New Rating System?

Hi everybody,

may I introduce myself as the responsable for some ideas behind the new ratingsystem,
so please blame me ! (compliments are welcome as well)
I occupied myself with the rating/navigation-system in the context of my thesis/diploma,
so that's way it appears more or less academic.
I spent some time analyzing the the database to answer questions like: How is the
distribution of the single marks? How is the number of votes
compared to the number of downloads? Is there a relation between those indicators?
In fact I got much inspiration of discussions of the past, i.e. to allow users just to vote for
their own ensembles or to forbid double votings to achieve a more honest result. (http://www.nativeinstruments.de/for...=&threadid=7170)
There is a great weight of downloads because in general there are more downloads than votes for a ensemble. In the long run this effect should go down as in the new version users are encouraged to vote after downloading. Still the number of votes and the number of downloads have the same weight thus they are a relative value to their corresponding maximum, LIME.lite v1.0 at present -> 10+10+10/3 = 10,04.
(That’s German engineering at ist best, we try to fix that bug
Yes I agree metric scales tend to split up the world in “good” and “bad” but everybody can judge on his own wether he wants to pay attention for in that case popular/high quality or unpopular/low quality? ensembles. In addition the new sorting features allows to completly ignore the new rating and to set up some criterias which make sense to oneself. In my concept I had four additional criterias “originality”, “sound quality”, “usability” and “performance” for the rating but there have been some opinions against it ... It is easy to answer the “what” with a metric rating but not the “how” still verbal decriptions are unbeatable. An alternative or addition could be criterias which apply (10) or apply not (0) to a certain feature or the usage of an ensemble, i.e. ambient, lead or rhythmic character.

Best,
Simon

 

Simon, you are MAD.

 

You would never have enough categories to describe everything properly. Even with categories like "sound quality, "usability", etc, often times none of those apply.

 

have a look at the first page of the top 100 most of those ensembles have loads of originality, they are all usable and
are quality uploads.

i think this is a far better reflection of the top 20 best ensembles
than before, although i think david coffin's tutorial should still be there.

MY main frustration is the rating marks. for me all the ensembles on that first page should have a rating of more than 9.

but i think the new system is better than the old. if i am correct,
all uploads in the top 100 had a rating of 9.3 or over, just before the new system was introduced. that is really stupid.

sowari

 

Absolutely.

Er, I think you misunderstood the plan here. Amazon has no forum. This would have nothing to do with the forum. It would be an optional thing that each builder/uploader could add to their *User-Library* user profile, not their forum user profile. As an example, right now you can set email options for your User-Library profile that have nothing to do with your forum profile. Once you set up this list, people could link to the list from all of your ensemble-description pages. Obviously it would require the designing and coding orf a bunch of forms and pages that don't currently exist.

- CList

 

Ah, got it now. I think that would be a great addition to the user library!

 

I agree. The best ones are more or less at the top of the pile but, as you say, the scale of the thing is wrong. Anyway, I'm off 'cos it's time for Eeenngggurrrlunnnd!

 

Oh well... :-(

A bit of light relief for football fans:

http://www.weebls-stuff.com/toons/35/

 

Ernest, you're a very good bug-fighter.

Thanks for all the hard work you put into your ensembles, and good luck with whatever it is you intend to do!



(He's probably going on a world cruise…)