Why isn't there a trillion dollars of funding and 100000 engineers working on SAFE?!

Ethereum has partners with governments. That itself is skeptical at best.

http://www.coindesk.com/the-united-nations-just-launched-its-first-large-scale-ethereum-test/

Eye-scanning hardware made by London-based IrisGuard, already in place to verify the identity of some of the 500,000 recipients currently receiving traditional aid, is being repurposed to grant access to coupons.

Too satanistic for my taste.

Nobody wants to inveset safe because it threatens the government establishment.

Meanwhile gov loves blockchain. Makes me wonder.

Edited: Add more more link

5 Likes

The problem with Bitcoin is that it has no way to incentivize increased capacity. You can pay the miners more, but the miners still cannot make blocks big enough to hold all of the transactions that want to be processed. The only mechanism they have is to disincentivize demand, Nothing to increase capacity…

Moreover, the issues at core are all political. The miners have cheap subsidized electricity, but lousy internet, thus the politics bend to their convenience. The problem at hand is silly, but it has been at hand for years now. 10 MB blocks are transmitted millions of times a second all over the world. Technology isn’t the bottleneck, a bad economic model, and bad political choices are the issue.

Good economic models need to have that capability. I don’t think MaidSAFE does.

Bandwidth may be plentiful some places. Electricity may be plentiful some places. Disk space may be plentiful some places, But all of that is at the whim of the provider. What happens when it’s not? What happens when China bans or blocks it, for example? And all of the sudden 40% of the capacity goes away? What happens when another provider offers more $$$$ for space, and hardware marches to profit? Will the network have the resiliency? Will it’s economic model be able to incentivize the replacement of capacity? Will it be fast enough? To keep up with all the changes? Once the code is in the wild, we don’t get to re-engineer it on the fly to handle these situations. It has to work, or it won’t work. It really isn’t a programming issue.

It’s complicated, and maidSAFE hasn’t demonstrated to me that they have adequately thought all of this through.

6 Likes

I agree… But every level of complexity adds a potential layer of failure, and failure isn’t an option.

Attempting to use blockchains for all distributed problems is not the answer.

Agree as well. If you read that into anything I said, I think you misunderstood me. My mention of bitcoin is only for learning from it’s failings and troubles.

Monothlithic may have been a bad word. But maidSAFE has always been a “eat the whole elephant” more than a “one bite at a time” there are advantages to both approaches, But the more common one is to string together existing trusted components to compose a viable product, not to write the whole thing. You have got to do what you’ve got to do, but the complexity and necessity of long development times, Is going to give a lot of people pause to diving in whole hog like Warren suggests.

8 Likes

This is where the customized hardware path has to be thought out.

1 Like

I was kind of with you until these questions tbh.

Banning SAFE would probably make it more popular, not less, given it is anonymous and useful and easily accessible. How could they ban it? As far as I know the great firewall is no match for SAFE. You can’t shut it down and banning it will only publicise it - there were more speak-easies per capita in NY during prohibition than there are bars today.

SAFEcoin is a utility first and foremost. The vast majority of users are not expected to ever buy or sell crypto. That’s one of the biggest barriers to mass adoption for all the other crypto projects. SAFEcoin is useful and will be desired because it is the only way to store or serve data on the network. In essence it makes the network free for all users who are able to contribute resources to it. People won’t flock in or out of decentralised farming because of fiat value of their 2 safecoins - which they need to dl their films, store their data etc. Very few users will know how to buy or sell their coins, nor will they much care about the $20 fiat value next to the utility value. There is no other way to interact with the network.

Agree that it’s complicated, but it’s not their responsibility to demonstrate anything to you, it’s your responsibility to find the answers to questions you have. They’re engineers, they’re not writing papers and making models to prove a theory, they’re proving it by building it. I feel like your worries about GET imbalance and the economics are a little misplaced. I’ve spent a year and a half playing these things out in my mind too and I do have foggy areas I’ll confess, but the things you mention above don’t worry me. The incentive and economic mechanisms are insular so the problems you raise are valid, but I suspect they’re far less important than you make out. And issues with politics and upgrades are all solvable and far easier to solve than most of the problems that have already been put behind us.[quote=“jreighley, post:23, topic:13498”]
I agree… But every level of complexity adds a potential layer of failure, and failure isn’t an option.
[/quote]

Actually, I would say the SAFE model is far less complicated than all the layers being built on other DLT projects or even the current internet. It is complciated because it so different and has a lot of innovation, The system itself is beautifully simple and elegant really isn;t it? Far more so than most of the systems we use today imo.

