![Ian Clarke](/img/default-banner.jpg)
- 5
- 27 193
Ian Clarke
Registrace 13. 05. 2007
Eclipse totality from 400ft, 8th April 2024, Austin TX
A view of the total eclipse from a drone at 400ft, totality begins around 3:15 mark.
zhlédnutí: 71
Video
Decentralized Reputation and Trust in Locutus
zhlédnutí 895Před 2 lety
Trust is fundamental to almost everything we do on the Internet, but how can trust be achieved in a completely decentralized system like Locutus? An introduction to Locutus: czcams.com/video/d31jmv5Tx5k/video.html Star our Github project: github.com/freenet/locutus Follow us on Twitter: FreenetOrg
How we will decentralize the Internet with Freenet (Talk + Q&A)
zhlédnutí 4,7KPřed 2 lety
On 7/7/22 I gave a talk introducing Freenet (formerly "Locutus"), a platform for building truly decentralized Internet applications, along with a Q&A. Github: github.com/freenet/locutus Support us: freenetproject.org/pages/donate.html Stay up to date: freenetorg Chapters 00:00 Introduction 02:33 History 05:54 The Oligopoly 08:23 The Solution 08:59 Fundamentals 11:21 Contract key & s...
3D Printed Sierpiński Pyramid
zhlédnutí 12KPřed 2 lety
UPDATE: Now available on Etsy: www.etsy.com/listing/1083500959/3d-printed-sierpiski-pyramid I 3D printed a Sierpiński Pyramid on my Ender 3v2, it took a day to create the model, and about a day and a half to print it. Model: www.thingiverse.com/thing:2573402 Filament: www.amazon.com/dp/B07PYDNQXS I generated this model myself using openscad to make it "order 6", which is twice the size of the l...
There's only so much an individual can control!
6:11 oligopoly: (n.) (economics) a market in which control over the supply of a commodity is in the hands of a small number of producers and each one can influence prices and affect competitors;
Would love to see an update on this
"I think Reddit might be the only large company that kept its API" Welp, that didn't age very well.
I've been using your service for a couple of years now and it's wonderful! It's good to know how e-mail addresses propagate and see companies squirm when you ask them how they got your e-mail address and confront them with their lies (turns out some companies even pay for illegally obtained data, such as data leaks/hacks). Furthermore, websites and webshops don't *need* to know my actual e-mail address (or even my real name for that matter) and 33mail is one of very few services that allow you to make aliases on the fly, without the need to create it in some portal first. I've decided to start paying for the premium subscription as I feel that good services that I'm really satisfied with deserve to get paid for it. It's not much, but I hope it helps to keep the service running.
TL;DW how does it compare with Qortal?
Are keys/contracts immutable?
I can't imagine they would be very useful if they weren't, since someone (I.e. the original creator) would be able to change them at will, same as the centralised web.
There are proof-of-work protocols that aren't energy intensive that could be used. Monero would be one example.
its old and wont pump
Wasteful either way and not needed
Interesting stuff
Any chance that some madman is rewriting freenet in a modern language (let's say RUST)
mz startup develops it in Basic and Fortran.
Would it be possible to utilize Locutus to make a decentralized multiplayer game?
It should be.
Will it be available on windows?
There are many similar platform-ideas like this; where you can build "dapps" and whatnot, but which one is the best? Shouldn't step one be a great comparison-movement; evaluating pros and cons of various approaches/implementations/use-cases incl feasibility for mass adoption/cross-app-functionality/security/speed etc?
None of those platforms are as small and OS like as Freenet
@@BotchedGod is that good or bad? how bodes it for cross-app-functionality?
Seems very similar to ZeroNet
Saw your interview with Louis Rossman, and now the almighty algorithm deems you interesting to me lol. It's not wrong... but I have a question: One of the advantages of IPFS over a client/server model or even traditional CDNs, is its awareness of network topology, which minimizes overall network congestion. If two people in the same house, or the same university LAN, want to access the same file, the first one caches it and serves it to subsequent requests. Does Locutus give any weight to physical routes like this when it assigns a node's location on the ring? Because of course, on the other end of the spectrum from IPFS, is Tor, which creates lots of duplicate traffic over many networks in the name of anonymity. But if anonymity is not a first-order goal of Locutus, is it necessary to limit/randomize/obfuscate the assignment of a location? Why not optimize the efficiency of the routing? Or, if one wanted to create/allow an anonymity layer, is there a way to emulate what Tor does and negotiate a new cryptographic exchange at each hop? Forgive me if I made false assumptions, I'm just discovering this project. I love the idea though, and I hope it goes well. Next stop, github!
Locutus peers monitor other peers' response times and prefer peers that respond faster, so the network will tend to self-organize more efficiently over time with peers preferring other peers that are closer on the network, all else being equal.
