Streaming large video files with a distributed network


#1

Continuing the discussion from Most wanted APP for the SAFE Network:


A Proposal: A New, Fair Internet Model to Restore Economic Balance
Dealing with copied content
#2

Just wanted to move this here. Curious to hear other’s chime on the technical feasibility of this, and if SAFE is actually a reasonable network to make this happen.


#3

I think treating content like a product needs to end. Pure and simple. You donate to the artist and support them if you want to see their art/project continue. Instead of using quid pro quo you leave it up to those with capital, the consumer, to donate to the creator. Artists need to forge relatinships with their audience not rely on copyright to force them to pay. Art should be purpose driven not profit driven. Look at maidsafe it was created with a purpose in mind and the crowdsale drew tons of donations. This is one example of how purpose motive can draw revenue. We need to rethink how things are funded.


#4

Not what we’re talking about. It must get exhausting being the copyright police police.

We’re talking about the technical feasibility of using the network to stream large video content, or if the structure of the network prohibits that. And if using a distributed system for content distribution is beneficial for artists financially (the cost of hosting, vs the cost of vimeo, etc).

We all know your stance buddy, don’t worry :wink: But I’d really like it if this thread wasn’t about your philosophical views on, well, anything.


#5

Thank you for the clarification there Russel. You must not have had your coffee yet or something. However

that being the case I believe I have as much of a right to express my ahem views as you do and present alternative sollutions. Whether you like them or not isn’t my concern.


#6

You sure do!

The “artists should only get donations” topic is just covered ad nauseam everywhere on the forum. I thought maybe we could not talk about it in just this thread, and instead talk about the topic. We’re already getting more meta than I’d prefer.

@janitor, so what were you saying about this? It sounded interesting.


#7

@russell with deference to you desire not to have this thread full of objections— I get the impression that you feel the only industry that could really be hurt by the transition, what ever it turns out to be is film. What about video games? And aren’t these two industries converging?

But while we are at it, I still don’t see why the strategy to save careers and livelihoods in the film industry must involve hobbling or dumbing down the potential of SAFE. The finance of film can adapt. We’d need some real numbers sifted intelligently but I’d suspect that the people in the film industry could reach a much wider audience post transition and have an even greater impact. Big budget films might be in trouble but I think some big film luminaries were predicting the implosion of the current system even without Pirate Bay or even bigger threats on the horizon.

Also if you start doing a film or documentary on this, I think there is a chance that by the time you finished your opinions about how this could go down and still have an acceptable ending might change.
Its hard to find people who are as for something as those who used to be solidly against it.


#8

This probably never happened. I’d like to point you to Exhibit A: Everything I’ve posted.

Big budget films will have their own system that they’ll keep using for decades: theatrical.

But yeah, smaller budget films will hopefully find a larger audience through online distro.


#9

From David Irvine:

Options for the consumer Internet

This is a particularly easy issue for the decentralised community. The easy wins here are:

Private data sharing (e.g. DropBox, Box.net etc.)
Video provisioning (youtube, vimeo etc.)
Social networking (FaceBook, Twitter etc.)
Email and messaging

There is a plethora of ‘low hanging fruit’ that would benefit immensely from decentralisation. Ask any of the businesses mentioned if they would like their infrastructure to be secure, reliable and free. I think the answer is obvious. Now ask their users if they would prefer to not have their identity taken and used in advertisements or have their activity tracked to advertise to. We know the answer to that one too, I think.

Well how do they profit?

Good question, maybe some don’t. Others will come up with ingenious mechanisms I cannot foresee. They should be innovators, right?

There are in fact many ways to profit. advertising is still possible. Build advertising into the app, select adverts from a network location that’s sorted on some categories, geographies etc. and serve these to the user. Significantly though there is no need to take user data, store it and mine it. All the matching for advertisements can be done in your application. Simple, cheap and in no way a harm to privacy of users. It’s still possible to measure advertising impressions etc. It just reverses the proposition, where advertisers get what they want, users get what they want at no cost to their privacy and security, now that’s great.

There will be a ton of opportunities to commercialise applications on the market.


#10

thanks chris. I think I saw that. I’m curious if the network will be able to stream content specifically though. I know it says it there, but what I’m curious about maybe is how the network returns data. So if its pulling chunks, in order to stream they’d have to be pulled in a specific order to buffer the front of the file. Right? Anyone know if this sort of thing is possible?


#11

So you would be referring to ‘Adaptive bitrate streaming’ which is achieved now by utilizing streaming servers like Flash Media Server together with Content Distribution Networks (CDN)

I looked into this quite heavily a few years back when CDN’s were a bit thin on the ground in Australia. I actually did some live streaming testing out various CDN’s and got good results with 480p.

In that client/server model, it’s all about having Points of Presence (POP) close enough as to keep latency within limits.

Distributed Streaming would probably utilize WebRTC

Have a look at PeerJS as an example…should be straightforward to implement.


#12

That is abstracted. The app (say, VLC Player) reads the file/object from the file system or over HTTP(S) or receives a stream (if content is streamed) and it doesn’t have to worry about what’s going on with bits and bytes on the network.

Let’s say you wanted to create an iOS app that allows one to watch a video that’s stored on the SAFE network.
You could use a traditional way (point it to your Web server app which could serve as a gateway to SAFEnet) so this would happen over HTTP(S) or a streaming protocol served by the Web site.
You could - to make this distributed and “native” build a SAFEnet client support into the video client (someone talked about using safe:// links) or you could use your client’s plugin API to build a SAFEnet plugin for it. Either way, once that is done your client would read the file from safe://net.wo.rk/video.mp4 (which would then fetch data in chunks as appropriate). Note that most clients would get chunks from the closest MaidSafe cache server located near them. The first request would go to vaults with nearest chunks, but if the video remains popular it will be cached (now for free) by nodes in between the client and vault (for each chunk, respectively).

So in short, yes, it’s possible to leverage this to get very good performance (considering the price), but I don’t think you can ever match a good CDN (which is understandable and I think fine - some CNDs are even free and still provide great basic (no encryption, for example) service).

But other parts of the puzzle need to come in place too.
It seems that in absence of a functioning copy protection and ways to effect “takedown notices” and such, one would have to focus on getting as much revenue from live streaming (the first broadcast) which wouldn’t benefit from stored data and perhaps selling “right to retransmit” to traditional Web sites (e.g. a regular Web site wants to stream that content over the Web, charge them per month or per view or whatever).
And then drive traffic to your Web site to make sure everyone knows which of (probably) many copies on SafeNet is served by the actual owner (which is useful as you can get revenue from ads).

Fortunately or not, it seems ad networks will probably become (and remain) one of better ways for content owners to monetize their content (assuming that “reputable” advertisers won’t advertise on sites that store stolen content).

(As requested I won’t go into philosophical battles about copyrights and so on).

EDIT: adding a link that’s relevant to the revenue issue:


#13

I am hoping that’s an old opinion by @dirvine David did recently post the clip from Bill Hicks on ads and how lovely they are.

I posted on the need for the end user to have total control over the end user interface and hypothesized ways to make to make ads and modal ads in particular impractical. David seemed to respond in the affirmative with regard to the crucial user interfaces and their connection to the SAFE infrastructure and the goal of keeping control in the end user’s hands. To me this means browsers, search/trending and community that block the spam and noise. But in the post we have above just looks like post SAFE we can expect the net continues unabated in its conversion to Cable TV and we continue lose our rights and quality of life to the crack down.

Because we are worried about artists, instead of children being lured into smoking or diabetes, we are going to allow the most important thing humanity has produced yet to be converted from a medium for open communication to a medium for barrel scraping and command and control exploitation? Cable was just as destructive and effective prior to the camera and them knowing exactly who you were. If it weren’t it couldn’t be clawing back now. The greatest threat to the future is not the fed or carbon energy. Its sponsorship and sponsored media platforms. If you allow that to continue you will have nothing and no way to get at solutions.


#14

Ohh I see what you’re saying. Thanks @janitor. I wasn’t aware of CDNs (reading up on them now). I was really hoping the network would be good for

Yeah, I’m thinking the same thing. I haven’t seen any good suggestions on ways to defeat the ad and sponsorship model for artists. It works too well and provides too much additional revenue. I’m probably fine with that model for now.