That’s true for sure. That’s why the internet and blockchain have become such a mess. Instead of holisitic design people have taken the easy route and build on top of shakey foundations, sticking plasters on top of bandages. Why does that make more sense for a system like a decentralised internet? I’d rather see 15 years spent on a holistic design myself. Does it matter if people pause or have good justification for doing so? All that really matters is ‘will it work?’

7 Likes

I see a lot of good alignment of incentives in MaidSafe’s proposed model, but I agree that it doesn’t seem to be given enough thought or discussion.

The economics of the network will be a huge factor in determining its long term success, so it’d be great to see some serious resource from MaidSafe put behind accountants and economists to examine the model thoroughly.

I hope there’ll be plenty of time to shift focus towards these issues prior to implementing test safecoin.

6 Likes

It really depends on how it plays out. I think everyone’s dream is a highly distributed system with individual computers providing all of the resources. That may work, but it seems more likely to me that big data will be used because Big data is much better at doing everything that needs done than individual PC’s on ISP and home networks etc. I can spin up instances on amazon, google, rack space in an instant and they are going to have faster connections, more reliability, pretty darn cheap. We see such in Bitcoin mining. It flows to the place where the political and economic incentives are strongest. (China mostly) Those companies are not anarchist rebels. If the government shuts them down they are very likely to shut down… And if they are a significant portion of the capacity of the network, the network will need the resiliency and automated economics to make that happen.

So in short - you may not be able to firewall out SAFE, but you likely will be able to control the major players that have the ability to provide capacity. These players will probably be both abundant and fickle. If they can make money they spin up. If they lose money they spin down. When the spin down, they need replaced, but if all like minded machines are also shutting down, you could get a mess in a hurry. They are paid only for delivering files. when a significant portion of their bandwidth becomes utilized by replication of files from other nodes going down, they wind up spending a lot of resources on activity that doesn’t make them money. (pay once store forever even though most files are rarely going to be used again)

The dream of MaidSAFE is very good. The realities are likely to be much more intricate and commercialized than the dream. The code is complicated, but the real world economics is likely to stress the system even more by utilizing it in ways that are not expected. If it is a good deal it will be over utilized. If it’s a not such a deal nobody will play. If people don’t play, then files risk being lost, and then you get a death spiral. The economic model needs to be capable of creating symmetrical back-pressure. Vast riches offered for free will be consumed, nearly without fail.

1 Like

I think you and me may have different visions of how the farming economics play out. I see a system which cannot become centralised like bitcoin. The economics of spare resources are pretty interesting. Whilst most users won’t care about their fiat value of $10 next to the utility of the coin, that price is the bottom line for any business. No business can operate in competition with spare resources, since their outlay, infrastructure and opportunity cost is always an expense not incurred by the spare resource. I do not see how any serious or threatening farming centralisation could happen in this context.

If farming is decentralised then you really can’t shut it down. You don’t have to be an anarchist to ignore the law, you just have to find it useful and believe you won’t be caught.

Safe farming is nothing like mining, it really doesn’t get you anywhere to use that as the basis of speculation imho.

The whole point and power of decentralisation is removing these central points of failure. I agree, if SAFE operated with obvious centralisation like that it would be vulnerable at those points, but I don’t see how that is possible given the economics?![quote=“jreighley, post:27, topic:13498”]
They are paid only for delivering files. when a significant portion of their bandwidth becomes utilized by replication of files from other nodes going down, they wind up spending a lot of resources on activity that doesn’t make them money.
[/quote]

Again, you’re looking at it as if people were primarily farming to make money instead of to make coins they need as a utility. I don’t see how you get there from the design tbh.

This really isn’t the case… if it were I’d agree with you. If files got lost because of a decent % of farmers coming online or offline suddenly then yeah, that would be a network killer. Even if they don’t go with ‘pay-once lasts forever’ in the final launch, it needs to be a pretty long time and reliable. You can probably just search the forum to read all about data republish and redundancy etc to get abetter explanation than I could give for how that will work and how it can be handled.

