True Self-Sovereignty: What Will It Take?
True Self-Sovereignty: What Will It Take?
Session: 13D
Convener: Doc Searls
Notes-taker(s): Scott Mace
Tags for the session - technology discussed/ideas considered:
Discussion notes, key understandings, outstanding questions, observations, and, if appropriate to this discussion: action items, next steps:
Doc: I saw before what Dave’s topic was. He moved the ball way downfield. Bruce Caron is a novelist, wrote a series of 3 books, one of his characters is encouraging development of new software that will change everything. He says to its developers I invite you to think fundamentally.
Here is the first one: <http://archive.org/details/junana/mode/2up>.
Iain Henderson sent out to many of us a link to the new MyData paper that lays out the field as it now exists. That field is still top-down. Screen shot of Dave’s slides. Fully decentralized system. Spotting economic opportunity slide. What we have now are distributed systems that don’t follow Dave’s principles. I think they are a high hurdle but not a wall for developers. I started Project VRM really here at IIW in 2006. Helped start IIW itself in 2005. Project VRM is a one-year project that is now in the 14th year. Joe Andrieu thought we should be a point of integration of what’s ours. Not just data. People are the real edge. Our theory, free customers are more valuable than captive ones. Gets us to something that works like magic and does better signaling than supply and demand. The book I wrote about this, the Intention Economy, came out in 2012. Intentcasting came out of that. None of that has taken off. But for lack of what we’re trying to do here. Project VRM begat Customer Commons, doing for terms we would proffer what Creative Commons did for licenses that artists might have. We’ve done some work on that. We created IEEE P7012 WG. A number of us in this call are on that. A standard for machine readable privacy terms. Here’s an important thing. The IEEE in the person of John Havens and others approached Joyce and I for giving the machine readable version of personal privacy terms. Scott’s our note taker there. We’re making some headway now. JLINC Labs is a P2P protocol to allow party A and B to have an agreement. An active WG. Our chair is David Reed. Authored end to end principle design embodied in SMTP, IMAP, other things we depend on. Also Project VRM begat the M2B Alliance. I see M2B as the inheritor of Project VRM, which is a wiki and a list.
Lisa LeVasseur: I come from a standards background in cellular. I thought all we needed was a spec compliant with the principles we hold dear. We are creating a certification criteria and will start testing connected products. Been doing this for about a year. Very close to having MVP of our specification. There’s a lot under the hood, so we’re breaking it off in bite-sized pieces by listening to what everyday people want. We do not reflect their understanding who think identity fraud and theft of credentials are the most important thing. Most people don’t understand there’s another kind of currency in the world. Maybe it’s ads. Will go on a journey together with people. Doc: What Lisa is doing is still new. The title of this is what’s it going to take. We need the invention that one look and you got to have it. I knew the original web from links, but it wasnt until I saw Mosaic demonstrated, the web in a graphical way, I learned HTML, owning searls.com. I had to be there. It was one of those things where it was a demo sell. That’s been true for many of us with a lot of other things. I couldnt get Joyce to use a phone until the iPhone. Timothy wrote a piece the other day about the wallet. We use in everyday world, could also use it online. Complies with all 7 of Kim Cameron’s laws he wrote in 2004. I have a wallet, haven’t used it. Dave, what gives you hope with that right now?
Dave Huseby: Six principles of user sovreignity that we apply to nine problems of distributed computing. My hope is that with these talks about principles and the problems, we could get on the same page, in terms of looking at existing systems and accurately judging them whether they are user sovereign. I am no longer hopeful that existing systems will be reformed. I won’t change Facebook. Every day at 5am I write more code or thoughts. Wanted to build a model where I could hold myself accountable. To design a system that resists corporate capture, I have to follow these principles. Also helpful to enterpreneurs to make money in a way that is compliant with the IIW conscience. We come to IIW all the time and express these wonderful values. Every time I dig into a company in my role at Hyperledger, companies compromise to make money. I wanted to put a stake in the ground and call bullshit.
Doc: You spent 3 long sessions on exactly this. Put together a bunch of principles and goals.
Dave: Just to hold me accountable.
