I really really really wish the FSF-haters all the luck in the world in their quest to thread the needle between bigcorp-hostile, GPL-style copyleft and user-hostile, bigcorp-friendly open source.
I'm sad they can't learn to love GNU projects. I for one do. They just... work. Whenever they don't, invariably there's some freedom-hostile hardware in the way. This makes me want to move closer to FSF-recommended hardware.
I wish them all the luck because the FSF just seems cursed at this point. They do absolutely essential legal and advocacy work and the world wouldn't be what it is without them, and virtually nobody has a deep enough sense of history to recognize that. They will still get my donation dollars for the foreseeable future, and I hope they can rehabilitate their image.
I do sorta think this whole thing will end up working out well overall for the FSF's mission of building a software domain that's solidly out of the clutches of proprietary predation. I've never seen the FSF, the distinction between open and free software, in the public mind more than the last year. I can get really snarky about how all these people who never cared before suddenly come out against the FSF. You didn't care before, you still don't care, good job expressing the total lack of impact you insist on wanting to have.
Are more free software foundations a good thing? Probably. It'll dilute the impact in the short run, these new projects will have to spend time and effort solving run-of-the-mill nonprofit stuff that the FSF worked out decades ago, but it also means more people will get invested. Eventually the new blood will learn to work with the old guard and we can all build a new order together.
Color me optimistic about the future of free software.
It was (still is?) really cool to hate on the GPL. I think the long lost root of the discontent is because it restricts developer freedom (in favor of user freedom, of course), but in reality it just became a meme and most people who critique it don't actually understand what its purpose is. It's like PGP, show me a better alternative and I'll gladly jump ship. But it's not terribly productive to complain without offering solutions.
As a developer, you have a choice to make when you encounter GPL bashing: jump on the bandwagon, or rebut. If you choose the bandwagon, don't come crying when you can't make a living building honest software that treats users with respect because you have to compete with large companies who ~~take~~ steal agnostic software and subsidize their products with user disrespect and line their pockets selling ad profiles.
As a user, demand honest software and new companies will rise to meet your demands. Instead of ads, pay $30 and support a project that puts your interests first. Subscribe for $3/mo when you find a product that brings you continuous value.
As a technical leader and someone in a position to make impactful technical decisions at your small company, educate yourself on the license landscape, examine your core product and target market. If you're targeting open source adoption by big companies and expect them to eventually pay for a proprietary overlay, probably avoid the GPL. If you're building a product for _users_, consider enshrining that position in your license terms and simultaneously defending their freedom while deterring predatory abuse by large companies.
It runs contrary to the FSF guidance, but I'd also suggest, out of practicality, patenting your (software) inventions. Ultimately it gives you flexibility and defense in the event that somebody else tries to assert their own patents. It also protects your GPL users because you explicitly license them to use your patented work, and it allows you to license that technology regardless of any software you've written to companies who may not find the GPL palatable but still would like to use your invention. Win win win.
But if you want to send secure messages it also matters if your friend has remember to enable --dont-be-stupid and disable --automatically-guess-its-20-year-old-insecure-format
I'm not quite sure if you're serious or not. The person you're responding to thinks that a special purpose app would be better than PGP, and your response is that the best solution is a special purpose app which calls into PGP. Why is the latter preferable?
The person I'm replying to says that PGP is hard to use because there's so many options you have to know and use correctly. I completely agree with that, its UX is most probably the main reason it isn't more widespread. But that doesn't invalidate PGP itself: the format is still useful, we have the tools to use it correctly. So I'm saying that using PGP crypto is still worth it, especially if all the bad bits (configuring, tuning) are managed automatically by the application.
The problem with a universal format is it's hard to know that everyone you're talking to does things right. If you have end to end security and the other end leaks, that's not very good. Whereas with a limited format I know a non-malicious counterpart probably has their client configured just like mine.
Here's a good quote from the article at the top of this chain:
> Take AEAD ciphers: the Rust-language Sequoia PGP defaulted to the AES-EAX AEAD mode, which is great, and nobody can read those messages because most PGP installs don’t know what EAX mode is, which is not great. Every well-known bad cryptosystem eventually sprouts an RFC extension that supports curves or AEAD, so that its proponents can claim on message boards that they support modern cryptography. RFC’s don’t matter: only the installed base does. We’ve understood authenticated encryption for 2 decades, and PGP is old enough to buy me drinks; enough excuses.
