TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Netflix Replacing Star Ratings With Thumbs Ups and Thumbs Down

242 点作者 gerosan大约 8 年前

63 条评论

pizzetta大约 8 年前
As I posted in the dupe:<p>This saddens me.<p>&gt;Users would rate documentaries with 5 stars, and silly movies with just 3 stars, but still watch silly movies more often than those high-rated documentaries<p>That&#x27;s not incongruous to me. The stars are not about &quot;enjoyment&quot; factor, they are about perceived quality. I may have a go to cheap ice-cream and rate it 3-stars but rate a good affogato 5-stars and only have it once in a while.<p>They are diluting the meaning of quality and instead are opting for a saccharine &quot;enjoyment&quot; factor. This binary choice does not sit well with me and I hope they abandon the idea soon.
评论 #13896231 未加载
评论 #13894415 未加载
评论 #13894410 未加载
评论 #13894968 未加载
评论 #13895740 未加载
评论 #13895116 未加载
评论 #13894497 未加载
评论 #13896396 未加载
评论 #13894531 未加载
评论 #13894428 未加载
评论 #13896573 未加载
评论 #13895856 未加载
评论 #13898141 未加载
评论 #13895962 未加载
评论 #13894412 未加载
评论 #13895118 未加载
yladiz大约 8 年前
I feel like star ratings, if just given in general, are pretty useless for anything that has multiple qualities. For example, maybe I like the cinematography of a movie but I think the acting is shit, or maybe I think the taste of the beer is nice but it&#x27;s too acidic or sour and wouldn&#x27;t work with specific foods because it&#x27;s overpowering.<p>Airbnb does this pretty much spot on, in that they ask for 1-5 stars for five categories, rather than just &quot;did you enjoy this Airbnb?&quot; I also think that Untappd does this wrong, because in general beers can&#x27;t be described in just a 1-5 star rating without deeper explanation.<p>I think that moving from a pure star rating to a thumbs up and down rating is better overall, if only because it makes me, as a watcher, not have to think as much and therefore give a rating where I might not have before. If I want to go more in depth, I can explain more too.
评论 #13895098 未加载
评论 #13895365 未加载
评论 #13897626 未加载
评论 #13894740 未加载
评论 #13900175 未加载
评论 #13895581 未加载
wvh大约 8 年前
Is this just part of the trend to dumb everything down to a polarising dichotomy? Because the trend to stop people from having a more complex and rational opinion than just picking sides doesn&#x27;t seem to exactly improve discourse or debate, or inform and educate...<p>Wouldn&#x27;t it be more beneficial to have fewer people give a more detailed review to describe what they like and dislike about something? What use is &quot;good&quot; or &quot;meh&quot; compared to for example &quot;I rate this product low because the shoes are very narrow for my normal feet&quot; so you can actual relate it to what you actually are looking for? One person&#x27;s &quot;meh&quot; could be another&#x27;s &quot;good&quot;...<p>From the article:<p>&quot;This makes sense – giving a five-star rating takes some thought, especially for something like a movie or TV show.<p>A binary “yes or no” option is much easier for viewers to commit to, [...]
评论 #13894576 未加载
评论 #13894449 未加载
ykler大约 8 年前
From the linked Variety article, &quot;However, over time, Netflix realized that explicit star ratings were less relevant than other signals. Users would rate documentaries with 5 stars, and silly movies with just 3 stars, but still watch silly movies more often than those high-rated documentaries.&quot;<p>But the signal isn&#x27;t just for Netflix; it is also for users, who might sometimes be in the mood for something silly and sometimes in the mood for something good. Also, people might rather get more suggestions of good movies even if they are more likely to watch bad ones. (Of course, people might also just overrate documentaries.)
