http://www.nytimes.com/2005/07/17/technology/17spy.html?page...
Yes, there are services which will promise the skies in delivering an ad-free or curated experience. In some small and narrow niches, this may succeed. In large part, these services will fail, with numerous results.
One is that the promise of ad-free paid subscriptions will give way to advertising-included pay services. See the long history of cable and premium television channels.
Or that a subscription-based service will simply prove unsustainable. If it's merely a publication, you'll find it folding. If it's a community, you'll find your community and relevant connections and activities (social, professional, intellectual, commercial) terminated, often with little or no notice, and grossly insufficient migration options. "Saving your own data" fails to accomplish the frequently desired goal of "preserving your community".
If it's a hardware-integrated service, you find yourself with both a terminated service and a closet- or garage-full of newly-useless kit.
If it's a hosted media access platform -- DRM-infested texts, video, or audio -- you'll find yourself cut off from access to content you have bought and paid for. This is starting to be addressed in some small parts, but remains largely intractable for many.
The biggest flaw in this article, really a classic "submarine" in the Paul Graham sense, is that it presumes that the only solution, or even a viable solution, to a market problem is to market-it-harder.
When broadcast technologies first emerged -- radio in the 1930s and later television in the 1950s, there was a large, and largely forgotten, very public debate over whether a private-ownership or public-provisioning model was appropriate.
In the US, the private approach was undertaken, and ABC, CBS, and NBC, among numerous other networks, now almost wholly consolidated, were born. In Europe, notably Britain, this was seen as a tremendous risk, opening the door to corporate and foreign interference in public discourse, information, and opinion (a fear which may have been well warranted), and instead a tax-supported Broadcast Service was created.
Each side relaxed somewhat. Public broadcasting didn't entirely vanish in the United States, and several independent stations joined forces in the 1960s. PBS, the Public Broadcasting Service (television) was formed in 1969, its radio sibling, NPR, National Public Radio, in 1971. C-SPAN, a paid-subscription, but public-service network, was created for cable by Brian Lamb in 1979. Its funding comes as a part of baseline cable subscriber fees.
There are other models in the online world.
Wikipedia and other Wikimedia sites support general informational content, including encyclopedias, dictionaries, books, news, data, educational content, and more, on a donor-supported basis:
https://en.wikipedia.org/wiki/Wikimedia_Foundation
Several years ago when looking at detectable measures of signal/noise ratios for various online services, publications, and domains, I found that the venerable Metafilter ranked exceptionally high. Though small (scale of any limited-access platform will be hugely reduced), the presence of intellectual rather than pop-cultural references, as determined by Google results counts for the Foreign Policy Top 100 Global Thinkers names as against the arbitrarily selected string "Kim Kardashian" was the highest of any social media site tested (FB, Twitter, G+, and others), indicated as the FP:KK ratio. Blogs, exemplfied by Wordpress domains, also did surprisingly well. See: https://old.reddit.com/r/dredmorbius/comments/3hp41w/trackin...
Metafilter's trick? A one-time $5 fee payed per account, forfeited in the event of abuse.
More generally, the problem of paid-vs-free, or even multi-tiered public-service provisioning, is that unless the quality of the free or low-cost service is abysmally poor, few will avail themselves of the higher-cost levels.
I'd first encountered that observation in the context of multi-class rail-service provisioning by French polymath Jules Dupuit. Upshot: the reason third-class accomodations are so miserable isn't that it would cost to much to improve them, but that improving them would cannibalise higher-class service levels. Those who've flown coach class in recent years can likely understand.
See: https://www.inc.com/bill-murphy-jr/why-does-air-travel-suck-...
Tim Harford's addressed this in the specific context of digital media services at FT:
Applied to information services, a similar inevitable logic is self-evident.
Put another way, there's a tremendous deadweight loss to any such proposal. Marketing-it-harder will deny access to all but the wealthiest members of the public. Though as a market-segmentation technique for identifying high-spend-propensity individuals, this is ambrosia for advertisers, hence the inevitable encroachment of advertising to such services.