There seems to be more projects coming and i was hoping people could advise me the differences and similar goals and technology
There seems to be more projects coming and i was hoping people could advise me the differences and similar goals and technology
Hmmm, I’m looking at page 8 of the Safe Network Primer where they describe that the safe network replaces the elements just below the application layer. Which is what I remember from before. Like us (and IPFS and others for that matter) they use a Kademlia DHT for distributed data storage, but unlike us it doesn’t seem that they include the ideas of the application, and validation of entries into the DHT, etc. Also, the Safe network appears to be monolithic, i.e. there’s just one, which is not the case in Holochain, there’s a separate network per application.
I see how because what the Safe network is for, is building apps that store their date on a distributed network it might seem like the same as what Arthur was saying. But this is the difference between what something is for and what it is. Essentially, from what I’ve just re-read, the Safe Network is a single (albeit distributed) space for storing data that applications can then access through the SafeAPI.
Holochain on the other hand, is actually quite different. It is a pattern for authoring and running distributed applications, i.e. where the applications (not just the data) is run and held in a distributed way. We don’t focus on the data storage space (though, as I said before, we have one) we focus instead on a shared pattern using that kind of space, each application of which has one.
In Holochain each application is it’s own network with it’s own shared DHT for the nodes participating in that application, and the data of which gets validated according to the rules of that application.
I’m not saying this as criticism of MaidSafe, it’s just not quite the same as what we’re up to. Perhaps there’s something else that I’m not seeing about patterns about how to create applications on top of the Safe Network that they’ve developed?
has there been any write up comparing Maidsafe also multi years in development with Holochain? Maid went with Rust Holo with Golang. Both mimic nature… @artbrock @zippy - have you taken time to look over maid ?
The shortest answer:
Maidsafe is distributed file storage.
Holochain is P2P/distributed apps.
You can’t run Twitter, Slack, Chess, Key Manager, Facebook, etc. on Maidsafe.
This is untrue, though I understand why somebody would say that. Two quick (I’m on holiday! ) responses to that:
Also, Art omits to mention other differences, several unique to the SAFEnetwork (autonomous, self encryption, truly anonymous, with an integral scalable efficient cryptocurrency etc)
I want to add that I like Holochain and Art Brock a lot. I’ve looked at their approach (a while ago now) and very much like their ideas, particularly around Ceptr. So I’m keen to see them succeed, though I think their story is stronger in the areas of Ceptr and “current seas” than as a secure, application and data storage platform. It looks to me that they had a look at marketing and begun to pivot towards decentralised Internet and I’m quite skeptical that they are going to deliver anything comparable to the #SAFEnetwork here because the issues we face in terms of data and communications vulnerability are so severe, and I don’t see a server based approach as credible in this respect (which also goes for Solid as much as Holochain).
Please do correct me if you feel I have the wrong idea, but I’m going to share what I think you mean here. I’m not sure if I’m understanding properly or if I’m under a misconception.
The Safe Network, as per your statement of “The SAFEnetwork will support various forms of decentralised computation in the future. The Primer doesn’t explain this because it focuses only on the beta/initial release.” is a bottom-level foundation which is intended for all other protocols and formats to be capable of sitting on top of, and thus enjoying the benefits of The Safe Network while still being capable of being their own systems.
Right now seeing as The Safe Network is currently within its humble beginnings, the focus is primarily on being able to store and share data in the absolute most secure way possible and therefore is not yet focused on the idea of data merely being passed through the network. I’ll provide some illustration as to what I mean by this.
Imagine a computer is running servers for common protocols such as telnet, ssh, irc, ftp or – whatever. In the future, Safe Network Server Applications which support these protocols could be created to allow these protocols to run inside of The Safe Network, allowing for the secure transmission of that data from the server to the computer requesting the data. The data itself may or may not exist within The Safe Network vault system (ie: SAFE-VFS). It can exist there, but would not absolutely be required to.
I am not suggesting that the example protocols be used to communicate outside of The Safe Network but merely suggesting that if server applications designed to run inside of The Safe Network were created to support these protocols, that though the protocols would for the most part work the same way as they already do now – they would be running inside of The Safe Network and thus allowing for greater security.
For example: telnet, irc and ftp as they stand, can be snooped, exploited and manipulated. Where as The Safe Network could theoretically offer protection to what traditionally are unsecure protocols.
So if my understanding is correct, HoloChain is something that could feasibly exist within and be transported through The Safe Network as a supported protocol, if developing the support for doing so is ever desired (not to say that it would or would not be).