评论 #13894390 未加载
评论 #13894076 未加载
评论 #13894364 未加载
评论 #13894889 未加载
kartickv大约 8 年前
If you have many reviewers, you don&#x27;t need to get more information than thumbs up or down from each person. How much people thumbed it up is a good indicator.<p>Star-rating can be too much detail, anyway: if you&#x27;re comparing two shows, and one has a higher fraction of five-star ratings, but the other has a higher fraction of ratings that are four or above, which is better? Star ratings can be too much detail and cause confusion.<p>If you want more detail from each person, you can ask specific questions with a yes&#x2F;no answer, like, &quot;Were parts of it boring?&quot; or &quot;Was it violent?&quot; That&#x27;s probably better than star ratings.
评论 #13894217 未加载
评论 #13894207 未加载
评论 #13894025 未加载
评论 #13894066 未加载
derrekl大约 8 年前
I&#x27;ve been in favor of up&#x2F;down for years over x out of 5 or x out of 10. The core reasoning is: it captures your overall emotional response to something very succinctly. Sure you might hate X more than Y or like A more than you like B, but you&#x27;ll return, revisit, recommend all the things you like and not do so for the things you don&#x27;t like. When you can rate out of 5 or 10 you get thrown into a sort of tizzy trying to rationalize a system of what is 1 star, 2 stars, etc. Do you subtract a star because you think something is overpriced even though you like it? Do you subtract a star because it&#x27;s hard to get it (say slow delivery) even though you like it? Basically you want to start to categorize every facet of a product&#x2F;service which becomes complicated quickly. Thumbs up&#x2F;down avoids all that and I think over many hundreds or thousands of reviews the ratio of up to down will provide a good probability indicator of liking something or not.
评论 #13896257 未加载
评论 #13896338 未加载
minimaxir大约 8 年前
It&#x27;s about damn time.<p>5-star rating systems are <i>broken</i>. A 1&#x2F;2&#x2F;3-star rating is effectively a dislike, and a 4&#x2F;5-rating is effectively a like. In the big data sense, a deviation from a 1-star for dislike and 5-star for like is statistically meaningless. (and this is <i>universal</i>; star ratings on Amazon Products and Yelp Locations have the <i>exact same distribution</i>).<p>The interesting part of this is what Netflix gains from the change, since their recommendation algorithms will become less granular. Maybe they came to the same conclusion?
评论 #13895584 未加载
评论 #13895393 未加载
评论 #13895089 未加载
jerf大约 8 年前
I think a lot of people are reacting here as if Netflix is moving from the only signal they have being the star and review system to the only signal they have being the up&#x2F;down system, and thus confused as to why Netflix would throw away the vast majority of their rating ability.<p>But that&#x27;s not even close to true. In addition to it technically being up&#x2F;down&#x2F;no rating, we&#x27;ve got how long people watch the show for, how the shows they watch form a pattern, how all the patterns of what people watch show global preference patterns, whether they rewatch a show, <i>when</i> they watch what sort of shows, what <i>individual scenes</i> are rewatched vs. skipped... Netflix is swimming in a sea of preference data, not sitting here trying to figure out &quot;Gosh, um, if the user likes this movie 3 vs. 4 stars, uh... what does that mean?&quot;<p>It makes perfect sense to me to optimize this one-data-stream-among-<i>many</i> to increase user participation and get more bits of information from more people engaging with the simpler system, rather than trying to squeeze bits out of the few people willing to use the star system and the even fewer willing to write useful reviews.<p>It isn&#x27;t really even as shocking as it may seem at first. The star system has 6 states, &quot;no rating&quot;, 1, 2, 3, 4, and 5 stars. That&#x27;s 2.6 bits, with some simplifying assumptions [1]. The thumb system has up, down, and no rating; that&#x27;s 1.6 bits [1]. To make up for the bits, you need only see ~40% increase in participation over the current star system... think that&#x27;s going to happen?<p>[1]: The simplifying assumption is that all outcomes are equally likely, but that&#x27;s not true. I don&#x27;t have the numbers to run a more complete information theory analysis, but it&#x27;s not hard to imagine the &quot;no star rating&quot; case is so common that it produces such a small fractional bit that a higher-participation-rate yea&#x2F;nay&#x2F;no rating (if you puth this UI in their face, &quot;no rating&quot; becomes much more meaningful, too) straight-up produces more bits of information on average, and is thus simply an improvement even <i>before</i> considering the superior UI experience. I rather suspect this is the case, some very sensible assumptions would suggest this, but I lack the ability to prove this; the &quot;assume all outcomes are equally likely&quot; is at least a concrete case I can discuss.
评论 #13894956 未加载
评论 #13894366 未加载
AlexandrB大约 8 年前
&gt; The streaming service said it had been testing thumbs up and down ratings “with hundreds of thousands of members” in 2016 – and it led to 200% more ratings being given.<p>I don&#x27;t understand why engagement is the right metric here. If someone isn&#x27;t sure how they feel about a movie, why is it a benefit to have them spew their half-formed thoughts into a like&#x2F;dislike rating?
评论 #13894707 未加载
评论 #13895839 未加载
评论 #13893954 未加载
评论 #13894195 未加载
评论 #13894042 未加载
评论 #13893955 未加载
wj大约 8 年前
I&#x27;m still disappointed they removed their friends feature. I valued my friends&#x27; ratings more than the ratings of the Netflix user base as a whole. I have noticed that Netflix tends to be more accurate with the &quot;users like you&quot; star rating than the whole user base ratings.<p>I no longer see a sum of my ratings but I believe it is well over 2,500. I know others that have rated a lot more (when there was the friends feature you could see your friends&#x27; number of ratings). Because of my number of ratings I felt that Netflix did a pretty good job with recommending me content. I will be sad to see this go as I definitely refer to star ratings when adding content to my queue.<p>Others have mentioned that the two factors of quality and enjoyment make the star rating more valuable and I agree. The only time I remember it breaking down for me was the film Rachel Getting Married (though I am sure there were others). I couldn&#x27;t stand Anne Hathaway&#x27;s character to the point that I gave it one star but at the same time I recognized that she gave a really strong performance in what was probably a good film.<p>Are they converting ratings to thumbs up&#x2F;down? What does a three-star rating convert to? Those are typically movies that I enjoyed but wouldn&#x27;t rewatch or recommend to others.
评论 #13895633 未加载
评论 #13895413 未加载
Tepix大约 8 年前
Bummer. I understand why they do it but I think it&#x27;s too coarse. I&#x27;d much rather know how many people thought a movie was fantastic and not just decent.
评论 #13893843 未加载
评论 #13893891 未加载
评论 #13896269 未加载
评论 #13894041 未加载
frik大约 8 年前
What&#x27;s next. Just a thumb up like on Facebook?<p>Rating is a hard problem. No system works universally well. IMDb (Amazon property) for example uses 10 stars, Amazon uses five stars. Facebook use thumb up (&quot;likes&quot;). Games ratings are often in percent 1-100% (summarized by metacritics.com and others). School systems around the world use A, B, etc or numbers like 1-5 or 1-6 for grades.
评论 #13893946 未加载
评论 #13894119 未加载
评论 #13895099 未加载
评论 #13893919 未加载
sergiotapia大约 8 年前
Amy Schumer&#x27;s comedy special bombed so hard they&#x27;re changing their ranking system? Total failure.
评论 #13896607 未加载
评论 #13896628 未加载
评论 #13895005 未加载
评论 #13894955 未加载
mr_tristan大约 8 年前
At no point do I recall Netflix promoting this rating system as anything but way to get better matches. The fact that so many comments are about losing a system for criticism, just confirms that it was the wrong metaphor to use in the first place.<p>Perhaps they could bring back a &quot;critical review&quot; mechanism, but I&#x27;d guess it wouldn&#x27;t affect your matches at all. And probably be mostly unusable on a TV anyway.
评论 #13895024 未加载
widdma大约 8 年前
A problem I&#x27;d like to see a good solution to is selection bias. Unless there is some kind of reward, only people that really care are likely rate something<p>I think Netflix&#x27;s move might help this. It certainly lowers the cost rating.
