Rating [1]

I palpably dislike NU's current rating system, and I feel that there are several ways to improve it. I have an ideal for where I want NU's ratings to go, and I'll roll out my suggestions incrementally.

This is the first and most basic suggestion. I am not cognisant of the technical details required for this — it depends entirely on the current implementation of the rating — so I wouldn't make any technical suggestions; for my future suggestions which are more technical in nature, I'll submit algorithms for implementation of them, to reduce the burden on @Tony when he is implementing them. I'm not confident in my programming skills, but if NU is implemented in languages I'm knowledgeable of, I'm willing to implement my own algorithms — though I suspect @Tony wouldn't welcome my assistance.

Okay, I've fulfilled my narcissism imposed quota of self aggrandisement, and now I shall get to the meat of the issue.

I am suggesting that we either use 10 stars, or you allow half star ratings.

Benefits:

1. 10 points has much greater flexibility in rating stories, and provides user's with a wider range of choice.
2. 10 points is implemented on most other sites, and as such we are used to it.
3. 10 points feels intuitive — most of us rate novels with a ten point system internally — my reading list for example has "5 star", "4.5 star" and "4 star" lists. There are several novels that I rated 5 stars here, that are in fact on my 4.5 star reading list.
4. 10 points is going to be very helpful in raising the efficiency of all my future suggestions.

Hope you consider it.


A 1 - 5 stars system with half star ratings IS a 10 point system.

A 10 point system does not imply numbers 1 - 10 — it simply means that there are ten different points of rating.

This was in the OP, but if it was not clear, I'll go edit to clarify.

I'm still working on the algorithm for one of the suggestions (either 2 or 3), so the next post may be delayed.

Good day.

Yours faithfully,
Dragon God. ​

 

I support a 10 point rating system. I like the one they have on MU, using Bayesian average. It helps to offset overrating/underrating series when they don't have many votes yet.

 

So many people either rate 5 or 1 without reading properly in the first place, which is a greater obstacle to having a relevant rating system in comparison to the system itself. Having 10 to 1 isn't going to change that. Only having 2-4 in the middle gives more concentrated information than 2-9, so right now, I think 1-5 is better than 1-10.

Regardless, I don't think the system will ever be that reflective of the quality of novels because the amount of people voting is either too small or too influenced by preconceptions or brigading.

 

Bayesian Rating is the algorithm I'm working on right now.

I'll suggest it once I've perfected the algorithm.

 

I have suggestions to deal with the latter points.

 

4/5 and 9/10 are almost the same thing, just like 2/5 is almost the same as 3/10 or 4/10.

Feels unnecessary.

 

There is a world of difference for me — hence why I have a separate list called "4.5 star" and one called "4 star".

Almost the same is not the same.

 

I think this is supposed to be the scene where you talk about them then, not just say they are there. The 'benefits' you have listed are not facts, they are just your own opinion, and honestly, they're not even really benefits. Too much flexibility isn't a good idea when the sample size is small, 'it's on other sites' is bad for the same reasons that the sample sizes are different. A 10 system may feel 'intuitive' to you but it feels like a waste of time to me because it's inefficient, and I don't even really want to talk about your 4th 'benefit'.

Still unnecessary. I think you are welcome to argue that a 10 point system is required but honestly you haven't really provided any real reasons to why a change is required.

 

Even if you make a 10 point system. Most people wouldn't be bothered to rate those series again leaving a lot of novele with weird ratings

 

All extant ratings would be doubled.

 

This rating system would be good if we didn't have people hopping on a new novel, and giving it a bad rate when only a few chapters are out...

Or people not voting correctly because releases are slow or fast

 

The sample size becomes irrelevant once a Bayesian rating algorithm is used.

Your main argument is that it is inefficient — how so? (Don't mention sample size as that becomes irrelevant later on).

My second suggestion is that you can't rate novels not on your reading list — you must have added the novel to your reading list and maybe reached a minimum of 5% of available releases before you can rate.

My fourth suggestion is that ratings are public — all an individual's rating information can be viewed. ideally this motivates you to give better and non troll ratings — such may be unadvisable to sites like RRL because you rate each other's content, but here you're not interacting with the authors of the stories.

 

Novel must be on your reading list before you rate it.

 

Firstly, can you please learn to either use multiquote or quote multiple posts in one reply. It's baffling how you have managed to be here for so long yet still not understand the basic concept of not consistently double posting.

Secondly, it's inefficient because your suggested changes to the system take effort for little gain. Your 'benefits' are literally just broad opinions like 'it's flexible,' 'it's initiative,' 'other sites use it,' or 'it's helpful.' These are just your opinions, they aren't arguments. You still haven't provided any real evidence why the system would be beneficial. You can write 'bayesian rating' all you like but you still have to point out why it would be useful.

Too easy to get around. You're already on the page with the chapters. Tick a box and rate away.

Just invites flaming, nothing else. Nobody is going to say to someone else 'wow you rated this novel that I also like highly, good work.'

 

I will completely disagree with you. Sure 2-4 and 2-9 may seem similar in the middle. But it makes a huge difference!

The current 1-5 system is the reason why I don't bother making reviews. If it was a 1-10 system, I would review a lot more stuff.

Why?

10/10 or 5/5 is perfect score. There are some novels which I'll give perfect score, but some novels I want to give a good score but not perfect. You say give the novel a 4/5 right? But I don't want to do that. Because that novel has a score of 4.6 for example. Then my vote of 4 hurts the average of the novel even though I like the novel and agree with the 4.6 score.

Get it?

Which is why I don't bother doing reviews or voting much, because I don't want to hurt the scores of novels by voting 4/5.

Why does it even show the average of 4.6? If these things didn't matter then they might as well just round and tell us 4 or 5.


Edit:

Also based on the code it looks like the possibility of rating in decimals is there, just the visual interface is not.

 

No, I don't really get it. Your stubbornness to not rate a novel 4/5 or 5/5 because it is not 'perfect' or '4.6' worthy is not a reason to overhaul the entire rating system. You rating a novel a 9/10 in comparison to rating it 4/5 is different but it is not different enough to have any significant impact on what people take from the rating system.

 

It is not stubbornness, the current review system is broken. Because it can't represent people's actual opinion in its rating.

You also don't need to overhaul anything. The rating system already has it implemented, all that needs changing is the UI.

Of course it has a significant impact on what people take from the rating system. Because the current system discourages people reviewing and giving scores. Not to mention the actual scores will actually reflect far better and accurately.

 

Your actual opinion can be put in a rating of 1 to 5. Just because it's not broad enough in your opinion does not make it 'broken'.

If you are going to throw out statements like that, you need to have something to back it up. People seem to have been using the rating system without a problem for almost a couple of years(?) now and there hasn't really been an issue.

 

You're fucking dragon god?
Thought you were nazgand for a few days.

 

Sorry, but it can't. All you get is a vague outline at best. Again, the fact that the average is shown in decimals and not just rounded proves that.

Your logic is completely flawed.

1) Humans have lived longer as cavemen than in civilization. Your logic of "we lived so why bother" is silly.

2) And for years there have been people complaining about the rating system, you ignoring them does not constitute no issue.

3) The poll in itself already shows that almost half see it as an issue. And there are many people who are discouraged but don't even bother to say anything.


End of the day, a system of letting you vote 4.5 stars or go to rating of 10 (whichever) is a net positive. That you can't deny right? Do you see any negative aspect to it? IF not, then why argue against progress?