>>
|
No. 36931
[Edit]
>>36930
>I would assume that band sucks.
That's a very narrow-minded and selfish view.
>It's how culture has always been
There's a reason the burning of Library of Alexandria is considered a tragedy. No one thinks "well it's ok, most of the good stuff was probably copied elsewhere anyhow".
>in terms of cost involved to reward gained
That's fine, no one's forcing you to start archiving random things. Only that if you do choose to archive things, it's a waste of effort to archive things that are already popular.
> >>36927 has the right idea in my opinion. Archive stuff YOU like.
That's me – and "archive stuff YOU like" isn't exactly what I said, the full formal statement coupled with its context would be more like the following: "In an ideal world where disk space is unlimited, we would just chuck everything into a metadata-indexed system to ensure all media has exactly N+2 geographically distributed replicas. In the real world however, we don't have a single unified system, we have a bunch of people all with their own independent archives and no communication between them. As such we don't know how many copies of a media exist at any given time. Nonetheless it is reasonable to make the assumption that the number of replicas of a piece of media is proportional to its popularity (and in many cases this assumption can in fact be verified by looking at e.g. number of seeders). Thus with the goal of ensuring redundancy, the best candidate media for a person to archive are those that are unpopular, as those have the highest risk of becoming unavailable. In the real world we also have the additional constraint that space, time, etc. are limited, so that it's infeasible for a person to just uniformly select a subset of the candidate media to archive. (Indeed, even determining whether something is a candidate for archival requires time and effort). Thus, pragmatically the best option is to try to at least select – out of the pool candidate (unpopular) media – a subset of those that you have/think you would enjoy. Then as time/bandwidth/space allow, sample from adjacent neighbors in that space." That is to say, it's not archive ONLY stuff you like, but "start with archiving unpopular/niche stuff that you have enjoyed/think you might enjoy, and if you can, try to branch out from there."
|