评论 #13894081 未加载
flaviojuvenal大约 8 年前
But the UX of the current rating system sucks. Rating stars are very small and you have to hover them to know you can rate. Only if you watch a movie to the very end you will see a clearer rating functionality, but even then you have few seconds to use.<p>I bet many new users don&#x27;t even know they can rate. I wonder if people aren&#x27;t rating much because the UX sucks, not because it&#x27;s a 5-star system.
评论 #13893866 未加载
评论 #13893900 未加载
Viper007Bond大约 8 年前
YouTube got rid of star ratings back in 2009 for similar reasons: <a href="https:&#x2F;&#x2F;techcrunch.com&#x2F;2009&#x2F;09&#x2F;22&#x2F;youtube-comes-to-a-5-star-realization-its-ratings-are-useless&#x2F;" rel="nofollow">https:&#x2F;&#x2F;techcrunch.com&#x2F;2009&#x2F;09&#x2F;22&#x2F;youtube-comes-to-a-5-star-...</a>
kefka大约 8 年前
And not only that, but also removing their existing collections of videos and replacing it with their &quot;Made by Netflix&quot; shows that mostly amount to shovelware.<p>I&#x27;m about ready to drop them, just like they&#x27;ve dropped MASH, soon to be X-Files, and many many movies.
评论 #13893993 未加载
评论 #13893826 未加载
评论 #13893821 未加载
评论 #13894058 未加载
评论 #13894411 未加载
评论 #13893825 未加载
评论 #13893836 未加载
danlindley大约 8 年前
I can&#x27;t imagine any film that conveys a message, cinematography, style, and meaning that could be encapsulated with a simple thumbs up or thumbs down.<p>What a disappointing simplification.
评论 #13893869 未加载
评论 #13893882 未加载
评论 #13894352 未加载
erikb大约 8 年前
I can only wonder why. 5 stars is the normal rating system for movies, and there&#x27;s a huge difference between each star, at least for me.<p>I basically rate like this: unbearable-cannot-finish (1), bad (2), okay-for-background-noise (3), what-I-expect (4), awesome (5).<p>Of course, some movies are somewhat between 3 and 4 or somewhat between 4 or 5. But if you just give me thumbs up or down there are like 80% of movies I cannot rate at all, because they are neither.
peterjlee大约 8 年前
I&#x27;ve always liked Rotten Tomatoes ratings more than IMDB star ratings. Rotten Tomatoes rating is basically, what percentage of critics gave a thumbs up.
adrianlmm大约 8 年前
Terrible idea, thumbs up or down rating only promote political reviews, a movie that makes critics of feminism? thumbs down despite the acting or contents inmediatly, a movie showing the good parts of religion? thumbs down by viceral atheists, some one just needs to look at youtube and see how this doesn&#x27;t work,
sgloutnikov大约 8 年前
I think it&#x27;s a great move. Maybe on a similar note, one of the reasons I really like and find more use out of Foursquare is the way their rating system works compared to the five-star Yelp rating system. For example, in Yelp I found that all &quot;good&quot; restaurants are in the above 4-star rating, and that&#x27;s all the information I can get out of that rating. With Foursquare, they have turned the upvote&#x2F;downvote&#x2F;neutral rating into a number between 1-10 that tells me more. Above average is 7+, something unique about the place 8+, truly exceptional 9+.<p>It just seems to me it&#x27;s a simpler way to rate something by a user, and at the same time classify it more adequately.
rebootthesystem大约 8 年前
What Netflix needs to do is improve their content, not screw around with ratings. The quality of the content has definitely deteriorated over the last few years. It is almost impossible to find good material to watch. I have found myself wanting to watch a good movie only to give up after searching for half an hour.<p>Amazingly enough non-premium TV channels have better movies than Netflix, so we record them and skip the commercials.<p>Frankly, the only reason we keep Netflix is because there are a couple of kids TV series our kids like. We are using them to improve their Spanish language skills. If it weren&#x27;t for that we would have cancelled a long time ago.
ggregoire大约 8 年前
I wish we could have the IMDB&#x2F;RT ratings directly in Netflix and a page with all the movies sorted by those ratings.<p>There are articles on Internet like &quot;top 100 Netflix movies of the month&quot; but it&#x27;s almost always for the US Netflix.
