Tldr; the author argues that the cloud aspect of current software is the bad part, the license of the software is irrelevant and hence the GPL is irrelevant, pr

Tldr; the author argues that the cloud aspect of current software is the bad part, the license of the software is irrelevant and hence the GPL is irrelevant, proposes we run software primarily on on our own machines, and doesn't seem to realize that this would result in the same situation as the 90s which necessitated the GPL in the first place. One of the author's big points is that the GPL doesn't solve the real issue of cloud services. (See the AGPL comment about an alternative). > Google and other cloud services may suddenly suspend your account with no warning and no recourse [...] Thus, you could suddenly find yourself permanently locked out of every document you ever created on Google Docs or another app. What he sees as the solution to this (and what he researches) is what he calls local-first software. "Local-first software runs on your own computer, and stores its data on your local hard drive, while also retaining the convenience of cloud software" > It is nice for local-first software to also be open source, but this is not necessary: 90% of its benefits apply equally to closed-source local-first software. He spends a lot of time on this, and outright states (see above) that whether it's closed-source or not doesn't have a big impact on what he sees as the "real" problem. What I don't think he realizes is that local-first software IS WHERE WE CAME FROM. It's closed-source software running on you machine which you cannot edit.