's happened to me a number of times now where the license for the features we were using went from open source to proprietary with 5-6 figure cost.
Developers gotta eat, I get that. But often the reason I'm using one of these components is it's a hobby or low value project where it simply doesn't bring in the income to justify paying for a license. If I had known this would happen, I would never have used it in the first place, used an alternative, or maybe just never bothered with the project. But now you're in an awkward position where the choice is either pay-up or re-do a bunch of work.
There was no rug pull! (And the term isn't even being used correctly) They talked in their community decided the set of features caused a support burden and for versions later on they would put them behind a pro tier to help pay for the extra costs for supporting them.
You can keep using your current version! You can even fork at that version. Calling it a rug pull is so entitled.
Like I said, I haven't heard of this project until now, so I don't know the wider context, but it may be that some of the people who are reacting negatively to it have been burnt in the past by the many other projects that have gone down this route: project starts as open source, then it goes open-core, then over time more of the dev effort naturally moves into the proprietary part, then sometimes eventually they change the license for the open part too.
Forking is often impractical in reality as a solo dev or small team rarely has the resources to keep up with security fixes.
I'm entirely happy to pay for things, do pay for many things, as well as donate to the authors of projects I use, and whatever this library is seems reasonably priced. Nevertheless, I'm pretty reluctant now to use open source libraries unless they're backed by a foundation, given how many times I've been badly burnt.
> There was no rug pull! [...] You can keep using your current version! You can even fork at that version. Calling it a rug pull is so entitled.
This is a dishonest perversion of the commonly accepted definition of a "rug pull".
I'll copy what I said in a previous thread:
When Redis changed licenses to SSPL/RSAL, users were also free to continue using the BSD-licensed version. Was that not a rug pull? Same with MongoDB, Elastic, HashiCorp, etc. These are quintessential examples of the "OSS rug pull".
The idea is that users were relying on a functionality to be maintained (the "rug"), and the Datastar developers decided to continue maintaining it behind a paywall (the "pull").
Nobody is claiming that developers physically took the feature away from users, as that would be ridiculous. But users of these features are now forced to either maintain it themselves, wait for someone else from the community to fork and continue maintenance (which has its own set of issues), or pay up.
You can argue how it's "only" a few hundred lines of code; criticize "incapable" developers who can't check out a Git commit or do maintenance work they previously didn't have to; that the features don't require maintenance at all; and come up with other defensive arguments. But none of it matters. The size of the "rug" doesn't matter. It's the principle and precedent it sets for any users who were potentially interested in the project.
To say nothing about putting essential features like a bundler and debugging tool behind a paywall. These are not "Pro" features.
Nope, at least following a common definition of "rug pull" outside of FOSS. A "rug pull" refers to a specific type of scam where the rug-puller absconds with funds. Wikipedia even redirects "rug pull" to "exit scam": https://en.wikipedia.org/wiki/rug_pull
If you're standing on a rug that you didn't pay for in any way, and you're nonetheless surprised and angry when the rug-owner eventually decides it is no longer practical for them to continue maintaining the rug for free, that's just entitlement.
Nothing in FOSS licenses says anything about future maintenance expectations, and in fact nearly all of them contain all-caps clauses shouting at you that the software is delivered as-is.
> These are not "Pro" features.
If you didn't create a piece of software, or pay for it, or directly contribute to it, then you certainly don't get to decide which features are "Pro" features. Instead, you can vote with your feet and your wallet and decide not to use it, that's fine; that's the model with all other products in life. Why should software be any different?
Simply using a piece of free software doesn't entitle to you have any control over the future activities of the software's creators!
> Nope, at least following a common definition of "rug pull" outside of FOSS.
The current context is within FOSS, however. And it has a widely known definition[1].
> If you're standing on a rug that you didn't pay for in any way, and you're nonetheless surprised and angry when the rug-owner eventually decides it is no longer practical for them to continue maintaining the rug for free, that's just entitlement.
So, let me get this straight. You're saying that because users didn't pay for the software, they're not allowed to be surprised and angry when they're forced to either continue using unmaintained versions, take over maintenance themselves or wait until someone from the community does, or pay up? And you're saying that paying for the software buys you these privileges?
So, essentially, if the author of cURL decided tomorrow that it's no longer "practical" for him to support ancient and obscure protocols like Gopher and Telnet for free, and he decided to put them up behind a paywall, that the people depending on these features don't have any right to be upset about it? And if they are, they're being entitled?
