Web sites and ratings on them. Attempt to rethink
Always considered himself a retrogradom. You do not have time to differentiate the growth of candles on the cake, and the monitors from black and green became colored, then colored and flat, the kilobytes were replaced by the same numbers in megabytes, gigabytes, terabytes ... sites. The young standard of web-dance, not having had time to grow and settle down, was recognized as a classic, imperceptibly moving into the category of an aging retro. It's time to stop and look through the photo album, or rather, that part of it, where ratings appear.
The concept of vebdvanolnosti revolves around the unification of extras in various categories. In any extras there is always the backbone, there are sympathizers, and, in fact, all the rest. The proportions vary, but, as a rule, the first ones are the least, the second ones are a little more, and the main mass is just “everyone else”. For example, in the “all programmers” sample there is a certain percentage of serious professionals, a larger percentage — those who are “well-versed”, and finally all the others — “do as I can” aka “bool b; if (ToString (b) .length == 5) {} ”- I specifically exaggerate for clarity.
First and foremost. If everyone is given the buttons for voting, the crowd will not vote at all for what the sympathizers will vote for, not to mention the backbone. And so it was and will be in any more or less wide sample, which is quite obvious.
By the way, if you have an association with democracy, then for good reason. Demos is not just a "people" , it is the very backbone and sympathizers, its most active and sensible part, which had the right to vote.
')
What will happen if fans of the Indian programming style, along with serious professionals, vote for something on the next web-dance website at the rate of one vote per hand? That's right, will be what the crowd vote for. I imagined this picture when I decided whether to make a rating system on my website. And, you know, did not. Refused the rating system at all. (However, I also refused to register, but this is a completely different story).
The second and last. Take for example the site of freelancers. Professional, registering on the site, has a zero rating. At the same time, the more deft Vasya Pupkin already have garlands of stars. Bypassing the discussed question “can the website adequately rate a professional,” I will touch on the topic of this inequality of newbies. The situation is worsened by the fact that the rating system is often made cumulative. That is, as long as a professional deserves a sergeant, Vasya will become three times a colonel (see the number of stars).
Does this increase the adequacy of the assessment? Or is it such an elegant way to increase the gap between the real state of affairs and the virtual rating? Is it too virtual, this non-democratic system, multiplied by the length of service? Look at the same freelance exchanges. There are rating tops freelancers who skim orders. But are they the ones who deserve it?
As a postscript. If the rating in one form or another is vital in the case of your site, then you should think about how to fundamentally change the approach to its calculation, to leave the averaging pattern “by ward”. On the forums, I am still urged to make ratings. What if you approach this wisely? For example, now my working version is to consider:
The counting system of the number of referrals (or what will replace it) will be left out of the brackets, it strongly depends on the specifics of the site.
In this case, "demand" newcomers have a chance to get to the top, and the old and inactive will go into the minus. Of course, this is just one of the options. I would like to see your ideas.
Alas, this solution is rather a “crutch” with an extremely limited scope. The whole question is whether it will be possible to find a universal solution that allows rating anything on the site from a truly professional point of view. And the second big question is whether it is necessary at all. In my opinion, it is still necessary, because it will allow us to strive for the best, and not for the “average over the ward.”
The concept of vebdvanolnosti revolves around the unification of extras in various categories. In any extras there is always the backbone, there are sympathizers, and, in fact, all the rest. The proportions vary, but, as a rule, the first ones are the least, the second ones are a little more, and the main mass is just “everyone else”. For example, in the “all programmers” sample there is a certain percentage of serious professionals, a larger percentage — those who are “well-versed”, and finally all the others — “do as I can” aka “bool b; if (ToString (b) .length == 5) {} ”- I specifically exaggerate for clarity.
First and foremost. If everyone is given the buttons for voting, the crowd will not vote at all for what the sympathizers will vote for, not to mention the backbone. And so it was and will be in any more or less wide sample, which is quite obvious.
By the way, if you have an association with democracy, then for good reason. Demos is not just a "people" , it is the very backbone and sympathizers, its most active and sensible part, which had the right to vote.
')
What will happen if fans of the Indian programming style, along with serious professionals, vote for something on the next web-dance website at the rate of one vote per hand? That's right, will be what the crowd vote for. I imagined this picture when I decided whether to make a rating system on my website. And, you know, did not. Refused the rating system at all. (However, I also refused to register, but this is a completely different story).
The second and last. Take for example the site of freelancers. Professional, registering on the site, has a zero rating. At the same time, the more deft Vasya Pupkin already have garlands of stars. Bypassing the discussed question “can the website adequately rate a professional,” I will touch on the topic of this inequality of newbies. The situation is worsened by the fact that the rating system is often made cumulative. That is, as long as a professional deserves a sergeant, Vasya will become three times a colonel (see the number of stars).
Does this increase the adequacy of the assessment? Or is it such an elegant way to increase the gap between the real state of affairs and the virtual rating? Is it too virtual, this non-democratic system, multiplied by the length of service? Look at the same freelance exchanges. There are rating tops freelancers who skim orders. But are they the ones who deserve it?
As a postscript. If the rating in one form or another is vital in the case of your site, then you should think about how to fundamentally change the approach to its calculation, to leave the averaging pattern “by ward”. On the forums, I am still urged to make ratings. What if you approach this wisely? For example, now my working version is to consider:
Element rating = lg ( multiplier * number of returns / time_to_point_publishing )
The counting system of the number of referrals (or what will replace it) will be left out of the brackets, it strongly depends on the specifics of the site.
In this case, "demand" newcomers have a chance to get to the top, and the old and inactive will go into the minus. Of course, this is just one of the options. I would like to see your ideas.
Alas, this solution is rather a “crutch” with an extremely limited scope. The whole question is whether it will be possible to find a universal solution that allows rating anything on the site from a truly professional point of view. And the second big question is whether it is necessary at all. In my opinion, it is still necessary, because it will allow us to strive for the best, and not for the “average over the ward.”
Source: https://habr.com/ru/post/53114/
All Articles