Rating System at SB

Used to freak me too....I have NEVER even seen a 7!

You should take a look here http://www.skinbase.org/msgboard.php?topicid=879 dear Fred ... <br />...and coming back to explain me :fool <br />LOL

Very well put Butch....primarily I make my walls for me. They are usually made to fit in with a desktop arrangement of my own so as you say they have to please me first. When I post my work it is to share that with anyone who wants it. I lose nothing if no one wants it because I have achieved my original intent of satisfying my need for it. And yes.....low ratings and derogatory comments usually spell jealousy. :congrats 😎 ;) HH :winter HH :winter HH :winter HH :winter

Well I usually dont like to comment on anything about rating systems as they mean absolutely nothing to me. My reasons are as follows..... <br />When I make a Windowblind first and fore most I make it for myself and if I am happy with it then who cares if some people wants to rate you with a 4,5,6, whatever because this is always gonna happen. You know you cant please everybody ! for me the most important thing is that I make one person happy, after all, isnt that what its all about ? You know, no one should actually take rating to heart. I have seen over the years many people rate your work very low, just because of jealousy and not based on its merit. On the other hand I have seen people get extremely high ratings because they are one of the Good Old Boys. So all in all ,and this is just my opinion if you like it, thats all that counts , accolades cannot replace self-satisfaction ! HH

All good points Iain. :) HH HH HH

I see your point on download figures Fred and do agree. I am simply saying I do not accept that the public always gets it right. At the end of the day if we did it commercially then yes they count. We can use them to guage what works and what doesn&acute;t so we can adjust our work accordingly. What irks me is people complaining of lack of downloads when they do nothing to interact or generate interest. Or they have been doing it for five minutes and expect an instant following. Commenting on others work is one way of getting to know others and getting known. It is a participation sport. The comments I get from people that count around here tell me all I need to know. Like you I know I am appreciated and being featured as I have said earlier is no longer a desire nor a need and that is not to say I do not appreciate it when I get it. I have simply learnt to have no expectation of what my work might bring then anything it does generate is a plus. As for this site....we can hold our heads high....we appear to be fairing better than others from what I have seen while touring other sites. We are as Nikos says, a friendly International site and in general, given the numbers here we get on bloody well together. Growth and change promotes that with an ability to accommodate and understand with a willingness to discuss. We have no elite here....we are all one and that is special in and of itself. :congrats 😎 ;) HH HH HH HH HH HH

Just one more point... I do think the two up now (Avy and Jacob) are well deserved. HH HH :congrats

Yes Iain - You speak candidly on this subject. Personally, I&acute;ve reached my conclusion as to the features and how and why they are picked. Its not a good one either. Politics is the BIGGEST motivator. With my experience here as of late - I had 6 features since I&acute;ve been here - my heart of hearts tells me it was much more to do with encouraging the new guy than thinking my work was the best at the time. If thats the intention here - then it is WRONG. They think they are helping people, when ultimately it leads to a false sense of accomplishment. And a BIG factor on how long it is up is how many downloads it gets... some are up for a day - others for over a week. RIDICULOUS. There are plenty of great works to keep it moving along better than that. And with that... its less likely that a deserved one would get overlooked. <br /> <br />Sorry all - thats how I feel. I see your point on download figures - but I respectfully disagree. Its about sharing with me (as with you) and a higher number there means more people have my work for themselves, regardless of their tastes. Funny - thats probably one (not all) of the factors that keep me here. If my work had no downloads, whats the point in sharing it. Yes - and the comments are great, I enjoy reading all of them. Another factor... I guess what I&acute;m trying to say is that I feel appreciated here - the feature thing is not going to change that (at least for me) <br /> <br />I&acute;m still gonna have a Merry Christmas, I hope you do as well. And everybody here. I just had to speak out on this one. <br /> <br />

Like Jacob said ...stats mean nothing as do download figures. I know from my twenty odd years in the printing industry that public taste is fickle and to the trained professional they sometimes seem lacking in taste totally. I recall when everything was done in pastel colours. That would have had to have been the most boring period in design history. Like today we have more 3D apps than you can poke a stick at and people insist on flat,bland designs. Makes no sense to me. What I think is of more importance here is to open a discussion about what constitutes a featured piece of art or skin. What criteria needs to be considered. Does it require being picked based on skill level involved or on improved talent and ability by the artist? Does it require to be politically motivated in an attempt to draw back artists who have not posted for excessively long periods or should it be drawn from those currently actively involved on the site in order to encourage their continued interest? Or do we give it out to the person who whinges loudest and ignore all what I have stated prior? I think it is a questionmark in everyones mind at the moment or has been. I also accept it is not an easy task. Every art site has its problems with features and not everyone can be pleased it would seem. Personally it does not bother me how often my own work is featured as it is not why I share my art by uploading it here. If we are to have a feature and ratings though I think it needs to be determined and clarified and probably explained to us all so misunderstandings and temper tantrums can be avoided. A simple explaination of policy in those areas would be helpful and then people just need to accept it and quit the antics. :congrats 😎 ;) HH :winter HH :winter HH :winter

