Sigh. The HackerNews standard of pushing some logic to its limits and going "HA, GOTCHA" is getting tiring.
x86, x86_64, ARM are used by billions of people around the world. It's easy to see that there is actual value for end users by supporting these platforms. John Doe from accounting can easily use your open source software and it will benefit him.
The clear benefits of supporting z/OS, aside from helping IBM push out more of their mainframes and making more money, are unclear. Supporting PA-RISC, POWER or any other esoteric architecture whose sole purpose is to make their vendors sell million dollar support contracts should not be anyone's priority, and if indeed they do want this software on their non-standard, non-widely used ISA, they can use some of these millions to fund LLVM development.
> Sigh. The HackerNews standard of pushing some logic to its limits and going "HA, GOTCHA" is getting tiring.
So is the Hacker News standard of ignoring the entire point just to nag about said supposed standard. Sigh.
Said entire point, specifically, being that there are far more reasons to develop and maintain software beyond "value for end users", and that even when targeting that specific reason, developing against esoteric cases does ensure the software is more robust even for "normal" use cases.
(And mind you, logic - like software - should be pushed to its limits, because that's the most surefire way to identify its flaws - like, for example, the failure to consider that nearly all computing platforms in existence and use today were and are the product of some large corporation that could be paying your bills but won't, or the failure to consider that it's users, not hardware vendors, who are generating support requests for their hardware)
x86, x86_64, ARM are used by billions of people around the world. It's easy to see that there is actual value for end users by supporting these platforms. John Doe from accounting can easily use your open source software and it will benefit him.
The clear benefits of supporting z/OS, aside from helping IBM push out more of their mainframes and making more money, are unclear. Supporting PA-RISC, POWER or any other esoteric architecture whose sole purpose is to make their vendors sell million dollar support contracts should not be anyone's priority, and if indeed they do want this software on their non-standard, non-widely used ISA, they can use some of these millions to fund LLVM development.