The utter dominance of English in the software development world is due only in small part to the interminable self-promotion of the US, and has quite a bit more to do with the fact that language is crystallised culture. It reflects the perspectives, priorities, prejudices, and most importantly, ontologies that make up said culture. It is also optimised for their expression, and this produces a feedback loop, with the language and culture being both cause and effect for one another.
A single culture established a clear technological lead for several decades. Every year of that accidental dominance, all the best tools (best because they were the only tools) reflected the language and culture of the people who created them. As a result, thinking in that language/culture conferred an instant technology boost on anyone adopting the language and mode of thought of that culture. It was (and remains) uneconomic to take any other path.
Thus, nearly all the good software tools are built by and for people thinking in English - even if they don't use English at home. If they aren't thinking in English then certainly they are structuring their thoughts with English derived ontologies because they don't have any choice if they want support from advanced tools. There is some serious cultural imperialism going on here.
Another reason for the dominance of English is that it is, unlike the Romance languages (and as far as I know, any other language) the only language that is also a meta-language: it explicitly supports the systematic extension of grammar and vocab.
My (now deceased) French wife observed that there were often half a dozen grammatically legal ways to structure a sentence yet only a few that wouldn't raise eyebrows or amuse people. She asked me how we know which are the "right" ways, cursing a language that seemed to her entirely composed of exceptions. After some days' reflection, my answer to her was that English is a creole, encompassing all the grammars of all the major language groups, and the "right" grammars are the ones that produce the shortest comprehensible sentences.
When a another language is more effective in some way, English speakers will brazenly borrow from that language whatever is good about it, and make it their own: English is a kind of linguistic Borg.
I mentioned ontologies. What made Delphi more than Object Pascal, what made Java useful and what made C# more than just another Java was the supporting technologies: the libraries and the IDEs. These necessarily reflect the ontologies of the culture of those who built them.
In summary: you can't stand on the shoulders of giants without getting cozy with them.
I have heard a French sysadmin cursing the fact that he didn't have the English Netware manuals: endless examples of technical idiom were erratically translated, making terms impossible to find in the index, and even when he did find the right section it was, he said, incomprehensible because l'Academie insisted on translating "bug" as "insecte" and similar foolishness, and the translation of ideas was inconsistent because the ideas themselves were alien to the translators who are generally «Diplômés des Arts, avec peu de science et encore moins du rigeur» (his words: Arts graduates with little science and less discipline).
All that said, three hundred years ago the trade language was French. Now, it isn't. Why? Follow the money. This logic says that - allowing time for the English-language leviathan to die - soon it will be Chinese. But not the chinese of today, and certainly not the chinese of Mao's fevered imagination.
They've been living in our minds for thirty years, copying our works, first cheaper and then better. A whole generation has lived and worked in a frame of reference that was the essence of us. Just as the Greeks remade the Romans, and the Romans both the Britons and the Franks, as our minds were forged in the embers of their zeitgeist, so have we reforged the Chinese in our own image. Give 'em 20 years and they'll be better at thinking like us than we are.
At one workplace I was involved in a lengthy assessment of the economics of internationalising our product in response to interest from Hong Kong. To paraphrase the outcome: "Not worth the trouble. Half of them speak English and the other half don't have any money."
Why is this? Maybe English will outlive us. In what other language can you invent the word "extensibility" and not need to explain it, spell it or justify it to l'Academie? Oh, and ideogrammatic written languages are less than keyboard-friendly.