Might be a good idea to keep an eye on secure messaging that is in SAFE. It will be interesting to see what this will allow as it is a way to transfer packets between two IDs. One could be your server and the other the machine communicating with it.
But of course incorporating servers is really a step backwards except for specialised reasons.
Compute is definitely on the plans but is not needed for the initial network.
I disagree that incorporating servers is a step backwards, because all a server is, is an application running on a computer. Every computer running The Safe Network Vault is running a Safe Network Server by pure default. So theres really no way around running servers because it is the nature of software. The step forwards I think, is to eliminate the divide between client and server and simply make it that computers are communicating with each other, period.
Every computer is always going to have local storage. In other words: files that are not on a vault and that exist on the hard drive of the computer in exactly the way files exist on any drive right now - be it a DVD, a USB keychain drive, or whatever. The capability of being able to copy and move files to and from these drives, is a client / server sort of process. You always have an origin and a destination, and you always have a piece of software performing the action.
On Linux, which is what I run – the operating system always views things in a client / server style way and it does not care whether or not the origin or destination is local or remote. All it knows, is that data is being transferred from one location to another by way of an application doing the transferring, and that the data is being transferred upon whatever port and via whatever protocol it is being transferred on.
Pending some extreme fundamental change in how computers themselves work – the client / server structure is the nature of the beast. Even BitTorrent / WebTorrent in all of its glorious decentralization, is still a series of peers acting as client (downloaders) and servers (seeds).
Beyond that, yes – the safe secure messaging system does sound by your explanation, like something that other protocols could ride on the back of.
Obviously. But on the internet servers is not just that since under your definition people’s PC’s are serving user data to the more commercial meaning of server servers. (client is serving user input to the server)
You gave a number of services servers which is the usage I was referring to. Servers like web servers, time servers, usenet servers etc.
SAFE is not making the computers servers in that sense, but more in the “NODE” sense where each computer is a link in a decentralised network. And obviously a NODE is supplying, passing on, storing data and responding to events. But they are not servers in the sense of internet servers that the SAFE network is replacing.
Also specialised servers on the internet are dwarfed by the multitudes of web/commerce/SMTP,NNTP,NTP servers out there. SAFE will allow the replacement of these multitudes of servers. Now my specialised IoT gateway server will happily work on SAFE using secure messaging.
Okay, then we’re just using terminology from two slightly different contexts then. I am merely suggesting that commonly used protocols can be adapted to The Safe Network, however I am not suggesting that those protocols would remain indefinitely in use forever by the majority of people.
Telnet still uses Z-Modem to transfer files but this of course does not mean that the majority of people on the Internet use Z-Modem or even telnet for that matter, for file transfers at all.
At the same time, there are still plenty of people who enjoy nostalgia and retrofitting and I believe that they should retain the freedom to do so. For example – I used to run a traditional BBS from 1994 to 2010. The only reason I shut the BBS down in 2010 is not for lack of my love of BBSing, but because the creativity and community support started to become lacking. Modern BBSing is alive and well to this day and is starting to slowly make a comeback, especially in the wake of all of this censorship and other BS. If the BBSing community ever becomes the supportive, creative, innovative community that it used to be, I’d set my BBS back up in a heart beat, using whatever the newest BBS Software is, with the best possible support for the 21st Century internet. I’d love nothing more than to be able to view that lovely ANSI art through a telnet connection made 100% secure by The Safe Network.
So I’m not suggesting taking any steps backwards, I am merely advocating for flexibility and freedom for people to be able to have their proverbial cake, and eat it too. There are many other technology genres out there which people are not going to be quick to give up their love and passion. So these things, as outdated as they might be – will need retrofitting. Just like how a hard core retro gamer is not going to be giving up their love of classic games anytime soon.
Hopefully I’ve made myself more clear.
There are various things that sound a bit like this. Two are where an app is running on SAFE Network (which will not be possible until decentralised computation is available), another is implementing a gateway between http and SAFE - which I think it’s what
@psecdocumentary is talking about. In the latter case you create a weak spot for security, anonymity, censorship, DDoS etc, so while I’m sure it will happen it is of little value wrt the goals of SAFE.
In addition to the above options, it is possible to run desktop applications and web apps which only understand traditional web protocols and for those applications to operate with SAFE as if it is a clearnet server - but they just operate with SAFE, and have no interaction with the clearnet.
The SAFE Plume demo was an example of this where an app written to work with a clearnet Solid server was able to use SAFE instead, using existing Web protocols (LDP in this case).
