Steembit: a decentralized storage marketplace for Steemland

in #steem6 years ago

Rationale

I introduced the idea of "steem - the blockchain and the community around - as a virtual country, dubbed Steemland" in a couple of previous posts:

STEEM the cryptocurrency is the internal currency of Steemland and the prosperity of the steemians - the inhabitants of Steemland - is proportional to the exchange rate of STEEM versus a basket of other currencies. The higher the price of STEEM with respect to $, BTC, KRW, etc., the more prosperous Steemland and its inhabitants are.

We are here into macroeconomics. And we are right to focus on this, for there are now hundreds of different blockchain networks. Steem is just one of them, competing against the others. It has at least two major advantages on the others:

  1. It's been working reliably for 3 years now, longer than most others. One of the reasons for reliability is that it doesn't attempt to be everything to everybody.
  2. It has, in the words of The Economist (the most prestigious global economic magazine) "The most elaborate working crypto-economic model".

In other words, crypto-economics is one of the most important unique competitive advantages of steem. A direct consequence of a well designed crypto-economy is the size of the community

Steemland macroeconomics

As explained in previous posts, when "importing" goods and services, Steemland pays in fiat currency. It does need to "exchange" (sell) STEEM against fiat, thus depressing the price of its currency.

On the contrary, when "exporting" goods and services, Steemland's customers need to exchange fiat for STEEM (buy STEEM) thus increasing the demand for the latter.

Corollary

Any goods or services which can be paid for with STEEM rather than fiat contribute to the prosperity of Steemland.

The main "industry" in Steemland is "content production and management". And one of the main inputs in this industry is ... storage space. Not for the text, which is stored in the blocks of the chain, but for the images which illustrate the texts.

Iulian.PNG
PowerPoint drawing courtesy of @smartiot

The main "content organizations" in Steemland are the content dApps, first of which is Steemit.com, along with busy.org, Partiko app, esteem app, DTube and its videos, SteemPeak.com, Appics and others.

Being able to pay for that storage space with STEEM rather than fiat (as we assume it's today overwhelmingly the case) would

  1. remove a source of "downward pressure" on the price of STEEM as Steemit.com for instance would not need sell as much STEEM as before in order to pay for AWS S3 storage space
  2. later on, if the Steembit project is successful, create a new Steemland export, storage space, hence a new source of STEEM demand

The Vision

There is no reason why steem could not serve as the "accounting back-end" for a variety of other applications where something else than pictures are stored and accessed. We have proven this by developing https://eftg.eu which runs on a clone of the steem blockchain.

We have modified the "imagehoster" module of the condenser to allow storing other types of files, such as PDF files, to our storage back-end.

PDF-insteadOf-picture.PNG
Nothing prevents steem from acting as a gateway to PDFs, Excel, PowerPoint files. Nothing ... except the cost of storage

The truth is that even today, we can easily see someone developing a competitor to https://scribd.com or to https://www.slideshare.net/ on the steem blockchain. The only brake is ... the economics of bulk storage!

The vision of Steembit is to internalize bulk storage in the Steemland economy, by offering steemians the opportunity to earn tokens by providing storage space.

Yes, this would be in a way a competitor of Sia, Storj, Filecoin and perhaps bitTorrent too. However, Steembit starts with a significant advantage:

  • the supply is already here. Unlike all the other competitors, which have to attract people to their single-use coins, Steem has a community of about 200 000 people, of which some have available storage space on their hard drives (seems like a reasonable assumption)
  • the demand is already here. Unlike the other competitors, which have to convince external customers to come consume the storage their users supply, we have already apps who should find it a much better idea to "source internally" or "buy patriotic" and pay for storage with STEEM, to their own users, rather than selling STEEM against dollars and then handing those dollars to AWS (or Azure or whoever else)

Economic orthodoxy

Steembit aims to deploy an additional competitive advantage: sound economics, in keeping with one of the main advantages of the steem blockchain. Many (if not all) of the competitors quoted above hark back to the anarchist philosophy of Bitcoin and put great store in "peer-to-peer" relationships in economic interactions.

