First, I am excited by the team’s work and confident about the success of MAID. But there still may be some risk resulting in the failure of it:
It is possible to do. There’s no reason why nodes can’t pass Chunks to each other. Bitcoin proved that a fully decentralized p2p-network can be done. Without trackers, without central servers etc. Freenet also proves that it’s possible. Self Encryption is tested, the Vaults are tested. No fundamentals why it couldn’t happen in my opinion.
Maybe people get in trouble with their upload speed. But let’s assume this happens and we don’t have enough Farmers… well, the reward for Farming goes up, maybe even to a point where it’s possible to make money using Amazon AWS. look at how much data Netflix is serving from AWS for these few dollars each month. So again, it should be possible.
Bittorrent, Usenet, Freenet and Bitcoin never needed any marketing. If technology is great it’s just great and people will use it.
Ethereum is working on their own sort of client for messaging and P2P called Swarm. They are going to support IPFS so that’s an option. But I think it won’t hurt SAFE.
There are no signs of that. And within a few months we might see a live network with Safecoin implemented where the devs get coins. Should be great.
So after all I think we’re good 
I think the biggest threat to SAFE is somebody building on top of IPFS… IPFS integrates rather nicely with the regular internet so it doesn’t have the “Walled Garden” problem to contend with…
In its present form it is highly insecure… But with a few tweaks you could easily take something like SAFE’s self encryption and using some ethereum contacts or other cryptocurrency rewards to make sure the data is scattered redundantly, and you could get something similar to SAFE without the complexity or overhead. It will not be as secure - but it will be more secure than the alternatives - and it won’t require the same overhead and dedicated environment.
The “Walled Garden” problem is definitely the biggest challenge… It is hard to overcome – although facebook did it, so it is definitely possible.
So basically implement self-encryption in IPFS, and it would be a GG for safe.
IPFS is lovely, and I feel like it is a threat to safe. Torrents are superior in every way, but might not better than safe when it releases. there’s new app called LIBS(I think), which enables content creators to earn btc via seeding. I love having people to share content from their harddrive as they see fit. If they don’t want to, they shouldn’t. Having 200 seeders with seedbox enabled, you would get insanity download speeds.
I don’t know about that. For newer files it works great. I get downloads up to 7mb a second. The problem is with the older files, like a file from 2 years ago. You only see 2 seeders and sometimes the calculated time get’s stuck at 3 weeks or so. The protocol looks at your upload and download etc. On SAFE you’ll always have 4 seeds that try to get you your Chunks as fast as possible. And look at Popcorntime when you try to watch an 4 year old episode of your favorite series. It takes a lot of time to load. And if you’re in Germany some law firm might come after you asking 815 euro’s. And where’s messaging in BitTorrent? Where are the P2P-websites? Where’s the incentive for you to share?? How do you log on to BitTorrent?? I really like the protocol, it served us well over the years. But it’s time for something new.
Do you fully get IPFS?
Basically it would move any website that chose to be into the Bittorrent world-- And it makes that Bittorent world accessible from the URL… They and others have currency incentives for seeding well in the works.
So you get scalability benefits of DHT with caching without the extra layers of watchers etc… It also provides the security of files matching their hashes as a default – so you cannot download a corrupted file.
SAFE will be better for things that need to be censorship resistant or highly secure -But that may be 2 or 3 percent of the content on the internet-
“How do you log in”? Why would you want to? That is a minus, not a plus in most people’s books. Most people just want to click a link and get what they what to get…
I love SAFE, but it does have some pretty high hurdles to clear to be what we dream it will be.
Isn’t is secure or insecure? Instead of highly secure and secure? I think 100% of the content should be secure if the uploaders wants it to be.
You mean the 2 or 3 percent of the most valuable data there is…
Most internet content is placed on the internet so that people can see it. You don’t need security for that…
You can put in on - but SAFE is overkill for the vast majority of sites.
I have always thought “High Security” has been SAFE’s best selling point. But most people want to sell it as “the next web” .
I actually miswrote what I said on the second paragraph. I was writing on mobile.
I think this is always an issue for me to explain, but for me security is two fold
Logical → encryption etc.
Physical
It’s the last part that is more invisible, but if data can be found it can be corrupted/blocked etc. So it’s important, also hiding who uploads in our case is important (Freenet does some of this). So a bit of bittorrent for physical security is also good, but it’s the mix of randomly where stuff goes in bittorrent, so this is needed in a dht as well. If we can identify locations then we can bring stuff down, or be forced to.
