Hacker News Reader

Comment by wodenokoto | Hacker News Reader

A librarian friend had the same complaint.

The problem is that libraries Dewey decimals are managed by librarians who want to sort things correctly. YouTube would be managed by uploaders who wants their stuff to be managed _incorrectly_.

YouTube recommendations and search is a super interesting problem not just because of the scale but also because uploaders are an adverse opponent, trying to keyword stuff their spam.

View on HN

The obvious solution is to actually have librarians correctly classify the videos. DDS focuses on the nature of the work itself, not on the keywords or spam in the content. Librarians understand how to class all kinds of works, and it should be relatively simple to build a DDS/MDS index (Melville Decimal System since it's open, see https://librarything.com/mds) for YouTube videos. Just like with books, disagreement on classification is inevitable and perfectly natural; there's no perfect classification scheme, though DDS/MDS does a generally good job.

Dewey Decimal is probably not actually appropriate but it would be nice to have a good and appropriate classification scheme be used.

> The obvious solution is to actually have librarians correctly classify the videos

Which videos? The 500 hours of video uploaded every minute?

All videos of accounts with more than X subscribers.

There is already auto captioning done by YouTube. It would be trivial to plug an ai that generates tags and classify each videos based on the whole content. I am sure they already do that.

Reply on news.ycombinator.com