> You can have backwards compatibility with the 1990s or you can have sound cryptography; you can’t have both.
I am very well aware of the criticisms against PGP-as-a-model and I actually agree with them. The premise that having an open protocol makes changes 10x harder to actually spread is very true. I still believe that it's better to have that than everyone doing the same thing over and over and over again, "but this time it's better".
HTTP, javascript and TLS have also shown that a sufficiently motivated set of actors can move the ecosystem forward. True, organizing the ecosystem is not the same job as actually building stuff, but it's still beneficial to all of us.
That's the problem. You have n*m possible combinations of clients, leading to many opportunities for bugs, and everyone has to degrade to the lowest common denominator because you can't upgrade in lockstep.
I understand pessimism of author and others in this thread when they notice that free software ideas are attacked from everywhere: companies subverted idea with fake "open source", many young people dislike GPL as "not cool", companies taking over projects such as with redis and so on.
But I noticed change in my personal perception (and I am not alone) towards more positive reception of actually free software and I believe that mass perception may change.
This is what changed my perception:
You see how often proprietary and "fake" open source software gets subverted by companies over time, especially when company is heavily funded by VC. After you see it couple of times you just get tired and there is no motivation to invest yourself into it so you just ignore it. Keybase is perfect example of such failure.
Now when I see something like Signal I just ignore it (their recent controversial addition of crypto proved to me that it was right decision).
And it applies to software that I use in my work too, if I get to decide what software to use I try to avoid such projects.
The irony of the GPL hate, is that without GNU we would just be using the usual UNIX flavours, with everyone taking whatever they felt like from BSD, as it was happening before Linux came into play.
I and others were there, with source available, PD, shareware, beerware, postware and whatever else was available as business friendly license.
I will give it two decades max, for everything to get back to the old way of selling software.
> The irony of the GPL hate, is that without GNU we would just be using the usual UNIX flavours, with everyone taking whatever they felt like from BSD, as it was happening before Linux came into play.
More like: a small part of the professional word would be using the usual UNIX flavours, and the rest of the World would use Windows both for desktop and for servers (add a tiny share of free BSDs here and there if you wish).
> I will give it two decades max, for everything to get back to the old way of selling software.
It would be nice to get back to Turbo Pascal-like and old Adobe business models, instead of the divide between subscriptions and free (as in beer) open source funded by a bigcorp making money from something else. So I'll take that.
That is easy actually, many just have to do like in other professions and pay for their tools, instead of expecting to be paid, while the authors of their tools need to survive from charity.
I think Free software has lost. Open Source is a far more popular brand (thanks to big tech, no surprise there) and it's designed to keep Free software ideas digestible by corporations.
Big tech has learned how to keep Free software at bay, to the point that loving or hating the FSF doesn't matter anymore. It's irrelevant.
> I'm sad they can't learn to love GNU projects.
I think you don't realize how this sentence comes across. In some ways, it's an extremely concise way of describing why the FSF has failed.
I can see how you think this way, if you're only thinking in terms of popularity.
Let me reiterate. The FSF does legal and advocacy work. You're only thinking in terms of advocacy. The legal work is just as important, without it, we wouldn't be able to use GPL licenses at all.
Big tech didn't accomplish it's goal of killing the FSF itself. The one org standing in its way of complete dominance. It's not a battle of ideas, it's a battle of people.
All it takes is a core group of people willing to put their money and time into it to keep it alive. Stallman himself said he never thought free software would ever become a thing and that the FSF has succeeded beyond his wildest dreams.
As long as the GPL continues to be a viable alternative, we will continue to have a bastion against the endless waves of proprietary software.
Do you believe they are making a genuine effort to do that? Or is it arrogance on the part of Stallman and co. which is still keeping the FSF closely associated with the Epstein scandal, while MIT and others have successfully moved on.
It's an old org with a lot of cruft. Stallman isn't stupid, and absolutely cares about the cause of free software, more than any of us could ever possibly care.
What's happening isn't 100% the fault of Stallman, but a continuation of the long-running chasm between the free software and open source communities. Though it certainly didn't help that Stallman chose to reclaim his board seat. There's just a lot of antipathy between the two camps. There always has been.
The he said/she said of RMS and Epstein scandal “misinformation” doesn’t need to be clarified. RMS said some questionable things (to some people). The corrected version is still questionable (to some smaller subset of people). This Epstein thing is not the only questionable thing he’s done/said. He’s a liability regardless of the truth, because the truth does not exonerate him.
The FSF is an advocacy org and you can’t advocate well for a moral position (free software) if your main advocate is ensnared in scandal over morally questionable behavior, regardless of the truth.
The FSF first now has to advocate for the morality of RMS to clear his name enough that he can advocate for the morality something else and have enough people listen. Frankly, his track record beyond the Epstein mess was already dubious, so it’s a losing battle.
You mentioned Epstein specifically so that's what I'm responding too.
What does it mean to be "questionable"? controversial? wrong? That the earth is round is questionable (to some smaller subset of people) - that doesn't change the fact that the RMS was misrepresented (without scare-quotes) - the truth of his actual comments wrt Minsky do 100% exonerate him - that a subset of people don't agree is irrelevant.
Questionable is that some significant group of people might not be ready to support or condone the views. The fact that its still in discussion means that even with his "exoneration" he's still not viewed positively.
He's viewed so negatively that this Epstein thing (which he is not exonerated entirely imo) is not the only reason to distance from him.
Sure, because of misinformation. A significant number of people may hold an insignificant view; a significant number of people don't wear facemasks either.
I really don't think it is just that. It is not just Minsky/Epstein scandal, those are just something people who disliked how much more low key much less impactful "uncomfortable" behaviors were enabled.
And it has zero to do with open source to wining over FSF. Like, nothing at all, popular organization survive sex-abuse scandals just fine. Partly it is business friendliness, yes. But also, personally I remember finding culture around FSF off-putting years ago, for reasons that have nothing to do with Minsky/Epstein kind of issues.
I found that off-putting because of hollier then tho, we are geniuses everyone else is looser and if you don't agree with us is bad programmer attitudes. And the massive thread wars over everything that kind of pretend to be rational, but are actually massively emotional.
IMHO most (if not all) of the problems described in the article (talking about other companies) started when the stakeholders sold their companies for the highest bidder. Is it a surprise that this money would corrupt the company?
It's just like when WhatsApp founders sold the company to Facebook. They got billions of dollars but eventually became disappointed with Facebook plans to shove ads into it (eventually).
The same idea applies to core developers accepting cushy high-paying jobs in huge companies.
The root cause is people become greedy. That's what happens when people optimize just for money. And these are exactly the cases that appear in the news, which further influences other people facing a similar choice...
Is that the root cause? Individuals being greedy? I think we can traverse up a few more parents to find the actual root of the tree. There's an entire economic system here actively encouraging and rewarding this cycle. Blaming individuals is not an effective strategy to improve society. We need to examine the system that is influencing the behavior of many individuals.
This is actually a discussion that has been explored here on HN too.
YC has as a sort of a motto "build something people want", or something along those lines, and I doubt that founders are all equally happy to see their ideas become a garbage can of hamfisted revenue streams, but that's where you end up when you get VC money, give up board seats, and can't show "sufficient" growth.
On this last topic there was an interesting discussion a while ago about where VCs and founder interest is misaligned.
The picture of software dev you paint in the article (great read btw) is of a dichotomy: it's either a Big Tech corporate-politics nightmare, or a startup doomed to be that.
I think there is another option. There are thousands of small software companies that don't have the growth or traction to get bought out, are happily making relatively small amounts of money in their niche (but still enough to pay decent salaries - not SV salaries, but enough to live on comfortably), and care about both their customers and their code. They tend to employ relatively few engineers, who stay for a long time, and have founders who know their market niche really well.
You won't hear of them, because you're not in their specialised niche and they don't market outside their niche because they don't need investment. They're not really interested in going public, or being acquihired, or even gaining a trade sale, though that is the ultimate destination for most of them (when the founder decides to retire). They're not "sexy", except if you like actually practising your craft and building rock-solid products ("software you can love" - good phrase).
That sounds like non-big tech, so normal tech jobs? Sure, but the section with the dichotomy (which admittedly is a simplification, true) was specifically about big tech. It's a counter argument to the idea that if you're a good developer then working for google or a hot startup is the best possible job for you.
Ah yeah, re-reading it it's clear you were talking about Big Tech, I missed that on the first go.
Though I kinda resent the industry's focus on Big Tech, like all we can talk about is these few mammoth dysfunctional nightmares that take up all the room. There's lots of good tech work out there too!
The Redis example particularly stands out for me. For the longest time Redis was hailed as a shining example of open source success – small group (mostly a single developer) working on lightweight, performant, easy-to-use, free (in terms of both cost and freedom) software which beat out several goliaths in the space.
Now the biggest call to action on its official website is to sign up for a trial subscription. New features are hidden behind an enterprise version that you have to contact sales to get the price of.
I remember when antirez announced that he was leaving redis in the hands of the community. Didn't expect it to take such a blatantly commercial turn, but guess it shouldn't be surprising in retrospect.
Are there any meaningful or developing alternatives to Redis that are truly free?
In the same vein I'm merely querying the funding of such.
Snark aside, I'm not aware of _any_ open source projects which are developed through altruism and not backed by full time sponsors from companies. Redis is no exception. It's still available as a complete open source offering, for free. Suggesting the maintainers have compromised the project because they're looking for a method to sustain development on a project that is depended on by many many developers is unhelpful, and naive.
sourcehut (https://sr.ht) is a good counterexample. It funds itself and is completely open source.
Anyway, I am not suggesting that the project has been compromised just because it has been commercialized. I don't do all my development in open source and don't expect anyone else to do so either.
You've dismissed my comment without giving any backing to it. What Sourcehut are doing is the same as Redis. They provide the core product at [0], which is completely open source, and redis labs [1] provides a hosted offering with support.
It's also on a totally different scale. With all due respect to sr.ht, it's a much more niche offering than redis, with far less demands.
The difference between sourcehut and redis is in the governance structure and in my perceived predictability of their roadmaps.
The governing committee of sourcehut is a single person - ddevault. He clearly states his motivations and intentions regarding sourcehut in his regular updates. It is easy to predict the direction in which sourcehut will be developed over the next year.
The article we are commenting on states that the governing committee of Redis consists of representatives from Redis Labs, AWS, and Alibaba. Their agendas are not as transparent as ddevault's and mark a potentially significant departure from antirez's plans for redis.
Most importantly, as a member of a team that is just setting up a redis instance, I naturally wonder what the alternatives are so that we can properly evaluate them. Because we are a small team and strapped for cash, we would prefer to evaluate free (as in freedom) options. This is what led to my original question.
Coming in late, but sr.ht also differs from RedisLabs in the sense that sr.ht is 100% FLOSS, no strings attached - and RedisLabs publishes a lot of source-available (i.e. nonfree) software.
Dolphin Emulator is open-source, devoid of corporate funding to my knowledge, and does not sell premium builds to my knowledge. I don't think any of the developers can live off their work on Dolphin. However many other emulators do sell premium builds, and I'm not aware of many other open source projects that haven't sold out.
Dolphin's a great example actually, thank you! I think you're right that nobody could live off of Dolphin. I know one of the "active names" has regularly stated that it's a hobby for them. I do wonder how they fund services like https://dolphin.ci/#/ - it's got to be someones passion. But clearly, funded by altruism, a very well deserved exception to my statement.
The comment I replied clearly indicated they weren't happy with the commercial turn of Redis after antirez left. Redis is still free as in speech (and also, beer).
"Hell is still other people" and the need to eat and keep a roof over one's head remains an aggravating pain in the ass for all idealistic people everywhere who wish they could do good works without having to sort out such inconveniences imposed by reality.
(Me included, so it's not intended in an ugly way.)
It's interesting that you put that quote forward as a reason to support UBI. I have the opposite reaction: I have no desire to be taxed to feed people "scratching their creative itches".
Yeah, I understand that. Took me a while to get past it too.
What helped was the realisation that almost all of the history of Science is made up of wealthy individuals scratching their creative itches. Art and Music less so, because patronage was a common method of funding that, but we lost out on a huge amount of both because patronage was rare. Imagine what we will discover, what vast wealth of new knowledge and talent we will enjoy, if we can free up those who can make a genuine contribution from having to pay rent? (and yes, 90% of everything is shit, but the 10% makes up for it).