Doc: Grace wrote, it’s self-immolating before our eyes. A majority of people don’t want contact tracing in their phones.
Adrian Gropper: I agree with Dave. Thinking this will be solved by private interests is a fiction. A few of us Rebooting Web of Trust, titled Will SSI Survive Capitalism? In the MyData white paper, I don’t see how we achieve the Huesby test in a commercial framework. Some of it was discussed in a session bvgfoere this. It’s the separation of concerns problem. If they want to be authorization and storage server, it’s making the authorization server independently chosen. In Europe, banks forced not to have a captive credit card. As far as I can tell, neither small operators nor big platform vendors have any interest whatsoever of damaging the business model that mushes authorization and storage server. The entrances can be sold for a lot more money than traditional surveillance capitalism. The benefit of being the gatekeeper and seeing the information. The Post Office and the telcos, those days are gone. This is the elephant in the room for what you and Dave are talking about.
Doc: I scanned the MyData paper. Looked like an incremental move in our direction, but more of the same. I think we’re going to have to be truly disruptive. We do them what the internet did to AOL and CompuServe. Will take a thing. The browser did it. Maybe taking the authorization server and ubiquitizing it somehow.
Cam Geer: Transaction analysis is valuable because it’s concrete, based in fact. At PayPal, we talked about that internally a lot. Kyle Den Hartog: Comes down to the economics of computation and the time to manage it. Why did email naturally centralize? It’s cheaper for a centralized service to manage all that, economics of scale. Second, the amount of time I have to spend to manage it. I’m also the lazy user. How do we develop around the lazy user so they can manage on their own?
Doc: Spam had an awful lot to do about it. Email was too open. In my case, I ran my own mail server from 1995 until last year. Bad guys set up something inside of it and used it to spam the world. SpamAssassin couldn’t do the job. Contracted it out to RackSpace. I’ve got a contractor who is handling that for me. But you’re right, the computation is cheaper. The big back end for all kinds of stuff. AWS. Zoom is using AWS. That’s how they scale.
Adrian: Yes there’s computation cost to those running their own auth server. But the problem of configuration, can it be inherited from someone who can choose. Choose your caucus and inherit complicated policies. Then the two dimensions Kyle mentioned can be separable.
Doc: Something Grace said is important. We need to work on these things together. Easy to gather people in conversation. Hard to get developers to collaborate. There is not one breakout thing yet. VRM is a terrible name in the first place. Someone else came up with it. I think M2B is better than VRM. Not just consumers. I’ve been troubled by the degree the term self-sovereign by Devin 10 years ago, the idea we should be in charge of ourselves. 90+% of the SSI talk has been about big companies. How about us working together? Grace Rachmany: I wrote a whole book about how we work together. We don’t even think of ourselves about being inside a community. Our language is competition. Open source has disproved that quite a lot. Women tend not to operate that way. The system is tilted away from collaboration and toward winner take all. Everything can be resolved in communication. Social distancing will further fragment our community. Recognize ourselves as part of a community, even with competitors. We’re not competing for anything anymore.
Doc: I know of at least two intractable conflicts in the last 2 days resolved exactly that way. I already know what you think. In both cases it got worked out because they talked. Self-sovereign, what Devin meant about that, as a teacher of mostly little kids, everybody learned on their own, and share it with each other. Wrote a piece for Linux Journal about kids teaching each other. The word fail to them means first attempt in learning, rather than competing for grades. In the first 25 years of the internet, we kind of ignored the individuals. Survivalists own this sovereign thing as well.
Grace: Our ontology is so prejudiced. Everybody learns alone. We speak that way all the time.
David Huseby: Adrian’s comment not able to build fully self-sovereign company. Timothy Ruff, etc., myself, we think the correct model is an edge service. If you are doing a function focus on what not who, skip identification and go straight to authorization, there is a way to make money. You need standard protocols and standard data formats. Spent political capitals before W3C in Amsterdam...if a company is dedicated to principles of user sovereignity, you don’t have to compromise.
Adrian: Where does separation of concerns come in?
David: If I don’t need to do anything about you...I’m bumping up against NDAs. There is a lot of theory about this, people building companies around this.
