> 1) It’s no longer trivial to build a “home” page that fits one’s diverse interests. People are lured into frameworks which prescribe a certain level of structure. How do we make it easy for people to freely add content to their “pages” , in whatever form they envision?
In the olden days, you were either affiliated with a university or you signed up to Geocities. Nowadays, you can go to WordPress and add your fill of content pages: it's not just a blog. I'm pretty sure the modern web is a win for ease of access to genuine user-controlled pages. Probably you had more control over the skin of a GeoCities site than a free WordPress page, but I'm not sure to what extent or how important that is.
I think the issue here is more that people don't just want to make content - they want to have readers. You're more likely to get and know about your readers if you post on Facebook or Twitter than if you post on WordPress.
> 2) The “early” internet (or what is left of it) is largely outranked now, as the methods by which we discover new information has been heavily commoditized by industry.
I think this is important. The modern web doesn't provide discoverability to independent content produced in good faith. Either you have to agree to donate it to Facebook or you need to do a lot of work drumming it up. Google become popular because it cared more about the content than the primitive SEO, and the web was small enough you could plausibly use dmoz.org to browse the internet. Now, advanced SEO is better than any search engine algorithm and there's too much godawful content to be excited to read content without some kind of active recommendation.
> A truly decentralized internet experience has the potential to foster diversity, but our best and brightest are figuring out how to capture this value with variations of digital signatures rather than novel and groundbreaking tech.
I recall another article posted here a few weeks ago, where the author argued that centralisation dominates over decentralisation. But the argument in this place (at least till I went to bed) was that because there was technical decentralisation, there was no centralisation.
I don't know which view represents the mainstream view of technical people today, just which view seemed to have the most advocates while I was reading the thread.
What I do know is that it seems hopeless to me to hope that our best and brightest will help decentralise the internet again. I think the Gemini project has it right when they do whatever they can. I wish it were more prosocial though; it mostly consists of isolated gemlogs without the opportunity to reply. (I mean, you can reply, but all you're doing is posting a new page. If the person to whom you are replying doesn't know you from a loaf of bread, it's more practical to just go wash the dishes. If someone wants to correct me, by all means: I'm all ears.)
Btw, I don't want to imply that the Gemini folk aren't the best and brightest: Just that the path to a decentralised internet is the same as the path to free software - wearing the straitjacket and doing the work. And even then, it's easily lost. How much centralised, non-free software do we all use even if we've never installed it!
I think the main problem to solve is interaction without inauthentic action (spam). Centralised systems allow one person to cut off the spammer once, and that stops them from affecting everyone. If you imagine a decentralised gem/weblogging network with a reply facility, then if you write a logpost (on a log hosted on your own computer) and send a trackback to mine (on a log hosted on my own computer), then I probably consider that interesting. But if now a spammer writes a logpost and sends a trackback to yours, you need a way to cancel that. And once you've cancelled them, they can still post to mine! Each spammer will annoy each logposter as many times as they can get around the spam protection: it's an exponential problem. With a centralised system, once you've cancelled them, they are much more likely to encounter difficulty trying to trackback to my log: it's a linear problem - or even less.
The other issues are probably surmountable - it shouldn't be impossible to create a decentralised network of posts and replies and threads that can be viewed together. It probably starts with something more like RSS than Usenet.