Fred the Barber Posted April 11, 2016 Posted April 11, 2016 @Ryechu why do we have to proceed on the basis of the assumptions you made? Isn't the point of this thread to question those assumptions and present alternatives? A lot of people seem upset precisely because of the "quick summary". If there's no willingness to change based on valid criticism of the system, then this thread is pointless. @Down, you brought up serious reviews. I'm on board with that. Here's the New York Times book reviews: http://www.nytimes.com/section/books/review. The masthead for each is apparently "'Title' by Author", and the first paragraph. No read/don't read recommendation. Certainly no rating. Looks good to me. @solidbatman I want to make it abundantly clear here: I have no intention of criticizing any individual review or reviewer, and in fact have gone out of my way to call out that all the reviews I've read have been good. I am trying to point out an unintended and frankly absurd consequence of the fact that individual reviewers have individual preferences, which the system puts into a ranking. The resulting ranking is, of necessity because of the subjectivity of reviews, irrational and almost certainly inconsistent with the reviewers' own views, not just my own.
Decay Posted April 11, 2016 Posted April 11, 2016 I am heavily in favor of keeping review scores. I think they are a big help in allowing readers to... how do I put it? Understand and come to terms with the reviews? Even after reading the entire review, the score somehow helps the brain in conceptualizing what it all means, it puts the text they just read into an easy to understand context. However, I don't like the ten point system. Especially if there are half points. I will now argue in favor of a 5 point star scale with no half-stars, similar to the video game website that is among the most respected for their integrity, Giant Bomb. First, the problem with 10 or 100 point scales (we ultimately use a 100 point scale, although I basically have been limiting myself to 20, only going in increments of .5), is that they're too specific. Too absolute. The scores almost discourage people from reading the text because they feel like they've learned all they need to know when they see something like 8.7. It also sounds too definitive to the average user. Like when we say something is 8.7, we say that this is exactly what the game is, no more, no less, like a mandate from Heaven. It's the kind of precision that does not accept any questions or dissent. But no matter what, reviews are a very subjective thing. We should be wanting our readers to think for themselves, to ask questions, and to dissent. This is where the pros of a five point scale come into play.The point values are fuzzy and broad. They basically go as follows: "awful," "bad," "average," "good," "great." This has a large number of benefits. First, the broadness forces users to read the review to fully understand the meaning of the score. They see a 4/5 rating, go "Oh, they think it's good... but how good?" and proceed to read the rest of the review. Instead of being an absolute, final conclusion, it's an invitation to read the actual review. And once they have read the review, they then have to think about how the text relates to the rating. The more they think about us, our review, and the more they think critically about the game we're reviewing, the better. Secondly, the broadness of the scale invites our reviews to use the entire scale more liberally. For whatever reason, one thing that has been figured out after decades of different review methodology is that the more compact the scale, the more people are willing to accept reviews on the extreme ends. It's a weird psychological issue, but that's how it actually works. When someone sees a 2/5 star rating, they go "Oh, so they think it's bad? I wonder why." And when they see a 4/10, they go "Holy shit, they fucking dumped on this game, didn't they?" Especially if you use stars instead of numbers. Two stars out of five is an idea, 4/10 is a number, and people interpret numbers differently, more concretely. This also means we can give games 5/5 stars. 5/5 doesn't mean perfect, just extremely good, and you don't feel like you have to justify a 5/5 review as stringently. But our reviews may be reticent to toss out 10/10s and such, even though I would encourage everyone to use the whole scale, and that 10/10 doesn't mean absolute perfection. The more of the scale that gets used, the more meaning it has. Take the age-old criticism of IGN (although they've been better about it in recent years). They've been criticized for using a 40 point scale, 60 - 100. Anything at 60 is dogshit, anything below 60 is still dogshit. The more of the scale you use, the better. Most of my other points are variations of the other two. Readers are a little bit more able to accept dissenting opinions when they're presented as being out of five stars as opposed to numbers on a larger scale. The biggest con of a five star scale is that a lot of people don't actually want to have to think. They just want to see a score, read the final paragraph, and have the conclusions drawn for them. I don't think should we care if we lose those readers. Another con that doesn't really apply to us is how metacritic reads your ratings. Giant Bomb has talked about how they've gotten a lot of flak for publishers and have even been blacklisted by some because they give a game a 4/5 when on a 100 point scale they might have given it an 85 or 86 or whatever, and metacritic lists it as an 80. They don't seem interested in bowing down to the pressure, and this pressure wouldn't even exist for us. I just mention it here because it's almost comical how dirty the video game industry is sometimes. Overall, I think a five star scale is the best compromise between no score and the system we have now, and is the best system available to us. Tenkuru, Eclipsed, Darklord Rooke and 2 others 5
Mr Poltroon Posted April 11, 2016 Posted April 11, 2016 A score out of three would somewhat remove some of my issues with inconsistent scoring across titles, as there's less room for petty bollocks. (Things like 4/10 instead of 5/10 sparking huge controversies can be replaced by a simple 2/3 "Recommend at discount") Not the best solution, but a step toward eradicating my personal dislike of the system. Even still, the issue remains and I, too, remain against any score at all. When I have time and access to a computer, I'll further elaborate on this matter, assuming I'm still heated by then. Until that time, I ask that you make do with my salt. As for sanahtlig, I am actually very much for two dissenting views on the same VN by different reviewers, something I suggested back on Little Busters' review, despite the impracticality/impossibility of such a system. But anyway, I digress. We're on scores right now.
solidbatman Posted April 11, 2016 Author Posted April 11, 2016 3 minutes ago, Decay said: I am heavily in favor of keeping review scores. I think they are a big help in allowing readers to... how do I put it? Understand and come to terms with the reviews? Even after reading the entire review, the score somehow helps the brain in conceptualizing what it all means, it puts the text they just read into an easy to understand context. However, I don't like the ten point system. Especially if there are half points. I will now argue in favor of a 5 point star scale with no half-stars, similar to the video game website that is among the most respected for their integrity, Giant Bomb. First, the problem with 10 or 100 point scales (we ultimately use a 100 point scale, although I basically have been limiting myself to 20, only going in increments of .5), is that they're too specific. Too absolute. The scores almost discourage people from reading the text because they feel like they've learned all they need to know when they see something like 8.7. It also sounds too definitive to the average user. Like when we say something is 8.7, we say that this is exactly what the game is, no more, no less, like a mandate from Heaven. It's the kind of precision that does not accept any questions or dissent. But no matter what, reviews are a very subjective thing. We should be wanting our readers to think for themselves, to ask questions, and to dissent. This is where the pros of a five point scale come into play.The point values are fuzzy and broad. They basically go as follows: "awful," "bad," "average," "good," "great." This has a large number of benefits. First, the broadness forces users to read the review to fully understand the meaning of the score. They see a 4/5 rating, go "Oh, they think it's good... but how good?" and proceed to read the rest of the review. Instead of being an absolute, final conclusion, it's an invitation to read the actual review. And once they have read the review, they then have to think about how the text relates to the rating. The more they think about us, our review, and the more they think critically about the game we're reviewing, the better. Secondly, the broadness of the scale invites our reviews to use the entire scale more liberally. For whatever reason, one thing that has been figured out after decades of different review methodology is that the more compact the scale, the more people are willing to accept reviews on the extreme ends. It's a weird psychological issue, but that's how it actually works. When someone sees a 2/5 star rating, they go "Oh, so they think it's bad? I wonder why." And when they see a 4/10, they go "Holy shit, they fucking dumped on this game, didn't they?" Especially if you use stars instead of numbers. Two stars out of five is an idea, 4/10 is a number, and people interpret numbers differently, more concretely. This also means we can give games 5/5 stars. 5/5 doesn't mean perfect, just extremely good, and you don't feel like you have to justify a 5/5 review as stringently. But our reviews may be reticent to toss out 10/10s and such, even though I would encourage everyone to use the whole scale, and that 10/10 doesn't mean absolute perfection. The more of the scale that gets used, the more meaning it has. Take the age-old criticism of IGN (although they've been better about it in recent years). They've been criticized for using a 40 point scale, 60 - 100. Anything at 60 is dogshit, anything below 60 is still dogshit. The more of the scale you use, the better. Most of my other points are variations of the other two. Readers are a little bit more able to accept dissenting opinions when they're presented as being out of five stars as opposed to numbers on a larger scale. The biggest con of a five star scale is that a lot of people don't actually want to have to think. They just want to see a score, read the final paragraph, and have the conclusions drawn for them. I don't think should we care if we lose those readers. Overall, I think a five star scale is the best compromise between no score and the system we have now, and is the best system available to us. Stars are a good idea for all the reasons you listed above, but most importantly, our plugin supports them already
Silvz Posted April 11, 2016 Posted April 11, 2016 2 hours 3 pages totally not reading it, although I wouldn't take the scores away. For lazy people like me it is actually pretty good to have them + the pros and cons. The biggest problem is sometimes people give biases scores, which scares lazy people away
