Pay to delete?


#1

I haven’t seen this question yet, perhaps because it’s too stupid…

I’ve been reading TFM and I thought whether it’d be feasible to accommodate in the code the ability to charge the owner to delete the file/object?

I haven’t looked at the code (and I probably wouldn’t be able to understand it anyway), but if a price could be attached to whatever maps to “delete” operation on immutable content, I guess that’s what I’d like.


#2

Sorry if I am way off base here. Maybe I have the subject matter wrong? Is this about allowing the owner to delete their own stuff or data? Would this be disabling a built in functionality to create artificial scarcity to drive profit? Kind of like the Adobe model of formerly having to deal with un-editable files unless you bought their word processor?

If so reminds me of having to pay to not have your number listed. This pay to delete will become too prohibitively expensive. A shame industry will blossom out of it. Soon only those who can afford to sue will have it work for them. They won’t even have to pay. The rest of us will pay all the market will bear. How about pay to revoke the charters of firms who try this?


#3

I deliberately didn’t discuss possible reasons or use case scenarios (partially because I didn’t think about them - I had in mind only one or two).

I’m just curious whether it’d be possible to implement it without major changes.

That wasn’t one of scenarios I thought about, but at least you could delete it… Compared to the current (default) state where you cannot delete my data at all (and then I can post it again, so you probably wouldn’t want to pay and I wouldn’t ask you to pay because I know you wouldn’t want to).

I don’t have any preconceived ideas about how the feature - should it be possible - would be used. By default you can’t delete anyone’s data anyway, so there probably will be (and it’s not necessary to implement that in MaidSafe) people who will demand payment to take data down.

The reason I asked was to prevent myself from deleting my own data. It’s admittedly a strange idea, that’s why I didn’t want to mention it and instead I asked a pure technical question.


#4

Sorry. I see the utility then I guess. In the fight between Google and the EU court, Google if I have this correct said being forced to delete on demand would open be prohibitive for them. For sure, it was a shot across the bow of privacy mining. But it would seem Google would incur some costs. With MaidSafe as an distributed automated corporation possibly in some of its forks it seems like where practical everything would be revocable.


#5

You don’t ever need to worry about this because nothing is deleted. The file system automatically retains all previously stored files and versions of files.


#6

Maybe I got it wrong, but are you not providing diskspace to get diskspace and pay for extra diskspace that you can’t provide? So basically if you don’t pay and don’t provide any diskspace your files will be deleted. I assume… Hmmmm the data is scattered accross the network so who would you approach to pay, even the people who hold your data don’t know that they hold your data (yeah that sounds strange)


#7

So even private/unshared data from ‘closed/cancelled’ accounts is stored on the network forever?


#8

I was wondering the same thing. Seems like a massive potential waste of resources. I thought you had to refresh to private data, or somethin’.


#9

a) What Luke said: does that mean, then, that the price of storing data on MaidSafe is not time-limited?
Weird.

b) I may not necessarily worry about data deletion, I’m just asking whether it can be done should there be good use case scenarios.


#10

I can’t answer details, its just something I remembered David saying in the early days I was learning about the project. It’s possible I misunderstood, but you can judge for yourselves because I was getting blown away each day as I learned something new about SAFE and started keeping notes…

On Fri, Apr 25, 2014 at 5:49 PM @happybeing wrote:

Is that you delete nothing ever, or you have the ability to retain all versions if the API is so instructed? I thought once no-one referenced a chunk it got deleted?

David:

Very old stuff is archived, hopefully never deleted, but maybe later if we can prove it’s deletable we can). We can reduce network maintenance to almost zero though. So it only takes up a wee bit of room.

~ https://groups.google.com/forum/#!msg/maidsafe-development/2r96Q4wmzK4/OmXBe8EiqR8J

EDIT: I agree this seems wasteful at first sight, but I think it probably works because a significant majority of the nodes on the network will end up being low rank archive nodes, so there’ll be quite a lot of space available for that kind of data. We’ll learn a lot when the network goes live and as it takes off.


#11

Why do we need to delete?
But if you want to delete then just never share it in public. It doesn’t exist if the private key is destroyed.


#12

I said above, I am hoping to get a technical answer whether it’s possible without an extreme amount of reengineering, not to justify it or get validation for a use case scenario.

But to come up with something that’s not relevant just for me, I’ll tie this with @happybeing’s comment: @nicklambert said the same in his post here , although perhaps he formulated that better:

At some time in the future it is envisaged the network will be able to detect such data and remove it from the network.

Let’s say I’d like that out of my 100 useless files (all 1MB large, to make it simple), certain files (say, 10) are less useless than the rest.

Now, since I already paid to store that stuff (and even if I didn’t) I may want to be able to ensure that when Autodelete kicks in, that the less useless files are not targeted first.
If I paid to store them and the SAFE net wants to delete them, I think it should pay me to delete any files.
Alternatively I could pay more for the storage of files that I do not want to have deleted even if they match criteria specified by Autodelete algo (say I’d be willing to pay a 5% premium per each additional year of non-deletion or whatever).

There are other scenarios I have in mind too.


#13

Is there a special advantage here to deletion of the data, as opposed to burning the particular ID associated, and thereby rendering the data inaccessible?


#14

Reclaiming space obviously, but were way ahead, due to de-duplication.

If deletion is allowed, would courts see fit to order self deletion in certain situations or would Governments intentionally destroy their own data.

Deletion tied to smart contracts would be interesting i.e delete certain types of criminal records after 10 years, delete health records at death, delete credit history older than 10 years etc


#15

I don’t think ID’s can be deleted.


#16

Interesting use of smart contracts @chrisfostertv


#17

I didn’t mean to delete the ID, I meant burning, like setting a random password that even you don’t know so that no one can access the ID again. Eventually, if the ID is not tended, and there is no safecoin in the accounts associated with it, the data would presumably be forgotten by the network, if and when the resources are needed. Right?


#18

Currently there is no mechanism for forgetting data, though it may be included at some point.


Introduction to MaidSafe: what it is, how it works, and how it compares to Bitcoin
#19

Who ever would be paid would seem to have absolute control over a part of the network or if it were some sort of aggregator we’d still have the problem of aggregators seeding their own businesses by lobbying for overreach overreach in privacy and encouraging it partner firms. Virus cleaner firms spreading viruses. We already have pay to delete services for celebrities. Its better to not to create incentives for firms having data they shouldn’t have or keep in the first place. Its easier to tell firms that they must bulk dump dated data without exception- their profit is again as usual irrelevant. As irrelevant as Telco arguments claiming their roe concerns should allow them to be something other than dumb pipes or allow them to profit from censorship.

These firms don’t get to profit from withholding or spreading data adverse to end users, it should be jail terms the officers and huge fines if they do. Its not blocking innovation its just keeping innovation from being confused with crime. .


#20

Uhm… wrong. The network has no way of knowing you lost your password (or set some random password).


Introduction to MaidSafe: what it is, how it works, and how it compares to Bitcoin