Yet the scientific truth is that "mediated", centralized way of interacting are decisively more efficient (if indeed prone to abuses of power from the "inter-mediator")

If it is to succeed (economically), Steembit needs to be a centralized service, like a wholesaler who would:

  • On one hand, buy and aggregate storage "chunks" of various qualities (and in various quantities) from the "small storage farmers" of Steemland (the steemians)
  • On the other hand, be the "go to" destination for the consumers of storage such as Steemit.com, just like most people do not buy their groceries directly from producers (very inefficient) but rather from Walmart, Tesco or Carrefour (infinitely more efficient, even if not bereft of issues)

Compared to projects such as Sia or Filecoin, which have designed the solution under very strong assumptions of "trustlessness" (which impose a high cost), Steembit shall be designed under much less stringent conditions.

The first product Steembit could offer is storage for very old pictures linked to old posts of steemians who have long left the platform. As an example to help you visualize, when I came to Steemit I wanted to use the account @sorin, but it had been already taken by some other Sorin who had discovered Steemit before I had. However @sorin has published his last post more than 2 years ago and seems to have left Steemit since.

Yet Steemit.com still pays to Amazon AWS for storing (admittedly beautiful) pictures such as this one:


Picture courtesy of someone bearing the same name as me, who has since left steem

I would contend this is rather inefficient, as it is highly unlikely many people will access pictures from my namesake (yet unrelated) @sorin (aside from this specific one, because now I linked it - not duplicated it - in this soon-to-be-trending post 😂 😂 😂 )

Wouldn't it be more efficient for everybody if Steemit.com would move such old pictures from AWS S3 (for which it pays with $) to the hard drives of steemians running the Steembit client (and who can be paid in STEEM for doing so, rather than in $) ?

Steembit as a "common good"

Of course this project idea looks great on paper but the "ugly truth" is that, in an by itself, it is probably not economically profitable. Storage costs are relatively low while the complexity of such a project and the cost of development work is rather high.

I strongly suspect Steembit is hard to turn into a profitable standalone business, at least in the short term. Steembit makes sense in the context of Steemland though - part of the development cost is compensated by the implicit promise that, thanks to Steembit, the price of STEEM will hold much better, or even increase.

This is where @utopian-io could and should, IMO, come into the picture.

I have already mentioned this idea in an "Idea Hub" broadcast hosted by Utopian, in a Steemalliance radio talk and in the latest Steem Aachen meetup (start at around 9'30", best viewed at 1.25 speed).

The main message of that talk is:

  • it is (likely) not economically rational for any individual actor of Steemland to embark on creating Steembit
  • yet it is economically beneficial for the community as a whole to organize and fund the development of Steembit

In practice, this could be done by a classical Utopian project, with some funding from Fundition, perhaps using the "worker proposal" system suggested by @blocktrades, with organizational help from @steemalliance or a combination of any or all of these.

Steembit is more than just a technical solution. It is also a concrete topic on which the community can come together to work for the common good. I am convinced that if we maange to implement this, the steem community would have made a huge leap forward in capability and integration.

Concrete proposals for taking this idea forward are welcome!


If you know what witnesses are and agree that people commited to keeping this blockchain ticking play an important role ...

(by simply clicking on the picture - thanks to SteemConnect)

Related posts

Steem ecosystem

Other posts you might enjoy:

Blockchain, Crypto and Society

Blockchain and Europe

Sort:  

That is because of solutions like this that I love this ever growing and improving blockchain. This is definitely the blockchain of opportunity.

Posted using Partiko Android

You got a 33.33% upvote from @voteme courtesy of @discernente! For next round, send minimum 0.01 SBD to bid for upvote.

Do you know, you can also earn daily passive income simply by delegating your Steem Power to voteme by clicking following links: 10SP, 25SP, 50SP, 100SP, 250SP, 500SP, 1000SP, 5000SP.

