Hacker News new | past | comments | ask | show | jobs | submit login
Significant performance and correctness improvements to the kernel (redox-os.org)
126 points by akyuu 7 months ago | hide | past | favorite | 22 comments



Has anyone ever done an architectural comparison between Redox and Fuschia? That would be quite interesting.


How usable is Redox as a daily driver? Is it sorta-kinda works territory or is it possible to run a significant number of programs without hoops? Firefox? VSCode? Flatpaks?

Can I run a VM inside Redox to fill in the functionality gaps?


From the FAQ[1], it sounds like neither X11 nor Wayland are currently supported.

[1]: https://www.redox-os.org/faq/#what-programs-can-redox-run


But they have their own windowing system, yes?



Why is it better? Asking an honest question.



Spoiler alert: people in the FLOSS space have different opinions on copyleft vs permissive licenses.


That's a real blast from the past. We're now in the post-open-source era where people don't even look at licenses (as long as they're OSI approved).


What lawyers are you talking to, because mine still care :)


Our legal department definitely cares.


From that thread:

> The choice of MIT is not an arbitrary one

...

> We wanted to encourage the use, modification, and packaging of Redox in absolutely all realms. Open source should be open, for everyone. There’s absolutely no reason for limiting the usage of the software. Therefore, MIT was the license of choice.

...

> We like to have a decentralized structure of the project, allowing people to do whatever they want, no matter how they intend to share it.

Based on their goals, the answer seems to be yes.


> The GPL is good for forcing people who make changes to the source to contribute back. [...] The GPL is upstream-centric, the MIT license is downstream-centric.

An interesting take. It doesn't match my understanding of the GPL though. AIUI, the GPL does not require that you send your changes "back"/"upstream". However, it does require that you make the source code to your changes available downstream, to the people you distribute the code to.

Conversely, MIT does not require that you make your changes open and available to downstream users.

> Open source should be open, for everyone.

...for everyone, except for users of modified versions of the project, apparently?

Look, it's their project. If they want to distribute it under MIT, that's their right. I just get a bit frustrated when the rationales for using permissive licenses don't seem to make a whole lot of sense.


> it [GPL] does require that you make the source code to your changes available downstream, to the people you distribute the code to

Many corporations don't distribute GPL derived software to users, but "package" it as a web service instead. That way they are not required to make their changes open and available to users.

> [> Open source should be open, for everyone] ...for everyone, except for users of modified versions of the project, apparently?

If modifications provide enough added value to users (so they use less open modified version instead of the open source product), shouldn't the developers be able to freely decide which business model they want to utilize? Not everything can be made sustainable using the GPL license.


> Many corporations don't distribute GPL derived software to users, but "package" it as a web service instead.

Then what benefits does the MIT license give in that case?

> Not everything can be made sustainable using the GPL license.

That's a fair argument to make. But it's not the one that was being made upthread.


> Then what benefits does the MIT license give in that case?

It allows development of commercial products where software runs locally, where users don't have to share their data with the corporation.

There are always some tradeoffs when choosing a license.


GPL doesn't requrie you to release all your changes publically. It requires you to offer source code to anyone you've given a binary to


> > > Many corporations don't distribute GPL derived software to users, but "package" it as a web service instead.

> > Then what benefits does the MIT license give in that case?

> It allows development of commercial products where software runs locally,

Those are not the same case!

FFS. Stop moving the gorram goalposts.


Explain how it is not the same case? There seems to be some misunderstanding.

If dependency is licensed under GPL, a commercial project cannot be distributed to users (to run locally), thus it is always offered as a web service instead.

If dependency is licensed under MIT, a commercial project can be distributed to users and run locally.


Oh, got it.

Sorry, when you originally wrote "Many corporations don't distribute GPL derived software to users, but "package" it as a web service instead" I didn't get the connection that they were packaging it as a web service because they were deriving from GPL'd base.

I was coming from the position that companies offer their products as web services, rather than distributing the program itself, because it allows them to lock their customers in, and track what their customers are doing, and hold their customers data hostage, and basically do all the shitty stuff that web services companies do. And some of them happen to base these services on GPL'd code, because the GPL doesn't prevent that - just like MIT.

It didn't occur to me that a company would write a program that they wanted to sell as a redistributable entity, but be so dependent on a particular GPL'd library that they could not find a commercial equivalent of, or rewrite the essential parts they needed themselves, that they'd change their whole sales/availability strategy and reluctantly make the program a web service instead.

Hence my misunderstanding. My bad.


Even if they persist in wanting a permissive license, I would still really insist they at least use the Apache 2.0 license, which has a patent grant.


Superior ( in my humble opinion )




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

Search: