> Amateur software developers dislike While the rest of the list is valid the author could have stopped at the first item Standards and still have made his poi

> Amateur software developers dislike While the rest of the list is valid the author could have stopped at the first item Standards and still have made his point. Perhaps the most perfect example of that is why JSON won versus XML for data serialization. Many people will claim that it's because XML is bloated. After all this was the primary argument of JSON's inventor Douglas Crockford. They aren't wrong, but its not why JSON won. JSON won because it's specification is tiny. When all the boilerplate is scrapped off the latest version is about 5 tiny pages of text any high school student can understand: https://tools.ietf.org/html/rfc8259 https://tools.ietf.org/html/rfc8259 Compare that to XML: * XML - https://www.w3.org/TR/xml/ https://www.w3.org/TR/xml/ * XML Schema - https://www.w3.org/TR/xmlschema11-1/ https://www.w3.org/TR/xmlschema11-1/ * XSL - https://www.w3.org/TR/xsl11/ https://www.w3.org/TR/xsl11/ Those are HUGE and they are written to conformance precision for implementation engineers. No high school student will ever read any of that for pleasure. I remember reading those when I first entered software development and absolutely everybody thought I was out of my mind and promptly discarded everything I said if they didn't already know it. I remember that being particularly frustrating because ever recommendation or proposal had to read like a pitch deck to a VC, but if I wanted to do that instead of engineering or product development I would be brand evangelism or merchandising. XML for a data conveyance medium is extremely powerful, but in practice this is completely irrelevant. XML is a world unto itself in all of its capabilities and potential, but that doesn't matter if nobody is using it. Most Americans, regardless of their actual educational attainment read at a 9th grade level (high school freshman). That applies to entry level software developers as well. Computer scientists and inventors are not the average of software developers. On one hand this is great, because I truly believe anybody can be a software developer and write really bad code. On the other hand this is horrible, because many people are code monkeys that write either bad code only or are fundamentally limited to a small subset of solutions limited by a small subset of tools.