why do companies waste resources developing programming languages

as the title says. what does google gain from spending $$$ for Go, or Oracle for Java or Microsoft for C#? seems like they're wasting resources and cucking out for other people, that's antithetical to how a company is supposed to act you could argue that they use them internally, but even then it doesnt seem worth the investment to me, especially when you consider that they could hand out the project to the open source community what am i missing?

Spyware post to short...

oop is like kicking dead whales down the beach and capital reigns supreme. we live in a society basically

>>38959 >hand out the project to the open source community If they did that the result would be an open source compiler - they don't want that. They want ever increasing amounts of IP they can control.

>google They've got a soykafton of money already. They support a lot of random project in the hope that one of them can offset the others' costs >oracle (and sun before that) They sell support to business. It isn't for nothing that java has a reputation of being for corporate bloatware >microsoft They hope to lock people into their ecosystem. It's more difficult to port software to another platform if there is only one compiler for your language >>38963 Aren't the go and java compilers open source (even if not fsf-approved)?

>>38964 this pretty much C# is microsoft wanting java without using java since that's competition and cross platform which is evil! So you make your own java+c zombie c# and tie it to windows. Go could be a subset C standard but nobody would care then - talk about what should be used is dime a dozen. So rebrand under a new name and people get hyped. Its like yolo replaced carpe diem - same thing but new=good! Oracles java model was having corp pay up while running it free for public use so the kids get into java and use it in corporations later. Same model is employed by photoshop & other corps when it comes to piracy - they dont curb it cause they know if they allow using ps you wont learn free competing products as a kid.

38964 >Aren't the go and java compilers open source (even if not fsf-approved)? OpenJDK is GPL licensed. How is that not FSF approved?

To establish market dominance and presence obviously. Sometimes it saves money on advertisement, is a good recruitment tool, can help them with their own development since a lot of open source works with big companies to get it developed.

Spreading out if something goes wrong Just imagine their main market getting attacked by a new technology and new competitiors

>>38976 Came here to say this as well The same reason why coke pays absurd amounts of money in advertising when they don't even need that much exposure To give an impression of brand dominance or whatever it's called and/or to try to impose it It's not a profit motivation directly, yes it looks good in the eyes of investors but it also enables them to have power over the ecosystem and communities

>>38959 Just as other lainanons said above, control is the primary purpose. Control over development process. Control over pipeline for newcomers (as >>38971 explained). Control over the community and the ability to bar undesirable elements from said community (everything is political now, so I think, in recent years this aspect of control is the most valuable to bigtech, with Rust being a terminal example of this). Meanwhile, if you want to do some actually useful programming, I don't see why you would need anything beyond C99/C++11 (and maybe IEC languages for industrial PLC programming).

>>38959 Java, Go and C# are not ISO languages, meaning that in spite of their seemingly open source nature, their specification and design trajectory is heavily influenced by the corporations behind them. There is a somewhat significant advantage over having monopolistic influence on a language, proposals are valued according to their practical utility in corporate applications, and if the language gets enough popular, you may even benefit from strategic power over rival companies. > especially when you consider that they could hand out the project to the open source community they have no interest nor incentive to be reliant on the opensource community

Doesn't every company want their own language that they can have complete cobtrol over? I think that ideas like this come naturally once people are annoyed that existing tools don't fit their taste perfectly and even if such a tool is open source, implementing a features and waiting for a positive review and a merge, that may never come, is hell. Doesn't every company that has some bucks spend a fraction on it for quality of life tools for their devs?

I may be talking out of my ass but I vaguely remember these facts. Java was created when there was a need for a cross-platform solution for deploying apps without the impractical hassle of recompiling for everything. They Sold their services for their free*** language (and other products like DB solutions) to companies and individuals. C# probably as a direct competitor to Java, but I think C# truly happened to be used as a way to create retro compatibility between Windows versions. Without paying Oracle. and Go... i don't fuarrrking know I don't understand Go. Malware dev i guess since its shiny and new... Jokes aside, Go is used internally and was used internally for a long time before it was released. Because the big tech company wanted his own language for reason. There's no reason for the common programmer to use Go other than you really want to interface with Google's products or/and want to work there. >>38984 >they have no interest nor incentive to be reliant on the opensource community >>38985 >implementing a features and waiting for a positive review and a merge, that may never come, is hell. this. I think the truth is that It starts as an internal tool, and gets released as a product. (With insidious intent or not.)

>>38986 Go was created for large-scale webapp backends written by very large teams of mostly mediocre programmers. It's intended to bridge the gap between the simplicity and quick turnarounds of dynanic scripting languages and the performance and safety of static compiled languages.

>>38983 >I don't see why you would need anything beyond C99/C++11 (and maybe IEC languages for industrial PLC programming) Many will fight you but I agree with you. The truth is that the big majority of developers working for these companies fuarrrking suck. This is evidenced by the excessive use of bloated libraries, they are so bad that they don't even understand how to solve simple problems and resort to libraries to "save time", and it explains the demand for libraries and "gurus" to give conferences and all that bs They have no interest in anything to the point they don't even check what their own code does, and use it to justify extra hours, QA, writing documentation, etc. Which also brings me to: >>38985 >Doesn't every company that has some bucks spend a fraction on it for quality of life tools for their devs? Good point but not necessarily true for all big tech companies, the "dev experience" can easily degenerate teams into becoming too comfortable (lazy and irresponsible) hurting the end product, users, making life more difficult for everyone else in the chain, etc

>>38983 >Meanwhile, if you want to do some actually useful programming, I don't see why you would need anything beyond C99/C++11 (and maybe IEC languages for industrial PLC programming). Mandatory garbage collection and strong typing, features C and C++ notably lack, eliminate entire classes of bugs before your program has even compiled, which is a massive boon to productivity. I've done in a day with Go what typically takes me at least a week in C. >>39002 >the "dev experience" can easily degenerate teams into becoming too comfortable (lazy and irresponsible) Seems to me that the root problem is that companies are hiring bad programmers, not that said programmers are using "bad" tools. People that genuinely care about programming will do good work no matter what language they're working with, be it x86 Assembly or Haskell.