The largest failing of the article is in the "Existing Approaches To Fix These Problems" where its ack'd that this is attempt number one zillion to fix some old

The largest failing of the article is in the "Existing Approaches To Fix These Problems" where its ack'd that this is attempt number one zillion to fix some old problems, and here is a slightly different approach.... the problem specifically being lack of analysis of past failures to improve the new proposal. We're just going to end up with POSIX / LSB again, except maybe less successful because its obviously going to be immensely more complicated. Its possible that analysis was done but not documented. The tone of the article is very strange, like someone who has no experience with distros or sysadmin work came in with a blank slate from an unrelated field, like perhaps sorting algo research. I simply can't identify with most of his postulates, so its no surprise I was all WTF at the proposed solution. Or rephrased, its an academically interesting idea in an abstract sense, but seems to be unrelated to, and not useful for, the existing real world. (edited to emphasize, that this is a particular instance of the general subject that new opinions from uninformed outsiders are often interesting to look at, sometimes even insightful to have a new pair of eyes, but are usually a truly awful source of advice)