5 Likes

The problem is you cannot force decentralization. We do have two different visions, but the vision that the market supports it the vision that will come to fruititon.

If it is cheap and profitable to farm I can spin up hundreds of instances in a few minutes. My instances in a data center with 99.99 percent uptime and redundant connections to the internet backbone are likely to outperform your nodes connected via an ISP optimized for downloads vs uploads. I am going to deliver more files, thus get more farming rewards. SAFE only controls so much of the environment -

If it is not profitable I can shut them down just as fast. I may shut down because I feel like it. I may shut down because the government threatens me. I may shut down because some other network offers me more. When I shut down you still need to replace me… And if you replace me with somebody who goes on and offline frequently or cannot deliver packets quickly, you will have to replace them again and again and again. All at no profit for anybody. Eventually the natural flow of things is going to lead to the most reliable nodes retaining the most data, simple because they don’t need to be churned constantly. But when they do churn it will be a much bigger burden.

It is what it is. We will see how it plays out.

My problem is that if we don’t know how it is going to play out, we cannot just trust that it will play out in the way we dream it will play out. The asymmetry of incentives leaves us pretty helpless to fix it if it doesn’t go according to our dream for whatever reason.

2 Likes

yeah there is no good solution, but economics always serve as a balance in a free system.
What would you want to do, allow facebook social graph and distribute safecoin as a lotery system to non anonymous users so that the safenetwork is properly decentralized ? no. it’s not going to playout…

1 Like

Firstly, you can’t just turn them on and off when it is profitable, you have to earn the rank with a reliable vault that runs for some time in order to get decent rewards. It will also surely never be profitable to switch on your data centre. It will cost you money to run it there… all the other resource providers are only paying electricity. All the other farmers will be farming more economically than you, so it should not be profitable for you to pay to farm if there is a surplus of resources to demand (which there should be). Farming payments should just about cover the electricity bill, no one would run it for profit in that kind of competitive environment vs spare resources, they will only run it because they want coins, that’s the real incentive, not financial reward. The idea of safecoin incentives is not to make a profit from farming, it is to gain access to a unique utility that people will want/need.

You are still looking at safecoin as money or a token of value as far as I can tell.

You raise a couple of good and relevant points about governance and the implications for updates to the codebase, but I still don’t feel like we are looking at the incentives and economics from the same page, so we’re bound to keep disagreeing.

1 Like

My Comcast internet has gone down at least 3 times this week. I have a web hosting accounts I pay for as overhead anyway that rarely go down. Where should I put my vault?

I realize that the money isn’t the goal… But the money is the only lever the network has to regulate itself to the goal, and it is in the position where if it is profitable it will be run for profit, and if it’s not it will lose its leverage to buy more capacity and resources that it needs. It’s a double edged sword.

Perhaps resources will always be abundant, and there is nothing to fear. But that is a pretty huge assumption in a world that hasn’t been invented or exploited yet.

1 Like

It’s an easy out to say “Market forces will fix the problem” It is much harder to plumb a artificial software market that actually siphons resources to the places that they need to be. And as it stands, i don’t think the SAFE model does allow for a free market.

The network only rewards one thing - GET lotteries. If MaidSAFE replaces Netflix and amazon prime streaming for example, The demand would be insane, Free is a good price, I will take everything I can get at that price. On the supply side though, The electrical usage, CPU usage, Bandwidth usage could grow significantly. There is no lever to slow down the GETs and all the processing and network costs that come with them. The networks only ability is to increase the cost of PUTs to generate enough revenue to compensate the farmers for their efforts – and PUTs aren’t the problem. Increasing prices make PUTing less attractive than S3, dropbox, IPFS or the like… Thus the network loses business to a more affordable option. Now it has no means to raise revenue at all.

If it is useful and has demand then people will farm, regardless of profit.

Some farmers may not care – but many connections are metered. If people computers slow down because they are too busy doing netflix job to run the user’s spreadsheet, they better be getting paid enough to compensate for that. Even if the user doesn’t care, the ISP might. SAFE is also only useful if sustainably works… The economic model is critical to it’s sustainability.

7 Likes

Are you assuming it has to be perfect on day one?

Nobody knows exactly what the endgame will be, and indeed, there is no static target because the world changes and SAFEnetwork will either evolve with it or be replaced.

