upvote
The corporate environments were here too, most companies used to run on Windows server. 20 years ago companies used to pretend they didnt use Linux, but they were, it was just introduced to places they didnt know about, as it was free so it didnt have to go through purchasing. The rise of the early web in the post dotcom years was the catalyst, Perl, PHP, Linux servers etc. Before mobile, that did bring back proprietary development to some extent, for clientside. That was the era when Microsoft said "Linux was a cancer". Many companies still have large Windows (dot Net pre dotnet core) codebases, but Java mostly runs in Linux now.

The language barrier is interesting, there is more Chinese open source now too, but yes so much is English. I remember using google translate for Nginx from Russian back in the day, and openresty from Chinese, but yes we are lucky,

reply
Gitee in China is certainly robust, but when you think about the sponsorship system, it's closer to an incubation environment for corporate ecosystems. There are a lot of public codes intended for national projects or large enterprise collaborations, so it's actually good for grasping Chinese tech trends. I sometimes find it fascinating how free the Western GitHub system can be.

It really makes me realize just how different cultures and values can be.

Sometimes I feel like I want to be as free as you all are, but I also recognize that my own biases are deeply ingrained. There's a line in Demian that goes, 'The bird fights its way out of the egg.' It makes me strongly feel just how narrow my world really is

reply
It is a distortion to frame the problem as corporate vs. open source. These corporations compete with open source but they are often sustained by it, and in any case they operate within a space that is impacted by it. A healthy open source community is generally to their benefit. The inverse is also true, to the extent that corporations support and integrate with open source, it benefits from a healthy commercial market. All too often, they take and do not give back, so many people in these comments here are pointing out the same contradiction that you have highlighted. But it is not so much a matter of predators and prey, we all share the same ecosystem.

The corporations themselves are not a monolith. Their leadership and engineering teams are made up of diverse perspectives on open source, and those perspectives can shift. These same questions are debated within the company and the balance is always shifting in a way that can either benefit or undermine open source. I'm personally skeptical for some of the reasons you described, but I wouldn't rule out the possibility of a better relationship to open source.

reply
That's a Rihgt However, what I'm curious about is that this project governance feels more closed than open source, rather than truly being open source. Your point is valid too. I admit my thinking might be a bit too binary.
reply
I think this is conflating opensource-corporate and english-non-english.

If you ask american/european/english-speaking developers about coding, it will mostly be about/in the context of corporate environments rather than open source too! The majority do not actively or primarily contribute to open source projects, but instead corporate environments as well.

In an alternative timeline where the lingua franca isn't english, I can still see open source culture exist; I don't think the desire to publish and cooperate in public is an inherently "western" culture. It will also run into the same conflict of interest between Open-Source and Corporate: one prefers transparency and full-disclosure, the other prefers control in the interest of minimizing risk.

reply
You're right. Ultimately, the absolute number of developers matters a lot. But when it comes to coding style and paradigms, it's overwhelmingly dominated by the English-speaking world.

For example, object-oriented programming or conditional statements generally follow a What -> Action -> Target order. But my native grammar follows a What -> Target -> Action order. So I have to translate SOV logic into SVO code syntax.

The reality is that English speakers are numerous, Japanese and Korean speakers are relatively few, and while Chinese speakers are quite numerous, there's still some cognitive load due to differences in thought patterns. It's almost like a difference in the sheer volume of accessible knowledge.

Due to the cumulative cost of translation, this feels like a bigger hurdle than people realize.

So sometimes language gives a sense of identity tied to 'ethnicity' and 'nation,' but when it comes to the competition for knowledge, I feel that the number of native speakers matters more. There are points where I agree with you, and points where I don't. It's complicated

reply
The Personal Computer culture in the SV developed from the counterculture in the valley: https://kbsm.org/technology/california-tech-culture-that-sha... It then hopped on the Free Software movement with redefining it as "open source" while the Internet was booming. And all of that is now being reaped by the corporations the "tech-hippies" themselves helped to create.
reply
If anyone's interested in this: "From Counterculture to Cyberculture" by Fred Turner and "What the Dormouse Said" by John Markoff.

Putting my nostalgia-tinted glasses on, it's sad how far we've strayed from that.

reply
Yeah, I had Fred Turner in mind when posting. A bit dry, but comprehensive.
reply
thanks, i will read it
reply
Great comment. Really interesting to see "scratch your own itch" described as an "aristocratic hobby".

If the language barrier disappeared overnight, would the situation still be the same, do you think? What would an Eastern open source movement look like, and why hasn't one developed?

reply
hmmm I'm not sure about Japan on this point, since I haven't communicated with Japanese developers very frequently.

But regarding Korea and China: in China, there's Gitee, which has a very robust open source environment, but it's not really 'Western style open source' it's more like corporate projects being made publicly available for free. In other words, companies release assignments and people gather to work on them. That's the dominant model. (And that becomes part of their employment portfolio. So it feels very much like an incubation system for corporate projects.)

For Korea, I think it's largely because the absolute number of Korean speakers is smaller than English speakers. As a result, Korea's tech infrastructure generally lags behind the English speaking world. It feels like: English trends emerge -> a few years later, once they stabilize, Korea starts adopting them!

The usual pattern here is that the people curating these English trends for Korea are Koreans who have worked at FAANG-like companies and come back, so they have a strong influence. But I don't necessarily agree with their perspectives, which is why I came here to see what the raw data from the West actually looks like.

On top of that, Korea's IT projects are mostly government-led (because the domestic market isn't that large), so the government essentially acts as a VC. And within this government-led incubation system, only the final winner takes everything. Given that kind of environment, I wonder if that's why open source doesn't really take off.

reply
[dead]
reply