Oddly enough I am not utterly opposed to this idea, at least generally. This would be a good mechanism to decentralize cloud storage for example. I also would like to see similar mechanisms for other data, such as facial recognition, surveillance footage, and etc., that civilians can employ to leverage their access to technology currently only available to institutional (tyrannical) actors.

Anything we can do to further resist the global panopticon, increase decentralization, and decrease censorship is a good thing. The question is, will this type of mechanism do those things?

If it will defeat Article 13, I'm in.

Thanks!

:-)

Come on, I doubt Article 13 has been written with the intention you think it has. But that is anyway a different topic. :-)

Coming back to Steembit, I believe it is quite fraught with (business, not political) risk ... Not sure the economics are easy to pull out ... As @sneak says, storage is really cheap ... If it doesn't work, it would be yet another revolution foiled because "the revolutionaries" had to get back to work in order to get a salary and put bread on the table ...

What kind of fault tolerance are we talking about with hosts? Will data be striped to many or will I be hosed if the image or images happens to be on a drive of a client that bites the dust?

I know a thing or two about Sia. In fact I think it could be used to bolster the Steem economy and the devs have a track record for continued innovation as they just released wallet 1.4.0 with many updates to show for it.

Video streaming capability is coming soon and in the words of Taek42 about what he is most excited about in the roadmap (Reddit user of David Vorick, creator of SC)

For the immediate roadmap, I'm most excited about seed based file recovery. For the 1 year roadmap, I'm most excited about the web content streaming that we are now working on. In the near future, websites will be able to stream heavy content such as images and videos to their users directly from Sia, which will save them a lot on bandwidth.

Source

The way SiaCoin works is it distributes the data to multiple hosts (up to 21 AFAIK) and up to 7 can fail with the data being recoverable. It's like a RAID on the blockchain which is kind of cool.

I would love for you to expound on any technical aspects that would ensure my stupid memes remain safe. I can't suffer to lose not one PEPE. I have also contacted them previously about a possible solution for storage for markdown style blogs but the video storage would be an added bonus.

I'll have to do more research into Steembit and compare.

Full disclaimer: I operate a host of the SiaCoin network.

Update: Reading this again and wanted to confirm this is just an idea you had or is there a product behind it? Want more meat to this bone so I can chew on it.

Is there any code or are we reinventing the wheel here?

This is, at this moment, just an idea, there is a small amount of manpower behind it but it's on backburner - as I said, our "back of the envelope calculation" tend to indicate that it is not profitable. The only way to "balance the equation" is to bring in a "social benefit" - the benefit it will bring to the whole Steem community.

I actually thought further in detail - yes, I think the data should be striped, as you say, a kind of "virtual RAID" is what I had in mind. Then I envisage different quality tiers - from the lowest (for the really old pictures from long abandoned blogs) to the highest (for ... I don't know what yet, it will depend on the quality of the implementation). At the lowest quality the data would be only copied 3 or four times but the customer of the service would be able to pay more and indicate that it needs a higher assurance for his/her PEPE meme

Anyway, the reason I published is to attract people like you who are interested in contributing (even if only by challenging the design, if not with actual code or testing or something else)

So, looking forward to collaborating on this (when my team has a bit more time and we can launch this in collaboration with utopian probably)

It sounds like a pretty elegant solution to me - very simple and easy to understand principle and tokenising it is a great idea.

I guess also if it's old photos we're talking at first then there's zero chance of storing a virus on yr system.

Very accessible diagram of the system there too!

Thanks but in fact it is quite complicated to implement well. You need to code a solid client (windows, mac, linux) that people would download and install and which would reserve a part of their disk and account in a fair way for them keeping that storage available.

But while the storage is available, there's not necessarily a demand for it ... so someone needs to take risk by "paying" steemians for keeping storage available even when there's not yet demand for it! Kind of a "chicken and egg" problem

The actual "payload" being stored is completely opaque to those providing storage, it has to be encrypted data. Even if someone wants to store "a virus", you don't care because "the virus" will be delivered within a "tight box" of encryption.

