Bootstrap data to enable instant large data upload

Let’s say that I want to upload 1TB of data right now, that’s not really possible (no wonder we are in Alpha, blablabla, hahaha). Let’s look at a future scenario, what if I want to upload 1TB of data and I don’t have SAFEcoins and don’t got enough money to buy SAFEcoins at an exchange, because 1 SAFEcoins is now worth $10 for instance. This setting makes it clearly impossible for me to upload my 1TB, how do we solve this problem? The idea is that the SAFE Network grows no matter what. A growing network is more valuable than the fiat price of a SAFEcoin IMHO.

I only had the idea of bootstrap data this night, while waking up, so please don’t silence this lamb yet. The idea behind bootstrap data comes close to Alexandria’s “pin to play” (it lets users trade something of value other than money for the content they’re looking for).

Imagine if you can upload data (1TB in my case) on the SAFE Network without immediately paying for it. Basically other people would provide storage and you can make use of that. Huh, but everything that’s uploaded to the SAFE Network costs SAFEcoin, hmmmm maybe not really. In our current Alpha test we’re able to upload data without a “activated” cost.

With bootstrap data people are basically buying time, getting something today that you would have got yesterday or not at all. How would bootstrap data work? You provide storage, an uploader doesn’t pay SAFEcoins immediately, but later on pays the storage provider through farming, pay to the producer or application revenue. Who knows this could be a hollywood blockbuster, that the storage provider could watch one time for free.

Bootstrap data can’t be deleted, unless the digital debt created through uploading it, is fully paid. The question is how do you determine the storage provider and pay out. This is where it gets over my head and I start to get a headache just thinking about it. Bootstrap data is a mechanism to counteract the sudden price rise of SAFEcoin which could make it expensive to upload files. Every file that is uploaded is network growth and if people can’t farm first or pay first, maybe they should have another option. To not have a total floodgate of digital debt, there could be milestone like 10TB bootstrap data. If the next wave of people want to upload bootstrap data, they have to collectively help pay the first wave’s bootstrap data’s debt.

I don’t know if this makes sense or is even possible, but every uploaded file is network growth. Once we have computing on the SAFE Network, the logic is that it wouldn’t make sense to pay a monthly fee + scaling costs to run your app on Amazon AWS, Google or Microsoft. What do you think?

4 Likes

Maybe there is an app here … with a model sort of like youtube. Free to put (the app pays for the put) in the hope that the app investors (who front the money for the put cost) get a return on investment based on the use of the app – use which pays the app owners and then gets divided back out to the app investors.

Keep in mind with regard to the cost of SAFEcoin and the cost of puts:

4 Likes

There is also a very debatable thing here as well. Pay the produce / uploader. We did have a lot of debate that is not finalised yet over this scheme. A huge issue was pay the copyright infringer or uploader and not the artist. An idea to prevent that would be give everyone a free ride to start with, no PtP for a period to allow all the current data to be uploaded. Slightly different scenario, but what if some of the Tb of data you want uploaded is previously known data and folk do not upload/share it, then that’s maybe not good.

However what if even the uploaders of all info were rewarded, even if they never created it? So Ptp form an early stage and copiers / uploaders who do provide a service (making data available) were rewarded?

I do not want to deviate form your OP as this is a tangent, but also is in interesting adjunct, perhaps?

4 Likes

This sounds a lot like a SAFEbank. People can deposit coins and the bank lends out coins. The coins have to be repaid. Depending on the model it could be a “not for profit” style where only the amount lent is repaid OR it could be a business where interest is paid.

Personally I think the idea of bootstrap would not be wise to have in the safenetwork core protocols but as an APP that could be the “lending” idea above or some other system.

Obviously farming is the best way to get coin if you have no other resources (eg money/bitcoin), and if the cost of PUTs is high then farming will be giving coins quickly, and if cost of PUTs low then only need one coins for TB’s

2 Likes

Investors only care about getting a percentage back for money they invest. This is about the growth of the network. It shouldn’t be that it’s impossible to upload 1TB or more on the SAFE Network.

How will we handle large data that want to immediately get on the SAFE Network? bitcoin is just money is priced at $623. The problem (advantage) is that we can do more with SAFEcoin’s. David couldn’t directly answer Ernie’s question “What if I want to upload 40GB?”. So how do we answer that question when people ask it over and over again?

[When you look at uploading data Amazon glacier for example it looks cheap.]
(https://aws.amazon.com/glacier/pricing/) Well if that’s cheap while the price of SAFEcoin (the thing we need to upload data) will go up, how do we manage?

Imagine a SAFEcoin at $100, people would provide resources like crazy, but what guarantees us that people will have enough SAFEcoin to upload files? A high priced SAFEcoin could also be an attack vector. IMHO with a mechanism like bootstrap data, people are incentivized to provide data because they are guaranteed to get SAFEcoins in the long run. So it’s almost like a savings account and the beauty is that there is network growth.

