Hi everyone
We released FileFileGo UI v0.3.0 with the following features:
- Backup/Restore uploaded metadata to storage providers
- Media viewer inside channels so a post can contain image in a p2p manner
- Dynamic fees functionality to allow storage node operators dynamically set their prices or delegate permission for setting their prices to other users
For changelog of both CLI and UI:
https://github.com/filefilego/filefilego/releases
https://github.com/filefilego/filefilego-ui/releases
Topic: ✅ [ANN] FileFileGo - Decentralized Data Sharing Network (Read 809 times)
Hi everyone.🥳 FileFileGo UI v0.2.2 is released. Please upgrade your software to avoid inconsistencies.
https://filefilego.com/#download
https://filefilego.com/#download
should we restart our storage node when new release of ui?
If you are running the CLI its a good idea to update when a new release is published. The UI uses the CLI binary in the background
Hi everyone.🥳 FileFileGo UI v0.2.2 is released. Please upgrade your software to avoid inconsistencies.
https://filefilego.com/#download
https://filefilego.com/#download
should we restart our storage node when new release of ui?
Hi everyone.🥳 FileFileGo UI v0.2.2 is released. Please upgrade your software to avoid inconsistencies.
https://filefilego.com/#download
https://filefilego.com/#download
As you can see its in the todo list. Will get it done soon, based on its priority
the tutorials are great i suggest you create 2 videos for the installation on windows and macos
Done:
https://www.youtube.com/watch?v=wTV_4QhZZKU
https://www.youtube.com/watch?v=XcjlDN7uvV8
Also updated in: https://filefilego.com/crash_course.html
great support thanks!!i am waiting for this https://github.com/orgs/filefilego/projects/3?pane=issue&itemId=32155322
maybe with a video tutorial?
the tutorials are great i suggest you create 2 videos for the installation on windows and macos
Done:
https://www.youtube.com/watch?v=wTV_4QhZZKU
https://www.youtube.com/watch?v=XcjlDN7uvV8
Also updated in: https://filefilego.com/crash_course.html
Sure, I will add them to my todo list
Appreciate your feedback
Added here: https://github.com/orgs/filefilego/projects/3/views/1?pane=issue&itemId=32851095
Appreciate your feedback
Added here: https://github.com/orgs/filefilego/projects/3/views/1?pane=issue&itemId=32851095
the tutorials are great i suggest you create 2 videos for the installation on windows and macos
Announcement:
Hey everyone! 🤩 We've just released a bunch of video tutorials on our YouTube channel. Be sure to check them out, and as always, we'd love to hear your feedback!
https://filefilego.com/crash_course.html
Hey everyone! 🤩 We've just released a bunch of video tutorials on our YouTube channel. Be sure to check them out, and as always, we'd love to hear your feedback!
https://filefilego.com/crash_course.html
ok tried a 20 gb file with 1% and without encryption and the processing speed is same
It's not exactly the same but the difference is very small. The encryption overhead is there so it really depends on how fast the CPU + Storage is.
In order to visualize it, you could increase the encryption percentage to 100% in the unit test files and you will see it takes more time to process because it encrypts the whole file.
ok tried a 20 gb file with 1% and without encryption and the processing speed is same
i verified the code in file utils.go by changing your unit test and i love it its a very smart way to encrypt and randomize the file data
btw is there performance impact?
btw is there performance impact?
Hi, a file is divided into 1024 segments (or less sometimes) before it's sent to the data requester. Currently, the network encrypts 1% of those segments so it would be around 10 segments to be encrypted using AES or Chacha depending on the hardware support.
The only overhead is the encryption of those 10 segments which is very small and normal these days considering most hardware of storage providers have AES instruction set on the CPU, so to conclude there is a very small overhead which is completely fine.
We also randomize those segments, so the additional overhead is random seek for 1024 segments on the HDD/SSD which is again negligible.
You can verify these at: https://github.com/filefilego/filefilego/blob/master/common/file_utils.go
Code:
func PrepareFileBlockRanges()
func FileSegmentsInfo()
func GenerateRandomIntSlice()
func FileSegmentsInfo()
func GenerateRandomIntSlice()
what is the price?
i verified the code in file utils.go by changing your unit test and i love it its a very smart way to encrypt and randomize the file data
btw is there performance impact?
btw is there performance impact?
You could say a channel is similar to a repository.
In filefilego, storage providers can store data which can be then retrieved by other peers using the file hash. This part of the system is completely off-chain meaning that file hashes are not stored anywhere.
Now here comes the concept of Channels. Let's assume that I am selling my own ML models which are 2TB of data, I could create a channel called "MLModels XYZ" and create different subchannels and categories. In those subchannels you can create entries which are basically posts, or upload files/folders.