Or, from a social perspective, if the people making meals at a community kitchen decide that it's no longer "practical" for them to supply eating utensils for free, and start selling them at $10 a pop, that the people relying on those meals are being entitled if they're upset about this change, because they're not paying customers?
Don't you realize how hostile this is?
Forget about open source. This has nothing to do with licensing. The libre software philosophy is orthogonal to monetization practices, and every author can choose how they wish to make development sustainable, if at all. As I've said before[2], I'm in favor of F/LOSS projects having commercial tiers or subscriptions, as long as it's done fairly.
What I am arguing for is having basic decency to treat all users of your software with respect. The moment you decide that users not paying for your work are not worth listening to, that they're entitled, etc., and that your attention and respect can only be bought, you've corrupted the entire ideal of why we build freedom-respecting software in the first place.
The sad irony is that I'm 100% sure that most people arguing against this, including Datastar authors, rely on tools like `curl`, `grep`, and other libre software whose authors never have and never will pull these shenanigans. So they're quite happy to "leech off" other people's work, but not happy to let others do the same off theirs. This double standard is what's ruining open source, not entitled users.
I won't bother responding to the other comments, as they boil down to this same point. And I'm really repeating myself here, so feel free to talk amongst yourselves.
> You're saying that because users didn't pay for the software, they're not allowed to be surprised and angry when they're forced to either continue using unmaintained versions, take over maintenance themselves or wait until someone from the community does
Surprised maybe depending on how long the software has existed, but angry? 100% they are not allowed to be angry.
Anyone who thinks that a maintainer of open source software is required to maintain that software for free forever is an entitled brat.
Don't you realize how hostile it is to have a bunch of non contributors demand and entitle themselves to things and burn out OSS organizers at an increasing rate so much so it's been cited as an ongoing critical concern to the entire software supply chain?
Your attitude ignores the real financial and time constraints that OSS has been facing forever and people are rightly starting to get tired of this shit.
We're well past the stage where its little well meaning communities trading cool bits of code with one another OSS is a completely economy defining thing now and people shouldn't settle for being worked to the bone, they should be setting proper safeguards, like Datastar has done here, to ensure their project does not burn them out and they can continue to grow and direct it in a way that accords with their own interests and desires.
> The current context is within FOSS, however. And it has a widely known definition[1].
Yes, and my argument is that definition is wildly inappropriate. It takes the perfectly-legal actions of software authors and equates them with an illegal financial scam.
> You're saying that because users didn't pay for the software, they're not allowed to be surprised and angry when they're forced to either continue using unmaintained versions
Correct, except no one is "forced" to do anything. That includes the software authors, who are not forced to continue maintaining their software, for free or at all, if they so choose.
> And you're saying that paying for the software buys you these privileges?
Potentially, yes. But that depends entirely on the terms of the license, product offering, payment method, etc. Just like anything else you buy, if the product is deficient (potentially within a set timeframe), you may have some recourse available but it varies.
> if the author of cURL decided tomorrow that it's no longer "practical" for him to support ancient and obscure protocols like Gopher and Telnet for free, and he decided to put them up behind a paywall, that the people depending on these features don't have any right to be upset about it?
Correct. If you are not the employer of the author of cURL, and you are not a customer of the author of cURL, why do you think you have any right to tell him what he can or cannot do with his software or his free time?
> And if they are, they're being entitled?
Absolutely, 100% this is unquestionably extreme entitlement.
> if the people making meals at a community kitchen decide that it's no longer "practical" for them to supply [...]
Open source developers are not even remotely similar to soup kitchens, and open source software users are not impoverished starving people.
> Don't you realize how hostile this is?
After the previous analogy, perhaps you should ask yourself that.
> I'm in favor of F/LOSS projects having commercial tiers or subscriptions, as long as it's done fairly.
The issue here is that your definition of "fairly" is completely subjective, and seems to involve you having a say in what other people do with their time and intellectual property.
> The moment you decide that users not paying for your work are not worth listening to
That simply isn't what I said, nor what any of the sibling comments said.
> So they're quite happy to "leech off" other people's work, but not happy to let others do the same off theirs.
I believe this is a complete logical fallacy, as I don't see the Datastar authors (or anyone else in this thread) complaining about leeching anywhere. But please link me to this if I'm mistaken.