Ryechu Posted April 11, 2016 Posted April 11, 2016 Okay, here we go. 1 minute ago, Fred the Barber said: @Ryechu why do we have to proceed on the basis of the assumptions you made? Isn't the point of this thread to question those assumptions and present alternatives? A lot of people seem upset precisely because of the "quick summary". If there's no willingness to change based on valid criticism of the system, then this thread is pointless. Please look at the title of the thread. Should we remove the scoring system from FuwaReviews? We are keeping the summary. That has been stated in this conversation. We are not deviating from that. That will not change. If you don't want to read the quick summary, you read the entire review, and decide whether or not you want to read the VN yourself. However, this question is more aimed for those who tend to skip to the bottom of the post, look at the score, the pros and the cons, and make their decisions from there. Maybe they don't want the game spoiled even in the slightest, maybe they're in public and can't read my nukige reviews because they're NSFW. What COULD change, however, is our method of "scoring". If people like the x/10, good. If people like Decay's idea of x/5 stars, which I'm on board for (and our system allows that, whether or not it works immediately might be a different story), great. If people don't want a number and they'd rather have "Recommend, Ehh..., Don't Recommend", that's fine too, but we need to know what people prefer. We don't need criticism of the current system: We know it's not the best thing in the universe, and we are asking people what they'd like to see. @sanahtlig, your idea is great on paper. However, getting the manpower for something like this isn't something in the cards for us. Sure we could create a discussion topic and go that route, but those always tend to turn into "Fan Threads", and rarely do true criticisms of these titles come to the forefront, unless the game has zero redeeming value. I could be off there, but that's how I see it.
sanahtlig Posted April 11, 2016 Posted April 11, 2016 4 minutes ago, Tiagofvarela said: As for sanahtlig, I am actually very much for two dissenting views on the same VN by different reviewers, something I suggested back on Little Busters' review, despite the impracticality/impossibility of such a system. But anyway, I digress. We're on scores right now. The problem with separate, dissenting reviews is that some information gets duplicated (or alternatively, new information is presented in each that should've been combined into one whole). In a full review, you're expected to objectively describe the story premise and gameplay (if any). This doesn't need to be duplicated multiple times. A dissent appended to the review would simply describe the difference in subjective viewpoints, and thus wouldn't duplicate information.
starlessn1ght Posted April 11, 2016 Posted April 11, 2016 15 minutes ago, Fred the Barber said: @Down, you brought up serious reviews. I'm on board with that. Here's the New York Times book reviews: http://www.nytimes.com/section/books/review. The masthead for each is apparently "'Title' by Author", and the first paragraph. No read/don't read recommendation. Certainly no rating. Looks good to me. Here. Washington Post. https://www.washingtonpost.com/goingoutguide/movies/batman-v-superman-is-so-desperate-to-be-taken-seriously-it-forgets-to-have-fun/2016/03/23/3fb28f10-f0f5-11e5-a61f-e9c95c06edca_story.html 1.5/5 stars Detroit News, also. http://www.detroitnews.com/story/entertainment/2016/03/23/review-batman-v-superman-exhausting-showdown/82151448/ Grade C.
sanahtlig Posted April 11, 2016 Posted April 11, 2016 12 minutes ago, Ryechu said: @sanahtlig, your idea is great on paper. However, getting the manpower for something like this isn't something in the cards for us. Sure we could create a discussion topic and go that route, but those always tend to turn into "Fan Threads", and rarely do true criticisms of these titles come to the forefront, unless the game has zero redeeming value. I could be off there, but that's how I see it. It wouldn't require extra manpower. You'd just need to select 2 other people (from the vast Fuwanovel community) who would play the game and could offer informed viewpoints. The only added time-sink would be the debate itself, which would take maybe 15-30min of the primary reviewer's time. This system wouldn't necessarily need to be implemented for every review, but for selected titles (e.g., major releases--which would likely also attract the most interest from secondary reviewers) the system could enhance the consistency and objective validity of FuwaReviews scores. The debate itself could be held in any medium, but voice chat would probably be most appropriate.
Ryechu Posted April 11, 2016 Posted April 11, 2016 1 minute ago, sanahtlig said: It wouldn't require extra manpower. You'd just need to select 2 other people that would play the game and could offer informed viewpoints. The only added time-sink would be the debate itself, which would take maybe 15-30min of the primary reviewer's time. This system wouldn't necessarily need to be implemented for every review, but for selected titles (e.g., major releases--which would likely also attract the most interest from secondary reviewers) the system could enhance the consistency and reliability of FuwaReviews scores. The debate itself could be held in any medium, but voice chat would probably be most appropriate. man·pow·er ˈmanˌpou(ə)r/ noun noun: manpower the number of people working or available for work or service. "the police had only limited manpower" We lack this.
Decay Posted April 11, 2016 Posted April 11, 2016 Just now, sanahtlig said: It wouldn't require extra manpower. You'd just need to select 2 other people that would play the game and could offer informed viewpoints. The only added time-sink would be the debate itself, which would take maybe 15-30min of the primary reviewer's time. This system wouldn't necessarily need to be implemented for every review, but for selected titles (e.g., major releases--which would likely also attract the most interest from secondary reviewers) the system could enhance the consistency and reliability of FuwaReviews scores. The debate itself could be held in any medium, but voice chat would probably be most appropriate. "It wouldn't require extra manpower, you just need two more people for each review" what. Look, playing VNs and reviewing isn't our full time job. It's hard enough to find ONE reviewer for a game a lot of the times. A very large percentage of our reviews are only played by one person on our staff. We cannot expect our reviewers to read even more VNs than they already do. We cannot plan for each new VN released to have at least three reviewers read it in a timely manner and then offer their opinions. We simply do not have the resources required to carry out your idea.
Down Posted April 11, 2016 Posted April 11, 2016 12 minutes ago, sanahtlig said: It wouldn't require extra manpower. You'd just need to select 2 other people that would play the game and could offer informed viewpoints. That's what takes 95% of the time in a review... Do you guys want me to split some posts in here into a topic about (non)objective reviews? I think batman is eating is keyboard atm. Edit: Here. Mr Poltroon 1
Fiddle Posted April 11, 2016 Posted April 11, 2016 3 hours ago, solidbatman said: Which you totally can. Just not in this thread. Any more critiques, sarcastic remarks, or posts comparing, analyzing, or otherwise discussing the content of FuwaReviews will be marked for removal. This thread is strictly to discuss the removal of numbered scores, and the pros and cons resulting of. If you have nothing to say in regards to that, do not post in this thread. Feel free to create your own criticism thread, or discuss reviews in their respective threads. Thank you. Oh, I didn't see this post before I made my comment that some may perceive as explicitly sarcastic. Now I have to back myself up. Stars are superior to numbered scores for several reasons. First off, numbers seem to be associated with a particular cognitive bias for reasons that would require exceptional contemplation to expound, but I shall offer thoughts. While websites offering numerical 1-10 scoring systems of any nature typically rank it along the lines of something like "10 = perfect, "5 = average," and "1 = oh god make it stahp," I'm quite certain you would find that, upon contemplation, the numbers do not accurately reflect the attitudes of the users when tied to the individual numbers' descriptions. For example, for a Dick Cheney fanfiction that induced a sense of raging a'ight-ness in the mind of the reader, that reader might be inclined to rate it favorably because he enjoyed the experience to some degree. Thus, the score may presumably be a 7/10 or possibly an 8/10 for somebody in a liberal mindset. However, when asked, I'm sure this reader would describe it as "average" relative to other Dick Cheney fanfictions he has read, because, indeed, the very statistical value of this system depends on relativity; if our ratings stemmed merely from our raw enjoyment of the medium, every consumer would rate all products of his respective medium favorably, and Dick Cheney fanfictions would soon manifest themselves as superior to Hemingway novels, which is an argument for another time. To offer a modicum of evidence for this bias toward higher scores, going against the proposition that they should generally be scaled to a 5, I clicked the random button on VNDB until I gathered 10 visual novels each with greater than five ratings, and the average of the ratings were as follows: 5.39 7.25 5.34 7.89 7.07 6.46 (this one had a disproportionate amount of 6's for some reason; I admire the company that instilled such a sense of mild contentment in the readers) 5.22 6.63 5.96 6.70 Evidently, none of them―and you will have to trust that I chose these indiscriminately―have a score lower than a 5. Though I don't have the time to conduct a more detailed analysis before my paranoid fear that my previous post will be deleted comes to fruition, I believe the margin of error is low enough to offer credence to the proposition that the bias toward higher scores, a bias which disrupts the statistical reliability of an individual product within a medium, is quite existent. If anything, the most common numbers would be higher than those listed above, because more popular visual novels tend to garner far more votes than less popular ones (for obvious correlative ones), while the randomization of my trial relied not on votes but on individual visual novels, of which there is a much greater quantity of obscure and low-quality ones. Stars, while not eliminating this bias entirely, significantly reduce it. On the assumption that the notions espoused in the above text can be regarded as more or less accurate, a rating of, say, three stars seems much more positive than a rating of 6/10. Allow me to offer a theory on this circumstance: Ratings out of 5 stars should generally allow for only 5 total ratings, while ratings of 1-10 should generally allow for only 10 total ratings. In the prior, this means that the lowest rating is 1 star out of 5; in the latter, it is 1/10. Naturally, one star would correlate to a rating of 2/10=0.2. What results from this? Average star rating: (1+2+3+4+5)/5=3 Average 1-10 rating: (1+2+3+4+5+6+7+8+9+10)/10=5.5 Average star rating scaled to average 1-10 rating: 3*2=6 Evidently, the average star rating is lower than that of the average 1-10 rating. The general view I espoused earlier is that a score of 5/10 is average, whereas the reality for the mathematically inquisitive is slightly different. However, the mathematical reality is what manifests itself in overall scores, whereas the false mathematical situation is what manifests itself in users' perception of individual scores. In the 1-10 system, this will create quite a perception bias. In the star system, however, we should find that our inherent perception of each star's value is more accurate, possibly accounting for why three stars seems to be more favorable than a 6/10. Another note: if an individual's opinion of a product is as average as possible, and also operates within a perfectly statistically sound mindset, he might want to rate the product a 5.5/10 in accordance with the system I used above. However, if his score must be expressed as a whole number, it will be a whole half a point off, and he must choose between 5/10 and 6/10―I surmise he will favor 6/10, which arguably offers additional difficulties because the average person will see 5/10 as perfectly average, and thus the score of all media which abide by this 1-10 system are scaled upward due to mathematical bias. Meanwhile, in the star system, the average is a perfect 3, while 3 also happens to be a valid score of stars. Apparently there've been 23 replies since I made this post, so I'm gonna stop. EDIT: And apparently Decay and I are on common ground (the difference being that I bullshizzled this argument for another purpose, though I've made such a compelling case that I'm starting to think I'm right). Ryechu, Decay and Darklord Rooke 3
Darklord Rooke Posted April 11, 2016 Posted April 11, 2016 4 minutes ago, Down said: I think batman is eating is keyboard atm. Pics or it never happened
sanahtlig Posted April 11, 2016 Posted April 11, 2016 11 minutes ago, Ryechu said: man·pow·er ˈmanˌpou(ə)r/ noun noun: manpower the number of people working or available for work or service. "the police had only limited manpower" We lack this. If it wasn't clear in my original post, I recommended recruiting volunteers from the community who would play the game and participate in the scoring process. You wouldn't recruit just anyone--only people who you think could give an informed opinion. The FuwaReviews staff member would write the review, so this wouldn't at all compromise the writing quality of the review itself. FuwaReviews lacks skilled review writers, but certainly not informed readers. And if you can't find secondary reviewers from the community? Then simply fall back on the existing system.
solidbatman Posted April 11, 2016 Author Posted April 11, 2016 I've been trying to. Last time I had recruitment open, I got two applicants. I accepted both. I've reached out to various otome writers as well and generally received no interest. Anyone who wants to write reviews right now, is already doing it. And the rest who want to, and want to join the team aren't quite at the standard I want. I absolutely love your idea, and in the future, I would love to do it. But right now it is limited by budget, manpower, and the fact most publishers are not going to give me three review keys.
Nosebleed Posted April 11, 2016 Posted April 11, 2016 If it all comes down to what do you replace the x/10 with, and "nothing at all" isn't an option, I'm sticking with my salt container idea. babiker 1
Kaguya Posted April 11, 2016 Posted April 11, 2016 9 minutes ago, Rooke said: Objective review attempt #1: The scoring! This is what it'll look like, folks, in all its glory. Hard core RPG fans will consider MGQ game a 1/10, but people interested in defeat rape might consider it a 7/10 game, people interested in both RPGs and defeat rape, well, you might consider it a 5/10 game, whereas people who just want a good story will consider it a 2/10 game until you get to number 3, in which case you might consider it a 6/10 game, but people who want both RPG goodness AND a good story will consider it a 3/10 game until you get to number 3 where you’re opinion will change and you’ll consider it a 5/10 game, but people who like RPGs and defeat rape and a good story will probably consider it a 4/10 game. People easily grossed out will not have a happy time here. -2/10 for you folk. You got that right, it's a negative, not a dash. People with morals and are easily offended … stay away. -2/10 for you folk. And monsters masquerading as humans in the audience? Man, you’ll fucking love this game, 10/10 for you guys. Keep defeating those humans over and over and over again. And yet it averages 7.6 on VNDB... Probably because of the lack of other games of the sort translated, lol. Alright. I prefer to keep the scoring as is. Why? Because while I'd like to have no score at all, as long as we have one, I prefer it to be as detailed as possible. The biggest alternative I've seen so far would be a star system... And that'd really suck. 1~5 stars would make me want to kill myself while giving scores to things. There is so much difference between games I'd rate 7 and 8 that I wouldn't feel comfortable rating things. If it were up to me, we'd go full 1/100, rather than the current 1/20 system we're running, too. Take my first review, Haruka, for example. I thought it was too low for an 8 and too high for an 8.5, and I wanted to give it something in between because neither would feel right, but evil bats didn't let me There's a world of difference between Rose Guns Days, Evolimit and Dies Irae in quality... And I'd hate the thought of lumping them together with the same amount of stars, for example. A "recommended, maybe and not recommended" section would be better than the stars, but it also brings its own set of problems. Ultimately, I prefer to keep things as-is. It's classy, calls more attention than stars and would prevent me a lot of headaches. starlessn1ght 1
starlessn1ght Posted April 11, 2016 Posted April 11, 2016 5 minutes ago, Nosebleed said: If it all comes down to what do you replace the x/10 with, and "nothing at all" isn't an option, I'm sticking with my salt container idea. I'd be ok with a salt container if it was sectioned in 10 equal parts.
sanahtlig Posted April 11, 2016 Posted April 11, 2016 3 minutes ago, solidbatman said: most publishers are not going to give me three review keys. Secondary reviewers would be expected to buy the game (assuming multiple free keys were not available). For reviews intended to be published on the day of the release, you raise a valid concern. Publishing before the review could be discussed in committee would sacrifice some of the benefits of the system I propose. However, one could still retain the scoring benefits by publishing the review without a score, then adding the score later after the committee scoring process had completed. Scores are intended for retrospective comparisons, so the time delay wouldn't have much impact on their utility. As for budget and manpower, such a system wouldn't have any monetary cost and would only present a small additional burden to the primary reviewer.
starlessn1ght Posted April 11, 2016 Posted April 11, 2016 6 minutes ago, Kaguya said: And yet it averages 7.6 on VNDB... Probably because of the lack of other games of the sort translated, lol. Alright. I prefer to keep the scoring as is. Why? Because while I'd like to have no score at all, as long as we have one, I prefer it to be as detailed as possible. The biggest alternative I've seen so far would be a star system... And that'd really suck. 1~5 stars would make me want to kill myself while giving scores to things. There is so much difference between games I'd rate 7 and 8 that I wouldn't feel comfortable rating things. If it were up to me, we'd go full 1/100, rather than the current 1/20 system we're running, too. Take my first review, Haruka, for example. I thought it was too low for an 8 and too high for an 8.5, and I wanted to give it something in between because neither would feel right, but evil bats didn't let me There's a world of difference between Rose Guns Days, Evolimit and Dies Irae in quality... And I'd hate the thought of lumping them together with the same amount of stars, for example. A "recommended, maybe and not recommended" section would be better than the stars, but it also brings its own set of problems. Ultimately, I prefer to keep things as-is. It's classy, calls more attention than stars and would prevent me a lot of headaches. Kaguya, you share all my thoughts. Three stars are way too limited and can make up horrible equalities. I already hate the MAL system of no decimals (like it means I like Hibike!Euphonium as much as I like Black Bullet). Japanese people usually prefer the x/100 as I've seen, and I endorse it as well.
