安装 Steam
登录
|
语言
繁體中文（繁体中文）
日本語（日语）
한국어（韩语）
ไทย（泰语）
български（保加利亚语）
Čeština（捷克语）
Dansk（丹麦语）
Deutsch（德语）
English（英语）
Español-España（西班牙语 - 西班牙）
Español - Latinoamérica（西班牙语 - 拉丁美洲）
Ελληνικά（希腊语）
Français（法语）
Italiano（意大利语）
Bahasa Indonesia（印度尼西亚语）
Magyar（匈牙利语）
Nederlands（荷兰语）
Norsk（挪威语）
Polski（波兰语）
Português（葡萄牙语 - 葡萄牙）
Português-Brasil（葡萄牙语 - 巴西）
Română（罗马尼亚语）
Русский（俄语）
Suomi（芬兰语）
Svenska（瑞典语）
Türkçe（土耳其语）
Tiếng Việt（越南语）
Українська（乌克兰语）
报告翻译问题
讨论规则及指引
举报此帖
Positive and Negative are like only offering a 0 out of 5 stars or 5 out of 5 stars. Honestly few games deserve either.
A 5 Star review system offers a more accurate consensus which is why many of the other big online stores use it (ie, Amazon and GOG).
And for the rare who do, the reference will be different each time : your 4-star may be the next user's 2-star, and depending on several criteria, the difference can be bigger, smaller or even reversed.
Both are tackled on with a simple binary system : the text box coming with the review can allow you to elaborate on your rating, mitigate it or supply your own.
This is why Youtube shafted its 5-star system for a like/dislike bar, and why the 5-star rating is unlikely to be considered.
If that's the case, that should be reason enough to strongly consider its implementation. If a lot of people are voicing something, there's a reason for it.
I beg to differ, take a look at GOG and Amazon's reviews. They give a clearer breakdown of review scores and review ratings are spread across the board evenly, respectively ofcourse (people use 2, 3 and 4 ratings also). Infact, having the 5 Star rating system encourages more accurate reviews.
Not only that, take a look at Metacritic's reviews and you'll quickly notice that it's the 10/10 and the 0/10 (binary) reviews which are the least trusting to go by and are best left ignored.
Offer the option and people will start to review more accurately, it won't be a rarity, it will happen always as it will be a requirement for posting the review. It's true, our views on things all differ but there is no mistaking 2 stars as being a positive review score or 4 stars as being a negative review score.
Some people don't want to spend too much time reading through reviews but would find a quick glimps at a 3.5 star rating on a game very informative. Also, refer back to what I said about Metacritic.
Irralivent as people on YouTube are not reviewing something they had just spent money on. You can easily load up a video and close it in a matter of seconds without any loss if it's not to your liking. Purchase a game and find out 2.5hrs in that, due to the overwhelming positive reviews, the game does not live up to your expectations and you feel cheated, tough luck.. either learn to enjoy your game or learn to enjoy your buyers remorse.
Valve already has Mixed, Mostly Positive, Very Positive and Overwhelmingly Positive (and I assume their negative counterparts). Why not just map those to particular star scores?
Mixed would obviously go somewhere around 2.5, while Overwhelmingly Positive would give five stars and Overwhelmingly Negative would give zero stars.
At first glance they both seem similar, but the defining difference between the two rating systems is the quality of the data being summarized. And crappy data = crappy results. I'll give you an example.
100 people review a game on both rating systems.
On a 10 Star rating system (10 stars illustrates the difference better than a 5 star system), those 100 people give the game a low positive score of 6-7 stars each which signifies a pretty average game that wasn't a disaster or plagued with issues. It's so so. A buyer could jump in expecting an average experience and get what they pay for.
On Steam's current rating system, those same 100 people have no choice but to vote positive. They clarify their reasoning in their reviews but regardless, the game now has 100 positive reviews and 0 negative reviews. The system compares the two and now rates the 6-7/10 game as being Overwhelmingly Positive.
An 8 or 9/10 game, on the other hand, with a few negative reviews from people who experienced technical issues and decided to review the game based on not even getting a chance to experience it(!!) may get an undeservingly lower rating.
In a star rating system, you would be able to clearly see that all of the 1 stars on a good game were most likely troll reviews or people who couldn't get the game to run propperly, so you could ignore those reviews.
Mind you I think Steam is pretty clued in on this already which is why the rating system hasn't changed despite many requests. This current system dishonestly glamorizes games which don't deserve high ratings as being overwhelming positive and hence sells more copies.. which is ethically questionable.
To be extreme: A lot of people advocate to banish Israel from the face of the Earth. Should we strongly consider it because there is a reason? Or should we just ignore them because it goes against what we stand for?
If Valve wanted a rating system, they had implemented it instead of the binary recommendation system we have now.
READ THE F*ING REVIEWS! The "one number to fit them all" approach is poison in this industry. Implementing a score system in something as popular as Steam would lead to the same dependency metacritic still holds on some.
