I understand the frustration with current recommender systems, but what are the alternatives? Adding a bunch of arbitrary buttons or sliders isn't likely to actually improve the experience.

"People who liked this also liked this" is already a pretty good way of discovering things. The problem is when the catalogue is crap. I get plenty of good recommendations on Prime Video and Criterion Channel, which both have a substantial number of great films available to watch.

▲

jjulius 2 days ago | parent | next [-]

>... but what are the alternatives?

I don't want "recommender systems", I just want easily-browsable catalogues that start out being sorted by their actual genre (looking at you, Netflix homepage with a bunch of bullshit 'categories'). I want to be able to look through things at my own pace and in my own way. I want to browse through movies/shows the same way I browse for records in a record store.

I'm happy to take recommendations from friends and family members who know what my nuanced tastes are, I don't want recommendations from some unknown algorithm built by someone for the purpose of keeping me on the platform.

▲

fasterik 2 days ago | parent [-]

The article is about improving recommender systems, not replacing them. I generally agree that the best way to discover things is through other information channels like friends and family, reviews, curated critics' lists, etc. Prime Video has a decent browse by genre feature, but I don't find browsing all that successful. For example, if I want a comedy, I'll look at a list by BFI or They Shoot Pictures and then pick one that looks good from those.

https://www.bfi.org.uk/

https://theyshootpictures.com/

▲

jjulius 2 days ago | parent [-]

>The article is about improving recommender systems, not replacing them.

But what's a "recommender system"? Are my friends and family who understand my tastes at a deeper level than an algorithm not a "recommender system"? Is your approach to finding comedies via BFI/They Shoot Pictures not a form of a "recommender system" for you?

If recommender systems truly depend on a deeper level of nuance that they likely won't ever be able to get without people giving up even more privacy than they already do, shouldn't we look at alternative forms of recommendations as a form of improvement, rather than sticking to altering what's already there?

I'd argue that using algorithmic recommendations limits exploration and agency and doesn't allow us to develop our taste beyond our taste's current horizons. Without them, serendipity can be genuinely serendipitous if not moreso, our effort can increase attachment/pleasure when it pays off, and we have more opportunities to build knowledge along the way.

I might be digressing just a bit here, but I fundamentally believe that the answer to broken algorithmic feeds, or recommendation systems, or whathaveyou, isn't going to be found in more algorithms.

▲

fasterik 2 days ago | parent [-]

>But what's a "recommender system"? Are my friends and family who understand my tastes at a deeper level than an algorithm not a "recommender system"?

Not really. "Recommender system" has a pretty narrow technical definition. It's an automated system that takes input from the user (clicks, watch time, ratings, etc.) and produces a list of recommendations.

Maybe what you don't like is the average user's preferences. People seem to like being told what to watch. If there's demand for a streaming service that dispenses with recommendations, then someone should be able to make a lot of money building it.

	▲	jjulius 2 days ago \| parent [-]
		Yeah, sure, that very narrow technical definition does exist, but it doesn't have exclusive ownership of the concept, and broader social definitions also exist. Why should we limit ourselves to thinking about it so narrowly? Must an algorithmic problem require an algorithmic solution?

▲

nemomarx 2 days ago | parent | prev | next [-]

I miss when you could rate how much you liked something after a watch and they could incorporate that. Now it's just kinda "how long did you watch it and who else watched it" which seems less nuanced?

▲

inigyou 2 days ago | parent | prev [-]

Run the same recommender on things I clicked dislike, as if it was like. Blacklist anything that algorithm recommends.