Hacker News new | past | comments | ask | show | jobs | submit login
Facebook Joins the Rust Foundation (facebook.com)
180 points by wongmjane 9 days ago | hide | past | favorite | 94 comments





This is a great step for the Rust Foundation and a sign that Rust is transitioning successfully to a foundation model rather than a Mozilla-centric organization.

It looks like some people are worried about Facebook's influence on Rust, but that doesn't seem like a huge concern to me. Facebook isn't going to want to influence Rust in any particularly "Facebooky" way since it's such a deep-in-the-infrastructure component. They will just have the same concerns that many other large tech companies have, like, can you use it successfully in an environment that has a zillion engineers and a zillion other programming languages to interoperate with. It's important for Rust to be nice to use at both big and small companies so that's a good sort of voice to have in the foundation.


It does bring up something we don't always like to think about, which is that by contributing to the progression of very general technologies we often empower people/organizations who are doing things that may be bad for the world.

But I'm not sure what there is to be done about that, and it isn't limited to Rust (or Facebook for that matter), and I don't think this event really makes that state of affairs worse in any particular way. It just remains an uncomfortable and (seemingly) unavoidable fact of our industry.


It is hard to imagine progress in any technology that couldn’t also be used for evil. Better food production could feed bad guys. Curing cancer could extend the life of a mass murderer.

I don’t think we will stop evil corporations by preventing them from using memory safe and performant programming languages.


The other post just raised it as something to think about. It's a good idea to do a sanity check once in a while.

More technology has generally been better, but not in all cases and it's not guaranteed to continue that way.

Bitcoin is cool, but the consequence is using a lot of energy and facilitating a lot of extortion. Open source seems great, but it also facilitated the rise of AWS which effectively locks people out of software more than closed source software ever did.

We already know the rationalizations, we tell themselves to ourselves every day. But remembering the costs is healthy.


I think the more appropriate analogy is not bitcoin, but underlying crypto technologies. Should stop researching crypto (or should we repudiate past crypto research) because it might result in new mining technologies? I think the answer to that is clearly no. But that's precisely the issue with these very general technologies: they enable a broad set of use cases that includes things you might not have thought about.

Roads enable bank robbers to get away from the crime scene but nobody thinks that's a reason not to have roads.

I don't think Rust has been a "Mozilla-centric" organization in a long time. Or at least the Foundation has no bearing on it.

Very early on Rust was made independent of Mozilla. And since the layoffs it's hard to argue they had much indirect influence left even before the Foundation was started. Well, no more than other companies using Rust.


I would be a bit more enthusiastic about it if the foundations wasn't the closed black box it seems to be at the moment. Either they are yet to have a board meeting or they have not published the minutes, both possibilities are not great.

Why take money, and allow influence from a company with a history of so many bad actions. It is all rather shameful that this is celebrated and IMO no different than if they would have welcomed Mindgeek, Parler or Gab to the table.

This will get downvoted into oblivion. But complaining (over and over) about how certain companies do "terrible things", or how RMS shouldn't be allowed on the FSF board, or Github/ICE, or Amazon predatory behavior against employees wanting to unionize, but then _not_ call this out and pretend it's fine ... is peak hypocrisy.


I agree, this is a sign of being highly relevant. The more companies that get involved, the better. The Rust foundation looks like it is set up for success here. Very similar to e.g. the Eclipse, Apache, and Linux foundations have been for ages. Mozilla did this right.

Basically Google, MS, Facebook, and probably loads of others are using Rust in their mutual tech stacks. They have a shared interest. none of them get to dominate. The Rust foundation is the neutral ground where they collaborate. Schoolbook use of open source as a means to collaborate between frenemies.


So Microsoft, Amazon, and Google being founding platinum members has little bearing on Rust transitioning to a foundation model which is not a Mozilla-centric organization?

Well those are great too. Every donation helps the Rust Foundation do its work, it's as simple as that.

The only thing that has any bearing on why Rust transitioned to a non-Mozilla-centric organization is the fact that Mozilla cut the team to focus on Firefox.

> to focus on Firefox.

Finally? It took them almost two decades!


Statement from the Rust Foundation: https://foundation.rust-lang.org/posts/2021-04-29-membership...

It seems there are a few new silver members as well.


The Facebook Member director's letter linked there is also very interesting:

https://foundation.rust-lang.org/posts/2021-04-29-introducin...


Thank you! And the FDISK story is absolutely true. I can still remember my father's surprise at how reckless I was with his computer.

Awesome. I've been keeping an eye on Zama AI, and in particular their Concrete[0] library. Glad to see they're a member now.

[0] https://github.com/zama-ai/concrete/


Can anyone speak to what power this gives them in the rust foundation, other than just good will and the dependence on funding?

The two things I generally think about with regards to sponsor influence on software foundations are:

1. If the sponsor is a software vendor, does the funding enable them to secure a marketplace advantage over competing vendors? For example do they gain a majority on a steering committee, or fund a critical mass of foundation-employed developers?

2. Does the funding enable the sponsor to unduly influence an official or de facto standard which impacts a marketplace outside of software? (For example, web protocols: see https://httpd.apache.org/ABOUT_APACHE.html#why-free )

I think you've properly identified "good will and the dependence on funding" as sources of potential undue influence. The structure of the foundation is supposed to guard against that. But more to the point, it's hard to imagine how Facebook would be able to leverage undue influence over the Rust Foundation: they aren't competing against Rust vendors, and it's unclear what public standards could be subverted via the Rust Foundation in a way which could benefit Facebook.


In the past, Rust release announcements have come with statements like this:

« The Rust Core Team believes that tech is and always will be political, and we encourage everyone take the time today to learn about [US-centered political issue of the day] »

I suppose this makes it less likely that Facebook finds themselves being the issue of the day.


Too much. It goads people in the Rust community to pretend to think Facebook is well-meaning company. I'm going to continue thinking Facebook doesn't care about people, even when they're being killed by the thousands, and I'm more comfortable doing that as someone who uses Vue and Python and not Rust.

https://www.bbc.com/news/blogs-trending-45449938

Edit: I realized I'm slowly getting into Deno, which is based on Rust, so I should probably just lean into Rust.


Facebook contributes a ton to the Linux and Python communities as well, by this logic you shouldn’t use either.

Thinking that infrastructure engineers have direct influence over the product is wrong. Instead they work at Facebook because it is a good means to being paid lots of money to contribute to open source.

Btw, getting out of us vs. them mentality will form a healthier community and result in more positive changes. You can call Facebook on shit without generalizing against the employees who work there.


> You can call Facebook on shit without generalizing against the employees who work there.

I disagree. It's this behavior and lack of disinfectant (sunlight) that allows companies to get away with the amount of crimes against humanity they do. Saying "Tech" isn't political, or "I'm just an engineer somebody else is in charge" is how horror happens.


I never said "Tech" isn't political, or that "I'm just an engineer somebody else is in charge". Both are true and we best take time to think of the moral implications of our work.

What I am saying, is that you should not generalize people just because they worked or currently work at a big tech company. This is as much a cognitive bias as thinking a company like Facebook only does good.

You have no clue why they work there, what they work on, or who they are.


Facebook is one of the sponsors of the Python Software Foundation too. https://www.python.org/psf/sponsorship/sponsors/

Somehow they've managed to stay out of the news, while Rust has been endorsing BigCo's regularly for a while now...

It’s more the other way around — big cos are endorsing Rust and joining the nascent foundation. It’s perfectly reasonable for a three month old foundation for a popular new language to be in the news quite a bit, and for its sponsors to enjoy the attention that goes with it

The Rust Foundation is really new, it just started in February. That's probably why you've been hearing a lot of news about corporate Rust sponsorships recently, or if not "a lot" then more than you've heard about, say, corporate Python sponsorships.

What does it mean for Rust to be endorsing BigCos?

Not much other than reflecting the reality that BigCos are using rust in production, and as a result, have a vested interest in the project's governance. This is true for pretty much every popular language of software with an open governance model.

It means any Rust dev will be even less likely to open their mouth when Facebook does some horrible shit than what they already are.

>It goads people in the Rust community to pretend to think Facebook is well-meaning company.

As someone who already views the Facebook brand as a toxic waste dump this won't do anything to change my mind.

If they want to sponsor Rust more power to them. It's not as if it's going to change anything for the average user.


https://www.python.org/sponsors/application/

Meet with the PSF Board of Directors Opportunity to meet with the PSF Board of Directors to discuss a topic of your choosing. The health and sustainability of our community is important to many organizations and we want to make sure community needs are heard and noted.

For visionary sponsors:

Meet with the Python Steering Council Opportunity to meet with the Steering Council to discuss technical aspects of Python or feedback you may have, or just a Q&A with the Steering Council. The health and sustainability of Python is critical to many organizations and we want to make sure community needs are heard and noted.

https://www.python.org/psf/sponsorship/sponsors/


I'm sure the "visionary" sponsors represent "community" needs perfectly! Translation:

"Buy yourself into an obedient hiring pool controlled by the Steering Council."


I'm sad to announce that I fucking called it on the day of the Rust foundation's formation [0].

Say what you want, but the problem here is that you don't need to give board seats to those who have tons of cash, rather instead have them sponsor it, like how the R foundation has done.

Facebook essentially bought a board seat to drive and prioritise which features they want in the development of Rust.

The wolf eats the same grass as the sheep and gets praise for it.

[0] https://news.ycombinator.com/item?id=26124316


> to drive and prioritise which features they want in the development of Rust.

The foundation is not involved in feature development, so this doesn't make a ton of sense.


what does a platinum seat get you then?

it doesn't make sense to spend a ton of money on a seat and not do anything. I am sure facebook has other motives than paying engineers to work on rust.

a regular sponsorship would have a been more enough in this case.


To be clear, I am not a part of the foundation. You can read about it yourself on https://foundation.rust-lang.org/ (with the bylaws being at https://foundation.rust-lang.org/policies/bylaws/ )

The foundation's stated mission is to steward the intellectual property of the Rust project, and assist it and its maintainers. That often means things like "paying for legal support" and "paying for CI" and such.

The Rust project is the thing that's existed for years, and makes all of the decisions about what happens to Rust technically. The foundation has no position in that hierarchy, and sponsoring the foundation does not get you a seat on the teams that accept RFCs.

> a regular sponsorship would have a been more enough in this case.

I'm not sure what distinction you're drawing here between "regular" and not. Maybe that they would only want a lower tier sponsorship? I don't know, you'd have to ask them.


The most important news to me, Patrick Walton is now working in Facebook?

Seems lots of people left Mozilla.


Mozilla laid off most of the people working on Rust last year, so there is a bit of a diaspora.

Does anyone know what the financial requirements are to be a platinum or silver level member of the Rust Foundation? I clicked around on their website but couldn't find the answer.

Rough guess: In the inauguration blog post[0], a yearly budget of $1mil was mentioned, when there were 5 founding members (so $200k/year/company). I am not sure if the distinction they make on the "Members" page between "Founding Platinum" and "Platinum" has any practical implications, but I guess that new members might have to pay a premium. So possibly +$200k/year/company for Platinum level.

[0] https://foundation.rust-lang.org/posts/2021-02-08-hello-worl...


So that would be just enough to pay the bureaucracy, like in other foundations.

It seems they just want some influence and use it as a hiring bait.

After having been hired, you'll be asked to work on a legacy C++ code base.


Days without a Rust headline on HN : 0. I almost hate Rust just because it seems like it's shoved down my throat.

Well, that makes me a tad less enthusiastic about Rust.

The kinds of things that Facebook most often receives flak for would be difficult to pursue though the layer of indirection that is sponsoring a foundation dedicated to the Rust programming language.

Similarly, since Facebook isn't a software vendor, the kinds of things that you might achieve by securing undue influence within the Rust Foundation don't really benefit them.


But what's in it for Facebook then? They're spending real money, there's got to be something for them to benefit from?

FB are large enough that they do a ton of their own very low level work - up to and including things like their own bespoke datacenter switches and SDN platform, stuff which is almost wholly distinct from what their core product is. (https://engineering.fb.com/2019/03/14/data-center-engineerin...)

It seems entirely possible that, even putting aside the PR, they're concerned about making sure a robust modern systems language stays around/thvives so that they have a robust modern systems language to use.


Ok, fair enough.

There are a lot of teams within Facebook that use Rust, and a lot of things that could be improved with Rust that would benefit Facebook as well as everyone else.

(I work at Facebook on a team that uses Rust)


Someone correct me if I'm wrong, but based on the Rust Foundation website, it looks like the point of the organization is mainly to allocate funds to pay people to do work relating to the Rust language and community. So it seems like paying for a board seat means you get to indirectly influence the direction of Rust by deciding which parts people get paid to work on, and which parts are stuck with unpaid volunteers. I don't necessarily see anything wrong with that, it's good to see people get paid and the people paying should have a say in it, but it seems different from people claiming that paying for a board seat doesn't give them any additional influence over Rust.

Making sure Rust works well is good for a company that uses a lot of Rust. It's pretty much as simple as that.

I think it's help with their image.

faster ads compiled from rust to webassembly. it's heading in that direction.

Facebook has been contributing to the Rust ecosystem for a long time; they've been sponsoring Rust conferences for the past few years, they've had employees helping do work as part of the teams.

EDIT: oh, they've also published another post detailing their history with Rust. Lots of very interesting things in here! https://engineering.fb.com/2021/04/29/developer-tools/rust/


Why? On the tech side, engineers at facebook have contributed some great open source work.

Just like how every environmentalist love the oil and gas companies because they are offsetting their carbon emissions

How does this comparison make any sense?

I'd definitely prefer oil and gas companies that offset carbon emissions over oil and gas companies that don't. (Not going into how much pr that is over how much it's actually worth anything)


Why? Facebook engineering is highly regarded.

Perhaps in capability, not so much by reputation or social utility.

If the Rust foundation only took money from morally upright tech companies, it would be bankrupt.

I agree, I was only meaning to add context to that other comment.

True but if someone would pay you 200k to develop rust things wouldn’t you say yes?

Sure!

I was just adding context to the other comment with my own. I've no qualms with whomever spending money on whatever language they choose.

