Global data archive

There are several application areas for the technology:
1. Consumer file sharing (non-piracy media commerce network without the heavy handed DRM)
2. Decentralised media diffusion (using the network as broadcast airwaves and for on-demand consumption)
3. Personalised ad insertion (ad insertion without the server farms)
4. Media production factories (broadcast and post production)
5. Enterprise document management (fuss free storage for large companies)
6. Home media networks (wireless media centres and players)
7. Independent media publication (independent media producers cut loose)
8. Media commons (creative commons for video and audio)
9. Global data archive (Baxter)

The Global Data Archive Scenario: Imagine being able to take the most important files on your desktop computer’s hard drive (or all of it, for that matter) and being able to back it up to a place that didn’t charge too much to store it away, was always available and where you could depend on the data being secure from prying eyes, yet robustly backed up against disaster.

The solution could be realised using the NeoPixSys servent. Vendors of backup services could either federate user computers, with each donating some space, or else run distributed servers for the purpose, charging users a nominal amount to keep their files safe and sound. That way, if you lost a machine (it broke or was stolen), reloading your data would be a simple matter of connecting your replacement machine, downloading and installing the NeoPixSys servent and simply calling your files back from the storage cloud, decrypting them with your special password.

Small businesses need never mess with backup tapes ever again. Files could be automatically backed up to the cloud at the end of the working day (or at each save). Disaster recovery scenarios would not see rehearsals involving the corruption of the only backup tape available. Partial restoration would become a reality. No fire could ever destroy your company’s precious data and no hacker could easily crack the encoding to decrypt your data, or find it easy to gather all the atoms that comprise it from the storage cloud. In short, small businesses would find it easier and cheaper to protect their data and sleep more soundly as a result.

The Media Commons Scenario

There are several application areas for the technology:
1. Consumer file sharing (non-piracy media commerce network without the heavy handed DRM)
2. Decentralised media diffusion (using the network as broadcast airwaves and for on-demand consumption)
3. Personalised ad insertion (ad insertion without the server farms)
4. Media production factories (broadcast and post production)
5. Enterprise document management (fuss free storage for large companies)
6. Home media networks (wireless media centres and players)
7. Independent media publication (independent media producers cut loose)
8. Media commons (creative commons for video and audio)
9. Global data archive (Baxter)

The Media Commons Scenario: Advocates of the rights of consumers to freely obtain and use materials that are out of copyright have run into a snag. To host media that is in the public domain costs a considerable amount of money, using the traditional central server farm architecture. There are some government-backed organisations like the BBC or RTL who are willing to contemplate this ongoing expense, but much media is in danger of evaporating once it no longer has a commercial value.

The NeoPixSys servent architecture provides a solution. Being a peer-to-peer network essentially comprised of user-donated disk space and connectivity bandwidth, there is a natural home for creative commons works. The marginal cost to each user of storing archives of out of print and out of copyright works is negligible, compared to the marginal utility of making this material freely available in the network storage cloud. At last, public domain media has a home that cost no single individual or organisation too much money to maintain.

The Independent Media Publication Scenario

There are several application areas for the technology:
1. Consumer file sharing (non-piracy media commerce network without the heavy handed DRM)
2. Decentralised media diffusion (using the network as broadcast airwaves and for on-demand consumption)
3. Personalised ad insertion (ad insertion without the server farms)
4. Media production factories (broadcast and post production)
5. Enterprise document management (fuss free storage for large companies)
6. Home media networks (wireless media centres and players)
7. Independent media publication (independent media producers cut loose)
8. Media commons (creative commons for video and audio)
9. Global data archive (Baxter)

The Independent Media Publication Scenario: Because the peer-to-peer network is inherently symmetric, you can publish into the cloud anywhere, from any node. No node is more powerful than any other. So, for independent media producers, they suddenly have access to a distribution and diffusion network that is no better and no worse than that which a big media company could provide. In other words, big media companies no longer have a monopoly over the distribution networks. This encourages greater diversity of content and means that media companies have to concentrate on the quality, originality and marketing of their content, rather than using their monopoly hold over distribution to control what content consumers can have.