Also, I changed my "scarcity" mindset. I enjoy what I do, and get paid enough to do it. Paying taxes is part of that. I resent the various governments I've paid taxes to for wasting my money in all sorts of ludicrous ways, including the vast subsidies to wealthy people and industries I flat disagree with. Paying my tax money to actual people seems like a great deal to me :)
I'd agree with you if it actually helped artists find their artistic stride - there is definitely artistic merit in forced constraints. But it doesn't; it forces them to commercialise to pay rent. To make what is commercially appealing (or appealing to a patron) rather than what they actually want to make.
And I totally agree with the argument that commercial appeal == value, and artists don't automatically deserve a living. The whole "Spotify should pay us more because we can't live on what they pay us" argument leaves me cold. But I think that's a different discussion. If you are going to force artists to commercialise then that argument applies, and artists must adapt their art to commercial reality, or starve, like everyone else. However that's not the same discussion as "if we paid everyone a basic income, I wonder what amazing art would be created?". Different starting premise.
is it really a benefit? seems to me like it inevitably leads to optimizing for earning money rather than helping people, leading to a lack of focus on solving problems which affect people but aren't profitable to solve (focusing more on a perspective of scientific research rather than art)
The question is not really if you want to be taxed to pay for UBI, but if you rather want to be taxed to pay for UBI than some other bullshit that you are currently being taxed for (of which there is a lot).
I think Qubes OS is a great example. It's a system that besides the security benefits also gives me the control to remove jank from other software (usually software that thinks it should take control of my computer to do what it feels is the right thing)
If you really want to win as a business with OSS, stop writing OSS. Use it, contribute back to it, support existing projects rather than running your own, donate to important development work. Build your business on it. Don't build your business on the idea that nobody should be able to compete with you using it.
There's not a lot of material to directly address here other than to congratulate Zig and its team for trying to chart a path that is both practical and principled. But yet the post touches on a dozen or so issues that one can commiserate with.
One is the notion that future stewards of a project may diverge from the vision of the founders. But this is unsolvable, unless you try to force them (with some combination of blunt and imperfect tools, such as licenses, patents, secrets, shame, money, power) or outlive them (as a chartered foundation, since humans do not live indefinitely). Instead, it's far better to rely on accidental ossification, where your work becomes so widespread and popular that some ancient subset of it becomes commonplace and impractical to fully displace by a future actor, friendly or otherwise. This sounds like sarcasm, but, it's thoroughly difficult to entirely displace C, Bourne shell, vi, grep, zip and deflate, TCP, HTTP... mostly-compatible independent implementations proliferate over time, but many of the design choices of the original creators nonetheless live on.
Another is the realization that for many startups, "open source" (in ironic quotes) is a recruitment/marketing/discoverability strategy rather than a genuine ideological stance. It's a way to get the attention of hip developers, and a way to hook people in a crowded landscape to try before they (or, more properly, their flush-with-cash company) buys, maybe 3-5 years down the road. And I can't fault people for wanting to get paid, but it's blatantly obvious that most of the open source products we all remember from the last 10 years have just been trying to do the plan of "1. give away, 2. ???, 3. PROFIT!". It's hard to feel sorry for them when a competent and well-known vendor comes along and offers a managed service of the open source thing, and they're left holding the bag.
One weird thing about the Oracle vs. Google lawsuit is that it made it obvious that mindshare and familiarity are almost mental pre-requisites of compatibility, even when the achieved compatibility is only partial. Clearly, Android was not the Java Platform, but yet a bazillion Java libraries worked anyway, and a bazillion programmers could pick up Java from a hodgepodge tutorial and crank out apps onto smartphones. In the end, this "programmer compatibility" was as key to Android's success as the fact that if you wanted to milk the gold rush of $1 games with ads, your only other comparable platform was Apple, who required you to own an expensive Mac. If they chose C++ or a brand-new language like (later) Dart, the barrier to entry would have been higher. In the end, everything boils down to "barrier to entry".
If Zig achieves lasting relevance, it's unlikely that the design vision of its founders will be invalidated in the future. But getting to that future is far from easy.
Thank you for sharing, I think you raise some good points.