Adrian: Your car can avoid other cars. As long as you make the car, applies to Apple/Google coronavirus search, Apple and Google want to control the app store. You can make lots of money on all aspects of this, including being the operator and data storage, but if you won’t introduce a protocol in the middle of your own platform and introduce substitutability in the middle. Nothing in cars.
David: Not true. Automotive grade Linux. We’ll see. I know what you’re saying Adrian. It is about protocols and standard data formats.
Adrian: I’m saying it’s about separation of concerns.
David: I would introduce it even in my own company.
Adrain: Change your credit card without changing your bank.
David: I do agree with you Adrian. Just using different words.
Nadar Helmy: The point Dave made about spotting economic opportunity, making it more lucrative to provide an edge service. Doc to your point on the human side, no barrier to entry, a no-brainer to join via social or privacy aspects. They have to work in concert. Without economic incentive it’s dead in the water.
Phil Windley: Sovereign raises hackles with some people. I talk about autonomy. We are autonomous by default. We give it out in certain circumstances to do things. On the Web, we are always in someone else’s administrative system. Autonomy starts with architecture. That is the key thing we have to get right. So it won’t get hijacked later on by even well-intentioned people who make a mistake.
Doc: Example of architecture?
Phil: Pure DIDs, an architectural feature, necessary but not sufficient, reduces our need to be inside someone else’s administrative domain. Structural design that makes us autonomous by default rather than administrative by default.
Kyle: The direction I was thinking as the counteraction for management. From a development perspective I haven’t figured out how to do that. I want users to be able to select a trust framework, like a CA, root certificates. We’ve gotten close many times, but there’s always one slipup that manages to push us in the wrong direction. It’s cheaper for a business to run software on my device rather than running it in the cloud. It takes a perfect storm and that perfect storm can go away very quickly.
Marc Davis: When the internet started, meant setting up your own server, creating your own domain. Delegating your autonomy to services that handle all that complexity resulted in us losing that autonomy. Replacement systems would have the simplicity we have today with the autonomy that was once lost. Dave would say a server that handles all that. That’s the big challenge. The complexity has led people to sacrifice autonomy for simplicity.
Doc: I haven’t really delegated my authority over my email. Just jobbing out something to a substitutable service. Substitutability is the key thing.
Marc: Gmail is not just managing your email but reading your email. That’s the tradeoff. The terms are unfair and don’t tilt toward autonomy.
Kyle: One of the things I’ve realized on this, my idea for the longest time, the router is the one thing that’s held up. People have to handle their own routers today. I would run my own cloud agent in the router, a $200 device that’s a Raspberry Pi. ISPs still want to manage this for people, still. I want to tackle the router market to make this easier.
Doc: I own the Santa Barbara router. The one in NY is rented from Spectrum. I don’t have a sense of running that one. I have a Eero (sp??) mesh wi-fi here, set to dumb. Eero owned by Yahoo. Hacking that is a real important thing. Moving from dumb endpoint out to rest of world is critical. How to do that I don’t know.
Adrian: The Freedom Box project is exactly that. A noncommercial thing. A bit of an ability to inherit from others. Will take another decade to make it simple. Has to be done as a non-commerical thing.
Doc: Johannes Ernst and Marcus Sabadello do this.
David: All my traffic is always over the Tor network. Throws away the sandbox at the end of the day.
Doc: Tor just slowed the shit out of everything.
David: I always turn the safety stuff off in the Tor browser. Cookies can track me from one page to the next, but there’s nothing persistent, forgets everything at the end of the session. I can browse the web unfiltered and still sleep well at night. Tor is the best solution we have now...Kyle said something about routers earlier. Some people are asking me about this 5am project. The hope is if it’s possible to use a fully decentralized solution for discovering other nodes, and also to design a protocols, the Last Known Wherabouts protocol, to reconnect in a mobile, disconnected world, without a DNS system in a decentralized way, imagine a router that looks at IP addresses are noise, picks out the signal. That will then get us fully independent of a naming system. Zucko (??) triangle flattes to a line.
Adrian: How is 5G going to factor into this?
David: Light the towers on fire . :)...I want to reinvent the Web from scratch. Maybe it’s something that grows out of the wallet.
Doc: JLINC has something outside the browser. Look at what they’ve got. John Wunderlich is in Toronto.
David: Iain is in Scotland somewhere. I said earlier in my presentation whatever solution we build should be app-based. App users are much larger than the web now. An existential threat to Mozilla. Maybe we should be looking at apps.
Doc: Maybe the Web is AM radio.
David: Network news groups still exist too. Whatever we build next, just to hold myself accountable as I write new systems. My biggest criticism of DID com is that it’s HTTP REST-based...how to route packets through a network in a privacy-addressed way that doesn’t rely on static IP addressing or naming. The biggest problem with TLS was it tried to be agnostic. I’m coming for DID COM. I’m going to light it on fire, we need to make it better. Slowly building a coalition. The new version of MegaOm or Group Noise (sp??) we would be on to something. A routing fabric that is anonymous and private by default.
Jim Fenton: The problem I have with apps, I feel I have so much less transparency about apps on my mobile devices than I do on the Web. I use privacy-enhancing browser plug-ins. I don’t have those opportunities with my mobile devices.
Kyle: One of the things worth mentioning here, the DID based on routing layer is where you ultimately do have to start. You do have to push down to that layer but you’re fighting protocol legacy aspects of it. As a protocol grows in size, it becomes harder for it to die. Harder to change, harder to replace.
David: I don’t disagree. There is a phone from Purism. I had a hand in designing when I was at Mozilla. A phone with a real security story. Decentralized, encrypted communication. It’s a step in the right direction. It runs stock Linux. It has an app ecosystem on it. Matrix, I like for a lot of reasons. The Libram 5 phone and Matrix infrastructure is a step toward decentralization and increasing user sovereignity. I didn’t even get a free phone.
Jim: What was the one a few years ago?
Doc: There were a lot of those.
David: Firefox tried to do it, but moved away from it. A $7 smartphone in India. Ran Firefox OS. Became Tor WiFi router.
Doc: It was loved by a lot of people in the less-developed world.
David: Google gave away Android phones for free for $100 per phone to keep us out of the market.
Doc: We haven’t talked about Inrupt or what TBL is doing with Solid.
David: I love what TBL is doing with Solid. What he’s acknowledging is we need standard protocols and data format around our data. But that’s just one slice of a pizza. I sent you a phone book of a email to you the other day. There are more problems to be solved. He’s solving the persistent data problem which is only one of nine.
Adrian: The session I did yesterday morning was about the paper about Inrupt and Solid. I put a link in the chat. It is philosophically exactly the right thing to do, but not doing the standards. I know from people that have worked there they could be doing. It’s a shame.
Doc: RINA is a forklift change that’s probably not going to happen.
Cam Geer: I sent you a link.
Doc: They’re in Boston
Saved Zoom Chat:
15:32:32 From David Huseby : LOL
15:32:42 From David Huseby : Phil Windly, the radical fighter
15:33:23 From Lisa LeVasseur : didn't know you were here Dave
15:33:40 From David Huseby : I live in Summerlin
15:34:38 From Lisa LeVasseur : huh.me too.
15:34:54 From David Huseby : 215 + Sahara
15:35:08 From David Huseby : Traccia
15:35:18 From Lisa LeVasseur : 215 & lake mead...
15:35:29 From David Huseby : ahahaha
15:35:34 From David Huseby : we're basically neighbors
15:35:38 From Lisa LeVasseur : yup.
15:35:50 From Lisa LeVasseur : I think joyce mentioned that to me last year, now that I think about it.
15:35:53 From David Huseby : awesome we have to give each other corona virus sometime soon
15:35:59 From Lisa LeVasseur : exactly.
15:36:59 From Lisa LeVasseur : love that background, Doc.
15:37:57 From Micah McG : http://blog.harvard.edu/vrm
15:38:24 From David Huseby : we're there now
15:38:54 From Timothy Ruff : Get a room ;)
15:39:06 From Micah McG : Oops sry blog link not working
15:39:46 From Micah McG : http://customercommons.org/
15:40:40 From Lisa LeVasseur : http://standards.ieee.org/project/7012.html
15:42:07 From David Huseby : : )
15:42:41 From Jim Fenton : SMTP is end-to-end? News to me…
15:42:55 From Micah McG : http://www.jlinc.com/technology
15:44:28 From Micah McG : http://www.me2balliance.org/
15:48:11 From Lisa LeVasseur : two pieces of kindling for that: (1) socializing the harms dictionary, and (2) getting the score or a label on products
15:49:54 From dsearls : Dave: "get on the same page"
15:50:01 From Grace Rachmany : it doesn't look like we will need to burn everything down-- it's self-immolating before our eyes. 15:50:19 From Lisa LeVasseur : yeah--sunlight is disinfecting.
15:51:26 From Lisa LeVasseur : (oh god....pls forgive the completely unintended connection to our prez.)
15:51:32 From Grace Rachmany : I think that you need to be careful about "making money" as a primary driver.
15:52:02 From Cam Geer : with a clear principles to build the right tools
15:52:09 From Adrian Gropper : q+
15:52:25 From Lisa LeVasseur : are those principles published? pls share link
15:52:38 From Grace Rachmany : This is also extremely difficult to build. I've been working with some of the projects that are trying to do this and from what I can see there are a lot of good reasons why this is incredibly difficult technologically.
15:52:44 From David Huseby : I am a purpose-driven purpose
15:52:49 From David Huseby : money means very little to me
15:53:26 From Dee Platero : @David are you planning on doing more sessions? I would like to hear more about purpose-driven development.
15:53:27 From David Huseby : @Lisa, they aren't
15:53:36 From David Huseby : I will be putting them in the notes
15:53:45 From David Huseby : and I'll drop them on my web server later tonight
15:53:49 From David Huseby : I will reprise my talks tomorrow
15:53:52 From Lisa LeVasseur : thanks!
15:54:09 From David Huseby : I disagree with that
15:54:14 From David Huseby : a commercial edge service
15:54:16 From Grace Rachmany : +1
15:57:12 From Lisa LeVasseur : I will also mention that no current product is poised to fulfill the Me2BA best practices.
15:57:49 From Lisa LeVasseur : Virtually all fail with not allowing individuals to manage the Me2B Relationship, or to specify the terms of the relationship.
15:58:43 From David Huseby : I can tell you exactly why it centralized
15:58:53 From David Huseby : my nine problems of distributed systems tells you
15:59:37 From Jim Fenton : Agree with Doc that spam had a lot to do with it. Dealing with spam is much more effective at scale.
15:59:43 From Grace Rachmany : @David, agreed. This is very difficult to develop. I think there are some profound efforts going on right now, including your work. These are difficult problems we need to work on together.
15:59:55 From Lisa LeVasseur : Going to jump to Kaliya's session. Great discussion--hope someone's taking good notes for the rest of it!
16:00:48 From Gabe Cohen : They just signed up for oracle cloud!
16:01:01 From Gabe Cohen : http://techcrunch.com/2020/04/28/in-surprise-choice-zoom-hitches-wagon-to-oracle-for-growing-infrastructure-needs/
16:01:27 From David Huseby : I bet Oracle use pricing to woo them
16:01:31 From David Huseby : free for one year
16:01:33 From David Huseby : or whatever
16:01:53 From David Huseby : with the CNCF and kubernetes standard, the underlying cloud is mostly irrelevant these days
16:01:58 From Micah McG : Adrian Grooper: if configuration is shared — great pint!
16:02:08 From Micah McG : *point
16:04:21 From David Huseby : my 5am project is about a fully user sovereign, fully decentralized collaboration tool.
16:04:46 From David Huseby : secure scuttlebutt + git + mailing lists + twitter + facebook + jira
16:04:51 From Kyle Den Hartog : @Dave, I need to follow up with you later in the gardens to get the low down
16:05:05 From David Huseby : @Kyle, I'll find you
16:05:08 From David Huseby : I know where you live
16:05:21 From Kyle Den Hartog : The bottom of the world :)
16:05:48 From dsearls : Grace: "Everything can be resolved in communication"
16:06:00 From Dee Platero : ^ +1
16:06:10 From Nader Helmy : I'd like to know more about the 5am project Dave. Fully supportive of the concept
16:07:57 From Tom Jones : montessori education
16:08:01 From David Huseby : @Nader it is largely theoretical at this point. But at least I now have a set of principles and an understanding of the problems that need to be solved.
