Safenetwork sustainability concerns - Bandwidth has an ongoing cost however Safenetwork is a pay once, benefit forever model


The chunks address is a function of its hash and if when the group goes to store the chunk finds it already existing then the actual store does not happen since its a duplicate. Remember self encryption will cause the exact same file to have the exact same chunks

Well you are wrong.

since when? That doesn’t even happen. The farmer is not retrieving cached data, its nonsensical since cached data is data that has been got before.

No. What do you think caching is?

That example was also a gross over simplication to attempt to see its effects. The reality is not quite so extreme and it varies between data types. See my post for a explanation.


Exactly and if people try to make it different then its not the SAFE network anymore, its another network. Rental means you don’t own the data but the one who rents it to you does. Bye bye ownership. Bye bye persistent Data, bye bye valuable data.

This is one reason I and others have so defended SAFE’s model since its important to the world, to change the way we are being railroaded into our data being owned by someone else.


If someone wants an example of people happy to supply free use of their computer, storage, phone line for the purposes of communications, downloading information and sharing information (messages & upload) they can look at the 1980’s BBS (Bulletin Board System) that were prolific around that time.

While it doesn’t directly support any particular model it does show how quite a number of people in each community were willing to donate their spare resources for others to use. The problems of SPAM in these systems wasn’t the same as for a fully interconnected network like the internet or SAFE.

A video from the eighties which is an episode of the “Computer Chronicles” focused on modems and Bulletin Boards. And Yes I often used them for many things.


I didn’t mean a spatial distribution, but a temporal one depending on age, something like data is accessed 100 times in the first day, 50 times the second one, 20 times the third one, … with real figures provided by a Maidsafe study.


I meant growth of data.

Usage is like a Gauss curve, meaning that a file can never be considered not used any more. But below a certain usage threshold it should be considered not lucrative enough by farmers. I consider this happens after 3 months and IMO this delay is very optimistic for farmers. You don’t think so, but the problem is that Maidsafe didn’t provide any studies about it.

Yes, my Excel file is too simplistic to model them. There will be MDs that will be permanently updated, but not all. Also to balance that, there will Immutable Data that won’t be accessible at all and so cannot provide any rewards: If you update a file then the previous chunks are still present in the network but are not referenced anymore by the MD entry that pointed to them. Previously this was true only for files > 3 KB but now this is true for any file size.

Data deduplication lowers the ratio popular data / total data. Also popular data will be fetched from cache. All this means that the number of GETs coming from popular data won’t increase farming rewards that much.

It should be Maidsafe’s responsibility to provide such a study.


As an example a popular video is downloaded by many people. The video data is stored permanently as chunks all over the network. When the video is downloaded many times the chunks get cashed, copied and held temporarily by many other nodes than the nodes that hold the permanent chunks. And I was thinking that those extra gets from the cashed data could generate farming rewards in addition to ordinary GETs.


I wasn’t even born until 85 so I totally missed this era of the net. :smile:


Sorry if someone already responded - I didn’t see someone referring to it within the following 50 or so posts… And I wasn’t capable of reading everything here…

This is exactly how I thought it is planned


You’ve got to stop thinking our data is being owned by someone else with other models, with the recurring fee model or other models, we are still the only ones who have our data. It’s not being owned by anyone. You could also have the data being forever there by pay for its access


And I consider this wrong for so many types of files. It may be true for some but not for most large files.

The Disney effect for one. This can be kids and timeless classics, or a limited effect for a TV series where people are just starting to be interested and start from the beginning. There are so many TV series that even 15% in a region on release is considered massive viewing, but the sales of DVDs and Netflix/Hulu/Stan/etc show a lot more watch TV series long after release (usually more than double the release figures). Yet because of the number of TV series “worth” watching no one can keep up to date with them all and as Netflix shows the old stuff is worth their disk and bandwidth costs to keep on file.

The evidence out there shows that while files are accessed less the usage of the files is still not small even after years and worth the cost to them to keep available. So the 3 months being generous is plainly false from both experience and commercial considerations.

File type usage is very important and movies being a major type of file that is both large and shows that a good portion are still used well above insignificant after years kinda of shows that the simplistic view that 3 months is way too short and your parameters need tweaking.

So while your examination is good the time frame and parameters are not right.

And as I said before I agree that overall file usage drops off as the file ages and is a important aspect of the payment model, it is certain not 3 months.