Home media networks

There are several application areas for the technology:
1. Consumer file sharing (non-piracy media commerce network without the heavy handed DRM)
2. Decentralised media diffusion (using the network as broadcast airwaves and for on-demand consumption)
3. Personalised ad insertion (ad insertion without the server farms)
4. Media production factories (broadcast and post production)
5. Enterprise document management (fuss free storage for large companies)
6. Home media networks (wireless media centres and players)
7. Independent media publication (independent media producers cut loose)
8. Media commons (creative commons for video and audio)
9. Global data archive (Baxter)


The Home Media Network Scenario: In the home, consumers increasingly want to obtain their content through a single transaction, via a single portal, and thereafter redistribute the media they have bought, either by physical means or via wired or wireless networks, to other players in the house. That means they ideally want to be able to detect if somebody else has taken their media and is misusing, but they also want few, if any, gateway restrictions on media portability. Content owners, on the other hand, want to know that selling media to the home means the media will stay largely within the bounds of the home.

The NeoPixSys condenser software marks the rendered file with the user’s unique ID. They can play it on any media player they like. However, if a friend were to upload a copy of the media or the copy the user bought to a file-sharing network, the name of the user being embedded would perhaps strain the friendship. If the friend tried to upload the media to the NeoPixSys network, not only would the person who originally bought the media be identified, but so too would the uploader.

In some cases manufacturers of media players may wish to include key matching in their products, so that media owned by a person can only be played on devices owned by a person. That would be more restrictive, but could prevent neighbours listening in on a wireless home media network, getting cable channel access for free, for example.

Again, because no two renderings are ever the same and the copy is a degraded version (albeit slightly) of the original, there is always an incentive to buy your own copy. However, the feature of the NeoPixSys network is that it tolerates low level swapping, lending and copying, with no draconian restrictions. The beauty of this system is that large-scale piracy can be traced, while permitting petty copying and fair use. This is a more realistic trade-off between user and content owner rights than the present proposals for digital rights management. Consumers are trusted up until the point where they breach trust, at which time they can be traced and the offending materials removed from the file-sharing network.

Enterprise document management

There are several application areas for the technology:
1. Consumer file sharing (non-piracy media commerce network without the heavy handed DRM)
2. Decentralised media diffusion (using the network as broadcast airwaves and for on-demand consumption)
3. Personalised ad insertion (ad insertion without the server farms)
4. Media production factories (broadcast and post production)
5. Enterprise document management (fuss free storage for large companies)
6. Home media networks (wireless media centres and players)
7. Independent media publication (independent media producers cut loose)
8. Media commons (creative commons for video and audio)
9. Global data archive (Baxter)

The Enterprise Document Management Scenario: In common with the media factory scenario, enterprises can store and serve all their corporate documents from a NeoPixSys peer-to-peer network, with all the advantages of self-configurability, high availability and reliability, automatic data redundancy, load balancing and self-maintenance, coupled with the advantages of decentralised networks, lower bandwidth costs, cheaper commodity servers, the ability to make better use of idle desktop workstation capacity, the ability to distribute the same data over a wide area network, between sites and so on. Gone are the enterprise storage systems, the huge central servers, the single enterprise database, the sophisticated IT staff that must be employed to provision storage space for new users and clear up data on machines of people that leave. Ownership of the data is without question, since the identity of the corporation is embedded in every file rendered, as is the identity of the accessor. In the case of corporate leaks, the culprit can readily be identified from the encrypted data embedded in the file.

Likewise, data pertaining to patents and copyrights can be automatically annotated, simply by uploading the file to the cloud, when saving the document. In this way, engineering and lab notebooks, traditionally required to prove active work on technologies prior to patent application, in court disputes over ownership, can carry distinct and unique marks stating when the file was uploaded and created. In extreme cases, the corporation may wish to automatically copy the data (source code, for example) held in their corporate cloud to the corporate cloud maintained by a trusted third party escrow company, in real time.

A further advantage is that obsolete desktop machines, no longer powerful enough for use by staff, can be left online in the network and used as storage. Each desktop machine sees the entire storage cloud as a single entity, irrespective of what types of physical machines are used to build it. When nodes eventually fail or wear out, there is no need to recover the data. Adding a new node with a clean disk drive will result in that drive being populated with file atoms automatically, over time. Similarly, when machines must be turned off, the shut down sequence ensures that copies of all atoms held on the node are available elsewhere on the network before allowing the user to log out.

Media production factories

There are several application areas for the technology:
1. Consumer file sharing (non-piracy media commerce network without the heavy handed DRM)
2. Decentralised media diffusion (using the network as broadcast airwaves and for on-demand consumption)
3. Personalised ad insertion (ad insertion without the server farms)
4. Media production factories (broadcast and post production)
5. Enterprise document management (fuss free storage for large companies)
6. Home media networks (wireless media centres and players)
7. Independent media publication (independent media producers cut loose)
8. Media commons (creative commons for video and audio)
9. Global data archive (Baxter)


The Media Production Factory Scenario: In broadcast television studios and postproduction facilities today, it is typical to find a room full of dedicated video servers, with tie lines and huge routing matrix routing content to individual workstations located throughout the building in real time. It would be cheaper to use IP networking. It would be faster to avoid the transfers and copies that are inevitable in the traditional architecture. It would be far more efficient to see all media assets as living on a single, huge, monolithic disk drive, than to have to track copies and versions, via sophisticated media asset management systems, throughout the building and the workflow of the production chain. It would be nice to be able to add capacity by simply plugging in a new server. It would be nice to trust that the data won’t evaporate due to failure of any one server. It would be great to be able to guarantee playout performance, irrespective of network, server or other machine problems or sudden surges in load.

Today, media production factories spend fortunes on mirroring their gear, paying highly trained staff to administer the servers and networks and in specialised, high performance routing systems. Yet, they could solve all the problems listed above with the NeoPixSys peer-to-peer network. With the NeoPixSys servent software and some generic commercial-grade servers, not specifically dedicated video servers, data could be distributed around the production facility by dint of the fact that machines throughout the building are on and storing some of the atoms in the cloud. Adding capacity is as simple as adding a new server and logging it onto the peer-to-peer network. Redundancy is built in. Load balancing is a property of the number of active nodes online.

The media production factory could have a cheap, simple, easy to maintain, highly reliable, highly available, intrinsically secure video network, built from commodity computing components, simply by adopting the NeoPixSys servent on a fast IP network. Better that that, copyright is embedded in the files and unremoveable, so that media cannot be stolen without a trace back to the thief. Users of protected media are warned and cannot be unaware of the provenance of the content. They can throw away their tie lines and dedicated routing matrices. The common apparatus room also vanishes, since the servers can be everywhere in the building (even under desks). If so desired, the plant can throw away the outmoded video servers, which were architecturally modelled as tape deck replacements, and fill their racks with generic servers.

Personalised ad insertion

There are several application areas for the technology:
1. Consumer file sharing (non-piracy media commerce network without the heavy handed DRM)
2. Decentralised media diffusion (using the network as broadcast airwaves and for on-demand consumption)
3. Personalised ad insertion (ad insertion without the server farms)
4. Media production factories (broadcast and post production)
5. Enterprise document management (fuss free storage for large companies)
6. Home media networks (wireless media centres and players)
7. Independent media publication (independent media producers cut loose)
8. Media commons (creative commons for video and audio)
9. Global data archive (Baxter)

The Personalised Ad Insertion Scenario: Much of the media we enjoy is cheap or free because we tolerate the ads that are inserted in it. The advertisers, ideally, want to display ads only to those that will respond to them positively, by taking action to buy the products or services they provide. So, in much of the world of advertising, there is a huge waste of impressions. Ads are displayed to people that can’t or won’t respond to them, annoying them and wasting money for the advertiser. What online advertising proved was that advertisers want pre-qualified audiences, delivering personalised advertising of the type and variety that will cause positive buying responses from each individual member of the audience. To do this for video or audio programming, whether broadcast or on-demand, requires that networks have interception points where ads are inserted specific to local areas. At each of these insertion points, there either has to be an ad server or a dedicated link to a remote ad server.

In our technology, it’s different. The ads can be uploaded into the cloud once and only once. They persist forever, floating around as atoms in the peer-to-peer network. Then, as users select programmes to view or listen to, they can be passed a play list, which consists of pointers to ads that will be of interest to them personally. As they view the programme, the player software condenses the atoms of the show they want to watch, as well as the atoms of the ads that are targeted just to them. Ads become just another form of on-demand content, allowing advertisers to offer reduced rates to downloaders that will watch or listen to the ads. The network can consist of PCs or hard-drive enabled set top boxes or both.

Once again, from the advertisers’ point of view, there are no huge server farms to deal with, no remote ad insertion kit to maintain in the field, precisely targeted audiences and ways to measure response directly, through interactivity. From the users’ point of view, they can opt for cheaper downloads, simply by watching ads that are of genuine interest to them.

Decentralised media diffusion

Today’s Market- Decentralised media diffusion (using the network as broadcast airwaves and for on-demand consumption)



There are several application areas for the technology:
1. Consumer file sharing (non-piracy media commerce network without the heavy handed DRM)
2. Decentralised media diffusion (using the network as broadcast airwaves and for on-demand consumption)
3. Personalised ad insertion (ad insertion without the server farms)
4. Media production factories (broadcast and post production)
5. Enterprise document management (fuss free storage for large companies)
6. Home media networks (wireless media centres and players)
7. Independent media publication (independent media producers cut loose)
8. Media commons (creative commons for video and audio)
9. Global data archive (Baxter)


The Decentralised Media Diffusion Scenario: In this implementation, ‘broadcasting’ is accomplished by distributing media by diffusion avoids large centralized storage and connectivity costs, while avoiding the risks of theft through vaporization of the content into the cloud constituted by the participating nodes on the network. Because all nodes are peers and the load is balanced by the servents, the network can support huge numbers of simultaneous users, while keeping file accesses confined to local network neighbourhoods, hence reducing bandwidth costs, latencies and network hops.

Compared to the massive central server farm architecture, there are no single points of failure in the network, no susceptibility to traffic congestion in particular branches of the network, no upper limits to load scalability and no need to run a central IT services department to keep it all running.

Compared to a traditional content delivery network, there are no risks of theft at co-located caches, no need to have staff maintaining remote caches and no need for specialised edge cache hardware. The servent software on each node, depending on local network conditions, manages network traffic and cache consistency. Management of the content delivery network is decentralised to the point that it becomes the responsibility of each individual user on the network to maintain their machine. Even if a machine fails entirely, the network survives.

Because each user effectively donates disk space and the connectivity to participate in the network, the costs of sending media to massive numbers of customers is next to nothing, for the content owner. Rather than paying a fixed amount for each additional viewer served, as is the case in media serving architectures today, the content owner pays only to upload the content once.

Consumer file sharing (non-piracy media commerce network)

Specific market overviews of implementation of this vision.

There are several application areas for the technology:
1. Consumer file sharing (non-piracy media commerce network)
2. Decentralised media diffusion (using the network as broadcast airwaves and for on-demand consumption)
3. Personalised ad insertion (ad insertion without the server farms)
4. Media production factories (broadcast and post production)
5. Enterprise document management (fuss free storage for large companies)
6. Home media networks (wireless media centres and players)
7. Global data archive (Baxter)

Today’s market-The Consumer File Sharing Scenario:
Typically the user is sitting at a PC and throws away his Napster, eDonkey or Kazaa servent, installs the NeoPixSys Cloud servent instead. Instead of storing entire files on their hard drive and making these available to others through federation of directory entries on the peer-to-peer network, the individual user uploads only those files to which he has ownership rights, since his unique identifier will be written into the file that he uploads. When the file is uploaded, the data is vaporised into atoms, which are coded in such a way that you can lose some of them and replicate them and still recover the data. The entire file is not stored on any user’s hard drive. Rather, the atoms of the vaporised file are sent out into the aggregate storage cloud comprised of the hard drives belonging to the users that are currently logged on. The atoms will be everywhere, diffused throughout the cloud.

Instead of a directory entry, each user will have a resolver that searches the cloud and finds atoms or alternates sufficient to reconstruct the file. These resolver entries can be replicated on other user nodes. As users come and go on the network, the directories and atoms are moved around the cloud before the user logs off. Once the file is in the cloud, it stays there, even if users log off. The only important condition is that there are a minimum number of nodes always on the network. When a user wants to download a file, they get information about it by aggregating resolver entries from peers in their local network neighbourhood first. The process of condensing a file from the cloud is one of gathering enough of the atoms to reconstruct the file, from wherever they happen to be stored in the cloud, rendering the file with the downloader’s unique identity encrypted and appended to the original file uploader’s identity.