16:08:15 From David Huseby : +1 Tom (my kids are Montessori kids)
16:08:38 From Marc Davis : The internet came out of government funded research, so one issue Doc seems to be addressing is the funding mechanism for developing a working self-sovereign architecture. The other question I have is if that self-sovereign architecture is layered on top of the web, which is layered on top of the internet, or is it replacing the web layer, or also replacing the internet layer?
16:08:39 From Tom Jones : My wife & I started a school
16:09:07 From David Huseby : burn down the web
16:09:12 From David Huseby : can't use the web
16:09:23 From Marc Davis : @David Huseby. I knew you’d say that ;-)
16:09:49 From Tom Jones : Just revert to ip - build up from their - trust over ip
16:09:57 From Dee Platero : As a new developer getting into the identity space - where do you suggest I start?
16:10:22 From Grace Rachmany : Internet came out of the government's need for an anti-fragile system and the concept of distributed computing was the way they wanted to do that. The internet has become fragile because of the centralization of a number of services. 16:10:37 From dsearls : http://www.linuxjournal.com/content/kids-take-over-0
16:10:52 From Tom Jones : The only centralization is the dns - what’s wrong with that
16:10:58 From dsearls : That's Devon Loffreto's project. Actually his family's project.
16:11:16 From dsearls : Key point, Grace.
16:12:04 From dsearls : I've always wanted DAs: disclosure agreements. You have to tell others.
16:12:13 From Grace Rachmany : Exactly, NDAs and profit-making come out of a system that puts self-interest in front of common good. Common good must be slave to the common good.
16:12:15 From dsearls : Open source is that.
16:13:10 From dsearls : I feel like I'm Adrian's Hors d'oeuvre tray. :-)
16:13:39 From Marc Davis : So does the development and deployment of a self-sovereign architecture need to be developed within structures and processes that are not corporate, but more of a collective or coop where economic incentives are also collective?
16:15:10 From David Huseby : @Dee you should start by getting radicalized and become an absolutist like me : )
16:15:23 From Kyle Den Hartog : LMAO
16:15:24 From Dee Platero : YAS!!!
16:15:34 From Dee Platero : I'm in :D
16:15:46 From David Huseby : just hang out with me and i'll wear you down until you agree just to get me to shut up
16:15:54 From David Huseby : : )
16:15:56 From Dee Platero : hahahaha!!
16:16:46 From Grace Rachmany : @Marc, I would say yes. We are at the edge of collapse of our existing financial systems and we have an opportunity to define what an economy is, what money is and what it means to provide value. Our current money is almost completely anti-value.
16:17:00 From dsearls : Phil: think autonomous if sovereign troubles you. It's not the default model.
16:17:21 From dsearls : Autonomy starts with architecture. Build it in structurally.
16:18:31 From dsearls : DIDs for example. Necessary but not sufficient. But part of a structural design that makes us autonomous by design rather than administrated by design.
16:18:36 From Grace Rachmany : To Windley's point, it is useful to think about the Cryptographic Autonomy License as a potential structure for that.
16:18:41 From dsearls : Phil, make sure I have that something like right.
16:18:45 From Nader Helmy : @Grace totally agree but I also want to avoid the idea that people on a human level are motivated by economic incentives. I think corporations and businesses obviously are, but I cringe a bit when people describe a "radical" future where people... can monetize their data. I think people are motivated by more organic reasons
16:18:57 From David Huseby : @Grace hodl bitcoin
16:19:35 From Nader Helmy : @Grace I should say motivated *only* or *primarily* by economic incentives
16:19:51 From Grace Rachmany : @Nader, I agree. Our data is not going to be worth much monetarily. We should not have to sell our autonomy for money.
16:20:09 From David Huseby : To Doc's point earlier, it wasn't the web that ended AOL. To be more precise, it was Mozilla Firefox
16:20:10 From Grace Rachmany : and I also agree that people aren't motivated by economic incentives nearly as much as we pretend.