Channels are public and if some files are uploaded to a channel, they are first uploaded to a storage provider and then the hash is included in the channels posts/files this way they point back to the file hash (the data will be on-chain)
Please read the docs here: https://filefilego.com/documentation/docs/
In filefilego, storage providers can store data which can be then retrieved by other peers using the file hash. This part of the system is completely off-chain meaning that file hashes are not stored anywhere.
Now here comes the concept of Channels. Let's assume that I am selling my own ML models which are 2TB of data, I could create a channel called "MLModels XYZ" and create different subchannels and categories. In those subchannels you can create entries which are basically posts, or upload files/folders.
Channels are public and if some files are uploaded to a channel, they are first uploaded to a storage provider and then the hash is included in the channels posts/files this way they point back to the file hash (the data will be on-chain)
Please read the docs here: https://filefilego.com/documentation/docs/
thanks for explaining
i cloned filefilego cli from github and i am exploring the code so far its well written. i managed to run the cli and the ui easily
i started a 48 tb storage node on my server what is the right fee to use? am i going to get paid automatically?
You choose your fees, i recommend for now 0.1-0.2 FFG per GB. High fees might be against you since users might pick someone cheaper.
About the payment, yes it's automated and part of the verification algorithm. When the downloader asks decryption key from the verifier, it sends a signature of the local file and by that time the verifier knows if the file is the requested file.
The fees will be unlocked immediately and sent to storage provider
Please check: https://github.com/filefilego/filefilego/blob/master/node/protocols/data_verification/data_verification.go
You could say a channel is similar to a repository.
In filefilego, storage providers can store data which can be then retrieved by other peers using the file hash. This part of the system is completely off-chain meaning that file hashes are not stored anywhere.
Now here comes the concept of Channels. Let's assume that I am selling my own ML models which are 2TB of data, I could create a channel called "MLModels XYZ" and create different subchannels and categories. In those subchannels you can create entries which are basically posts, or upload files/folders.
Channels are public and if some files are uploaded to a channel, they are first uploaded to a storage provider and then the hash is included in the channels posts/files this way they point back to the file hash (the data will be on-chain)
Please read the docs here: https://filefilego.com/documentation/docs/
In filefilego, storage providers can store data which can be then retrieved by other peers using the file hash. This part of the system is completely off-chain meaning that file hashes are not stored anywhere.
Now here comes the concept of Channels. Let's assume that I am selling my own ML models which are 2TB of data, I could create a channel called "MLModels XYZ" and create different subchannels and categories. In those subchannels you can create entries which are basically posts, or upload files/folders.
Channels are public and if some files are uploaded to a channel, they are first uploaded to a storage provider and then the hash is included in the channels posts/files this way they point back to the file hash (the data will be on-chain)
Please read the docs here: https://filefilego.com/documentation/docs/
thanks for explaining
i cloned filefilego cli from github and i am exploring the code so far its well written. i managed to run the cli and the ui easily
i started a 48 tb storage node on my server what is the right fee to use? am i going to get paid automatically?
Hi, Collect and monetize users data? Not one single piece of personal info is stored anywhere on the system including the website so you are wrong.
Luckily for us everything is open-source and people can verify and see what the project does and how it does it!
Show proof otherwise you are showing your lack of knowledge about the subject.
For those who want to understand the algorithm with a simple example:
1. Node 1 stores a 20GB file, and it asks for a fee as a storage fee+retrieval fee
2. Node 2 wants to get that file
There are 2 problems here:
Node 1 sends the file, but node 2 denies it got the file to prevent paying the fees
Or
Node 1 actually doesn't send that file instead some random data which is not the file requested, and collects the fees.
In this case both nodes can not be trusted, so in order to prove that the requested file was successfully transferred we developed the proof of data transfer algorithm which can be found in our repo
Luckily for us everything is open-source and people can verify and see what the project does and how it does it!
Show proof otherwise you are showing your lack of knowledge about the subject.
For those who want to understand the algorithm with a simple example:
1. Node 1 stores a 20GB file, and it asks for a fee as a storage fee+retrieval fee
2. Node 2 wants to get that file
There are 2 problems here:
Node 1 sends the file, but node 2 denies it got the file to prevent paying the fees
Or
Node 1 actually doesn't send that file instead some random data which is not the file requested, and collects the fees.
In this case both nodes can not be trusted, so in order to prove that the requested file was successfully transferred we developed the proof of data transfer algorithm which can be found in our repo
There is no consensus for transferring data from one node to another. There is a dumbsensus. Data travels around the network like purling water in the doomsday river, it passes from estuary and ends up flowing into a billabong with aglae. There is nothing more to be said about this project. It was an attempt to collect and monetize user data, selling ur information to CIA affiliates for their unchecked access to identities and private details of people in the broad crypto space.
Jump to: