• NOW LIVE! Into the Woods--new character species, eerie monsters, and haunting villains to populate the woodlands of your D&D games.

Reviews 'beta' site

Actually, I wouldn't mind weighting at all. I'd use a simple rule: staff reviews count as 2 normal reviews. I'm not sure if it'll actually be used, though.

As for the formula, I'm all for sqr(mean * median); it was discussed on this forum for quite some time and picked as the "best" method. Still, whatever Morrus decides to go with is fine with me.
 

log in or register to remove this ad

Blacksad said:


I'm not sure on the weighting of staff reviews, If I want the staff opinion, I can clic on simon collins or psion (as they have a link on the main review page). If I want an average of some gamer opinion, I prefer that the staff review aren't weighted, especially when there is only two review on a product.
Otherwise, the new look is very nice!

Unlike Blacksad I am very sure about being against any weighting in the reviews. I would go with something like not taking highest and lowest scores in account, at least before a reasonable number of reviews. However, by assigning a higher weight to the staff you are implying that, a priori, their reviews are better than anybody else. Don't get me wrong, both Psion and Simon are good reviewers, but I don't think that there is reason to believe that any piece by them is, a priori, superior than a review submitted by any other member. As I have deffended before, bad reviews can be taken care by carefull management by the moderators and use of commentaries. Additionally, as someone who actually do statistics for a living, I am a great fan of the theory of large numbers. There is no way you can refute the opinion of many. Quite frankly, this particulary part of the annoucement left me very cold about the future of the reviews section. Psion and Simon Collins do a great job, but I cannot expect that their views are superior to anyone who collaborate with this site.

Regarding the new design. It appears to be good, but it did not worked properly in IE 5.1 to Mac OS X. I hope it was only a temporary problem.
 

Nobody said they were superior. That came from you alone. I'm sure you can figure out the reasoning behind it.

Incidentally, the forumula I went with was sq rt (mode * median).
 
Last edited:

Blacksad said:
If I want an average of some gamer opinion, I prefer that the staff review aren't weighted, especially when there is only two review on a product.

You won't get the average gamer's opinion from just taking the average of all the ratings. Trust me. This will give a slightly more level-headed reflection of a product's quality.
 

Morrus said:
Nobody said they were superior. That came from you alone. I'm sure you can figure out the reasoning behind it.

Incidentally, the forumula I went with was sq rt (mode * median).

Morrus,

Quoting you from the EN World home page:

"Also a weighting in favour of staff reviews."

As you plan to weight them heavily than others it was natural to me to reason that you believe that, a priori, their reviews are superior than ordinary ones. If they should be considered as good as any other there is no justificative for a differential weighting scheme.
 

Morrus said:
Incidentally, the forumula I went with was sq rt (mode * median).

This is an interesting choice. With only 5 ratings to choose from (1-5), and most ratings in the upper portion, many products will have the exact same ranking.

I'd love to just use the median, but it would give too many products the same score, so I suggested adding the mean (normal average) to the formula.

Has the formula been implemented? If so, I'll check on it and see if it's the way I think it is.
 

Ron said:


As you plan to weight them heavily than others it was natural to me to reason that you believe that, a priori, their reviews are superior than ordinary ones.

Interesting. It's not natural to me to assume that. I guess we look at things differently.
 

CRGreathouse said:



Has the formula been implemented? If so, I'll check on it and see if it's the way I think it is.

Not yet, no. It's probably the last thing that'll get done.

BTW, everyone - check out the page for the PHB for a demo of how a product should look once everything is up and running. It has links to errata, articles, extra reviews etc.
 

Originally posted by Morrus
You won't get the average gamer's opinion from just taking the average of all the ratings. Trust me. This will give a slightly more level-headed reflection of a product's quality.

I do not think that ENworld average is the average gamer opinion, but the ENworld average is still usefull, just look at dragonstar ( 7 reviews, Psion 4, Simon Collins 4, average 4.57 )and rokugan ( 2 review, Psion 5, average 5).

If you weigh the staff review, the dragonstar book will drop, while if someone post a score of 4 for Rokugan, it will drop but the staff review will help keep a good score.

That way it would seem that Rokugan is better appreciated than Dragonstar on ENworld, while in fact Dragonstar get more support on ENboards.

(Currently, products within the 4.5, 5 margin seems to me as being of similar value)

This is usefull if I want to discuss on the setting with others, while their is many people who like dragonstar, their isn't as many who like rokugan. Popularity play a role, sure, and your system might help or hinder some products to get popular on ENboard, I am not sure if it is good or not (it will only affect ENboard).

So I would like to know a bit more on your reasoning to weight staff review, so I can be sure you get the right decision with your website:D

BTW, everyone - check out the page for the PHB for a demo of how a product should look once everything is up and running. It has links to errata, articles, extra reviews etc.

It is very nice! but one thing is lacking, the author name.

[edit]
Is it possible to have the score that was given in other reviews website next to the link?

Also, if you want every D&D/d20 reviews linked from here:

the major french review website has a D&D section:
http://www.roliste.com/jeu.jsp?id=1322

and a d20 system section:
http://www.roliste.com/jeu.jsp?id=2021
[/edit]
 
Last edited:


Into the Woods

Remove ads

Top