4 ms·

This is the same reason I like libraries and book stores. Especially used book stores. You can find some interesting books at "random". Generally random of cour

This is the same reason I like libraries and book stores. Especially used book stores. You can find some interesting books at "random". Generally random of course. But these are books that would, 9 out of 10 times, never be recommended to you by an algorithm. Taliban by Ahmed Rashid was one such book. I found it late last year by accident at the library. Totally changed my perspective on the war on terror. A way better understanding on how 9/11 came to be (the book was published in 2000). That led to the past year of reading history books on various subjects. And I'm happy. While my amazon account only ever recommends tech and business books to me. Which all generally sadden me. Oh plus, this was a few years ago, I road tripped through the Oregon coast. I went into used book stores and bought a metric fuckton of old school pulp fiction crime noir books for like a buck each. All published in the 20s and 30s. Holy crap, so awesome. Again, never would have been suggested any of those if I was walking around and picked random books to look at. Random and physical is nice. Digital and "recommended" has its place. But the real world can't really be replaced.

yepthatsreality 7y ago
I have come to the conclusion that digital/algorithmic recommendations will always be marketing influenced and lack the context to make an absolute guaranteed recommend. Random-physical and friendly suggestions are more accurate and reveal much more as they come from sources rather than a machine "guessing" what will work. Algorithms are also subject to poisoning. Friendly suggestions are subject to poisoning but the risk is of course much less. TLDR; I enjoy the work required for discovery. For example, I'd rather cancel Netflix streaming and returning to Netflix physical for missed releases, hidden nuggets/gems, hard to find, and better consumption management.
- corodra 7y ago
  I agree. But again, I'm not 100% against algorithmic recommendations. Same way I'm not always 100% for human curated recommendations either. Both have pros and cons. What I liked back in the day was the original netflix recommendations when they first started. From what I understand, they did it where they matched movies you like/dislike compared to other people that roughly liked/disliked the same. Then they recommend movies and guess how you might feel over a movie you haven't seen. Those I have watched, but haven't rated yet, the ratings were pretty close to how I felt about said movie. Then, it always seemed like the end of the list of recommendations was an oddball film. I loved those odd ball movies. Always out of my "known territory" of films. 70% of the time, yea, never ventured to watch. But the 30% that I did end up watching were always worth it and never regretted. That's the system I like. A decent mix of the pros and cons from both worlds.
randcraw 7y ago
My greatest regret with the rise of Amazon is the fall of used bookstores. No pastime compares to browsing a hole in the wall bookshop with creaky floors and narrow passageways where books of every variety climb the walls as far as the eye can see in all directions. Though Amazon's policies are a mystery to me, I very much doubt they will never recommend a book to you that isn't still in print and/or available as new. I also suspect they'll downplay publishers or authors who have lost favor with them for some commercial or legal reason. Personally, I hate having my view of the world curtailed by giant corporations whose motive is only profit.

yepthatsreality 7y ago

I have come to the conclusion that digital/algorithmic recommendations will always be marketing influenced and lack the context to make an absolute guaranteed recommend. Random-physical and friendly suggestions are more accurate and reveal much more as they come from sources rather than a machine "guessing" what will work. Algorithms are also subject to poisoning. Friendly suggestions are subject to poisoning but the risk is of course much less. TLDR; I enjoy the work required for discovery. For example, I'd rather cancel Netflix streaming and returning to Netflix physical for missed releases, hidden nuggets/gems, hard to find, and better consumption management.

corodra 7y ago
I agree. But again, I'm not 100% against algorithmic recommendations. Same way I'm not always 100% for human curated recommendations either. Both have pros and cons. What I liked back in the day was the original netflix recommendations when they first started. From what I understand, they did it where they matched movies you like/dislike compared to other people that roughly liked/disliked the same. Then they recommend movies and guess how you might feel over a movie you haven't seen. Those I have watched, but haven't rated yet, the ratings were pretty close to how I felt about said movie. Then, it always seemed like the end of the list of recommendations was an oddball film. I loved those odd ball movies. Always out of my "known territory" of films. 70% of the time, yea, never ventured to watch. But the 30% that I did end up watching were always worth it and never regretted. That's the system I like. A decent mix of the pros and cons from both worlds.

randcraw 7y ago

My greatest regret with the rise of Amazon is the fall of used bookstores. No pastime compares to browsing a hole in the wall bookshop with creaky floors and narrow passageways where books of every variety climb the walls as far as the eye can see in all directions. Though Amazon's policies are a mystery to me, I very much doubt they will never recommend a book to you that isn't still in print and/or available as new. I also suspect they'll downplay publishers or authors who have lost favor with them for some commercial or legal reason. Personally, I hate having my view of the world curtailed by giant corporations whose motive is only profit.