How can you have relative reputation? A single entity may have rep A in context A but rep B in context B.
Different contract for each context
Very interesting project. I'm only 2 minutes in, but you said something like how primitive the internet was in 2000, no social media, etc. I was on IRC and Usenet in the early 90s. Was there just not an interest in implementing these kinds of communication channels for Freenet?
Discord is what we have instead of IRC. Reddit is what we have instead of Usenet. IRC and Usenet were decentralized, Discord and Reddit are not, and are owned by companies. I think these two are vital to talk about. (Sorry this is a running commentary, still not done with video.) :-)
Monetization is what killed everything. It was what changed the Internet from a relatively collaborative space for people to share in many different ways, into a gaudy mall with a few extremely rich and powerful, obnoxious mallcops. It's hell.
Yep you are exactly right. The public square/commons has been bought, with the goal of making money. It is also used by the ideologues in media and government to only allow the "right kind" of opinions. I fear however that the truth you're spitting is largely wasted on the current generations, fat on decades of both political and economic propaganda. Nanny/daddy state, etc. It's scary to have to be responsible for your thoughts and actions, and to figure out who to listen to/believe on your own. They want a cable channel/web site to do that for them. People are lazy, and that is easier.
I have no doubt you're familiar with Matrix. I ran a server for awhile... it's not "easy." I could get one strung together with Elmer's glue and string because I'm a UNIX/Linux guy, but it's kind of laborious.
I think you have some very interesting ideas. I will try to join another Q&A if you have one. Some alarm bells went off in my head at points, but I'm just a lowly BOFH... a lot of this is over my head. But hey, I subscribed to your CZcams channel and will be keeping an eye on your project. I think you've bitten off quite a lot, and much theory needs to be sifted through. :-) Good luck sir.
Have you considered creating relative reputation scores based on clustering instead of an absolute reputation for all users? How are you going to prevent users from being identified/spied on through the feedback interactions and reputation scores, maybe a time delay and a differential privacy algorithm?
Could have a different contract based on the interaction, maybe like a buyer one, seller one, commentator one, then you could control access to your rep depending on interaction
Is it fair to say that the key being made out of a piece of web assembly code is a process that is unique in both its identity and its function simultaneously? Is this function an efficiency advantage distinct from the small world benefits? Does the key being web code mean the material costs in maintaining an authorship node, scales revolutionary based on the simplicity/efficiency of the decentralized apps created for it? Can apps, if coded succinctly enough, live sustainably across relatively fewer keys than a less efficient application? (I'm assuming applications are distributed across children instead of clones) If the key, itself, carries information, can it be checked by its validity, utility, and its identity? This sounds like unimaginably tiny painters being traded as currency to coordinate resource distribution.😁❤️👍 Please sir, may I trade you my tiny painter, for your tiny painter, so we may trade on forever more tiny painters? They accidentally paint us useful pictures, but they will make good couriers on their way. Oh the wonders, of tiny painters.
Yes, because the key is the webassembly code, its functionality is immutably linked to its location in the key-value store, which can be verified by any participant in the network. It's a generalization of an idea in the original freenet called "signed subspace keys".
30:30 I love a decentralized email idea! I also would love to see decentralized VOTING if possible? How can we create a system where we know that our votes are cryptographically counted? These hackable Windows XP voting machines need to be tossed in the trashcan if we want to maintain election security. I'm hyperbolizing, but you get the idea that I am trying to convey against closed source voting machines...
Decentralized voting will be possible. It would be great to implement something like liquid democracy: en.wikipedia.org/wiki/Liquid_democracy
This is very impressive and promising. Could enable "web2.5" use-cases where the underlying data and infra are decentralized, but companies can build centralized indexers on top to compete with tech monopolies.
The observable model is fantastic for things like social media feeds and subscribing to content creators, but for things like a search engine, I don't see how you'd do it any other way than a centralized indexer. I'm curious how you would implement a crawler for such a search engine. I guess you'd have to make a portal where people could submit their contracts to be searchable. But then you still rely on someone's algorithm to rank you. The biggest challenge for p2p technology is not distribution, it's discoverability.
@@DFPercush That's what I meant by web2.5, its ok to have centralized indexers as long as there are many to choose from and the underlying data that is being indexed is not in a walled garden.
sounds like GUN with smart contracts and also react is soy
It will work with any browser web framework.
Thanks for the incisive comment about what's "soy", "Fishy Pug Bruh".
Looks like AR.
You're real life Richard Hendricks and I love that
😂
Excited for August 2022
Question: How does Locutus differ from Holo? Would the differences be a benefit or detriment to either projects?
Not so familar with Holo, but Locutus' approach of using cryptographic contracts as keys in a distributed key-value store, I don't think Holo has any equivalent.
This is beautiful