Both people storing data and those hosting it need to be absolutely sure that the barrier between them is perfectly tight: people storing data (in general, not for old Steemit pictures) do not want to risk anyone "peeking" at their data either.

Just saying that in fact the design will by necessity have to be quite complex

"Even if someone wants to store "a virus", you don't care because "the virus" will be delivered within a "tight box" of encryption."

I strongly encourage you to consider that no encryption exists for various state level actors. While that may not be an issue for ordinary folks availing a cloud storage mechanism for pay, for some folks it will be.

The Intel Management Engine simply exists at a layer below the OS, and therefore has access to any data prior to encryption having effect.

Just a note regarding encryption and security, not a denigration of the idea itself.

In this context the question was : "if I offer storage space, do I run a risk that whatever an unknown customer of the service stores might harm my computer?"

The answer is yes. However, we run that risk anyway. It's not an increased risk, simply another vector that can be used to do harm. Cars can be hacked by simply shining LEDs on the vehicle's lights. Computers used to scan DNA can be hacked by presenting DNA that will cause the computer to suffer a malware injection.

Actors able to access IME level of processors can do similar things with the encrypted data. They can do similar things in countless ways, however, and don't need you to use the storage capacity of your system to create the necessary vector for their attack.

I just want to clarify that security of computer systems actually doesn't exist given hackers of nominal skill and requisite motivation. From me your system is safe regardless of any measures you fail to adopt. From some people, no system anywhere is safe regardless of what can be done to secure it. Security applies in between those extremes, and knowing this can be useful in considering what to implement, and what costs to undertake.

For most folks, good encryption will suffice.

LOl Yeah man! Steem land was also an idea by the late @adept to mix steem and decentralland but nwo we can do it with @dlux-io by @disregardfiat hpopefully using games like @hashkings for proof of concept #hashkings

IU think hashkings could even let people rent REAL land in California where people will grow plants indoor and outdoor, with live webcam or updates, and it could end up giving people teh chance to buy libve weed plants that they grow in REAL lkife, andthen allow them top pickup at a legal dispensary in calfornia D

stimulating the imagination of Steemland's inhabitants was one of the main points! :)

I'm sure there's plenty of devs on here that would love to get their teeth stuck into a project like this! It sounds like it would be a real asset to steem!

Makes a lot of sense and we’ve seen quite a few decentralized data storage/CDNs in the works like SIA and Storaj

I think it’s a practical approach to a real problem and not just solving it but creating possible new revenue streams and support for steem

It will also help dapps like dTube and like you said alternative file storage! This is a huge market for both amazon, google and Microsoft and a market that needs disruption

Posted using Partiko iOS

I read the Storj whitepaper a while ago and don't recall the details. But I read the Sia one recently and the cryptoeconomics of Sia suck big time - it attempts to match storage providers to storage consumers in a peer-to-peer network ("bitcoin for storage").

Which can be useful if you try to store controversial material ("censorship resistance") but it is hugely inefficient.

On an unrelated note, you only approve 4 witnesses - it costs exactly 0 (zero) to pick another 26.
Highly appreciated if you'd pick @lux-witness (it's only a couple of clicks to do it). Thanks in advance

Sure, I've voted for @lux-witness now. Good luck wit steemland and

Have you heard about Gaia which is a decentralized storage system used by Blockstack? What do you think about it?


You just planted 0.10 tree(s)!


Thanks to @ucukertz

We have planted already
8145.18 trees
out of 1,000,000


Let's save and restore Abongphen Highland Forest
in Cameroonian village Kedjom-Keku!
Plant trees with @treeplanter and get paid for it!
My Steem Power = 24651.02
Thanks a lot!
@martin.mikes coordinator of @kedjom-keku
treeplantermessage_ok.png

Storage is not very expensive. Extremely large hard drives are very cheap.

