I’m trying to see if Anytype have the ability to run like a selfhost service. Similar like Joplin with nextcloud. But i don’t know how to do it. Also i don’t understand that this “backing node” stand for, where is that “node”? is there any way to know this?
Thank you very much for this amazing tool
Hi @literaldani, thanks for posting! The self-hosted component of Anytype unfortunately is not yet avialable. It is on the roadmap, though! I believe it is scheduled for around the public release in about half a year, but that is off the top of my head…
There has been some discussion about the unclarity of nomenclature/terminology in general, but the backup/cafe/anytype node is the node where a copy of your data is synced to. It is encrypted so the Anytype team cannot access your data. This is important for you to ensure your data is secure, but it is also important to understand that if you lose your passphrase, you lost the access to your data! The backup node (in my opinion the best name for it) is hosted in AWS and is basically another instance of an Anytype application setup to sync your data to and from.
Awesome explanation, im playing with the format of the objects and its really fun. I will wait for the selfhost solution no problem.
So one question regarding privacy, what is the difference with the data that i have locally (but also sync in the cloud) with notion, from this? is it the same communication?
Thank you very much
You’re welcome! The main difference for me is that if Notion decides to quit or kick you out of your account for whatever reason, you lose access to your data. In short: you don’t truly own the data stored in Notion.
In contrast, if I store data in Anytype, my data is accessible even if the Anytype team quits, goes bankrupt, or otherwise shuts down their services. I won’t be able to use their backup node anymore, but I don’t need that for the tool to work and to access my data. There are two very important notes here:
- As the software is in closed alpha stage, sync to the backup node is mandatory to give the developers as much sync metadata as possible to optimize the sync service. This requirement will go away with (as far as I know) the public release later this year. Until then, you will need to be connected to the Anytype backup node in order to login. In that sense, if the Anytype backup node goes down, you cannot access your data (properly) as can be seen in the various topics about sync issues in the last few days due to issues with the Anytype backup node infrastructure. You can already use other devices on your network as an additional backup node, which allows local sync of data.
- When the Anytype backup node is no longer a requirement, you should be able to start hosting your own (backup) node to become completely independent from the Anytype backup node.
I just wanted to add also, when this becomes available,
Can you make it really easy to set up a self-hosting Node?
Like it would be cool to set this up on various different operating systems!
Mac OS/Windows/Linux - this would be great to just use any OS as a node and if they needed a lot of storage they could just point the storage to external storage. like USB Hard Drives or SMB share (which would be like NAS storage). No complex configuring - just install and some simple settings and be done. (No Code or command line stuff).
NAS - there are so many different Server OS options out there. I would not have the time to list all of them but maybe a few that are well-known or important to professional levels. Also, a lot of these make it easy to install via package installers so no command line stuff.
– Qnap NAS
– Synology NAS
– 45 Drives - Houston UI (Linux OS)
– or any other Linux OS for that matter.
The current plan we have right now is to open source our nodes code and write proper docs on how to setup then on your servers. And we will add UI in the app to change default nodes to custom ones. This is planned for this quarter
I hope a docker image is the way (or one of the ways) the team will go for selfhosting. Can be cross platform and relatively easy to setup.
For Desktop (Linux, Windows, MacOS) I think installing Anytype, leaving Anytype running in the background, and allowing users to set a custom data location should be enough, right?
Yeah, that was what I was thinking also! It’s a great option.
Is there a big difference between the app and node?
It’s a little bit confusing to me when I first used Anytype, my main purpose is to find a tool like Notion but can store any information on my own device, it said Anytype can meet my requirement, but the weird thing is you need to login to use it and the content you created is always synced to somewhere.
Now I can understand a little, this is alpha stage and you need to login in to use backup node as sambouwer said above sync to the backup node is mandatory at this stage. Maybe later after the public release, there is option not use backup function and no need to login to use Anytype.
To be honest, not sure whether there is a (big) difference between client (node) and backup node. @ignatovv is this something you know or should I ask Roman?
It’s already confirmed that once Anytype has publicly launched, you are no longer forced to sync to their backup node. It would be a requirement that you can self host as an alternative, but I think the team already started exploring the options for self hosting.
Thanks for your reply, one more question, do you know if it is possible to use Anytype without login after it is publicly released?
Most likely not. There will be a moment that there are public spaces (similar to a website) so I can imagine that you won’t need an account
It depends on what do you mean by login. Your recovery phrase is basically a encryption key used for well encryption. In this way there are no ability to use anysync protocol without encryption at the moment. But maybe you meant smth else
If you meant will it be mandatory to connect to Anytype network in order to work the answer is no. If you will setup your own network you can even register new accounts directly in new network.