|
I have noticed recently, that the main factor in calculating popularity of an article (or other type) is votes. This is all well and good but, while not taking much notice of views is sensible, I would like to suggest that the new feature Downloads be added in. I have several articles that have only two or three votes, so have low popularity at the moment, but have a comparatively high number of downloads, several hundred. Therefore, I would like to suggest that Downloads would be a better method of judging popularity or quality of an article.
Thanks, Ed 
|
|
|
|
|
How do you cater for those articles that have code downloaded lots, but the code really sucks? All of a sudden, they'd seem to be better than they really are.
|
|
|
|
|
Yes this is true, however, perhaps do not make downloads of equal weighting to ratings but perhaps half or one third. If the code really sucks and is too popular, people will gradually be more disappointed by it and so then downvote it (you would hope). It is true to say, I think, that upvoting is less common as it seems to have less of a purpose (for some odd reason this is what people think  ) however, downvoting is common (when it is necessary) as people vent their frustration with an article and can see an obvious purpose to downvoting. On this reasoning, I would say that an inclusion of downloads in popularity/rating is justified and worthy of considration.
|
|
|
|
|
I have known some very popular people who also really sucked.
|
|
|
|
|
Pete O'Hanlon wrote: All of a sudden, they'd seem to be better than they really are.
I think you are mixing "popular" with "good." Perhaps the voting system, since it takes less effort to reward one's friends and lash out at one's enemies, is more vulnerable to being skewed than downloads, but both are capable of manipulation.
The 3-legged stool of understanding is held up by history, languages, and mathematics. Equipped with these three you can learn anything you want to learn. But if you lack any one of them you are just another ignorant peasant with dung on your boots. R. A. H.
|
|
|
|
|
Nope. It's just that I have read a few articles that seemed really good, but once I have downloaded the code I have realised that the article has had more care taken with it than the code. Generally, in this case, I try to give feedback on how the code could be improved.
|
|
|
|
|
Pete O'Hanlon wrote: It's just that I have read a few articles that seemed really good, but once I have downloaded the code I have realised that the article has had more care taken with it than the code.
Yes, that's happened to me once or twice. But haven't you also encountered articles where the voting system has given them a much higher, or much lower, score than they deserved?
Pete O'Hanlon wrote: I try to give feedback on how the code could be improved
Of course you would. I wish I felt you represented a norm rather than an outlier.
The 3-legged stool of understanding is held up by history, languages, and mathematics. Equipped with these three you can learn anything you want to learn. But if you lack any one of them you are just another ignorant peasant with dung on your boots. R. A. H.
|
|
|
|
|
Oakman wrote: But haven't you also encountered articles where the voting system has given them
a much higher, or much lower, score than they deserved
Yup. Bill Sergio's spring to mind - he's actually produced some really good code but his articles have attracted a lot of vitriol. Even though he and I have had some fairly public spats, I have always tried to be fair to his articles; it would be demeaning to play voting games just because we've butted heads a few times.
|
|
|
|
|
I would have to disagree - sometimes you can't judge the value of an article without downloading the code. Then you find it was written by a chimp. Since you can't un-download code, that would inflate the apparent utility of the article.
Real men don't use instructions. They are only the manufacturers opinion on how to put the thing together.
Manfred R. Bihy: "Looks as if OP is learning resistant."
|
|
|
|
|
Please read my response to Pete - don't weight downloads equal to votes, then people can downvote an article and thus counteract the increase in popularity.
|
|
|
|
|
But why give it an artificial increase in the first place? Isn't it better to avoid a mistake, rather than bodging round it once it's made?
Real men don't use instructions. They are only the manufacturers opinion on how to put the thing together.
Manfred R. Bihy: "Looks as if OP is learning resistant."
|
|
|
|
|
But there you are assuming that the code is always bad and so will always be a mistake. More often than not the code is good and so no mistake has been made and an increase is deserved.
|
|
|
|
|
No, I'm not - I'm reserving judgement until I have enough information to make an informed decision.
Real men don't use instructions. They are only the manufacturers opinion on how to put the thing together.
Manfred R. Bihy: "Looks as if OP is learning resistant."
|
|
|
|
|
Okay but unfortunately people tend not to upvote articles even if they like the code they download. This is why I think downloads should be included in popularity calculations. If people start finding popular articles with rubbish code then they will do something about it - downvoting appears much more common than upvoting.
|
|
|
|
|
People don't always upvote answers in Q&A or the forums either, but nobody is asking for views to be taken into account!
You just have to accept that some people are too rude to be worth bothering with...
Real men don't use instructions. They are only the manufacturers opinion on how to put the thing together.
Manfred R. Bihy: "Looks as if OP is learning resistant."
|
|
|
|
|
That example is a very different situation, but if you are totally against the idea then never mind...
|
|
|
|
|
It's not that different: it's all about correct ratings for information. The rating of both depend on votes: code downloads are analogous to answer views. Logically, if you count a download as a "good vote" for an article, then viewing an answer should count as a "good vote" for that.
Real men don't use instructions. They are only the manufacturers opinion on how to put the thing together.
Manfred R. Bihy: "Looks as if OP is learning resistant."
|
|
|
|
|
Not so - you view an article, if it is convincing then you download the code. Q&A, however, you view and may or may not like what you see. With views it is impossible to tell whether the user likes what they read but with Downloads, it suggests that the user liked what they read enough to download the code. Clearly then they are not equivalent examples.
|
|
|
|
|
There is a lot of merit in what you say, but it's too simplistic.
Luc will be along shortly to tell you and me why we're both wrong.
|
|
|
|
|
I agree that any implementation I have suggested is too simplistic - sort of why I suggested it, for CP people to work out the real solution - but the concept is more what I was suggesting.
|
|
|
|
|
Luc's on his way, he will get this sorted.
|
|
|
|
|
SA's following on behind to point to his post and explain why it's the correct solution.
|
|
|
|
|
There are lots of factors that can be used to measure the popularity and I would argue that bookmarking an article is just as, if not more valuable than a download. I could also add a timer to measure how long an article was read, or whether the entire article was read via scroll tracking, but in the end the popularity value was designed to simply be a way to differentiate two similar ratings with different numbers of votes.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
|
They are probably using a server-side link in a masterpage. Server-side stuff likes to butcher relative links (so "?display=Mobile" becomes "http://www.codeproject.com/Masterpages/?display=Mobile").
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