The costs are power, redundancy, HA, retrieval, bandwidth; even with a proof-of-storage challenge to nodes that claim to store certain information, paying them to solve "what is the hash of this range of bytes in this huge file?" is not equivalent to "will you please send me this huge file in its entirety at high speed with five-nines reliability guarantees?"

Exactly and for many of the reasons you enumerated. I personally believe SiaCoin is on its way to be a decent provider of this technology.

For example, they recently enabled a feature where your data may be distributed to whitelisted hosts which one may limit to a certain geographic area such as CONUS if you happen to be in the states. This should help reduce latency for retrieval.

The host algorithm penalizes for downtime so this should translate to HA as there is an economic incentive.

For redundancy, the data is distributed to up to 21 hosts and up to 7 can fail and the data still be recoverable.

There is so much more the Sia network has to offer and have had my eyes in it for Steem as a markdown blog storage solution.

I think this would be more ideal than building from the ground up personally and the benefit should be explored.

I looked recently at their whitepaper and for me the crypto-economy of Sia (which has nothing to do with the technical prowess involved in providing cheap reliable storage) sucks big time.

On the contrary, the Crypto-economy is Steem's strong suit. So in my opinion combining the Sia technical solution with the steem crypto-economics could prove a real winner

Indeed, this is one of the main challenges, storage is otherwise not very expensive ... but it has to be paid in USD nontheless ... and you get into a chicken and egg problem - if you acquire storage and nobody comes to use it, you have wasted your hard-earned USD ...

Excellent ideas. I will consider it, as I have also identified this, as a huge barrier to success and yet, should be possible to contend with...

Thanks, @surfyogi. I'll discuss this further this evening on @pennsif's radio show

Seems like you could combine this with a stock photos service so that profits on photos sold help pay for storage costs. Just an idea. Would probably need to implement some kind of AI to check for copyright infringement.

very good idea!

Isn't this a more centralised version of what BitTorrent are now doing (having been bought out by TRON)?
The BitTorrent token is close to Steem's level now and they haven't really even launched the service fully, I think. They have millions of users already and they will be rewarding seeders for their storage and bandwidth. I suspect that BitTorrent should at the least be carefully examined here.

Actually one of the main advantages in terms of efficiency over other existing competitors is precisely the "centralized" aspect.

Trust is hugely expensive and powerful - to replace trust fully you need the 10 Twh which Bitcoin and Ethereum spend in PoW. Filecoin and (I think) BitTorrent are also attempting to be more or less "trustless". I predict that the price they'll have to pay in efficiency will weigh them down seriously.

Steembit positions itself as an utility of Steemland - a bit like the road network or the electricity network in a country - trust should be provided by something unique to Steem: the community, the spirit of belonging to a community members care about. My assumption is that if it works, it will practically give wings to the whole Steem eco-system (more trust grows on trust)

I'm not fully aware of all of the technical details of Bit Torrent, but I've never once heard of anyone complaining of receiving the wrong files using Bit Torrent.

That was not a risk I had in mind. I was thinking more in terms of latency and throughput

I see, ok. I haven't seen it in operation yet, but the whitepaper from Bit Torrent for their new token describes them solving any speed issues by tokenising the seeding process.

Very interesting read, I couln’t Help but jump from one place to the other but yes, the future is here and it is now. I heard Steem had something to do with Amazon but today, I learned that it hosts all of our pictures.
I see your point about converting Steem to Fiat, it would be way more efficient to pay in Steem.

Do you think there will be a way to purge the old useless photos at some point or are they just going to pile on...

I think they shouldn't be purged - one of the goals of Steembit is to build, on the steem backbone, an archiving service. People come to Steem for the idea of "forever persistence", among other reasons. They assume everything they post is on the blockchain and will be there as long as the blockchain runs. They will feel cheated if they discovered their old pictures had been purged

Sure but the photos themselves are not on the blockchain, only the writting; correct?