I’ve used VS Code for a long time, but have recently grown weary of Microsoft’s approach to OSS. I’ve checked out VS Codium which seems like it might be a great option.
What text editor are you using?
I’ve used VS Code for a long time, but have recently grown weary of Microsoft’s approach to OSS. I’ve checked out VS Codium which seems like it might be a great option.
What text editor are you using?
VScodium to VScode is as Chromium is to Chrome
Yeah, I know that. But much like with Chromium, it’s only technically open-source and you’re still eating whatever changes Microsoft/Google decide to include.
So, if it’s that what they mean with “Microsoft’s approach to OSS”, then VSCodium is no different.
Not sure I get the nuance, isn’t one always eating whatever changes the main developers decide to include?
Yeah, admittedly there is alot of nuance here.
The thing with VSCodium or Chromium is that the bulk of the development work comes from Microsoft/Google, and they provide additional infrastructure around these projects, built up a brand etc…
This means that these projects practically cannot be hard-forked (taking them into a different direction), you can at best soft-fork them to remove the worst of the worst (like e.g. ungoogled-chromium does), and such a fork would likely not gather many users either, even if it’s objectively better.
That gives quite a lot of power to Microsoft/Google. For example, Chromium with its extension API ManifestV3, that wouldn’t even be up for discussion, if that was a community-lead project. The repo of the maintainer would get abandoned and everyone would contribute to a fork instead. With actual Chromium, though, not a chance of that happening.
I don’t understand the implication, what is it that makes a hard-fork impossible? In fact, isn’t Brave a hard fork of Chromium?
In my understanding, while the freedom of forking the project is certainly determinant in the question of whether it is open source, I don’t see any relevance in the one of creating a fork that can get popular enough to strip the original project of its users.
Well, my definition of a hard fork is that you take the state of a project at a certain point in time and then you largely carry on development on your own. For example, LibreOffice was a successful hard fork from OpenOffice.
So, I don’t see Brave as a hard fork. They are very much dependent on Google continuing to open-source the Chromium code. And they don’t really have the capacity to make larger changes to the code base, or even just maintain the status quo, if Google decides to make changes that go against Brave’s interests.
Yeah, that’s why I wrote that they are technically open-source. They fulfill the official open-source definition, but they don’t match up with the subjective expectations that people often have for open-source.
So, for example, if you think of VLC, GIMP, KDE or other community-driven projects, they may be shit in one way or another, but they would never make a change with which the majority of the user base disagrees.
That’s what I personally think of when I hear “open-source” (although I have started calling it “community-driven”, to disambiguate it from those shitty open-source projects).
Ok, thanks for clarifying, I definitely don’t know enough about Brave’s developement to comment on that. Do you mean that Brave’s team wouldn’t even have the manpower to mantain security updates if they want to harder-fork? I still don’t really understand what is the difference between Chromium and say KDE about the possibility to hard-fork.
I see, thanks again. Indeed “community-driven” is a better fit label for that state of affair !
I mean, I don’t have deep insights into Brave’s development either, but they only have around 150 employees (according to https://craft.co/brave-software ).
I doubt there’s numbers out there for how many Google employees work on Chrome, but the Mozilla Corporation has around 750, which they only just cut down from 1000 ( https://www.theverge.com/2020/8/11/21363424/mozilla-layoffs-quarter-staff-250-people-new-revenue-focus ).
As far as I’m aware, Brave doesn’t really tweak much about the rendering engine (Blink) either, which is probably the biggest chunk of code in Chromium. So, they would need to build up a lot of expertise.
And finally, well, long-term security may also require feature development. Chrome and Firefox both have developed forms of sandboxing. Mozilla published a new way of doing sandboxing via WASM not too long ago. So, yeah, you kind of need to continue innovating, which requires manpower.
As for forking Chromium versus a piece of KDE software, the difference is that with Chromium most of the development work comes from one entity, Google. You would pretty much need to poach these employees, or build up a huge team of your own, to be able to move the project in a different direction.
With a piece of KDE software, there’s usually multiple entities involved, i.e. multiple independent people. So, the core maintainers are not the only people who do development, or you may even win over some of the core maintainers for a fork.
And even if that is not the case, a handful of frustrated community members can probably outperform the current maintainers.
You just have a lot less development manpower tied to the maintainers.
LibreOffice might again be a better example. OpenOffice was maintained by Sun Microsystems and they provided a bunch of infrastructure and held the brand name, but most of the development work came from the wider open-source community.
So, when Oracle bought Sun Microsystems, those developers organized, created The Document Foundation, and set up their own infrastructure. So, the maintainer changed, but the developers largely stayed the same.
Now, this seems to be more about the manpower needed to mantain a browser that about the open-source or free (or in fact even community-driven) natures of Chromium. Is there anything that makes Firefox more opensource than Chromium?
Again, even departing from the strict opensource definition, I think that how much a project is free come from the possibility to have a fork that works, not necessarily one that can compete with the original project. Is it necessarily a good thing if the development team of a projects can be outperformed by “a handful of frustrated community members”?
In fact, is your point that it would be better if there were no professional developers and only projects run by communities of hobbyists?
I think VScodium is more like ungoogled-chromium than the actual chromium