You don't get to decide what they want to maintain. They could have just deleted the features. This entire saga has made it so clear to me people that can only see what is in it for them versus being able to step into the shoes of a maintainer offering something for free.
There is literally no obligation to support you at all. Sucks if it puts you in a bad position that's not on the maintainer, there's no legal or social contract they broke in fact that's entirely in the spirit of what OSS is supposed to be. It's not so a bunch of mediocre free loaders can demand continued access to a library they never contribute to and only consume.
The correct way to look at this is you have the privilege of consuming the hard work of another person for free and you should be greatful they put in work to make your life easier in an open way when they did not have to ever and at all and understand when the burden on a small team makes it hard for them to continue to support a feature set. There isn't a sacred duty to people who consume something another group opened because they wanted it to exist.
That's what the licenses and all the legal bits say. They do it because they want to and if they stop wanting to you can all get bent. Instead they looked for a sustainable middle ground.
At least the controversy is making them even more popular.
> The idea is that users were relying on a functionality to be maintained (the "rug"), and the Datastar developers decided to continue maintaining it behind a paywall (the "pull").
Why were the users so entitled to free ongoing maintenance that its end is worth describing using a term from financial fraud?
> It's happened to me a number of times now where the license for the features we were using went from open source to proprietary with 5-6 figure cost.
In those cases, prior to the project going commercial, did you contribute nontrivial code to the project and/or financially sponsor the project?
I could see being upset in those situations, but in most cases I find the answer is no.
> If I had known this would happen, I would never have used it in the first place, used an alternative, or maybe just never bothered with the project.
If you had used an alternative, the same scenario could have played out with the alternative.
Realistically, what are the "never have used it" / "never bothered" scenarios? Presumably you chose the project because you needed it for something; that implies the never-bothered alternative is essentially just writing something from scratch instead. Which you can still do now. And you can use the last FOSS version of the project as a starting point, which saves a tremendous amount of time. So how exactly were you burnt by a supposed "rug pull"?
If the rug isn’t there, how can it possibly be pulled?
All free licenses make each commit free - forever. If a library does what you need today, use it! If the terms become unacceptable in future, fork it and maintain it yourself, or hope someone else will. Note this can even happen with free software (GPL2 to 3, for example).
No one is entitled to the future work of someone else without paying though. You very definitely are the entitled one here.
I think it's totally normal and correct to have a license where a company like Amazon can't come in, steal the volunteer work of hundreds of developers, slap their logo on it and sell it.
I'm sure open source purists do not like this, but the world is the 1980 anymore. It's been 45 years. Things need to adapt. Open source needs to adapt.
Yeah we really need to normalize licenses that protect against that. Even GPL doesn’t because everything is SaaS now and companies will just isolate the GPL code to one micro service. AGPL might prevent this but there aren’t a lot of cases that have been litigated. And if they don’t modify the source then then it doesn’t do much IIRC.
But “open source” was in control of big business from the start. The open source consortium was a late 90s attempt to co-opt the free software movement and turn it into something business friendly.
Tim O’Reilly funded it to start and now it’s funded by big tech companies.
> I think it's totally normal and correct to have a license where a company like Amazon can't come in, steal the volunteer work of hundreds of developers, slap their logo on it and sell it.
Why shouldn't this to apply to every company - including the one ostensibly shepherding the open source project? I would argue that employing a bunch of core developers doing 10% of the work doesn't entitle you to be the sole entity to monetize the work of the other 90% of the community, but I don't think anyone has come up with a proper license to defend against that yet.
Open source indeed needs to adapt, but I don't think the source-available or open-core models we are seeing these days is the right solution. If you really want to prevent third-party entities to profit off your work you'd need to go for something like the AGPL, but that is for obvious reasons not exactly a popular choice.
> Why shouldn't this to apply to every company - including the one ostensibly shepherding the open source project?
Because that's simply not how copyrights and trademarks work. The licensor doesn't need to abide by the terms of the license, by definition. The purpose of a license is to grant rights from the licensor to the licensee.
> employing a bunch of core developers doing 10% of the work doesn't entitle you to be the sole entity to monetize the work of the other 90% of the community
Very few of these cases are 10% company / 90% community. If anything, it's usually the other way around. Not to mention the huge amount of time spent on code review and ongoing maintenance of third-party contributions.
> I don't think anyone has come up with a proper license to defend against that yet.
That wouldn't really make sense; a software license isn't going to remove rights from the licensor. More realistic solutions are things like intentionally not having a CLA (effectively preventing the project creator from relicensing) and/or reassigning copyright and trademarks to a foundation.
I have to work with some closed source external tech at my job right now and I hate it so much. Probably more to do with the specific tech than the open/closed factor, but it feels really bad whenever there's an issue or an improvement that could be made but I'm not allowed to fork or directly change the dependency.
I honestly feel so burned by it that I will think strongly about ever joining another company that isn't using more open source tech.
I have no opinion on datastar, and I support things like tailwind selling pre built tailwind components to make money (not that I use either, but idea wise I'm happy for them). But sometimes working with closed source is a real pain.
I will post again what I wrote ( slightly eddied ) earlier in the Datastar thread.
There is a mainstream generation divide in open source ideology over the past 15-20 years.
The modern one is what lots of younger generation agree upon. It should always be open source and continue to be supported by the community. For the interest of Public Good. Some of these project where basics living cost are met money mostly driven by VC, or zero interest rate phenomenon. Preferably GPL or APGL, no body owns the code or even rights, the project belongs to the community. Everything is or should be OSS. When money is involved, donation is the preferred method. At one point Open Core were fine, I believe Sidekicq works on that model for at least 10 years. But I guess now even Pro version of anything is borderline unacceptable as it is "bait and switch".
The old folks are basically and mostly take it or leave it. Fork it into my own while taking the maintenance burden too. Sometimes they charge for it, sometimes they dont. They are just sharing something they did to the rest of the world in the hope some may enjoy it. Mostly started out as just a hobby, not something big or professional.
Yeah, it is weird to me how many people expect free stuff, and when they get it, they act entitled to all of the developer’s time and get angry if they don’t get it. No other industry operates this way.
And the whole “the developer gave me so much for free for so many years that I am now entitled to never pay for their work in the future” attitude. It’s bizarre.
It's true for things other than developer tools. I built a free utility that became popular and people started asking for more features. I spent several months working full-time on building out improvement and put 10% of the new features behind a paywall. I took nothing away and made 90% of the new features free too. The stuff I put behind a paywall is stuff I thought would only apply to professional business users. Just the fact that I had paid features made people upset. I got a ton of angry emails asking why the tool was no longer free. I also used to make some money via donations through BuyMeACoffee and those nearly completely stopped.
If I throw code up on GitHub, I try to be real clear that if it's good enough for me, it's done as far as I'm concerned[0]. I'm also licensed under WTFPL. I figure if that scares off any commercial interests, so much the better.
You're probably a lot more fully baked than I am, so this path may not work for you.
Sorry to hear that man. It does really sour the ideas of the MIT license especially. It's a social problem and won't be solved by tech. My response now is a pretty hard line, cool... then fork it. Anything softer doesn't matter cause if they really cared they'd already be actively helping.
I agree in theory, but the problem is that I have been burned over and over again by proprietary tooling. The interests of the dev tool makers often do not align with my interests as a developer, and that's a problem.
Proprietary tool vendors are just trying to create shareholder value. If that means firing the entire dev team and never doing another release, they will do that. If it means switching from "pay once, use forever" to "pay $20 / month", they will do that. If it means going from $20 / month to $2000 / month, they will do that. If it means putting ads inside my IDE, they will do that. Just look at what Broadcom is doing to VMware to see what this can look like in practice.
OSS developer tooling is usually made by and for the community. The interests of the dev tool makers align with the interests of the developers using it, because they are the same people. There are no incentives to enshittify my developer experience, so I can safely rely on the tooling without worrying about whether it'll still be usable next year. And even if the core team decides they want to make some wildly unpopular changes, the rest of the community can still fork it and continue on their own direction!
I really wish it wasn't the case, but there are very few proprietary tool vendors I'd be willing to believe if they promised they wouldn't ruin my day a few years from now by doing a rug pull. Some small just-started firm I've never heard of? Probably not in that list. I would love to cut them a break, but trust has to be earned.
>OSS developer tooling is usually made by and for the community. The interests of the dev tool makers align with the interests of the developers using it, because they are the same people.
That’s the case for the vast majority of projects. Most of them are made and controlled by 1 person. And many that are controlled by more than one person are de facto or even de jure controlled by a single large company.
99.99% of the “community” will never contribute meaningfully to a project.
Sure you can fork OSS projects, but for anything remotely complicated doing so will turn into a full time job.