With regards to the OSS point, I am forced to lash at both Redis Labs and AWS every time I want to say something about Redis to avoid easy exploitation of my words by either party, but I think the problem with open source goes much deeper than what companies like Redis Labs are doing.
AWS is in many ways being a good OSS citizen, and yet it too has devastating effects on a lot of OSS projects and not just because of conflicting business models.
Open source is fine, single corporate entities controlling and owning the copyright is not. That was always a problem and Elastic, Mongo, and Redis are forcing the issue by unilaterally changing the license and doing a bait and switch on their loyal users and contributors. They have the right to do that. That's the problem.
Linux, Mysql, Postgresql, etc. prosper and thrive with many companies collaborating and competing at the same time. These projects survive pretty dramatic changes to the companies backing them including bankruptcy, acquisition, hostile take overs, etc. There are many healthy open source projects like that.
What they have in common is that individual developers rather than a single corporate entity owns the copyright. This makes changing the license impractical. The best you can do is fork the code base and then hope developers continue to work with your code base. IMHO distributed copyright ownership is far more important than the license text because it makes changing that text impractical. So Linux is fine despite the legal pitfalls (perceived or real) associated with GPL2. It's a take it or leave it deal. Most of the industry seems to have taken that deal and is actively contributing and has found ways to bundle proprietary code regardless. So Linux is fine. It's used by mutually hostile companies and even countries. That's the whole point. The OSS community is a neutral place where you can choose to collaborate.
That makes the AWS fork of Elasticsearch interesting because they forked but don't insist on copyright transfers. They are going to do some legal checks to verify who you are and that you are legitimate. That's basically the nuclear option in OSS. It happened to Oracle when they took over Sun. Forking is a dangerous weapon and it tends to only be used when companies mess up the stewardship of their code bases.
That's basically what just happened to Elastic. This means that over time, the ownership of the forked code base will get more distributed and include lots of Elastic, AWS, and (mostly) third party developers.
Elastic employees will likely not contribute (or even be allowed to contribute) to the AWS fork. At least not while they remain employed at Elastic. However, Elastic might see some of their external contributors, former employees, and people from e.g. the Solr and Lucene communities contribute to the AWS fork instead of Elastic's "we own everything" non OSS copy. My guess is that this is exactly what AWS wants to happen. Drive a wedge between their contributors and Elastic. Never mind their motives, I actually think it is the right thing for the Elastic code base. And I say that as somebody who has been a long time fan, occasional contributor (and mildly disgruntled about Elastics recent moves) that has several friends, former colleagues, etc. working for them.
It's an interesting experiment that doesn't cost much to run for AWS. I don't think AWS is going to do much to make a lot of change happen but I'm sure they are happy to accept pull requests. If that start happening in more volume, the right move might be to eventually move control to the Apache Foundation. Amazon wins, users win, developers win, oss wins, Elastic & its share holders lose. Might be unfair but I prefer using open source and not betting my career on proprietary code bases. If successful, there are other companies that need to worry about losing control in a similar way.
had me in the first half ngl.
then i was really taken aback when he just went on to totally dismiss FSF by saying it has failed so utterly and completely because...
I'm sad they can't learn to love GNU projects. I for one do. They just... work. Whenever they don't, invariably there's some freedom-hostile hardware in the way. This makes me want to move closer to FSF-recommended hardware.
I wish them all the luck because the FSF just seems cursed at this point. They do absolutely essential legal and advocacy work and the world wouldn't be what it is without them, and virtually nobody has a deep enough sense of history to recognize that. They will still get my donation dollars for the foreseeable future, and I hope they can rehabilitate their image.
I do sorta think this whole thing will end up working out well overall for the FSF's mission of building a software domain that's solidly out of the clutches of proprietary predation. I've never seen the FSF, the distinction between open and free software, in the public mind more than the last year. I can get really snarky about how all these people who never cared before suddenly come out against the FSF. You didn't care before, you still don't care, good job expressing the total lack of impact you insist on wanting to have.
Are more free software foundations a good thing? Probably. It'll dilute the impact in the short run, these new projects will have to spend time and effort solving run-of-the-mill nonprofit stuff that the FSF worked out decades ago, but it also means more people will get invested. Eventually the new blood will learn to work with the old guard and we can all build a new order together.
Color me optimistic about the future of free software.