Ohh, thanks @chrisfostertv, i’ll take a look at this.


#15

Ohh, thanks @chrisfostertv, i’ll take a look at this.


#16

Except that its well known that consumers prefer the Netflix model and ad free music systems and ads are not value added but value minus. And the only issue is content strike that a working government and law where anti trust would prevent these content strikes. And if the content where, which it would be without the content strike, Netflix would mop the floor with the useless ad industry and kill if off. But then came all the enclosure stuff with bandwidth caps, proven to be fraudulent and now the lack of neutrality where advertisers will determine who you can talk to and at what price and for how long- much larger prices by the way, possibly prohibitive for many.


#17

Oh yeah? How interesting… I’d love to see the viewership statistics on that! Would you mind linking me to them?


#18

Are you kidding.

Netflix killed Blockbuster and was killing Hulu despite it being backed by Sony etc… The Hulu people kept bragging that their ads were value added but didn’t they despite the backing, didn’t the spin off go under something like 2x against Netflix. Netflix would be on top now if it had a bigger on line library when it tried to kill backwards disk.

People don’t want ads, they don’t want to pay for them and certainly wouldn’t them eating into unnecessary but expensive caps. As for the music system when you can get the universe of music apparently now even the Beatles and Led Zeppelin for $9.99 a month why would you ever want ads? You’ve already got everything. Again no value added. The main reason for Netflix success on digital was no ads. And the reason they are in a fight with useless cable is because the ad enclosure people want you to pay more to be force fed ads. The ad situation is always an enclosure situation and you always end up paying way way more one way or another That goes for the zero rate stuff in the developing world. The cost there is incalculable.


#19

Just to reiterate, you said consumers prefer the Netflix model (subscriptions) and ad free systems. I’m asking to see the viewer stats on broadcast/cable/youtube vs Netflix. Or the number of people using free Spotify (or just the radio) vs paid Spotify.

I agree. People don’t want ads. Why would anyone want ads? We want free food and no mortgage too. That doesn’t matter, though. It’s how much advertising their willing to have shoved down their throat before the go somewhere else to consume. And according to the numbers, the vast, vast majority choose ads over paid-and-no-ads models.

Not saying its better or worse, just the facts.

Also, where are the numbers on Netflix destroying Hulu? I’m not sure that ever happened.

EDIT:

Completely different thing. Netflix offered a subscription rental system that mailed you discs, and eventually offered streaming. Blockbuster was renting physical items. Also, Redbox was a big player in Blockbuster’s demise. Advertising dollars actually managed to have nothing to do with that demise.


#20

I remember Netzero in the states. It failed because it was ads per minute and pure spyware- your browser as an advertisement. One of the programmers her was saying the thought MaidSafe would be slow imagine that being made worse by spam. AOL failed for equivalent reasons. With AOL in 96 or so they were trying to charge people 3.95 per hour to access stuff that resided on their own hard drives that was downloaded from the general internet.

We don’t need numbers on this we have the well known market history and the preference of public for Netflix over cable- it was Netflix radically gaining share on cable not the other way and pretty much without limit. That people are willing to pay for ad free system says everything vice suffer with ads that only diminish quality. Really ad services produce content that becomes all about what the sponsors want because they are he customers. If people weren’t willing to pay for ad free Netflix wouldn’t have succeeded at all and become dominant for a while. It should be clear that even the Netflix hobbled by a content strike being ad free at 1/15th price of cable would kill Cable and that was part of cable trying to play games with its internet access. It succeeded and beat Hulu with better content for a while prior to the stumble brought on by the content strike which, which only intensified the strike afterward. Tons of people paid for ad free total music services. But the issue isn’t whether people are willing to pay for ad free the issue is its possible to provide a more competitive service without ads.

I don’t like this model but much more apt than the ad model would be one that offers access to pretty much all content forms ad free for a nominal fee- that will always beat an ad model in the developed world and ethical governments would spare their populations from the enslaving zero rate models. Ads never win when the markets aren’t enclosed and people have access to what a level playing field can offer.