评论 #13895569 未加载
评论 #13897147 未加载
fmihaila大约 8 年前
The way I understand their explanation is that users tend to view the act of rating as a stronger endorsement than it actually is. In other words, users think their ratings are more predictive of what they would enjoy in the future than the implicit signals that can be derived from actual behaviour. Netflix concluded that it&#x27;s better to boost the weak(er) signal by encouraging users to rate more. They will continue to use it on top of the stronger implicit signals to provide personalized ratings, which they will display as a percentage match as long as it&#x27;s greater than 50%.<p>So their recommendations may actually improve.
DaveWalk大约 8 年前
Question to the Machine Learning folk: are five-star ratings &quot;better&quot; than thumbs up&#x2F;down, or is it just a matter of algorithm design?<p>I know ratings&#x2F;prediction has long been studied by the MovieLens.org scientists.
评论 #13894995 未加载
评论 #13894572 未加载
shdon大约 8 年前
I get why they do this and it&#x27;ll probably work out fine, but for me personally, it&#x27;s going to mean I will rate a lot fewer movies&#x2F;shows. In the past, I&#x27;ve rated pretty much every single thing I&#x27;ve ever watched on Netflix. On YouTube, I hit the thumbs-up or thumbs-down buttons on a very small fraction of the videos that I watch. That&#x27;s just because there is no middle ground (and no 2 or 4 star ratings either). There&#x27;s plenty of stuff that I enjoyed, but didn&#x27;t love, plenty of stuff that I disliked, but didn&#x27;t really hate.
openasocket大约 8 年前
I don&#x27;t have particularly strong opinions about this change, I don&#x27;t actually rate many things that I watch (I think the recommendation system can go mostly by what I watch).<p>I think the fundamental problem with different rating systems is that they are all one-dimensional. Maybe a two-category system, like 1-5 stars for &quot;enjoyment&quot; and another 1-5 stars for &quot;quality&quot;. Or a 1-5 rating paired with a label, like &quot;campy&quot; or &quot;serious&quot;, to prevent apples-and-oranges comparisons.
Pxtl大约 8 年前
In most cases I would agree with the simple up&#x2F;down rating, because we&#x27;re seeing the lunacy of &quot;less than 5* is failure&quot; in various platforms. But in the case of Netflix where it&#x27;s primarily being used to manage my own personal preferences on content, some nuance of &quot;more like this&quot; and &quot;well I kinda enjoyed it but it was badly flawed&quot; and &quot;meh&quot; and &quot;never ever show this to me again&quot; is useful.
muninn_大约 8 年前
I wish there was a &quot;meh&quot; option. Sometimes things are ok, but not good or bad. This will leave me not rating a bunch of things unless they&#x27;re good or bad.
评论 #13894274 未加载
评论 #13894211 未加载
eiopa大约 8 年前
I understand people who want more expressiveness in their reviewing, but the reality is that the 5-star system just leaves too much ambiguity.<p>It&#x27;s like how in Uber&#x2F;Lyft, most people just default to 1&#x2F;5 star ratings. In that case, the average rating a driver has to maintain gets skewed to a a pretty high number, like 4.5, and people who think &quot;oh, this was a pretty good ride! 4 stars!&quot; end up unintentionally boning the driver.
andy_ppp大约 8 年前
You could come up with an algorithm that instead of showing an average of 5 stars, biases the weighting towards people who rate things in a similar way to you. We all have an internal set of assumptions about what a 5 star rating system means to us; this is a case where the filter bubble leads to better understanding. People would then see the ratings they expected rather than a mix of different methodologies averaged?
drivingmenuts大约 8 年前
I don&#x27;t rate things on Netflix (and rarely on any other service). If I made it thru the movie, I probably enjoyed it; if not, I didn&#x27;t.<p>What I would <i>like</i> to see is a collection of all available movies by genre, and I&#x27;ll pick what I want to watch. That&#x27;s it - I don&#x27;t need or want recommendations. In fact, if it&#x27;s recommended, my first thought is &quot;who&#x27;s getting paid for that?&quot;
ultrasandwich大约 8 年前
I&#x27;ve found some solace using Mubi, which has a pretty devoted film fan base that really puts thought into written reviews + star ratings. I almost always read a handful of these before following-through with a viewing. They also have a nice 1-in-1-out movie per day system which lets them theme films by event (Cannes, director birthdays, etc). Not for everyone, but it&#x27;s out there.
评论 #13896330 未加载
Houshalter大约 8 年前
Normalize star ratings. If something has an average rating of 4 stars, but four stars is in the lowest 10 percent, it should be given 1 star. This solves the classic issue with star ratings where anything under 5 stars is below average.<p>Personally I use star ratings badly myself. I only ever rate titles 5 stars. But that&#x27;s because I mostly only watch stuff I am fairly sure I am going to like.
yitchelle大约 8 年前
I have always wanted to get ratings from a selected group of viewers who has similar tastes as me. For example, I like SciFi movies that involves planetary travel, ie Star Trek or similar style movies. If I can get ratings from folks with similar interests as well, then the rating system would make sense for me. Otherwise, it is still a hit and miss for me.
评论 #13894450 未加载
cpeterso大约 8 年前
I already push my Netflix ratings higher or lower than I really feel just to more heavily influence Netflix&#x27;s suggestions. Otherwise, Netflix thinks I would give a 3.5 stars to pretty much all movies, which is not helpful to me.<p>That said, a simple thumbs up&#x2F;down doesn&#x27;t allow me to tell Netflix which movies were not bad and which to never show me again.
garrickvanburen大约 8 年前
I&#x27;ve been a Netflix subscriber for a decade.<p>To me, 3-stars means: this movie has all the characteristics of something you&#x27;d enjoy - but we don&#x27;t think you will.<p>Unfortunately, it&#x27;s been more and more challenging for me to find 4- &amp; 5- star stuff in their streaming interface.<p>Perhaps the thumbs-up will help expose those things, or at least help me hide the thumbs-down things.
lukeschlather大约 8 年前
Netflix&#x27;s star rating system is really poorly designed. Of course people clicked on a thumbs up&#x2F;thumbs down widget more. I used Netflix for months before I even noticed the star rating was interactive.<p>I wish they would invest in a quality star rating system like Amazon&#x27;s, rather than dumbing down the UI.
sugarygrind大约 8 年前
My Previous ratings already in Netflix, was the reason for being there. This dumbed down version of up&#x2F;down voting is ridiculous. Most companies become sophisticated. Netflix is going backwards in time.<p>All in all this is a RIDICULOUS move. I wish I could download all my previous ratings before they remove them!
coding123大约 8 年前
I&#x27;m going to guess they looked at things and found that 80% of the ratings were either 5 stars or 1 star. (edit - not that this is a good thing.. I&#x27;m guessing they could have carved out the 2-4 star ratings and did its own data analysis that probably provided much more nuanced information)
bjarneh大约 8 年前
What I really want to avoid is the <i>mediocre</i> content i.e. 2.0 - 3.5 stars, very good and very bad can both be entertaining..<p>Now that mediocre content will be scattered across the two categories I really care about; I guess IMDB&#x2F;Rotten tomatoes will steer clear of this ridiculous binary rating systems.
评论 #13893951 未加载
评论 #13893983 未加载
relaytheurgency大约 8 年前
There&#x27;s always a better option if you like metrics: instantwatcher.com sorts Netflix&#x2F;Amazon content with a variety of external metrics like IMDB score etc. It&#x27;s pretty great if you are just browsing (rather than seeking recommendations).
chrisper大约 8 年前
After almost 5 years I finally cancelled my Netflix subscription. I just realized I was watching more movies &#x2F; shows on Amazon, Youtube, or any of the other paid ones more than Netflix. I used to enjoy it a lot more back then.