Unless it's VB. Yuck.


I would not say yes. You can't buy ethics.

It makes me the opposite. I hate the idea of Facebook and the damage it's fine, and I don't have an account. That being said Facebook has done some absolutely amazing open source projects (rocksdb, open hardware) and papers detailing the exact technology they use (seaweedfs is based on a research paper they published) so I'm sure a lot of cool tech will come from their adoption

It had to happen, the only way to make a programming language successful is to have it backed by most big tech companies.

They can contribute without having any special status. Rust already has a robust RFC process and good technical leadership. My worry is that these things weaken and rust very gradually becomes your typical corporate software product where more and more independent features are just bolted on without care to any larger vision. That won’t happen if FB, etc keep contributing as ordinary contributors. But it could happen if they’re elevated to a position of leadership.

Note that the Foundation doesn't have any impact on the RFC process or the project's structure or governance, so while that worry may exist, this announcement isn't a path towards doing that.

Python, C, C++, Perl, Bash were pretty successful before the modern FAANG culture.

Same for projects other than programming languages, e.g. Linux.

I don't feel comfortable with corporate-driven FOSS. I'll stick with community languages.


As for C, sure, AT&T is so much better than FAANG. C++ was also heavyly corporate. Python won over Ruby because of FAANG support (I personally like Ruby more, but as it's not supported by big companies, it remained behind in deep learning for example). Also NVIDIA has great Python support, which is really important for high performance software development.

As for Perl, I just look at it as a stepping stone for more modern scripting languages.


A lot of companies also contributed to Linux, but that's not the issue.

The concern comes from having a *small* number of companies with a notoriously aggressive behavior controlling a project.

Google is especially known for using products as weapons against competitors and drop them suddenly after a while.


Sure tying Rust development cycle to Azure+VS code, AWS or GCP would be quite bad (run rust-analyzer on Azure for example). At least Facebook doesn't have this kind of infrastructure yet.

I wouldn’t say so. A language like Haskell is successful, even if their motto is “avoid success at all costs.” Books have been written about the language, and the ecosystem and community are thriving. Even still, few corporate sponsors exist.

This is what adoption and maturity looks like. As the language starts to see serious use, you see companies start to consider the language's future and direction.

Why? Facebook is not the sole large corporation in the foundation. Microsoft, Google, and Amazon are all founding platinum members.

Starting with the next version; when you compile, you will see a facebook ad.

"rustc" is the facebook ad. They've linked themselves to the name.

I doubt Zuckerberg will be personally involved. It will be all engineers.

Exactly right. After all engineers are just working for salary, eating free meals, cashing their RSUs. They have no clue or hand in how Facebook makes money today. Moreover they promise to learn about FB's business and write a strongly worded letter on "Why I left Facebook" when leaving after vesting most of company stocks

Rust should include in their license: "The Software/Language shall be used for Good, not Evil"

https://en.wikipedia.org/wiki/Douglas_Crockford#%22Good,_not...


Who decides what’s good and what’s evil?

Real life isn’t kindergarten, world isn’t black and white. Even for things that most of people can agree are good, you’ll have negative externalities.


It was just a suggestion without the details carved out.

You could make it very explicit: X shall not be used for advertising purposes. In legalese, of course, so a judge knows what is meant.


Define advertising purpose. Rust affiliated organizations are advertising (for example conferences are being advertised).

Is only serving ads or also using ads not allowed?

Given that almost every single company in the world is using advertising then it becomes a hobbyist tool.

With serving - how about if I have a page/app that includes ads served by third party (like most of the world does). Does it make it also off limits?

Statements like that sound pretty cool, but ignore real world complexity.


This is why you ask a lawyer to write your license.

Lawyer won't be able to make those basic decisions for you. They'll be able to help write it in a way that has chance of being enforceable, but if you don't know where you want to draw the line, they also won't

Why? What effect would this have, in your opinion?

Also, who decides what is "good", and what is "evil"? For proper licensing terms, a judge can decide whether those terms have been violated within a given jurisdiction. Would you have philosophers take their place? What a bunch of nonsense.


> Also, who decides what is "good", and what is "evil"?

Huh, Rust developers have plenty of opinions about "good" and "evil" on everything under sun. They peddle it non-stop on social media. So they seem perfectly capable to decide on that.


The Rust developers don't have the power to enforce a license – that's up to courts.

They have power to create license. How about doing that much.

Core team member here. There's no desire to change the license.

Also, practically, we do not have a CLA, and have had just under six thousand distinct contributors. We cannot unilaterally change the license. This is a feature, not a bug.


That would make Rust no longer Open Source according to the OSI definition, and incompatible with most other OSS.

Since that line could be interpreted as just "The Software/Language shall be used for Good", any software that is not explicitly "Good" would be forbidden to use Rust, effectively making Rust useless in 90% of companies.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: