You pick the right language for the job, like Cobol for databases and Ada for space suits. NASA can afford to spend some of that money on programmers who are capable of learning. They spend millions of taxpayer dollars on space suits and billions on launches and then hire JavaScript weenies to "cut costs." These UNIX weenies pick JavaScript because C and C++ are worse. Perl and PHP were created because awk was worse. If your language makes you consider using awk, it really has to suck.
I stumbled across this little item in the current (October1992) issue of Communication of the ACM:"BANKS UNDERDRAWN... The banking industry spent over abillion dollars on technology last year, yet they are noteven close to employing leading-edge tools. A new survey... indicates that over 75% of bank computer programs arestill written in Cobol and 84% of banking software isdesigned for mainframes, not PCs. Moreover, 80% of thesoftware used by banks is over six years old and only 37% oftheir locations are networked. The report reveals mostbanks are simply not investigating new advances in computerapplications." [Communications of the ACM, Vol 35, No 10,p.9]A translation:BANKS CONSERVATIVE... The banking industry spent over abillion dollars on techology that works, rather than thelatest glitzy play toy. A new survey ... indicates thatover 75% of bank computer programs are written in a languageappropriate to the task as opposed to trying to force theirmodels into the latest Object Oriented fad and 84% ofbanking software is designed to run on systems that have lowmean time between failures, juggle hundreds of users, handlehuge databases, and push megabytes at high rates, not tinylittle machines that crash with great regularity, aredesigned for a single user, if even that, have minisculedisks, and have bandwidth the approximating that of asclerotic soda straw. Moreover, 80% of the software used bybanks has been fairly well debugged and only 37% of theirlocations are open to attack by thirteen year olds withmodems and a lot of time on their hands. The report revealsmost banks are simply not chasing the latest fad in confuserscience and piddling their money away on recoding workingapplications unnecessarily.