Not at first, at least, in my opinion, if crowdsourcing is what makes it useful. If the network effects are what makes a product useful, a subscription fee pushes against the growth of that network.

Yeah, what I'm foreseeing is public and private links, tagging and the ability to search for links by page description, tags and user. The subscription fee would mainly be for hosting costs and an eventual commensurate salary. Even if the site were to gain traction I wouldn't be interested in selling ads based on user base size.

I'm willing to incur some hosting costs for a beta period but I won't fund the service for long without revenue if the hosting costs are prohibitive as I'm not really interested in selling it later for someone else to monetize with ads.

Depends what you get beside storing links I suppose (cached version of content, super smart search function not limited to og: and metadata tags, super privacy, bridges to-do app maybe ?, android app + ff ext. + chr. ext. + apple thingies for ubiquitous access to links, etc.). I could see myself dropping ~10bucks a year like for bitwarden.

How is it different from https://pinboard.in (besides $1/ mo)?

Beyond some differing features and interface, there wouldn't be much difference.

They pay pinboard enough for idlewords (HN user) to live off it, and buy Delicious when they went broke.

Delicious once sold to yahoo! For 15 million, bought for 35k.

https://news.ycombinator.com/item?id=14463102

Delicious spending twice as much as pinboard in running costs in 2017.

There are some detailed blog posts from the time about running costs somewhere.

I already do: raindrop.io