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
该主题的作者已标记一个回帖作为问题的解答。
讨论规则及指引
举报此帖
How exactly is adding a scale between two points *removing* ambiguity? Especially if you add an option right between two end points?
Hell No, No, maybe, yes, hell yes
The discussion may continue if others are dis-satisfied but i think this suggestion is a reasonable compromise that reflects both sides of the opposing arguments that have been presented. So I'm pleased with the suggestion
The opposition to the numeric argument is that it doesn't remove ambiguity because different people have different meaning to the numbers assigned. Although it has not been established, it is reasonable to argue that recommend or not recommend has little impact on people buying or not buying. We all agree that the textual description in the review is what is most important. The suggestion gives intuitive metrics that establishes a spectrum of options without numbers and which also incorporates the recommend/ not recommend scheme. Hell no is a definite don't buy while hell yeah is a buy. Maybe is a its good but the decision is up to you. Yes and no are buy but its not perfect while no means don't buy but its not the worst.
The discussion may continue? PUBLIC thread not your personal thread.
Kargor suggested 3 values you upped it to 5.
Hell No, No, maybe, yes, hell yes?
What exactly is a hell yes? - The one you play and endorse
What exactly is a hell no? - The one you will not play and reject.
2 choices - play, don't play, recommend, don't recommend and they remain subjective NOT universal. After all one mans trash is another mans treasure but keep on ignoring that.
Valve use Recommend and Not recommend to leave the reader with the final decision to purchase or not and reject a review. So simple, yet effective.
And finally - NEUTRAL - yeah, maybe, I don't know - ermmm.
I do not need to see your response to Cinedine.
You both assumed and presumed that Kargor's idea was a compromise for both sides when in fact I and others including Cinedine rejected your idea because the current system does not warrant changing.
Are you now under the impression Valve are going to implement it when the current system is also tied in with other elements of the store or that Valve did not think it through when they implemented it.
Ideas, suggestions, wants, needs share one thing in common - they require acceptance and implementation otherwise they remain in limbo or fade into the obscurity.
Besides. Steam is already taking the approval rating, converting it into a value of 1 to 10, rounding it up or down when needed, and then providing this value as metadata for google to ultimately display as 5 star rating, You can test this yourself by googling any game from Steam and google will show you the approval rating converted into a 5 star scale.
This is however, problematic because a game with 90% approval recommendations will be displayed as a 9 out of 10. Or a 4.5 out of 5. Even if the people who recommended it wouldn't have rated it a 9. We simply have no way of knowing what they would have rated the game at this moment, and there's a world of difference between a 6 and a 10. But in Steam's current review system, they are both the same, because they are both positive.
This is what you get when you collect data from a two point scale and convert those results into a 10 point scale. You get unreliable data. The same cannot happen the other way around.
Finally Steam already uses the 5 star system. For rating mods, guides, etc. So it makes perfect sense if they extend that system to games too.
The data is not unreliable, it just doesn't mean what you think it should mean. An approval rating of 90% means just that 90% of people that played the game recommended it. The strength of that recommendation isn't measured (but also isn't relevant it was at least strong enough for them to recommend it and if 90% of people recommend it that means I have a 9/10 chance of enjoying the game assuming I like the genre), but if you switch to a 5 point system you lose the other value instead (how many people recommend the game) as a 4.5/5 could mean anywhere from approximately 80% to 100% of people recommend the game which means my odds of enjoying it are more varied. If you are making a purchase in my opinion knowing the odds I'm going to be happy with it is far more important than knowing the average rating is 4.5 stars.
Also a 6 isn't a positive rating in general, that is quite subjective but is usually restricted to 8-10 typically on a 10 point scale where anything less than an 8 is considered not good.
That's what it means yes, but Valve itself is taking that 90% and telling google to display that as 4.5 out of 5 Stars which is a completely different thing.
This is what years of bought reviews from IGN and Gamestop conditioned you to think. I suggest you go on IMDB and check out the surprising amount of 5's and 6's that you actually enjoyed, mate.
Btw, I recommend "It Follows" if you like horror. It's a 6.8 on IMDB and I agree with that score.
If you haven't seen it, give a go and then tell me it's not good. :)
A system that breaks down to 3 - No's and 2 - yes'
So why not just have the single 'Yes' and 'No' and you can communicate the magnitude in your commentary? Whi add superfluous Yes's and No's?
And that would be perfectly accurate.
That depends entirely on who's doing the rating. The differences can just as easily be quite small.
Not really. A person can give a game 9/10 and still 'Not Recommend' it.
How hard is it to click the quote button and not completely mess it up? Like, multiple times throughout this thread alone.
Anyway, please answer my question: how does adding more options in between, including a neutral one removes ambiguity from yes/no?
It doesn't. Because it adds. Instead of a havign a "(not) recommended" you now have "(not recommended)", "slightly (not) recommended" and "whatever". The last one being completely pointless and nothing but ambigous.
Anyway, yeah "the discussion might continue". You are aware that this discussion was held at Valve's long time ago and they explicitely decided for a binary system. Nothing will change that.
Fine? Still not satisfied?
This is a discussion forum, not cast your vote and it becomes reality.
You asked for thoughts in the OP post then object when others say no and believe you can simply dismiss those posts by posting "a solution has been found to satisfy both sides", well lets be honest it has NOT satisfied one of the sides who are still saying no.