This is happening with a lot of companies that don't really have to care about "startup value"; they're tiny (sometimes even one-person) shops that aren't likely to ever go public, be bought for millions, or so on. I think you've got the cause and effect reversed for a lot of these -- the App Store pricing model has driven the price people expect to pay for apps down substantially, which made the "buy outright and then buy a new version in a couple years, at a discount, if it appeals to you" model much harder to sustain. I don't love subscriptions for apps that don't have a server-side component, but I get it.

(The example of the Pomodoro app with the subscription is pretty dubious, though.)

> Why go that far back?

Because this is the era before really any major software was sold by subscription. Even 2008 was a hybrid era where some Web software was already subscription-based. Consumer behavior had already started to shift by 2008.

> for year apps were sold for a flat one-time fee.

This model also makes a lot of assumptions about where user data is stored and who is paying for that storage. Many apps have non-trivial backend requirements to support very normal use cases like "I also would like to access this from a Web browser sometimes." Those requirements cost money on an ongoing basis. One-time purchases are not a good fit for this either. We do not see many one-time purchases of Web software; many mobile apps are essentially parts of a larger whole that includes other modes of access to underlying services.