Comment Concerns with the Permanence Distributed Storage (Score 2) 192
In the article, Shawn says "Obviously, a natural extension is to include other types of file formats, things like that. I think it's pretty obvious to most people that Napster is not media specific, but I could see a system like Napster evolving into something that allows users to locate and retrieve different types of data other than just MP3s or audio files."
This struck an odd chord with me. While distributed storage undoubtedly makes sense, what does it do for the long-term preservation of data? As we move towards distributed storage, don't we place the content more and more under the control of trends and fads?
The web is, in a manner, a type of distributed storage too (I think?), and can maybe serve as a weak demonstration of my concern: When I conduct a search and find a myriad of invalid links, it indicates to me that what I'm searching for is an unpopular topic, and has therefore been eliminated to make room for the newest download that, like the data it has replaced, will also be forgotten in a matter of weeks. One can't blame webmasters for this problem - it takes time and effort even to maintain an archive. If the majority doesn't want the data, why put in all that effort? This extends to commercial sites where that time and effort translate into dollars lost.
One might argue that by bringing shared storage down from the level it is on the web to the level of the individual user in fact makes it less susceptible to these trends, as there has to be at least one other person "out there" that wants to share the data with you. But can I really count on this the next time I want to know about the mating habits of Pseudoceros Bifurcus?
Is any work being done on this problem - if, in fact, it is a problem at all?
This struck an odd chord with me. While distributed storage undoubtedly makes sense, what does it do for the long-term preservation of data? As we move towards distributed storage, don't we place the content more and more under the control of trends and fads?
The web is, in a manner, a type of distributed storage too (I think?), and can maybe serve as a weak demonstration of my concern: When I conduct a search and find a myriad of invalid links, it indicates to me that what I'm searching for is an unpopular topic, and has therefore been eliminated to make room for the newest download that, like the data it has replaced, will also be forgotten in a matter of weeks. One can't blame webmasters for this problem - it takes time and effort even to maintain an archive. If the majority doesn't want the data, why put in all that effort? This extends to commercial sites where that time and effort translate into dollars lost.
One might argue that by bringing shared storage down from the level it is on the web to the level of the individual user in fact makes it less susceptible to these trends, as there has to be at least one other person "out there" that wants to share the data with you. But can I really count on this the next time I want to know about the mating habits of Pseudoceros Bifurcus?
Is any work being done on this problem - if, in fact, it is a problem at all?