16:20:33 From David Huseby : it was open source software designed to be 100% standards compliant and respected users' interests.
16:20:38 From David Huseby : that's a model we can replicate
16:20:45 From windley : Our default model for the web was one of location.
16:20:48 From Kyle Den Hartog : +1
16:20:50 From Grace Rachmany : People are only motivated by economic incentives up to the survival level. When people want to get rich, it's usually about something like security or status, not about economic incentive.
16:20:59 From Nicky Hickman : @ Nader - most people didn't used to be paid for their labour in the feudal system, then we had the black death and the scarcity of labour and then people finally got paid for their labour, isn't data similar?
16:21:04 From Micah McG : @David didn’t DSL kill aol?
16:21:23 From David Huseby : the web's reliance on DNS is a major weakness
16:21:36 From David Huseby : don't need DNS for discovery
16:21:39 From David Huseby : or coherence
16:21:41 From Jim Fenton : DNSSEC FTW
16:21:53 From Grace Rachmany : +1 David about DNS
16:22:05 From Micah McG : +1 David about DNS
16:22:26 From David Huseby : @Micah, no. AOL was dead by 2000. The first DSL/cable was ubiquitous in 2001
16:22:32 From David Huseby : Firefox was released in 1998
16:22:42 From David Huseby : AOL's decline started at almost the same time
16:22:44 From David Huseby : sure sure
16:22:52 From David Huseby : correlation is not causation
16:23:09 From dsearls : Note: letting Rackspace handle searls.com mail is not delegating autonomy. It's just handing off work to a substitutable service.
16:23:24 From David Huseby : but I'm almost certain that all web pages working in Firefox was a major piece of AOL losing any value add through providing a curated walled garden
16:23:30 From David Huseby : it was finally safe to wander the hinterlands
16:24:24 From mahod mah : Yes.. AOL fell apart due to walled garden
16:24:24 From Micah McG : @david: sometimes it is;)
16:24:27 From Grace Rachmany : I've seen at least 2 attempts in the DLT space to try to create self-hosting in a distributed manner.
16:24:27 From Marc Davis : @Kyle yes!
16:24:28 From dsearls : Kyle: we all still have our own routers.
16:24:32 From Jim Fenton : @dsearls The fact that you have your own domain gives you a lot more options than the people that use gmail addresses or addresses provided by their ISP.
16:24:34 From Adrian Gropper : That’s the Freedom Box project
16:24:44 From David Huseby : @Kyle stop thinking in terms of IP and DNS
16:24:59 From Marc Davis : http://freedombox.org/
16:25:14 From David Huseby : what if your home router used Last Known Whereabouts protocol to maintain statistical connections to the nodes you need to talk to
16:25:26 From Grace Rachmany : The transport layer makes us vulnerable as well, and we haven't talked too much about that.
16:25:27 From David Huseby : IP is just a transport but it is fully dynamic from you perspective
16:25:47 From Cam Geer : +1 grace — on the transport layer
16:25:59 From Micah McG : @david - what about mobile?
16:26:09 From David Huseby : @Grace Tor 100% of the time and sleep better
16:26:22 From Kyle Den Hartog : @David I think that would be amazing to handle it
16:26:38 From Kyle Den Hartog : @Adrian yes it is, that’s what make it obvious to me
16:27:16 From dsearls : Adrian: the Freedom Box is a good example of a router with bonus autonomous features.
16:27:38 From dsearls : Tor is another one, using the tor browser with first party isolation.
16:27:40 From Kyle Den Hartog : Wait how is your video so clear over tor?
16:27:41 From Grace Rachmany : The Holoports are also an attempt at making it easy to be a host.
16:28:23 From Micah McG : TOR has become a lot faster lately
16:28:42 From Kyle Den Hartog : Hmm this is convincing me to at least try it again
16:29:30 From Micah McG : The biggest pain of TOR was all the cloud flare captcha2. That’s been greatly diminished too
16:30:17 From Micah McG : Anybody using Matrix Protocol?
16:30:55 From Grace Rachmany : That's a bit like how IPFS works
16:31:09 From dsearls : Dave: imagine a router that looks at signal amongst noise. We can be independent of a naming system. 16:32:56 From Marc Davis : @David Huseby and @Kyle Den Hartog, what layers of the current internet and web stack can be used as is vs. have to be replaced/redesigned?
16:33:20 From Kyle Den Hartog : I would assume you keep to the IP layer and redesign from there
16:33:42 From Kyle Den Hartog : I could even see IP layer being replaced with DIDs
16:33:43 From dsearls : Dave: the app world is bigger than the Web world now
16:34:03 From Grace Rachmany : +1 on creating a new infrastructure
16:34:27 From Grace Rachmany : Web 3 is trying to address that and the first attempts are extremely clunky, but that's to be expected of first attempts on anything.
16:34:37 From dsearls : The web won't die soon.
16:34:39 From Grace Rachmany : Yes, also agree with the replacement with some kind of DID
16:34:52 From Grace Rachmany : AM radio didn't die yet either.
16:35:01 From dsearls : Sam Curran had a thing about privacy enhanced routing.
16:35:03 From Nader Helmy : DIDComm is designed to be transport agnostic. it’s not there yet but that’s the intention
16:35:13 From dsearls : Agree about AM radio. I'm still into it. :-)
16:35:31 From Nader Helmy : please do Dave!!!
16:36:09 From Grace Rachmany : @David, is Secure Scuttlebutt looking towards that type of implementation?
16:36:30 From Phil Windley : wait. format agnostic is ok. but the transport should be opinionated?
16:36:41 From Kyle Den Hartog : To a degree they are. We’ve been working with Dominic Tarr a bit and he’s taken a look at JWMs and DIDComm
16:36:59 From David Huseby : opinionated means that we pick one encryption algorithm and one protocol and don't allow for "negotiation"
16:37:05 From Grace Rachmany : I also worry about the mobile devices themselves...
16:37:12 From David Huseby : like the new wireguard in linux
16:37:21 From dsearls : Jim Fenton on apps vs. Web.: apps are opaque.
16:38:17 From Marc Davis : @David Huseby are you familiar with http://inrupt.com/ and http://solid.mit.edu/ and do you think they don’t rearchitect enough of the existing web stack?
16:38:20 From Phil Windley : feels like we’ve rabbit holed a bit better
16:38:27 From dsearls : Kyle: the DID based browsing layer is where you have to start. A protocol grows in size and becomes harder to die.
16:38:41 From dsearls : BTW, I have friends who are all about replacing TCP/IP with RINA.
16:39:34 From Grace Rachmany : @marc do you have more info on inrupt? I was not really able from their website to understand what they are doing as far as tech stack
16:39:42 From Micah McG : Yes Matrix!
16:40:09 From Micah McG : Sold me
16:40:26 From mahod mah : phil zimmerman’s black phone
16:41:35 From Micah McG : Don’t be evil
16:41:39 From mahod mah : Standard Oil in action
16:42:30 From Adrian Gropper : Inrupt: http://docs.google.com/document/d/1KX6Xcm_jAzj_CWMhoYjFBOX7KXE8vVEgYHua6Kj_AKo/edit
16:42:45 From Kyle Den Hartog : This is SSB’s take at DIDComm messaging format: http://github.com/ssbc/envelope-spec
16:43:34 From David Huseby : @Marc solid is a possible solution for persistent state. But there are 8 other fundamental problems of distributed systems.
16:43:43 From Kyle Den Hartog : @Doc is this this RINA stuff you’re referring to? http://csr.bu.edu/rina/about.html
16:44:13 From Cam Geer : http://puri.sm/products/librem-5/
16:44:21 From David Huseby :
Nine Problems of Distributed Systems
Discovery
Introduction
Coherence
Public Service
Trust
Privacy
Coordination
Membership
Persistent State
16:44:42 From Micah McG : 👍 16:45:04 From Grace Rachmany : Thanks! 16:45:11 From Wenjing Chu : Thanks! 16:45:15 From Marc Davis : @Grace this website perhaps: https://solidproject.org/
