The writer presents an ideal vision for the future - one where people can "build software" purely through abstract thought, without needing to know the semantic

The writer presents an ideal vision for the future - one where people can "build software" purely through abstract thought, without needing to know the semantics of specific tools and programming languages. If such a future is possible, that would be great. I would be all for it. But the people who are already in the field, working in the trenches, don't think it's a realistic vision for the near future. All attempts thus far to produce "layman friendly programming" have been either failures, or relegated to non-functional toys. Hence why we don't want to waste our scarce time and resources on such moonshots. If the author and his peers disagree, they are free to found/invest in such ventures. And if they're right, they can make a fortune for themselves in the process. But just sitting in the sidelines and armchair quarterbacking is a pointless waste of time for everyone involved.