|
|
|
Ah, Missed/forgot about that one.
|
|
|
|
|
|
Ah well...
cheers
Chris Maunder
|
|
|
|
|
I believe that we need to do something about down-voting CP articles. It appears that any member can do it. I just read Work around Windows 10 WiFi reconnection failing[^]. I thought that it was a reasonable approach to a problem its poster had encountered. But it was voted 1 by 2 voters. I do not know who did the down-voting and I would like to be assured that the down-voters had sufficient cause to down-vote. Either they had obtained a sufficient reputation (maybe 5000 or more) or they offered a reason in the comments. I believe that any vote less than 3 should meet some requirement.
It takes a lot of effort to write an article or tip/trick. I think that CP is doing a disservice to its authors by not doing something about down-votes.
Gus Gustafson
|
|
|
|
|
This comes up on a regular basis. The previous experiment which required a comment for every down-vote didn't end well:
Article Voting: The dangers of all-good news - Chris Maunder - Professional Profile[^]
The current system does a pretty good job of filtering out down-votes from low-rep members, so long as higher-rep members also vote. For example, your "5" on the article you linked to has effectively wiped out the two "1" votes, bringing its total up to 3.91. The details of how this works are explained in the FAQ:
Code Project Rating and Reputation FAQ[^]
NB: That's not to say the system can't be improved.  Just that the obvious solutions have already been tried, and found wanting.
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
I appreciate the effort that CP put into assigning valid scores to articles. However, that does not answer the question as to how much I should rely on existing votes. I would have voted 4 for the article but voted 5 just to offset the unexplained low votes of others. It is imperative that CP fix this issue. I believe that the very reputation of CP is on the line.
Gus Gustafson
|
|
|
|
|
You're asking, in essence, for every article to have a definitive score.
That's like asking for a definitive review of a book, or painting, or movie.
Some people will like them. Some people won't. Some people will put lots of effort into thinking through their vote. Some will vote from their gut. Some will vote because they absolutely positively disagree with the methods used in creating a solution, even though the article was beautifully written and simple to use. Some will vote because it was the only thing they could find that solved their problem and let them get their work done.
The options that have come up over time are:
1. We show "Expert votes" whereby only votes from those with a certain rep are counted. This reduces the fly-by votes, which means far less votes, but probably more thoughtful ones. Or maybe not.
2. We split voting into categories: Presentation, Technical correctness, Easy of use. That way you could focus on just those articles that have high "technical correctness", while voters could just give high presentation scores and maybe abstain from the technical correctness vote.
Neither of these address the issue of spurious votes, neither address the "eye of the beholder" issue either.
I'm open for other suggestions.
cheers
Chris Maunder
|
|
|
|
|
Just thinking loud here, but why not both?
"Quality votes" on categories but allow anyone to giva a general vote. Members with sufficient rep gets a "popup" with the possibility to give categorical votes.
|
|
|
|
|
When I was young and impetuous this is exactly what I would have done, however doing so merely increases the complexity of the system for debatable returns.
Ultimately the system works best when it gets lots of votes, as in enough to drown out the small percentage (and it generally is a small percentage) of spurious votes. Hence the system needs to be as simple and inviting as possible.
I'm currently leaning towards showing an overall rating and an "experts" rating. When I get the chance I'll run some experiments and see if there's any actual difference.
cheers
Chris Maunder
|
|
|
|
|
Experts rating would definitely be much simpler and it is certainly more important that it just works. And you're probably right about the debatable returns.
But since I'm into wishing, I would like to mention, that one of the things I'm missing is to be able to search for whether the article is a "Code Sample", a "Walkthrough" or a technical background article.
Or combinations thereof. And while that easily could be done in the form of tags or similar, the articles usefulness in these areas should be asserted by the users I believe.
Ah, scrap that. It probably wouldn't work.
|
|
|
|
|
do you really think the future "authors" will get that sorted correctly?
And what happens with the millions of past articles without that cathegorization?
M.D.V. 
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
That's why I finished with "scrap that".
|
|
|
|
|
Jörgen Andersson wrote: to be able to search for whether the article is a "Code Sample", a "Walkthrough" or a technical background article
Yes! And it's really easy. If the articles are tagged correctly. This has been an issue for ages but one we're addressing in the coming weeks. It's important.
cheers
Chris Maunder
|
|
|
|
|
Nice! 
Chris Maunder wrote: If the articles are tagged correctly
There's always an if, isn't it?
|
|
|
|
|
Chris Maunder wrote: Some people will like them. Some people won't. Some people will put lots of effort into thinking through their vote. Some will vote from their gut. Some will vote because they absolutely positively disagree with the methods used in creating a solution, even though the article was beautifully written and simple to use. Some will vote because it was the only thing they could find that solved their problem and let them get their work done.
Don't forget the "(un/)friendly votes" which IMO is very extended in parts of the community.
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
You have to remember that anyone can read an article here, just as anyone can write one. And that means Joe Moron from Kansas, Ohio cna read it expecting it to be exactly what he needs to hand in as his homework. And when it isn't, he is free to vote it as he pleases. That's democracy! The system has algorithms in place to exclude "spurious" answers once sufficient "real" ones are in place, and each vote it "weighted" by the reputation of the source. So a 1 from Joe doesn;t make a lot of difference compared to a 5 from me for example. If the article gets sufficient 5's, the spurious 1's will not only have minimal effect but will eventually be discounted completely.
Equally, if Joe writes a poor article and gets several 5's from his mates, they can be countered by high-rep 1s, and they will eventually be discounted as well.
For example, look at this: List<T> - Is it really as efficient as you probably think?[^] - it got 16 1's because people didn't really read past the introduction, and 83 5's. And those 1's prompted me to rewrite the intro to "persuade" people to read further. It's sitting there with a solid 4.38 and someday the 1 will be discounted.
And this: Using struct and class - what's that all about?[^] won "best article of the month", and still got downvotes from Joe and his mates! 
It's not a perfect system, but it's pretty good - it works out in the long run. If you like an article, vote it up. If you don't, walk away or vote it down. If it's spam, plariagrised, or abusive, report it. But don't take any of it personally!
Sent from my Amstrad PC 1640
Never throw anything away, Griff
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
modified 11-Oct-18 7:26am.
|
|
|
|
|
OriginalGriff wrote: ... gets several 5's from his mates, they can be countered by high-rep fives, ...
High-rep ones, surely?
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Fixed.
![Java | [Coffee]](https://www.codeproject.com/script/Forums/Images/coffee.gif) required...
Sent from my Amstrad PC 1640
Never throw anything away, Griff
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
I needed to attract Sean to a problem: How do I fix image links[^] but @sean-ewington doesn't get changed to a link in QA answers, or in the comments - so presumably Sean doesn;t get notified. It works in the lounge (and presumably here)
I'm pretty sure it used to work in QA, but it's quite possible I'm losing my mind...
Sent from my Amstrad PC 1640
Never throw anything away, Griff
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
I don't believe it worked in QA answers since they are meant to be answers, not discussions, but I could be wrong and am certainly open to changing this.
I'll add this to the TODO list since it actually fits in neatly with the massive refactor we're currently working on
cheers
Chris Maunder
|
|
|
|
|
|
|
We already use the syndication-source meta tag for blogs for exactly this reason.
cheers
Chris Maunder
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