How differs a 3.5 star game from a 4 star game? You can't measure this per se arbirtrary score which is applied by thousands of anonymous users. As has been said: one's 2 star is another one's 4 star. And yet it will create an environment where users ignore everything below a certain treshhold and call it utter junk. And developers will miss out on bonuses because they're missing 0.05 points to the contractual bonus.
And with everything which is wrong with the review system now - joke reviews, 0.1h bashes, non-content oneliners, ... - a finer score system would bring absolute nothing. You'd still have "5/5 - IGN" or "0/5 thought it be something else" reviews.
It is proven that people will either go to the extreme to make a point or just choose the middle option going "meh".
It could also be like game trading : lots of people asking for the option, with few people considering and dealing with the counter-arguments, does not lean that it will get implemented.
Metacritic's 0 and 10 reviews still get counted on the general score tally, so they are not quite ignored per se.
I could admit that Amazon's and GoG's ratings are evenly matched, but then again, why are Metacritic's ratings not ? This seems to imply that the 5-star rating requires the reviewer's responsability to use it, a thing that seems unlikely due to the joke reviews being numerous enough to ruin the system for some.
Except when it is : If one person rate a game 2 stars because of a deature you don't hold a great deal of, the game might great. This goes the opposite way too, so it's still an issue.
Also, I'd be wary about users self-regulating themselves : as joke reviews showed, this is not the case.
Nothing stops the reviewer to supply their own rating and make it readable to cater to this group of users. The others are free to read the review to know more.
If the reviews are overwhelmingly positive, it's translate to a 4-star rating at worst. Hence, the issue is not being tackled on.
If it would score below, that leans the reviewer wrote something to counter-balance the positive rating, which solves the issue as soon as one reads it.
It's just a remapping of the existing percentage to a 5-star rating, so another question would be "Why generate a 5-star rating from the percentage of positive reviews ?"
Seems like additional effort for no practical gain.
That's comparing Oranges to Bananas buddy, bad example. Though what you said about Valve having already implemented it is true. Why go for a more honest review approach when you can deceitfully paint a so so product as being mostly positive.
You nailed it there in your last paragraph. Reading through the reviews often feels like pulling apart your toilet to find a ring that was accidently flushed down there.
The dependency on getting a good review is called quality control. It sets a standard for developers to reach before putting out their games for sales. People are paying money for games and deserve quality. The rating system is ment to be there to judge that, not to give an overwhelmingly positive score to game simply because it was a catalyst for thousands of 1 line joke reviews. To view the current binary system as being more accurate than a star rating system is completely ludicrous.
The currents system allows reviews to be as informartive as the reviewer wants to be. There's already a defacto star system. , it's actually 7 star. Look at the overall rating .
1- Overwhelmingly Negative
2. Mostly Negative
3 Negative
4. Mixed
5- Positive
6. Mostly Positive
7 Overwhelmingly Positive.
So ther eyou have a 7 Star system. Words can be as meaningful as numbers. If not more so. And since you're taking an aggregate of binary responses... yeah it works out being pretty accurate. Meh.. is not a very helpful statement. It's just an excuse for teh reviewer to ramble.
I don't really have the time nore intended on arguing for it as it was just something that occured to me while I was browsing the shop and realized that other places (Google, Amazon, GOG, etc, etc...) seem to do it better and give a better breakdown of user feedback.
I just wanted to put the suggestion out there as I was dumbfounded as to why Steam's rating system was the way it was - less accurately depicting the general consensus and requiring fishing through the proverbial sewerage to find a review that accurately describes the quality of the product in question.. assuming there is one. Though a bit messy at times, I suppose it works. It has its charm.
Food for thought.
Cheers to all who posted for your input and sharing your views!
I know where you're coming from but check my earlier respons to Roxor as to why it's not even close to being as accurately depecting as a star rating in terms of the quality of its results. The current system is a bit of a wolf in sheeps clothing.
That said, it still works and does what it does, it's just different and less accurate. It's more of a like or dislike rating which I feel is not enough because there are games which fall midway and a lot of opinions, which such a rating system would rely on, might not get voiced because they think it's unfair to vote on a particular game as being negative or the game doesn't entirely deserve a positive score.
You're forgetting the same thing many others who come up with this suggestion. The system isn't asking the reviewer how good the game was. It's asking. WOuld you recommend it to someone else?
So the binary actually makes it more accurate. Sure there's midway but the potential buyer is already at midway. meh reviews don't help anyone. A solid yes or no, is always 5x more valuable than a maybe. and if you can't reduce to a binary then you clearly haven't reviewed the material enough.
You're right. It explains why valve hasn't impemented the system. As opposed to reviewing the product, they offer the user to recommend or not recommened the product and then give a one line joke... i mean explanation as to why.
Neither is entirely accurate as a gaming experience is subjective but kudos to Valve for going down the more fun path... which jells better with gamers and keeps the shop a bit more light hearted.
Nothing that measures humans can be entirely accurate. Valve just asks the quyestion. You you recommend someone buys it. If it were a food critic... the quuestion would be 'Would you eat there again?' The answer to this question will tell you all you need to know. especially if you know a fair deal about the critic. They don't need to say much more than that. Not saying well worded reviews aren't helpfuul or desireable but in the end =steam's system allows for both.