How solid is Tim’s plan to redecentralize the web? • Hacker News Board

Re: How solid is Tim’s plan to redecentralize the web?

#171

Post by mark_l_watson » Sat, Oct 06, 2018, 1:19 PM UTC

gibsonf1 wrote: post #39

gnfurlong wrote: post #33

Earlier quoted context omitted.

Node-solid-server is just one implementation of the spec. There's nothing stopping another implementation from having a cpanel like management interface.

I'm going with SBCL Lisp and Allegrograph and Caveman. Anyone interested in teaming up on the Solid Lisp library? cl-solid

Count me in. My email is in my profile.

As much as I like AllegroGraph (and I wrote a book using AllegroGraph) if you want people to be able to install your project easily, then using SQLite, or something similar, might make it easier for people try try it out.

Re: How solid is Tim’s plan to redecentralize the web?

#172

Post by mark_l_watson » Sat, Oct 06, 2018, 1:26 PM UTC

I enjoyed listening to Tim and others talking about systems like Solid at the 2016 Decentralized Web Conference.

I liken projects like Solid to be similar to permissioned blockchain where small groups (people or organizations) use a private platform with lighter weight consensus algorithms than proof of work. Success can be had for small focused communities if there is enough value for users.

After trying to get friends and family to use a private Apache Wave instance for shared communication, and failed due to lack of interest on their part, I now don’t underestimate how difficult it is to move people off of centralized platforms.

Re: How solid is Tim’s plan to redecentralize the web?

#173

Post by gibsonf1 » Sat, Oct 06, 2018, 4:45 PM UTC

mark_l_watson wrote: post #171

gibsonf1 wrote: post #39

Earlier quoted context omitted.

I'm going with SBCL Lisp and Allegrograph and Caveman. Anyone interested in teaming up on the Solid Lisp library? cl-solid

Count me in. My email is in my profile. As much as I like AllegroGraph (and I wrote a book using AllegroGraph) if you want people to be able to install your project easily, then using SQLite, or something similar, might make it easier for people try try it out.

Great Mark,

Agreed - the cl-json library will handle all the Solid standards and then developers can layer on top of that their preferred db etc. I'm not sure if we need a db for the core library or not...

Re: How solid is Tim’s plan to redecentralize the web?

#174

Post by yason » Sat, Oct 06, 2018, 7:37 PM UTC

indigochill wrote: post #135

yason wrote: post #102

People are lazy and don't care. They're happy to let Facebook host and own their data if they can comment and like friends' videos for that. That's basically fine. From what I can learn from the history the web has mostly been centralised and will continue to be so. To think, all business domains tend to converge into few large companies/services, and finally into duopolies and monopolies if left to their own devices…

>From what I can learn from the history the web has mostly been centralised and will continue to be so. How do you define "web" and "centralized" here? The internet (not necessarily the world wide web) was created to be decentralized because the US military recognized that having a single point of communication failure was a bad design ( http://ccr.sigcomm.org/archive/1995/jan95/ccr-9501-clark.pdf ). In my estimation…

How do you define "web" and "centralized" here? The internet (not necessarily the world wide web) was created to be decentralized

Web as the world wide web of HTML pages served over the HTTP protocol. And centralised as served by the handful of biggest players in each market.

Even early on big centralised portals were attracting users. Internet started to attract common folks as soon as some services grew big enough to become "go to sites" at which point word of mouth was crushing smaller players. People didn't use Neighbour Joe's Web Crawler for searching, people used something like Alta Vista. The market for shopping on the internet increased directly as a function of big commerce sites gaining popularity. The small players never enjoyed more than a fraction of the number of users the big players had.

It seems to me that this suggests common people who do not have an interest in technology want to rely on a single (or at most few) brand(s) of service providers which basically leads to a winner-takes-all market and that just adds into centralised nature. If your mother wants to buy something on the internet isn't the only clue she has called "Amazon" ? (Or some applicable local market leader, depending on the country.)

Even if internet itself was completely decentralised and people were using RFC 1149 to reach the net I'm pretty sure they would still be connecting to Amazon, Ebay, Google, and other centralised hubs.

Re: How solid is Tim’s plan to redecentralize the web?

#175

Post by lovemenot » Sun, Oct 07, 2018, 12:17 AM UTC

mLuby wrote: post #157

lovemenot wrote: post #82

Earlier quoted context omitted.

>> If somehow each piece of data on a user were exponentially more expensive to keep, that might be possible. But that would be a very strange law… Dave and I agree with your diagnosis, and we would like to propose a solution. Who is Dave? Well you know him, and I know that you know him and I know that when we both talk about Dave we must be talking about the same person. (mLuby I probably don't really know you, this…

I don't understand. If Eve (who we don't know) adds data like location, age, interests to her profile on our Dave, people will want to access her profile on Dave. This creates an incentive for Eve to aggregate as much data on as many people as possible. Are you saying that nobody can know which Dave we're talking about unless we identify him as Dave-with-cell-5551234567? I don't think that's true, since human social…

Yes. That is what I am saying. And I agree that the protocol must be so designed that more data does not increase specificity, but rather increases noise.

Re: How solid is Tim’s plan to redecentralize the web?

#176

Post by gibsonf1 » Sun, Oct 07, 2018, 3:52 PM UTC

gibsonf1 wrote: post #173

mark_l_watson wrote: post #171

Earlier quoted context omitted.

Count me in. My email is in my profile. As much as I like AllegroGraph (and I wrote a book using AllegroGraph) if you want people to be able to install your project easily, then using SQLite, or something similar, might make it easier for people try try it out.

Great Mark, Agreed - the cl-json library will handle all the Solid standards and then developers can layer on top of that their preferred db etc. I'm not sure if we need a db for the core library or not...

https://github.com/gibsonf1/cl-solid