Over a year ago I put some things in an older thread: <br /> <br />Under stats (in your profile section or base page) you find: Activity ratings - a general log of anything you input from comments to uploads or ratings, I think, and such, no matter what quality or how good you were rated. Just amount. This results in a position solely calculated on the ratio to other artists figure for that in comparison, also ending in our monthly statistics clickable in the left sidebar when working (sometimes in maintenance because of design overhaul, system changes, or calculation phases when months or years change) <br /> <br />Uploads - the amount of skins or screenshots you have submitted. May or may not contain uploads deleted in the meantime but will mostly count the amount in your gallery. <br /> <br />Page views - the amount of times your profile pages like gallery, comments, etc. was called up intentionally by any user including yourself. Note that seriously working artists making big suites or using legal copyrights of other artists need to test links, downloads, in stages giving some more views to that figure without their wanting to distort their own popularity. Unfortunately, it seemed in the past entering a comment also added a DOWNLOAD to the stats which is dl, by the way. Likewise, there is intricate interaction with our Slovenian pages all added into the stats which are a correct total but might also result from clicks and downloads on pages you are also displayed on but do not even read or think about... <br /> <br />So, average dl, is a bit like the popularity ratio - how many downloads per days it is up which appears under Skin Statistics or personal Skin statistics. You submit Monday, watch stats on Wednesday means up 3 days. When you view other or your own galleries you can sort by total downloads (over the years) or current popularity (what is downloaded most from current section or has best download ratio when held next to amount of days up) <br /> <br />Downloads itself is of course how often your submissions are downloaded, including by yourself. If you abuse this too much it becomes obvious and might be logged by some of the top admins. But some testing can even be good. AND it ALSO includes how often you download skins by OTHER users, I think. <br /> <br />In Top Author etc. ratings in left sidebar you also find MOST COMMENTED which I think is amount of comments your submissions get, not how often you comment others. END OF QUOTE from my Nov. post in 2005. <br /> <br />HH DONT LET IT KEEP YOU FROM A NICE HOLIDAY OR 2 ;) HH <br /> <br />Most systems have a general way of working. That also means being playable and not perfect. (Example Slovenian pages and comments and things can add to downloads). Things have changed a bit, but Serge´s link still explains SOME things rather well. Although I must admit I am not that much of a pro to automatically understand everything it means. <br /> <br />I dont think much is biased here on purpose. If certain users overdo things, it shows and can be corrected. In fact, testing your own downloads might be a good bump if its only once or twice. HH <br />

Point well taken, Jacob. 😁

Forget the ratings. I really wonder if it reflects anything at all. Mostly it is all biased as it seems with a lot of comments as well. Or it is political. The one thing that you can measure is popularity.

Thanks Iain - but no PM&acute;S - half the time they do not get answered anyway. Probably because the PM system is less than reliable. :,

Like I said... I won&acute;t rate anything else here. Nothing but a band of confusion. And I&acute;m good at math, too. <br /> <br />Thanks everyone!

Thats true Jacob. There will always be people here rating low for the sake of rating low. Popularity is driven by the number of downloads vs amount of time posted, so I see your point. Unless someone out there keeps downloading their own work - to drive up the popularity. Anything is possible. 😁

Rating is a bad idea. It will always be a biased one, unless all people who download the work are enforced to rate. Maybe then you get a somehow impartial view on ones work. Since enforcing needs policing it tends to become dictatorial. And when staff has a bigger say on what&acute;s worth being rated higher for whatever reason, that&acute;s plainly ridiculous. So, forget the rating, I stick to the popularity vote, which doesn&acute;t mean a work is better. It only means popular - non-popular.

K.I.S.S. <br /> <br />Keep it short and simple... <br /> <br />Just so no one is left in the dark. Thats all folks!!