Another HUGE area to consider with other projects is all the work we are doing in crust (still some to do) and this is NAT traversal, a complete PITA for p2p projects but required for real p2p, otherwise you need servers and manual port forwarding and possibly even static addresses. Worse many of these will also use a standard protocol and port, leading to more issues with banning/censorship.
I like them all and want them all to bring something to the table and I hope ipfs/bittorrent/bitcoin/ethereum for instance can use crust at least to make their nodes run from behind NAT/ routers as it will help them a huge amount. So this is not a we tick this box and they don’t it’s there is more to consider and if we tick a box we do so in a way that others can use that lib and also tick the box. Hope it comes across like that 
I should state crust currently uses a load of threads and will internally switch very soon to a more OS level scheduler via the MIO library. It’s a task and a half but will make it super efficient.
I wonder how something like this could ever fail?
I can’t believe that you guys are even mentioning IPFS, even the creator of IPFS says that it isn’t secured. Filecoin their currency will probably be another blockchain based currency. The best hope for blockchained based currencies right now is Openchain. But even that is not really a blockchain and still depends on bitcoins blockchain.
Sorry that my question sounds rude, but are you forgetting that there is something else coming to the internet. Something like the IOTs devices? It’s not Maidsafe that will fail, it’s everything around it that is failing right now. When it arrives we just have to be creative, we’re the absolute lucky ones to know about this technology at this stage.
Yes, but as I mentioned in up front - but more layers can be added… And for a ton of things high security is expensive overkill.
I have played quite a bit with IOT… I see very little need for super high security on my greenhouse temperature sensors etc. MQTT serves most IOT apps just fine. To run such things through SAFE would be rube-goldbergish. Not to say there isn’t a place for IoT within SAFE – It’s just going to be a vast minority of data that needs SAFE level security.
As far as the currency goes – That isn’t even written yet- let alone put into practice… Time shall tell… But I don’t think returns will be nearly as significant as some people dream. Money that is handed out without sacrifice isn’t really money that anyone is going to value. Computer resources are abundant, and thus not a significant sacrifice. Add on top of that that most of the General public has zero enthusiasm for cryptocurrencies, I don’t know that SAFEcoin is going to lead to mainstram adoption
Your optimism is great – But like I said a lot of high hurdles getting the whole world to climb into a new “Walled garden”
I love SAFE – but overoptimism is not going to help with adoption. Strengths need to be marketed and weaknesses ought to be mitigated – Competitors are up and coming, and if you take them on head to head optimistically putting your weaknesses against their strengths, you will tend to lose…
SAFE is going to beat nearly everyone on Security… That ought to be the primary thrust of the message and the marketing.
Are you guys forgetting things like the TPP, TIPA, ACTA and other assorted trade deals that will massively restrict and censor internet freedoms? Nevermind federal programs like Canada’s C-51 (now becoming C-58) and the states notorious patriot act. Then there’s the whole GQHQ thing. I don’t think maidsafe is overkill at all. I think it’s essential for one’s day to day internet usage.
SAFE is going to beat nearly everybody on security – that is what ought to be marketed.
The money used within SAFE’s unparalleled secure system is the new value, along with its INSTANT speed, untraceable true cash nature, new technology, lack of blockchain, ripe time in history to be released…
I very much liked your reply and largely agree, but here I need to point out that a safe_client is automatically started as an anonymous client. So also in the safe_dns example we released in august; there was no need to log in to visit a url. You need to log in to put data to the network, or change a domain you own.
I’d also just like to add that those of us who are old enough to remember dialup will also recall one did have to “log in” to dial the internet back in the day. So the concept isn’t all that strange.
The thing is, as you say, “more layers can be added.” This means afterthought.
SAFE builds security and privacy/openness options in at the base level. Latency may be an issue at first, but as the network scales, it should improve and further tweeks will also improve it.
We see all manner of problems proceeding from the fact that control of one’s communications and data are not secure at a default level. That that is handled from the get-go is the feature of SAFE that may not be appreciated, but is the feature that will make it a real foundation to build on.
It also means simplicity vs. monolithic design.
Consider that their stuff works today (to a large extent).
But more importantly, this is another of those funny topics. There’s already 19 comments and the OP didn’t even ask a question (!). Of course anything can fail (and in the extremely long run, invariably does). What’s the point of discussing this? David and the team are working full time on making sure the project succeeds. That’s the answer.
They might fail, but the IPFS and Bitcoin and everything else might as well.