Is there anyone who goes through everything uploaded to J2O and looks through reviews? If there isn't, I believe there should be.

The topic came up in #jj2 for a little bit. Basically there is one main problem that many people are facing which is "n00bs" coming in and giving a download a very poor grade based upon really trivial or unconsequential ideas, or really for no reason at all, thus needlessly driving down the rating for the download.

Here's an example: In SXR there are a few entries from reviewers no one has hardly ever seen that give it a 5.0. I think one of the people's entire review was "So much hard I think". The poor English aside, does this person really have the right to lower a downloads' average if they display little knowledge of reviewing at all, or even a good reason? I know way more than the SXR entry have faced this. I know people are entitled to their own opinions, but in some cases, frankly, their opinions are senseless. Not to say this person's opinion that SXR is hard is dumb, but giving it a 5/10 because of it is.

Blurred said in #jj2 that if he requested of for it to be deleted by any J2O admin it would be deleted it promptly, but I'm sure they don't like doing other people favors all day. That's why I think someone should be given the job to delete (or make the rating N/A) some reviews, certain downloads, etc. They would have to be available for such work frequently, and someone who has a lot of experience in reviewing levels. Be trustworthy, and crap like that...since, wouldn't they have be admins themselves?

Another choice is for reviewer rating weights. I know that has been discussed in the "Suggestions for J2O 2.0" thread, so I won't go into further detail.

Comments and critisism?

I agree. Although a lot of my reveiws would be deleted if they have to be in detail.

I agree totally!

I would agree, but that's where mainly all my reviews come from: N00bs and vets alike saying it sucks and giving bad reviews.

Originally posted by Xion
I would agree, but that's where mainly all my reviews come from: N00bs and vets alike saying it sucks and giving bad reviews.

That's because you do suck. There is really no way to solve this issue. Everyone reviews in their own way. I think the only way we could solve this problem is only letting designated people review things (which would suck). Though I do think that the number of reviews should influence a score, since one person rating it a ten can instantly make it the highest rated download. Prehaps less common scores (e.g. everyone rates it a 7 and one person rates it 5) should not affect the overall score as much.

I strongly agree, and can also give an example. My latest, though already quite old level, Knightmare Castle got an 8.2 by EvilMike (!!!), which I am very proud of, of course. Though, a guy called "Magic Card", who at all has 4 reviews, though it is not better than a cheap 5.2.
What the moderators should do there: of course they should not just delete such reviews, but perhaps talk to the reviewers and, if they dont respond or whatever, talk to top raters like EvilMike, Aiko, Trafton etc, if it would be justified to remove such a review.

Moderators can't just remove all low reviews. Like I said, people have different opinions when reviewing, and getting something liked by all opinions is difficult, but that's why everything on J20 isn't rated a 10. Though I still think seriously outnumbered ratings shouldn't count.

Non word-perfect quote from EvilMike: "The rating itself does not matter, if I look at the review and it's bad I will delete it".
Thus, if you think much less/more of something then everyone else, make sure to give a lengthy review so your review won't be deleted.

It's an interesting idea, but it raises several problems. The first of which would be what should be considered an unfair rating. Some cases, such as rating a level 1.0 that previously had a rating of 9.0, are plainly not reliable reviews. However, there are blurrings in the line, such as the SXR incident you pointed out. Some may consider SXR to be a 5.0 level pack. I'm not sure how, but some people have different ideals for levels. What the rule up until now seems to have been is if the review does not provide ample information on why the rating was given, it is not a good enough review. However, if the review does in fact provide sufficient reason to rate the level as such, then I suppose the review deserves to stay. Administrators do watch the downloads section, and remove unfair reviews. However, there are so many reviews, and it is hard to see when new reviews are posted, so not every single review can be checked and critiqued. Still, I agree that it would be nice to see increased administrating in the J2O downloads section. I think that this could be improved with a "Report this Review to a Moderator" function, which will probably be in J2O v2. And, until then, if you think a review to be unfair, I suppose you have to wait until you can report it to an administrator. I can't see many other solutions now to that problem.

The Trafton has spoken.

Originally posted by Radium
The Trafton has spoken.

Eh?

Another thing: while some of the mathmatical formulas for J2O rating calculation are interesting, I think that leaving it alone for now would be fine. After all, the formulas make some ratings completely ignored. People would stop rating because they would feel like what they rated it wouldn't matter. Until I see a really good formula, I support just sticking with averaging out all of the reviews.

Trafton is right. I have to agree on Evilmike's quote Unknown mentioned there.

Originally posted by Trafton
Eh?

Another thing: while some of the mathmatical formulas for J2O rating calculation are interesting, I think that leaving it alone for now would be fine. After all, the formulas make some ratings completely ignored. People would stop rating because they would feel like what they rated it wouldn't matter. Until I see a really good formula, I support just sticking with averaging out all of the reviews.

Onag's algorithm doesn't ignore reviews, because for example a very high rating does make the rating higher even if it's within the top 5% of the ratings, so your review always has influence.

J2Ov2 will have download admins and a report review fucntion.

Yay!

*sings some random song that dosn't make enough sense to be relevant*

Originally posted by Fquist
Onag's algorithm doesn't ignore reviews, because for example a very high rating does make the rating higher even if it's within the top 5% of the ratings, so your review always has influence.

J2Ov2 will have download admins and a report review fucntion.

OK, I stand corrected. Onag's formula actually makes very good sense. Still, all formulas are most effective when there are a lot of reviews, and with two or three reviews (about average) it wouldn't make much difference.