The balance between Betting on Standards and betting on future Standards

08 Jul 2015

A few years back , I worked for a company that was big on C++, for many years despite the existence ofthe C++ STL, the company had decided to bet against, the STL , and continue to go it alone. We had our own instance of Vector classes, String classes ( which at that time were legit faster), and finally we had our own double, float etc… . After well over a decade of going it alone deviating from the standard interfaces, hundreds of thousands of lines of code,better technology emerged, that utilized standards. We were at a stand still. Without a crazy herculean effort, we weren’t able to take advantage of the advances that were made to the standard, from boost::std to the emerging Cx standard. I’m sure no less than an a few years of engineering hours were spent porting, and standardizing.

Currently, we’re at another such impasse with JS. Simply put we have alot of silos , from RequireJS , Browserify,Webpack,Grunt,Gulp,Typescript, Elm, and finally the ES6 and ES7 specifications. My money is on backing the horses that respect standards, the most. If no standard exists or the standard is subpar, then my money is on the custom library that is without dependency, and is flexible enough to be extended.

comments powered by Disqus