I wonder if privacy is a strong enough motive/need for such a paradigm shift in technology. @guptaneil, what are your thoughts on this? Is there any other benefit of having a decentralized web?
Great question. I wasn't trying to list all of the benefits of a decentralized web, just some reasons why people are more ready for the shift. There are many other benefits. Some that come to mind:
* lack of government or corporate control, making it difficult to censor
* increased security by distributing data warehouses so hackers can't steal all your data in one place
* cheaper infrastructure needs for startups and businesses by spreading the load out across the network
Some more aspirational benefits that have been discussed before are creating a self-healing and self-archiving network since distributed systems require some level of redundancy.
There are many other benefits, that's just the tip of the iceberg.
> Like I said above, we can’t even imagine what the economics will look like, just like Tim Berners-Lee couldn’t have imagined the current web 30 years ago. If we stop to worry about fitting the decentralized web into our current economic models, we will trap ourselves in a box.
Oh bollocks. This is just side-stepping the elephant in the room. For the web to work you need machines computing the requests. Either this is done via dedicated servers that someone has to pay for or it is done by the clients, which is unreliable: See Bittorrent, where everything depends on the goodwill of a few people paying for seedboxes (personal computers eating electricity included) or the extreme popularity of the content, content that is not easy to update across the network. And this problem is what drives many of those who care about the network into private trackers.
As for people running their own services, this is at the moment, delusional. John Doe will pick up every time of the day free GMail or Facebook before hosting his own e-mail (even assuming you "fix" the administration overhead) and his own Diaspora node which he has to pay for.
If we want to change this, we need to make personal servers attractive to the average joe with a "killer app" so that the first major block is cleared for users to start installing other things on it. An integral, attractive, and functional IoT solution with a huge fear mongering campaign about privacy is something I've thought of that could be a driver for such a thing, but the likelihood of success is extremely low.
3 comments
[ 3.3 ms ] story [ 14.1 ms ] thread* lack of government or corporate control, making it difficult to censor
* increased security by distributing data warehouses so hackers can't steal all your data in one place
* cheaper infrastructure needs for startups and businesses by spreading the load out across the network
Some more aspirational benefits that have been discussed before are creating a self-healing and self-archiving network since distributed systems require some level of redundancy.
There are many other benefits, that's just the tip of the iceberg.
Oh bollocks. This is just side-stepping the elephant in the room. For the web to work you need machines computing the requests. Either this is done via dedicated servers that someone has to pay for or it is done by the clients, which is unreliable: See Bittorrent, where everything depends on the goodwill of a few people paying for seedboxes (personal computers eating electricity included) or the extreme popularity of the content, content that is not easy to update across the network. And this problem is what drives many of those who care about the network into private trackers.
As for people running their own services, this is at the moment, delusional. John Doe will pick up every time of the day free GMail or Facebook before hosting his own e-mail (even assuming you "fix" the administration overhead) and his own Diaspora node which he has to pay for.
If we want to change this, we need to make personal servers attractive to the average joe with a "killer app" so that the first major block is cleared for users to start installing other things on it. An integral, attractive, and functional IoT solution with a huge fear mongering campaign about privacy is something I've thought of that could be a driver for such a thing, but the likelihood of success is extremely low.