I think before we begin designing these aspects it is a bit early to judge. Discussion is fine, but I think it’s just a substitute for patience :wink:

7 Likes

I think you are not taking into consideration two dimensions here:

  1. The big strength of SAFE in the early days in my opinion is that it is not a democracy. Somehow, we have a leader (david irvine) and governance (Maidsafe). It is important because there is no doubt market forces can be destructive, and could divide a community (see bitcoin). Being able to see that early, and takes steps unilaterally to patch & correct imbalances is so important. Once the balance has been found then Maidsafe can step down and progressively let the network live on its own.

  2. When there is success, there is generally a way to monetize it. Think what you want, but the day the Safenetwork takes over Netflix in popularity, I am ready to bet that there will be several advertising platform fighting for views. You don’t know what the future holds, maybe advertisers will be as much as a plague for SAFE that miners are in the bitcoin environment.

4 Likes

I do share your concerns here, but I don’t think the are insurmountable problems, relative to the complexity of what has been solved. The economics have to be viewed against all other systems as you rightly point out.

It is ambitious to make all GETs free, but we can all see why it is important for utility value. Can it be done? We shall see.

Clever technical solutions which ends up burdening the client more than the network may make the question moot. Caching locally may result in very little network load, but the consumer may be draining their own bandwidth trying to abuse this, for example.

From observing development so far, it is evident that some of the work is exploritory. They know approximately where they are heading, but figure our the details on route. I am confident they will continue to do this and be pragmatic when they cannot.

4 Likes
  1. If MaidSAFE overtakes Netflix, You have a Sawn Fanning netflix problem if you have centralized control. The nefarious uses of MaidSAFE is

  2. How would that work? MaidSAFE stores and retrieves files without knowing what they are etc. How would you know what ads to send to who? Why would anyone want that?

I don’t think it needs to be perfect out of the gate – I just think that if it is decentralized it is going to be very difficult to fix, and if it is not decentralized it will not be politically sustainable.

There isn’t a lot of room for failure. People are going to be pretty fickle about trusting their files to something known to lose data or be unreliable or slow.

Time will tell. I want it to work. I just raise these issues because having been a devoted fan, and stepping away for a bit, I have some pretty serious doubts. Hopefully people have thought them through. I am not hearing a lot of dialog that indicates people have though… We shall see. Even if the project fails, a lot of good will come out of it in the long run.

4 Likes

If MaidSAFE overtakes Netflix, You have a Sawn Fanning netflix problem if you have centralized control.

—>Maidsafe will have stepped down long before it happens.

How would that work? MaidSAFE stores and retrieves files without knowing what they are etc. How would you know what ads to send to who? Why would anyone want that?

---->what if big content providers turns into advertisers (by embedding ads within the content) ? what if established killer apps decide to monetize later on when they have a big userbase ?

I beg to differ. I’m against that idea. In programming world, I think benevolent dictator is the best sword on the table. Prime example, Linus. Linus has been attacked many times, been tempted to step down, so others can take his place. It is to mess up the code even further. Imagine some MS boy managed to have authority of the Linux code. That’s a goner for sure.

SJW, and culture marxism attacked at Tech Giants CEO, to paint them whatever they come up with, to make excuse, and use the state to enforce it. Once achieved, they prop up new dictator which destroys the code, and the community. Bitcoin. Github. Mozilla.

This stems out from other areas, such as arab spring, and 2011 revolution in the middle east. It is all designed by control opposition, to replace good honest leader with corrupted leader. To do so requires double speak.

The Catholic/Orthodox Church has it right for passing down the highest order to the next bishop. Bishop can’t lie, and if one did, it is a grave sin. Other bishop would know too, and they’ll act accordingly.

Just like how Satoshi passed his code down to Gavin because he highly trusted him. Gavin didn’t want to maintain it. Gavin had no trust in the circle. He gave it to Adam Back who work with axas, which has ties to Soros, and banking elite. Bitcoiners got back stabbed.

To maintain honest highest order is to have absolute authority of these matters. Code shall not be corrupted. I trust Irvine pass his authority down to one of his highest choice than democratically voted because majority are liars, manipulative, and unworthy.

2 Likes

But what of trustless systems, I always think that is part of what we are after.

1 Like