The user sees a service, just like Kazaa say, except that when they take a copy, their identity is appended to the identity of the original copyright owner. They can take the media to other players they own, copy it and back it up freely, quote from it, make a single offline copy for their best friend, add it to a compilation and so on, but if they were to mass replicate it, by uploading it to any file sharing network, they would be giving away information about themselves with every copy – a powerful deterrent to piracy. DRM solutions can be added, so that commercial owners of content could charge users for each initial rendering they make from the cloud, when they condense the atoms into a file. After that payment has been made and the copy condensed (i.e. rendered), subsequent copies made from that unique rendering are traceable. The condensing software can also be made to deliberately add a small amount of digital noise to each rendering, so that copies are never identical to the original and copies of copies degrade. This provides a further incentive for buying a copy from the source, instead of making a copy of a copy.

From the content owners’ point of view, there is no need to build a huge on-demand server farm or install network links having massive bandwidth. As long as enough people are online, the files vaporised into the cloud remain in the cloud. Also, because no single node ever has all the atoms that comprise a file, stealing media by copying it anonymously from a single hard drive cache is never possible.

Imagine Trust in the Virtual World

The goal of most content distributors is to deliver their content in exchange for something of value- cash, viewer attention, or some other information that has value. The current model of achieving this exchange involves creating awareness, leading to intent, and then an action- a purchase or giving attention to the content. Today, the digital networked marketplace is bogged down in intermediate steps aimed at reducing the risk of theft by a public that is presumed not trustworthy. Exchange has been stifled instead of opened up. Fear is in the tone of the content distributor’s voice, and the public recognizes that the open hand of the distributor is not held out to them.

What would a world where the customer’s intentions are presumed good be like? Where the power to choose how and where to spend one’s money is what vendors want for consumers? Where satisfying specific needs and wants is the focus of the sales conversation? What if technology made a customer welcome in the virtual entertainment store and content library?
Let’s say you could visit the entire world of possibility in content, and filter it for your tastes. Say you liked a little nudity in your content, but not swearing (something Hollywood doesn’t provide today). Or you enjoy historical battle accounts but not spurting blood. Let’s also say you live in a town that doesn’t get the full offerings of entertainment in your local movie house or on the one independent station you can get on cable or through the air. And let’s say you want to be able to take content with you while you travel, whether it is around the world or from your living room to your bedroom. Let’s say you want to do that with a wireless network.
How could technology do all this in an environment of trust?
It could if the distribution network was built so that one couldn’t steal a product, because the product was never assembled until it was watched or listened to. And if every time the product existed as a complete file, it was modified in a unique way, with an identity specific to that time and that customer. Just to make it even safer, let’s say that unique modification also included a random imperfection negligibly disturbing to the customer, but rendering mass duplication and distribution difficult, while also signaling the source.
In this ideal distribution scheme, any content at every stage during its existence is much like water. It flows freely from place to place, except instead of gravity, it is the wants and needs of its owners responding to the wants and needs of their customers that directs the flow. Like water in nature, the content isn’t put in containers that must be guarded, but rather is dispersed, or vaporized into clouds that float in the ether of the internet instead of the air above our heads.
Each bit of the content spreading itself to multiple locations in the ether is waiting to be called to condense wherever the content is desired. The customer who expresses such desire must satisfy the conditions of the content owner who disperses the content, be it for free, fee or some other exchange, and when the appropriate conditions are met, the customer has the choice to put the content in a container with their own label which will permit them to do the things that have traditionally been the usual and customary behavior of those who have possession- move it around, share it with friends and family. And if such a customer were to attempt to replicate the content in mass, or spread it broadly they would do so knowing that their condensed instance was uniquely and indelibly marked with their identity.
The agent of such magic is a distributed network, utilizing Grid computing concepts blending with the file fragmentation principles of RAID technology. The software to vaporize and condense, or disperse and assemble, as well as manage the constituent parts of the content is analogous to the heat of the sun in this digital content water cycle.
This solution to the challenges of digital content distribution gives content distributors the technology to welcome the public to their stores. It also dramatically reduces the hardware and infrastructure costs of high volume distribution across the network, and essentially makes piracy visible.

Next- specific market overviews of implementation of this vision.