Beh, I guess i'm getting old then, because I spent ~10 years of my life (over 10 years ago) learning dozens of scripting and programming languages, and programming professionally in quite a number of them.
The takeaway after all that was, rarely is one language better than another, its really about trade-offs.
And frankly the older ones are a lot better at following the KISS principal and actually providing an uncluttered framework for writing _EFFICIENT_ code. For example, everyone knows that Java is a "better" and more productive language than COBOL, yet when you start talking to people trying to port old COBOL apps to java, what you discover is that it turns out that COBOL might actually do a better job in the business/transaction space. Same with python and Fortran in the HPC space, and strangely enough the core infrastructure all these more modern efficient languages are running on tends to be "C" based. Which despite the laundry list of issues that people like to parade around, seems to solve systems programming problems better than anything else.
None of this would really be a problem except for the fact that the new up and coming language that is cool this year changes from year to year leaving a wasteland of abandoned poorly maintained projects using languages that no one really knew, and are despised by the people who are hired and have to learn them for that one job before moving on to the next one.
So, yes bash is a POS, but its universal at this point, isn't really that bad at solving the core problems its designed to solve, and any developer with 1/2 a brain can write a sys V init script (or a lot of similar things) with little more than a couple hours and without the baggage of yet another 1/2GB language/framework/etc pulled in to sit alongside the dozen others already in the project.
The takeaway after all that was, rarely is one language better than another, its really about trade-offs.
And frankly the older ones are a lot better at following the KISS principal and actually providing an uncluttered framework for writing _EFFICIENT_ code. For example, everyone knows that Java is a "better" and more productive language than COBOL, yet when you start talking to people trying to port old COBOL apps to java, what you discover is that it turns out that COBOL might actually do a better job in the business/transaction space. Same with python and Fortran in the HPC space, and strangely enough the core infrastructure all these more modern efficient languages are running on tends to be "C" based. Which despite the laundry list of issues that people like to parade around, seems to solve systems programming problems better than anything else.
None of this would really be a problem except for the fact that the new up and coming language that is cool this year changes from year to year leaving a wasteland of abandoned poorly maintained projects using languages that no one really knew, and are despised by the people who are hired and have to learn them for that one job before moving on to the next one.
So, yes bash is a POS, but its universal at this point, isn't really that bad at solving the core problems its designed to solve, and any developer with 1/2 a brain can write a sys V init script (or a lot of similar things) with little more than a couple hours and without the baggage of yet another 1/2GB language/framework/etc pulled in to sit alongside the dozen others already in the project.
</rant> Now get off my yard.