Stopping the Internet of Noise – A useful internet back again

Re:Stopping the Internet of Noise – A useful internet back a... by TeMPOraL on Wednesday July 05, 2017 @04:04AM (#14700652)

Thinking of such a tool as a service is a symptom of the same disease that got us where we are in the first place. If such tool would, instead, be open-source and self-hosted (or a component of trivially self-hosted) application, websites could do exactly shit about it. Traffic will be coming from everywhere, and the tool will get updated by people to overcome upstream interface changes.

Ultimately, it's SaaS webapps that's the problem. The model has legitimate benefits (lack of installation, cross-platform), but it also is the reason the Web centralizes so much.

[ Parent | on HN ]

Re:Stopping the Internet of Noise – A useful internet back a... by alanfranzoni on Wednesday July 05, 2017 @05:04AM (#14700893)

Most probably the github repo will be suspended, even though it's self hosted, I think. And you still need to account for upstream changes.

[ Parent | on HN ]

Re:Stopping the Internet of Noise – A useful internet back a... by TeMPOraL on Wednesday July 05, 2017 @05:25AM (#14700974)

Forget Github. It might be just a regular git repo on someone's server. IANAL, but could companies really go against a person hosting the source code of a software that scraps a website?

Even if, that's probably solvable (by turning it into cat-and-mouse game, TPB/SciHub-style). Have a non-turing-complete set of conversion rules that do scrapping -> standardized-data-format via a general-purpose scrapper. Distribute those through P2P. There might be problems with malicious actors, but generally this should be possible to implement in a way that's mostly transparent for end users.

> And you still need to account for upstream changes.

Upstream changes will never be too extreme to handle manually, if we're talking scrapping solutions - because any such extreme changes would also piss off the regular browser users.

[ Parent | on HN ]