Now its improved... but still misleading. You know, when I was in school - it did not matter how many grades on tests I had in a given term. The scores were averaged out and that was that. The same principle should be applied here. How is an artist supposed to corrolate an 8.0.... with an actual &quot;10&quot; that someone gave him/her? The rate should read exactly the same, as in an averaged grade in school. A 90, 100, and an 80 would gain an average of a &quot;90&quot; on a report card, just as a 10, 9, and an 8 should earn a &quot;9&quot; as a rate. <br /> <br />This BS of trying to turn something that should be so simple into something so complex is ridiculous. I appreciate the effort of the change that was made - but I say lose this system once and for all. Did you ever hear of the K.I.S.S. concept? <br /> <br />I&acute;ll leave it at that. And I&acute;ll say no more of this subject. Honestly, its not even worth the effort. <br />

<br /><b>The rating system has been changed </b> , it is still based on the BAYESIAN ESTIMATE, but now it has higher base. <br /> <br />Who knows maybe in the near future we may get rid of the rating system ;) <br />

Good idea Keeko! If this site wants to attract a lot more traffic, I would suggest a very different approach to what other sites offer. Get rid of rating is one. Involve people with interaction is another. Not the everpresent chats and messaging, but real involvement with the site. Get rid of hierarchial structures is another. Make a difference in the skinning society, do not follow the same path as all the others, you become one of many and frankly, it becomes boring.

Why don&acute;t we just get rid of the rate system and have a weekly vote for the for the Featured Skin, Walls and Photo, that way it would change every week. I really don&acute;t care about the rates I am just happy that someone likes my wall enough to download it.... 😂

Wow! Seems OTHER systems internally might ALSO be more intricate or have more cons than I thought, then. 😮 Thanks for the details, Gregor. So it sort of autoadjusts for various good reasons. And also considers the amount of total traffic related to the subject. <br /> <br />Maybe politics should work the same way :p Does changing a year reset any counters? ☕ Aside from that, you DO have good traffic, might be another thing involved... 😂

Did someone change the system? I noticed a rating of 8.2... much more in line with most people&acute;s understanding... if that is the case, hats off to Skinbase for making that change! :congrats

Thats ridiculous. Give me a break. I&acute;ll never rate anything here. What a bloody waste of time. Jesus, you might think we are in a chemistry class. I&acute;m all set with this. Thanks grave for taking the time, though. 😮

Am I glad that the formula for creating a design is far simpler than rating it! That definitely makes my head hurt! 😂 😂 😂 😂 😂 :congrats

For rating we use Bayesian estimate rather than standard average rating suma_votes / num_votes. <br />We dont use this formula because some article have 100 votes of 4 and one article have 1 vote of 5, will be this article with ONLY one vote higher rated! <br /> <br />So we deciede to use <b>BAYESIAN ESTIMATE</b>. which also counts how many votes article gets .... <br /> <br />Detailed description: <br /> <br />We are using the same true Bayesian estimate formula used by the Internet Movie Database, for calculating average ratings. The only difference is that our rating system uses multiple dimensions (presentation, graphics, usability, technical, and personal slant), so we apply the formula to each dimension by using the respective mean value of that particular dimension (C), take the sum of individual results, and divide this sum by the number of dimensions, which is five. <br /> <br />weighted rank (WR) = (v / (v+m)) * R + (m / (v+m)) * C <br /> <br />where: <br />R = average for the design (mean) = (Rating) <br />v = number of votes for the design = (votes) <br />m = minimum votes required to be listed in top 25 (currently 6) <br />C = the mean vote across dimension <br /> <br />This formula normalizes scores, that is it pulls a particular score (R) to the mean (C) if the number of votes is not well above m. In other words, if a particular design has only a few votes above the minimum required votes to be listed in top 25 (m), the average score is decreased a little if it is above the mean, or increased a little if it is below the mean in accordance with the normal distribution rule of statistics. <br /> <br />Here is an example for a single dimension: <br /> <br /> WR = (6 / 10) * 5.33 + (4 / 10) * 7.18 = 6.07 <br /> | | | | | | <br /> v v+m R m v+m C <br /> <br />The formula normalizes the average rating of a relatively low rated design from 5.33 to 6.07 since the number of votes (v=6) is only slightly above the minumum required votes (m=4) and the mean across the dimension (C=7.18) is quite high. If, in the future, this particular design gets more votes, the difference between R and C will increase as the number of votes increase. The idea is that the more the votes, the more representative the average rating is. Please note that this example is only for one dimension. Different dimensions have different means (C). <br /> <br />Or find more about this formula on google :) <br />

Thanks Nikos. I respect that. :)

The rating system in Skinbase has no power to give any bad results. <br />Believe it or not, I never see it or pay any attention to it. <br />It is not even standard; Gregor has explained to me some years ago, it was complicated then as it is still now in my head ;) <br />Rating systems never were fair and never will be fair; I wonder why we still use them... :sen <br />Who knows, maybe the next year we will erase this option from Skinbase. <br /> <br />It&acute;s OK Fred, I respect your point of view, as we all should do for each other ;) <br />The only politics in the Featured works are when we have 2 good works, one from a &quot;newbie&quot; and one from a &quot;veteran&quot;, <br />we choose the &quot;newbie&quot;, for obvious reasons, no other politics! <br />