1 Like

We don’t want to incorporate the “I lend you this, you pay me this percentage on top”. When you provide storage, you’ll only get the digital debt that is create at that moment. We should assist the SAFE Network to grow, bootstrap data would be a reason to always provide resources, because you’ll get a return.

Sorry I can’t discuss further I got to help my moms clean :stuck_out_tongue:

The way i would see it is that the network sponsors your PUT, and retains ownership of the data, giving you read rights. Then if you don’t repay it gets deleted. And when you repay the ownership is transferred to your account.

But event his mechanism would be better at the app level, since it need some kind of verification to prevent spam. A major e-mail account or similar.

2 Likes

What is the quota? 1TB, 10 TB? What will prevent misuse of the settings? I can simply create another account and make a new upload.

If somebody uploads populair content and that suddenly gets deleted, I don’t think the network will make a lot of friends.

With bootstrap data three people could pay the bill

  • The first uploader

  • The second uploader

  • People who already got SAFEcoin

  • The first uploaders would pay incrementally

  • The second uploaders would pay, so they can have their turn (because this might be a big group, every person could pay 1SAFEcoin to upload data). If you pay but can’t upload anymore in the second wave data pack, you’ll automatically get in line in the third wave.

  • It might come as a surprise why somebody who already got SAFEcoin would want to pay for a bootstrapper. It’s simple, you pay so that the SAFE network grows. People who already own SAFEcoins could set a time when they need their SAFEcoin again (ofcourse your not paying some else’s bills and loose your precious SAFEcoins). I’m a true believer of SAFEcoin, but it’s idiotic that my money hasn’t move since the crowdsale of 2014. Money should always be flowing and instead of it being asleep in your wallet, it could help stimulate network growth. So what if your not getting a extra percentage on top of it, the SAFE Network will grow if SAFEcoin can constantly keep coursing through it’s veins.

This is what SAFEcoin should be doing

I don’t mean HFT, but it should constantly be moving at high velocity. :stuck_out_tongue:

We don’t want this dinosaur believe of saving money on the SAFE Network.
The Network would just see
That the first wave owes money and will pay back as time past
The second wave paid money and can now upload etc
The people with SAFEcoin will get their money back, because they supported bootstrap data

IMHO the pay the producer problem can be solved by the producer providing data that they are really who they say that they are, but you’ll need human input. On a bitcoin sub reddit NSFW, they have to verify with pictures.


For instance if I say that today I’m @fergish I would have to provide a NSFW :kissing_heart:picture of @fergish before people believe it is me who uploaded the SAFE crossroads podcasts.

Maybe we can also use login methods from the legacy internet to determine somebody’s id.

Previously known data can still be uploaded using another format. For instance I uploaded Safe crossroads mp3, @fergish could upload wav’s, giving a different spin to the uploaded file. Even cutting 1 second from the original file might do the trick, since it’s not in it original form anymore.

It would be fun if we can somehow see a date when a file is uploaded to the SAFE Network, that way fergish could make a picture on that date and say “Hi this is the real legendary voice of the SAFE Network”.

LOL maybe it would force the arrogant Youtuber with over a million subscribers to pay attention to the SAFE Network.

One genius way to incentive people to pay for stuff is in this Yours demo

4 Likes

Yes I think a version of “yours” may work nicely in SAFE actually. Be great to get the source code and try an example as soon as the low level API is available. We can simulate safecoin using any structured data above 10,000 tag. A nice example app for sure and who knows where that would lead … of course if Ryan and co did it themselves then it would be easy and perhaps they can earn dev rewards. So win all round all going well.

6 Likes

Yes, SAFEnetwork would be a great fit for Yours! Nice find :slight_smile:

4 Likes

How would this be possible? Force this into the launch version, hope people don’t change the open code, then release a later version and hope everyone updates?

Seems pretty messy to me. I think freedom from day 1 would be much more powerful for the project.

I also think that it’s a problem that doesn’t need solving, because people who have created things in the past have for the most part, already earned their money from it already.

Technically very simple, do not switch on Ptp till version X. Politically, well that’s a different matter :wink:

3 Likes

With the currently single payment model, there is a business opportunity for a company to hedge the sale of low cost storage.

A company could break the cost into monthly instalments, where they make money if the payment period extends beyond the original PUT cost. This is essentially what hosting services already do, as their infrastructure costs will be relatively fixed, but subscriptions are not.

Using multiple signatures, the provider could give the consumer full control of their files, with the provider auto-signing transactions, as long as payments are made each month.

This would give those who do not want a large up front cost the option of much smaller monthly costs. It may cost more in the long run, but that will be acceptable to many.

