5 ms·
I also really hope to see the open web make a resurgence, but I have no faith in regulation as the answer. I'm not actually sure how regulating the open web wou
by flagrant_taco 3y ago
I also really hope to see the open web make a resurgence, but I have no faith in regulation as the answer. I'm not actually sure how regulating the open web would work, seeing as regulations require laws to be written by someone deciding what can and can't be done on the internet.
Bottom-up solutions are often the hardest, and a massive undertaking compared to top-down alternatives, but if the goal is am open web I can only see that working bottom up.
Not least of all, top-down regulations that actually improve the open web would be extremely difficult when companies like Google have nearly unlimited funds and most regulators and politicians are already well accustomed to being captured. Even on a good day I would expect well-meaning regulators to ask the "experts" at Google how it's supposed to work and what the regulations should be. The FAA allows Boeing to regulate itself as does the FDA with pharmaceutical companies, why would this be different?
- Sparkyte 3y agoSelf regulation only works in a top down solution that can be held accountable. You can't hold anyone accountable with the Internet of Things and regulations that establish accountability allow self regulation.
- flagrant_taco 3y agoNot sure that I follow, but the open web is a bottom up approach rather than to down. Top down would be a controlled system with regulations and punishment, that isn't open at all. Where does the internet of things fit in here?
- Sparkyte 3y agoWell the Open Web is one thing in the Internet of Things. Imagine what changes if we move closer to a closed web environment how that changes other things in the umbrella term. Open Web is desirable but the clock is ticking to a closed web.
- flagrant_taco 3y agoAh OK I follow you now. Unfortunately I think that clock on the open web has already ended, we've effectively had a closed web for at least a decade. Google sure seems to want to hammer in that final nail, but it's been a long road to get here. Honestly I can't help but think we'd be better of revisiting early decisions made in Http itself, especially cookies though long-lived connections are also risky. The gemini protocol is really interesting in that area, but I can't help think they rolled it back to far with odd decisions like not requiring cert verification, utf-8 encoding URLs, and only supporting GET requests.