评论 #13895488 未加载
yashksagar大约 8 年前
Why not keep the rating system, and have an additional thumbs up&#x2F;down rating? That way you get to vote on the &#x27;objective&#x27; quality and the &#x27;subjective&#x27; enjoyment factor of the content.
vthallam大约 8 年前
This is what even Google does now for the movies and tv shows on the search. If you are logged in while searching for a movie or TV show, it gives thumbs up or down and shows the approval rating in total.
neuronexmachina大约 8 年前
Uh oh, now they&#x27;re going to need to re-film the Black Mirror &quot;Nosedive&quot; episode (s3e1) using thumbs-up&#x2F;down instead of 5-star ratings...
j_s大约 8 年前
The best way to rate a movie for me would be: when will I watch it again?<p>They kind of already know this though. (How many times have I watched a particular movie on Netflix?)
评论 #13894699 未加载
nrki大约 8 年前
Should be &quot;five star-rating&quot;...&quot;five-star rating&quot; indicates they are replacing ratings of five stars :)
PhasmaFelis大约 8 年前
I don&#x27;t have Netflix, so I &#x27;m not clear if the rating shown to users is changing. Or does it not show a rating, just use it to decide what movies to surface for you?<p>In any case, simplifying the ratings users can bestow makes a lot of sense, unfortunately. We like to imagine people carefully considering the merits and flaws like a professional reviewer, but the fact is that the <i>vast</i> majority of users only ever use two ratings anyway: if they like it, 5 stars; if they didn&#x27;t, 1 star. There&#x27;s maybe a third option where if they&#x27;re ambivalent they just don&#x27;t rate it.<p>So star ratings aren&#x27;t actually very useful for evaluating products. This xkcd <a href="https:&#x2F;&#x2F;xkcd.com&#x2F;1098&#x2F;" rel="nofollow">https:&#x2F;&#x2F;xkcd.com&#x2F;1098&#x2F;</a> made me realize that an Amazon product which has a 20% chance of exploding when you open the box, and otherwise works normally but unexceptionally, is gonna average out to a four-star rating for a product you really shouldn&#x27;t buy. I always look at the one-star reviews before buying stuff now, no matter how few of them there are; knowing a thing&#x27;s failure modes is much more useful than a bunch of praise.
评论 #13894724 未加载
kochandy大约 8 年前
Whenever I would browse the Netflix reviews it appeared most people voted either 5 stars or 1 star anyway.
Fej大约 8 年前
Interesting. YouTube made the same switch some years ago, I wonder how it&#x27;s worked out for them.
nailer大约 8 年前
YouTube used to have a star based ratings system too. I wonder if they drew the same conclusions.
wiseleo大约 8 年前
Great. I can&#x27;t stand 5-star systems. Anything other than 5 stars is negative signal.
ed_balls大约 8 年前
Netflix recommendations are quite bad.<p>You&#x27;ve seen this standup? How about watching it again a day later.
oculusthrift大约 8 年前
this makes sense to me considering most people only leave bother to use the extreme star amounts anyway. anyway, you&#x27;re kind of a sap if you take netflix ratings seriously and don&#x27;t use something like rotten tomatoes
jschwartzi大约 8 年前
What they should be asking is &quot;would you watch another movie like this?&quot;
sodapopcan大约 8 年前
Well, if no one else is gonna link it I may as well: <a href="https:&#x2F;&#x2F;xkcd.com&#x2F;1098&#x2F;" rel="nofollow">https:&#x2F;&#x2F;xkcd.com&#x2F;1098&#x2F;</a><p>I&#x27;m all for this!
mrmondo大约 8 年前
Life is not black and white, how you feel about something isn&#x27;t terrible or brilliant every time. This is idiotic.
instaheat大约 8 年前
What does this say about their algorithm? Forget UX&#x2F;UI.<p>I feel as though this devalues their recommendations to you.
Markoff大约 8 年前
so how exactly i distinguish between liking something a little and loving something?<p>this seem like pretty retarded decision<p>at least we have TMDb where we can rate and discuss movies, unlike IMDb or Netflix
评论 #13895987 未加载
true_tuna大约 8 年前