A Storage Solution for MISSION ETERNITY

Symposium:


Presentation Title:
  • A Storage Solution for MISSION ETERNITY

Presenter(s):


Abstract:

  • We present a low-cost, long-term storage solution which is currently being developed for the MISSION ETERNITY project presented at ISEA2006 by the etoy.CORPORATION. The persistence of data over time-scales much longer than that of typical storage media is achieved using a collaborative distributed network, similar to e.g. freenet or the OceanStore projects. In this paper, we provide an overview of the technical, social and usability challenges encountered during the development process and our approaches taken to solve or circumvent them.

    MISSION ETERNITY, a project launched by etoy.CORPORATION at ISEA2006 aims to provide ultra-durable interactive portraits of selected individuals.

    A central problem encountered in the implementation of this project is data storage over long periods of time (ideally hundreds of years). Currently used storage procedures typically involve storing a single copy on hard-disk or cd, and tacking on the occasional, often ill-defined, backup policy. The lack of reliability of this procedure is unacceptable in the context of MISSION ETERNITY and is even gradually developing into a real problem in digital day-to-day life. We have therefore looked for a storage solution which (i) would afford the required reliability, (ii) avoid the high cost of acquisition and maintenance of specialized hardware, (iii) be sufficiently simple to be easily usable and maintainable by everyday computer users (i.e. every single member of etoy.CORPORATION).

    Collaborative software and the peer-to-peer paradigm as implemented in projects such as the freenet, OceanStore or BitVault packages clearly showed the way forward in solving (ii), but were, unfortunately, not themselves suitable solutions for MISSION ETERNITY: freenet, being focused on anonymity of information sources and consumers, deliberately forsakes storage reliability (i), OceanStore, a research project under progress at UC Berkeley seemingly has no ambition to develop into an easily usable desktop software package (iii) and BitVault, being a closed-source commercial product would be impossible to deploy by a third party (i.e. etoy.CORPORATION) in an open, social environment (iii) in addition to the risks inherent in basing a long-term archival solution on a product one has no control over (i).

    Making “The Network is the Computer” come true imposes a healthy mix of constraints, which seemingly Sun itself (or anyone else for that matter) has so far not been able or willing to fulfill: To provide (a) a portable, easily usable software package which implements the required peer-to-peer functionality, as well as (b) implementing a trust and security mechanism which would enable arbitrary collectives of users to pool their storage resources easily and safely and (c) make this system available to everyone for free, essential for the virality of such a system, but implying that since the users then provide both the infrastructure and the content, there is no (direct) way to make money off it.

    Clearly, etoy.CORPORATION, being a high-tech think-tank dedicated to the production of viral products, and willing to do so without necessarily making a (commercial) profit is in a unique position to launch such a system at the interface of the legal, social and technical problem complexes, and has started its development one month ago, with a working prototype expected to be available in april/may and a finished product presentable at ISEA2006.

    This paper will be mostly technical, consisting of an overview of the storage problem encountered in MISSION ETERNITY, and then proceeding to discuss implementation details such as networking, encryption and trust models employed in our system, as well as detailing the integration with standard desktop systems (thereby affording an essentially flat learning curve). After the description of the employed data structures and algorithms we will discuss the tools, languages and libraries used in the development process and conclude with an outline of possible future developments and applications.

    etoy.VINCENT, Zurich, Switzerland & etoy.MONOROM, Zurich, Switzerland. etoy.CORPORATION is art and invests in art. The firm represents the core and code of the corporate sculpture, and controls, protects, promotes, and exploits the cultural substance (intellectual property), particularly the U.S. trademark “etoy” and the etoy.ARTCOLLECTION. The firm shares cultural value and intends to reinvest all financial earnings in art—the final link in the value chain. By sharing risks and resources etoy.AGENTS, art collectors and fans invest time, knowledge, and ideas (or simply capital) in etoy.OPERATIONS, which focus on the overlap of entertainment, cultural, social and economic values. etoy.SHAREHOLDERS participate in a dynamic artwork that takes place 24 hours a day in the middle of society—on and offline.

Category: