This article is not making its case particularly well, because it's pointing to two examples of weak security and suggesting that they were weak as a deliberate

This article is not making its case particularly well, because it's pointing to two examples of weak security and suggesting that they were weak as a deliberate choice to make it easier to develop with those tools, when it's rather more likely they were weak simply because the developers were incompetent with or unconcerned with security. (The third example, ElasticSearch, certainly lacks security as a design choice, but it's fairly clear to me that it lacks this choice simply because the developers believe it should never be a globally accessible service, rather than to make it easier to use.) Furthermore I'd like to repeat the now pained challenge to writers of programming topics: please, please, please retire "considered harmful". At best it is high-handed and smug, at worst it is meaningless puffery.