I think it could be done without the reduced security. All that would need to be done, is the protocol would need to be emulated instead of being literal. For example, like how with a virtual machine, an operating system thinks it is being stalled to a physical computer when in truth, it is being installed inside of an application. Well, for things such as ssh, telnet, irc, etc, etc to work – the server only needs to think its using the protocol. When in realty it is not, but the protocol is being emulated through some digital trickery.
That is the beauty of 21st Century technologies. You can trick software by simply making it think it is using one format when in truth, it is actually using another.
Back in the day we would trick phone modem based softwares to use telnet by making it think it was accepting an incoming phone call but in truth, it wasn’t.
I don’t see how you can have a gateway to a clearnet server (eg for email outside of SAFE) without reducing security etc. Certainly it would be very hard and less secure etc than not doing so.
What you describe in your last post is is the third scenario in my post just above, which I agree does not affect security.
It isn’t clear to me that you are making the distinction, but if you are we agree
I am speaking of using common internet protocols INSIDE of The Safe Network and am absolutely NOT talking about using them outside of it. Imagine Safe Network as a cup. Everything I’m talking about is in the cup, never being spilled out onto the table at any point.
If The Safe Network wants people to have incentive to switch over to it, then all of the protocols and platforms they know and love need to be able to run on it, because when someone loves something enough – they aren’t going to be willing to give it up. You need to make sure they can take it with them.
Plus, seeing as other projects such as HoloChain exists – The Safe Network is going to need to allow other new networks and new protocols to run inside of it. There is no point to dividing the Internet up in to 100 small internets. All of the projects that gain traction and become widely used NEED to be able to communicate with and inside of each other. This is a must whether any of us like it, or not.
I would also like to add that whether we like it or not, privacy is a secondary issue and security is the primary issue. Without the freedom to express grievances as it is stolen from us by censorship, privacy issues are a moot point. You can’t tell people about how your privacy is being violated if there is no way to talk about it.
First and foremost, and again regardless of who does or does not like the truth of the current state of the world – making The Safe Network unable to be taken down and unable to be censored should be PRIORITY #1 and the option for Anonymity should be secondary. I don’t mind if our increasingly totalitarian government knows that I am Dave Kelso from Chicago, Illinois when I voice my thoughts about how the morons who we the ignorant masses have allowed to run this place, should go take a long walk off of a short cliff. Or that the Unazi Kingdom might ban me from their country for disagreeing with them, like they’ve done to Lauren Southern and others. I do however care about my ability to face all of that stuff and to not have that ability stolen from me by shadow bans or having my content nuked because the morons that be have seen fit to claim that I violated some fake community guidelines standard that doesn’t actually exist.
With that said – though I have come to understand why any aspect of Safe Network interacting with the outside Internet directly would be a bad idea for both security and privacy – I do feel that Firewall Gateways are going to be needed if we want to provide people incentive. I am going to explain what I mean by this.
As it stands, if I have a Local Area Network and I setup an Internet Server of some sort, naturally I am going to have a firewall. The outside Internet won’t be able to see anything on the other side of the firewall that I don’t want them to see, which is obviously the whole point of a firewall.
So imagine for example if a regular Internet site were to also run a Safe Network node. Well, that website is already making itself known public. Its ip address is known. It has no desire to hide the fact that its domain name links to its ip address. So that is a choice that the owner of that site is already making. Now imagine that a site such as HiddenService.net decided to add Safe Network content to their service. They already allow browsing of TOR and I2P.
The regular outside internet would only be able to see that Safe Network content is coming from HiddenService.net. The outside internet would NOT be able to see ANYTHING that is happening INSIDE of Safe Network. That would be cloaked. There is literally no way that could be seen. Privacy for people inside of The Safe Network would be maintained.
Such gateways as far as I know, are always read-only from the outside internet. So with such a gateway, people on the regular Internet would not be able to participate on The Safe Network. They would need to use The Safe Network directly in order to do that.
So yes I understand that it would be bad for the Safe Network to touch the regular Internet, but it would not violate security or privacy simply for the regular Internet to be able to see content through a gateway, because the only traffic they would be able to see (using the previous hypothetical example) is that traffic is going to and from HiddenService.net. The regular internet can already see that as it stands. HiddenService.net exists and people can see that data flows to and from it. But that is all they can see. They can see who is contacting that address, and that the address being contacted is providing TOR and I2P content. But they can not see the flow of traffic inside of TOR or I2P.
Still all your picking me up on server terminology my response still had what could be the answer to your desire for SMTP, NNTP, etc style of servers running purely on SAFE.
And yes I get your idea and desire.
The thing to watch is secure messaging. It has the potential to carry payloads between to IDs securely. So in effect a secure layer carrying http(s) packets across the SAFE network between your old time server and the person using that server.
And as long as that server machine does not have a way to communicate apart from the SAFE protocol then it will remain secure but with all the old time insecurities of centralisation of data and non-user owned data, etc.
Now for NNTP (usenet) & bittorrent, I think SAFE will be a drop in replacement for files being transferred across those protocols. And secure messaging will replace SMTP, text messages on NNTP, etc. But still there may be a need for some old style of servers in SAFE which maybe able to use secure messaging to transport the packets.
I get you and I don’t disagree with you. I was merely clarifying.
There is also one thing I need to make a more detailed post on at a later date, an idea that I’ve had for many years now and decentralized networking technology would make the idea possible, where as before then I was busting my brain trying to figure out how it would work.
Brief backstory: before the Internet as we know it, cyberspace consisted of BBSes. Inevitably, the desire to link these systems together by way of communication networks came about, so the first ever Echomail Network – FidoNet – was born. Echomail is comparable to NNTP, but both Echomail and NNTP each have functions the other does not have. Several other Echomail formats were also created and as the server softwares evolved – inevitably, servers were created that were able to support more than one format at once, while storing to the same message archive database. This added vast flexibility and a great many advantages.
I think a new type of BBS Software is needed. By BBS Software I am not referring to a web message board. BBS Softwares include the web, but are not limited to the web. They are their own sophisticated animal, as it were. So just making a clarification.
As Echomail was born from BBS Softwares, a new 21st Century BBS Software must be born in an inverse way. It must be built on top of the network, instead of the network being built on top of it.
I’m not going to go into specifics in this post, I’ll keep it general for now. My goal in this would be the eventual creation of an Echomail Network as a Communications API, which would be inspired by and borrow ideas from the original Echomail Network structure – but would be infinitely more flexible, scalable and openly compatible with anything that wants to plug in and use it. Where as the original Echomail Network formats were so incredibly proprietary that it was difficult to include support for it, which is why it took much longer than it really should have for multi-format softwares to emerge.
Also in the old structure, each Network (capable of both syndicated public message forums as well as syndicated file archives) was its own separate CENTRALIZED Network. In this new format, each separate network would be a part of the collective whole – and instead of using a centralized HQ <–> Region <–> Host <–> Hub <–> Node system that though good for what it was back in the day was still clunky and problematic – each Echomail node would read messages from and save messages to the vault system, just like everything else works on SafeNet.
Echomail Networks still exist and though this may not be the best possible example, here is a screen shot of a web interface for a BBS that carries several Echomail Networks.
As you can see, it looks similar to what most people are used to with regular web message forums, but now imagine that a multitude of websites could share in the conversation in a similar but different way as one might imagine Usenet.
My idea for a 21st Century Echomail Network expands well beyond the traditional capabilities to also include features and aspects that are common on social media, and so on. And I mean the good aspects, not the ones people complain about. Tee-hee
Though it may or may not be difficult to imagine exactly what I’m on about without me making the fully detailed post on the subject that I wish to do at a later date, my point in mentioning it is a simple one: The Safe Network through decentralized networking is going to REVOLUTIONIZE THE LIVING SHIT out of EVERYTHING! It is going to give all of the stuff we’ve always known and love greater power, flexibility, scalability, efficiency, security and more! It is not re-inventing the wheel, but more like going from wooden wheels to rubber, to be sure. I feel that The Safe Network (and any other projects that do similar things) are going to make the Internet a wonderful place for those who love freedom and liberty and creativity – and a nightmare for those who would seek to subvert such things.
One other thing about Holochain – having looked into it a little more, it feels like a more advanced version of my “CHAINMESH” idea that I suggested awhile back. Holochain looks like it could be a good transitional go-between platform from Blockchain Cryptocurrency to Safe Network Cryptocurrency to help people more easily fully transition into Safe Network format. So instead of having to quit one “cold turkey” to go to the other, you let Blockchain and Holochain communicate until the currency in question fully migrates over to Holochain. And then the currency from there would begin the process of migrating over to Safe Network.
So is this correct that Holo is the coin and an app on Holochain and that is a fast release of the larger Ceptr. That they feel is a new versioning of a stack that is unique. Thus programming methodology and servicing of users will be different - If i understand it is how low they go into the semantic tree.
This will be running over the current internet…
Can this be brought into running under maidsafe?
Or as mentioned as a route onto maidsafe?
Holochain - Ceptre on the normal net - also encrypted running in maidsafe?
Im interested to see if maidsafe is a layer under holochain/ceptre.
Or any ideas from Ceptre are just part of the current maidsafe basics.