EDIT1: In my modeling I never considered websites and wide variance of their use of files and/or MDs. This it seems is also going to be a very large usage of data/storage and does not follow your model of fit into the study I read.

EDIT2: @tfa I am trying to find the source of a study that read a while back about file access over time. In the meantime can you give me the source of your all significant access in 1st 3 months please


Oh you want to change reality now. Rental means the person does not own the object being rented. Its a definition of rental and a experiential fact. If someone else holds the delete/don’t delete strings (and determined on your paying rent included) on a file then the other person does not own the file. They may have created it but they don’t own it on that system. Even if an admin can delete a file on you then you don’t own the file, pretty simple really. But I guess that you are so used to not actually owning stuff that you feel having use of something is owning it. Use != ownership

You have a strange view on ownership, or else this is another one of those misinformations to keep the topic going?


And there is another dimension to that. The coin creation attempt which occurs once on average every 1/FR GETs only succeeds if the generated (pseudo random) coin address is available. The effect of this is to reduce the successful coin rewards as the number of coins reduce.

So for a given GET

  • a function determines if a coin attempt is to be made. This occurs on an average of once every 1/FR GETs
  • a function determines the coin address to try and create
  • If the address already has a coin then no coin is created
  • If the address doesn’t have a coin at it then the coin is created and given to the farmer
  • The %age of coin creation successes is going to be (“Coins_Available”/“Max_Possible_Coins”) * 100%
  • Thus as the available coins reduce the successful attempts to create coin reduces.

The effect is that the number of attempts to create a coin from farming is determined by FarmingRate and is always this independent of available coins. But the successful coin creations reduce as available coin reduces.


No you do… Does anyone own the blockchain? Or data stored on sia? I consider if the data is maintained in a decentralised way then the data isn’t owned by someone else. It has no owners, only controllers, and you’re the only human controller, the other is the smart contract which isn’t owned by anyone.

Like I said, it’s not rental, I simply call it recurring fee but you’re not “renting” per say


Strange when did you redefine your desire to rent, and call it by its operation. That is recurring fees while the person is using the things being rented. I don’t care what you call it, it still is not true ownership like what SAFE provides. BTW that is what rental is “recurring fees to use something” (including storage containers)

And that isn’t ownership is it.

You know in my era only children equated use with ownership, but I guess times have changed.

Well SAFE gives us real ownership not the pseudo ownership you are peddling.

Oh BTW SAFE only curates (for want of a better word) your files and does not control it in the sense you were using it (as a controller)


You don’t own something you need to rent. If you have to pay to access it you don’t own it.


I have a proposal for this issue of files becoming stale over time, we disagree on that time period but hey this isn’t about the actual time or curves they follow.

David has mentioned often the concept of archival nodes that some people may run. These would be huge vaults with some archival storage methods and these can survive because people will GET enough chunks to cover the cost.

Well it would not take much for chunks to have a #transfers count since last request for the chunk. And once this reaches a network calculated value the chunk is stored on a archival node rather than a normal farming node. (and yes keeping to the target of 8 copies to be kept.

This way when a vault is powered off and the chunks are moved to other nodes then when the chunk is retrieved from one of the other vaults its count is incremented and if over the network calculated value it is sent to one of the archive nodes. That way farmers will have the freshest chunks while the archive nodes end up with older chunks. If a chunk is never access again then all the 8 copies will end up in 8 (maybe 12 for safety) different archive nodes.

Also the effort required is a meta data count that is only incremented when a vault is powered down and the network is creating new copy for each of its chunks Or zeroed when the chunk is requested from the vault… This way it is extremely small additional load for the network.


No, rent should be defined as, recurring fee to use something that you don’t own. And the fee is set by the owner of the property and they can take the property from you after you use it, and that the owner has to be a human. You can own a vacuum cleaner, but pay electricity to use it, hence recurring fee for usage, you can’t say that you’re renting the vaccum cleaner now can you?


By that logic, You’re paying electricity to access your phone therefore you don’t own your phone.


Whatever, you live in a different reality to many of us here. Electricity is not a fee on the vacuum, get real. Electricity is electricity and is another thing you pay for.

It must be easy when you can just change the meaning of things to fit your argument.


I still don’t see how you keep saying people will GET enough to cover the cost without mentioning that all the money paid when people GET is from when people store new data! And sometimes people may not get paid enough depending on if there’s enough people storing new data