Hi folks `n devs,
I just stumbled upon an app/protocol for a backendless social network: https://scuttlebutt.nz/ .
Discovering that it’s also based on node.js
I asked myself oder it has more in common with TW, and whether we even share some ideals.
Could this be a possible way to store and deploy TW? Or vice versa: Could TW enhance scuttlebutt?
@JanJo “SCUTTLEBUTT” is a fascinating word.
(Water fountain now;
previously a cask of water on a ship.)
Now a networking site in NZ. (NZ = New Zealand, near Australia)
With BACKENDLESS we enter a matrix of whats?
What does “Backendless” mean?
How could it help me?
Just an innocent question.
TT
It is a social network that stores the data decentralized on the devices of the participants and exchanges peer2peer with a pgp based protocol.
I think this video explains quite good what it is an how it works:
1 Like
I’m working on an implementation using the hypercore-protocol as a peer to peer “backend” for TW. It’s a somewhat similar concept but a completely different base. The former Beaker browser was based on the hyper-xxx idea, as it still was named: DAT
4 Likes
I watched. 33 minutes later I’m none the wiser.
I truly have no idea WHAT this is about.
I’m maybe an idiot. But I would like to know what I am missing. 
TT
At the moment, you are missing nothing which could be used with TW for now. I wondered whether this could provide a base for developers to build an app (or enhance for example their app manyverse) to be able to store show and share TW files via P2P.
1 Like
A backend that acts like a server. But I’m still not clear yet if it should work file based or if it should look more like a single database. TW users like files but they do have some disadvantages. Especially if syncing between different users and differen OSes is involved.
The hyper- system has some very nice features out of the box. Eg. Version control is part of the system itself. That’s very nice
On the other hand there is a movement, which is named: “The right to be forgotten”. So it needs to be possible to delete something that should have never been published.
3 Likes
I remember how Noteself did all this using couchDB and PouchDB one local and the other could use a service on the internet, to store tiddlers and their versions. The problem was the free IBM Database was restricted and no longer available.
- It seems to me we need something similar that connects to common and freely hostable databases such as MySQL and add in the ability for Wikis to “subscribe” to more than one tiddler store.
- This would allow one or more online single file wikis (even themselves read only) to save and share selected tiddlers resulting in the ability for open federation,
- avoiding the difficulties sharing node servers online
That’s exactly the problem. The dependency on 3rd party resources or services. With a peer to peer system, there is no server. With hyper-core, syncing is part of the core system
couchDB was used because it is capable to sync as a “multi master” system. It has a built in “error resolution” mechanism. Every other database is significantly more complex to maintain and sync, especially if 1 of the peers is a browser only setup.
2 Likes
btw: That problem could be resolved by setting up your own couch or pouch db server. It is one of the features of yunohost:
2 Likes
In effect scuttlebutt is named a root of beaker on its github-tombstone
And Paul Frazee names the AT-Protocol as the future. Bluesky, the company involved, is hiring. Maybe someone can sneak in and make it a place favorable for TW
1 Like
This is a screenshot of Manyverse the transplatform scuttlebutt app. The format already is .json the machine below node.js. Now imagine a link to wikis on the sidebar would open some sort of tiddlydesktop, which automatically syncs to your phone-app via p2p.
I found another video a video which explains for what scuttlebut is , what the focus of current development is and how it might be used in the future.
https://cinemaphile.com/watch?v=1iL43jj6IM0
It is interesting that the focus of the video shifts to our usecase: usinig ssb to serve apps and how this is done.
Further info on the work of the ssb-ngi-pointer group which works on this topic can be found here
1 Like
I think this discusses the real trends in distributed connections and would be a great partner to TiddlyWiki. Some thoughts;
- Many of the advantages of such a solution are already inherit in TiddlyWiki
- Using this as a backend may be helpful +
- It need not only be the backend although, it could be a messaging system on top of tiddlywiki
- Including where only “changes are messages” so Online TiddlyWiki’s can be “edited” online, because they are the published wiki + the changes. Only the wiki owner can commit the changes to the wiki.
- The distributed app approach may allow easier adoption of tiddlywiki.
Has anyone reached out to the team and told them about tiddlywiki?
I stumbled into the manyverse development chat a bit accidentally… they were nice but not as euphoric as I am…
In January we discussed the hyper-x DAT and AT protocol as base of a backendless deployment.
The company we stumbled upon there bluesky seems to have become the next big thing being discussed to be the new Twitter.
Maybe TW can also make use of the hype…