1 Like

The ideal solution is for the user to (run a vault) until they earn enough Safecoin to pay for a 1TB upload. This may be an inconvenience because…

  • How long does it take to earn 1TB worth in Safecoin? 1 day? 1 Week? 1 Month?
  • How much resources (storage space) are needed to earn 1TB worth of Safecoin PUTS?
  • What if I store a ton of data that is rarely accessed, therefore earn very little Safecoin?

Fortunately, there is a simple solution looking back at some old ideas. See below…


The original POR Tokens solves not having any Safecoin.

What is POR Tokens?
A user gets POR Tokens for promising storage space on their machine. So if 10MB of storage is promised, they receive 10 POR (1 Token per MB). This allows them to immediately upload (PUT) 10MB onto the Network.

Why did we disable POR Tokens?
There were concerns about gaming the system, by allowing a vault to promise lots of space, then flush whatever was stored and start over. This basically gives them unlimited storage.

I was a very vocal critic of this as well. But after running vaults on every Test, I think there maybe a solution to this old problem.

Instead of giving POR tokens for “promised” space, just issue POR for actual PUTS. Here’s why this works…

  • It takes time for a vault to collect chunks and my average chunk collection was about 100MB over 7 days. It would be pointless for me to flush my vault in a short time period because I’d lose my Safecoin earning opportunity from GETS.
  • The key component is time which is hard to exploit for much benefit.

Imagine a POR token balance that grows as your vault collects more data chunks. How cool is that!? It would be exciting to see your POR balance grow as your vault works in real time! This also encourages people to keep their vaults running longer, collecting more chunks.

Doesn’t this conflict with Safecoin’s purpose?
I’ll debate with others because Safecoin is really a bootstrap currency. We should designed the Network to function even without Safecoin. Keep in mind, Safecoin can and will be used for other goods and services anyway… as long as POR tokens doesn’t compete with Safecoin. We just need make sure POR Tokens are non-transferable.

Isn’t this a rehash of your SAFE GB idea?
No, SAFE GB was meant to be transferable alternative currency. POR Tokens are only for storage (PUTS).

What happens to the POR balance when the user resets or turns off their vault?
Well… that is up to the brilliant dev minds. I do recall they don’t like to micro manage which causes more work for the Network. I say don’t bother. It takes time to build up a POR balance and the cost of policing is moot. The loss of Safecoin earning from servicing that data (GETS) is more than enough penalty for the farmer.


TLDR:

Reward POR Tokens on actual PUTS not promises, allowing people to trade (time) and (resources) in order to upload data in a Quid Pro Quo manner. This isn’t immediate gratification but it can work for those with out money.

Reward Safecoin on GETS. This encourages farmers to retain data long term.

“The wealth of a Network is not fiat value of it’s currency, it’s the productivity of it’s people.”

2 Likes

I have been wondering how much the safecoin will be worth. What will determine its value? Shouldn´t it follow the storage, ie hard disk, prices? Though, I asume many of the current maidsafecoin owners hope that it will be much more worth in time.

That’s not what I noticed when I had some nodes running. I don’t remember the exact figures but I remember clearly that after a few minutes the directory size had already reached the plateau corresponding to storage of all data a node was responsible for.

I am curious about your setup. It doesn’t seem optimum for network health and a proposal that helps this kind of setup has not my favor.

With bootstrap data there is a limit how much you can upload. Even if your content would earn little SAFEcoin, whoever wants to store data after you data block would have to pay. It’s almost like a blockchain.

If we don’t have mechanisms in place to anticipate large volume of data, we just wasting time. Our competitor don’t say wait a day, week, month. When people want something they want it right away or they’ll go somewhere else.

The networks need will determine the value, but outside the network people will manupilate the fiat price to make it do whatever they want. If I could buy $1 M today, that alone would set a buy frenzy, fun thing is that I can sell tomorrow and as easily crash the price. This scenario is real for all the little marketcap crypto’s.

On the other hand the fiat price could also be something that get people talking about the SAFE Network, so it’s all good.

@dyamanaka Devil’s advocate…

I run a lot of vaults, none of which store any of the data they receive as PUTS, and just use the POR tokens to get free storage. By not storing any data it’s easy to run a lot of vaults - enough to generate a worthwhile amount of POR tokens?

Regular resets keep these vaults receiving PUT data, and getting free storage.

How to stop this?

Instead of getting a token on every PUT make it a lottery, like a Safecoin farm attempt. And set the probability low enough to ensure that a vault has to stick around - ie not get kicked off for failing to retrieve data it claimed to PUT.

We need someone to work the numbers on that. Hmm, who’s good at that? :wink:

1 Like

Agreed,

If we don’t offer a better service, many will not adopt.