Kaguya Posted April 11, 2016 Posted April 11, 2016 To expand upon it a bit further, I ended up giving Haruka an 8.5, even though I wanted to give it an 8.3 So, the next time I review a nukige, I need to keep in mind that I gave Beat Blades Haruka an 8.5. Then say there comes a game like, Erect, which I'd give an 8.7 to. Well, normally, I'd just leave it be... But I can't do that. And so I'd usually end up taking it to an 8.5 - that's reasonable, right? However, I gave Haruka an 8.5, and I'm reviewing Erect as a game I believe to be significantly better (closer to excellent than it is to pretty good.) So I'd either have to settle it with the same score as Haruka (which I really wouldn't want to do) or I'd have to give it a 9. Giving it a 9 would temporarily solve the problem, and it'd be more or less in line with the differences between the scores I'd give for those games... But the next time I review a nukige, I'd have to keep in mind that I gave Erect a 9. This is already unsatisfactory to me, so I'd pretty much commit sudoku with a star system instead
Decay Posted April 11, 2016 Posted April 11, 2016 8 minutes ago, Kaguya said: To expand upon it a bit further, I ended up giving Haruka an 8.5, even though I wanted to give it an 8.3 So, the next time I review a nukige, I need to keep in mind that I gave Beat Blades Haruka an 8.5. Then say there comes a game like, Erect, which I'd give an 8.7 to. Well, normally, I'd just leave it be... But I can't do that. And so I'd usually end up taking it to an 8.5 - that's reasonable, right? However, I gave Haruka an 8.5, and I'm reviewing Erect as a game I believe to be significantly better (closer to excellent than it is to pretty good.) So I'd either have to settle it with the same score as Haruka (which I really wouldn't want to do) or I'd have to give it a 9. Giving it a 9 would temporarily solve the problem, and it'd be more or less in line with the differences between the scores I'd give for those games... But the next time I review a nukige, I'd have to keep in mind that I gave Erect a 9. This is already unsatisfactory to me, so I'd pretty much commit sudoku with a star system instead Deal with it. To be honest, that sort of immense attachment to scores is the thing a star system is supposed to prevent, and I honestly don't think it's a healthy mindset for a reviewer to have. If you want to be able to get across the subtleties of the differences between one game and another with the same rating, then you have to do so in the text of your review. You can't rely on score as a crutch. And the readers won't be able to easily tell the difference at a glance? Good! They will now have to actually read the review.
Darklord Rooke Posted April 11, 2016 Posted April 11, 2016 8 minutes ago, Kaguya said: To expand upon it a bit further, I ended up giving Haruka an 8.5, even though I wanted to give it an 8.3 So, the next time I review a nukige, I need to keep in mind that I gave Beat Blades Haruka an 8.5. Then say there comes a game like, Erect, which I'd give an 8.7 to. Heh, whenever I see someone scoring something with those type of decimals, I can't help but recall Penny Arcade xD 28 minutes ago, sanahtlig said: Secondary reviewers would be expected to buy the game (assuming multiple free keys were not available). In a perfect world all reviewers buy their own copy of the game, but that's definitely unrealistic. As in, costs way too much money along with other problems. For some reason, I get a creepy, insincere feeling when developers hand out free keys to be reviewed. Maybe I'm just far too cynical.
Kaguya Posted April 11, 2016 Posted April 11, 2016 10 minutes ago, Decay said: Deal with it. To be honest, that sort of immense attachment to scores is the thing a star system is supposed to prevent, and I honestly don't think it's a healthy mindset for a reviewer to have. If you want to be able to get across the subtleties of the differences between one game and another with the same rating, then you have to do so in the text of your review. You can't rely on score as a crutch. And the readers won't be able to easily tell the difference at a glance? Good! They will now have to actually read the review. Half the people probably only check the score and glance through the review, so of course I care about it. It's also why I'd rather have no scores at all. The star system is basically a simplified score system as far as I see it, so it's even worse. To me it's also related to how many VNs I've read... If I didn't separate them like that, the scores would pretty much be pointless (like my current VNDB scores. I don't even want to look at that thing.) The difference between 8 and 9 when you have to rate 10 or 20 games isn't anything to be worried about, but it's the difference between the sky and the earth when you read a couple hundred of them, though I dunno if most fuwans can relate to that.
Recommended Posts