You're right in that I try to take an upbeat view about the value of the content. I have a real fondness for CP, and will fight to try to preserve what I see are the great points of it. If you talk to Chris, the area of CP he is proudest of, are the articles, so I think he will be open to suggestions on how to improve them.
In order to understand what the solution is, we need to first understand what the problem is - and this is my personal opinion, I could be wildly off course here. I guess the question is, why do people write articles? To some, writing articles is something that will look good on their CV - so they really don't care what the quality is, quantity counts more than relevance or quality. Others write articles because they have identified a niche that they are perfectly suited to fit, and they do so out of a sense of community - sadly, these articles will only really be of interest to those who are searching for that niche. Next, there are the superstars who have a yearning to share all the cool stuff that they have learned or are interested in. Finally, there are those who have written something outside CP and have come here to write articles to promote these things (be it an Open Source project, or a commercial product). We'll call each group A, B, C and D respectively.
Group A - the article approval process was meant to fix issues with quality. There are two areas of concern here; first, there were many, many dross articles posted before the approval process was put in place, so they need to be tackled. Secondly, it's plainly obvious that the approval process doesn't really work - partly because people have wildly different views of what counts as a quality article and partly because others are too zealous.
Group B - there needs to be a way to highlight these niche articles. Quality may be great or quality may be terrible, but if people aren't reading them, then the author will never improve (or even have an incentive to improve them).
Group C - the superstar authors are just as capable of turning out rubbish as everyone else. This needs to be recognised. Fortunately, people take the time to read their articles, so this is one of the highest probability of issues being raised and fixed - the superstars generally really care about the quality of the work they produce.
Group D - be it open source, or be it commercial, if an article is poor, it will only serve to damage the reputation of the product; as long as they are challenged on the quality, these are also likely to fix issues with the articles.
So, how do we deal with A and B? Well, group A is effectively a three stage process.
First of all, authors should be prevented from producing rubbish; this means halting articles in the approval process - if something's a tip and is posted as an article, the approval process should let the community automatically move that to the appropriate type, right now the process isn't quite fit for purpose.
Secondly, rather than automatically rejecting articles, more use should be made of the mentor process - most people aren't aware that we have mentors here on Code Project. I'm pretty sure that we could put something in place where the votes to remove could automatically involve the mentor team (and only let an article die if it can't be fixed in this process). Let's give budding auhtors a chance to "up their game" and produce something they are proud of.
Finally, the old dross articles need to be tackled. They should either be removed or, if possible, editors could fix issues with them. This would require plumbing being put into place to allow editors to know what articles need looking at, and to allow editors to mark them as being improved.
Something I would put in place as well - if an editor has had to make substantial changes to improve an article, I would remove the original author details. That way, they will not get the credit for other peoples hard work, and they will hopefully get the message that poor work is not rewarded.
Now, as to group B - that's a trickier one. I do feel that we could have a section on the home page that highlights a selection of random articles, and these could be selected based on number of views/time present on Code Project. I also feel that they should be language independent; if your filter is set to C#, for instance, you will miss out on an interesting article on extending C++ STL.
Again, these are my personal thoughts, and as always, the groupings have a million shades of grey in there, so there is no one size fits all category.
*pre-emptive celebratory nipple tassle jiggle* - Sean Ewington
A most eloquent, and thorough, analysis, thanks ! I consider your words in this message a virtual "tutorial" on the dynamics of article and tip/trick content, authorship, and quality, on CP. I hope elements of what you said, that, potentially, could be of value to authors, will be inserted into the CP author's guide
My comment was not focused on articles, and tips/tricks; rather, I was expressing concern about some rarer, truly great technical content that appears in the Lounge, and in comments on articles, and tips/tricks, even in certain QA answers: content, I feel, now, gets "lost."
However, I can also see that you could make a powerful argument that by using the search facility expertly, and filtering by rating, etc. that if you are after a specific topic, you have everything you need.
Perhaps what I am wishing for is kind of multiple "halls of fame:" an equivalent, in some way, to "pinned" messages on a forum ? ... but that's as far as I get with the idea.
You need to vote. Everyone needs to vote. Both up and down as appropriate. And, where appropriate, report.
It's only through voting and reporting that the top articles rise to the surface and the bottom...go to a place where only Google and those who can eek out a gem of wisdom from a potentially underperforming article will see it.
This weeks news letter had an article titled "WPF-Drawing Canvas Control". It looked interesting so when I had time on another machine I opened Code Project and tried search - tried Drawing Canvas - Nothing tried Drawing Canvas Control - Nothing tried WPF Drawing Canvas Control - Nothing tried WPF-Drawing Canvas Control - Still Nothing So I gave up on searching in Code Project and went to Google tried Drawing Canvas Code Project - - - Bingo! It would be nice to see the search engine in Code Project improved just a little bit. This example is just a sample of my frustration trying to find articles with the internal search within code project. Can it be improved?
Search for "Drawing Canvas wpf", set search filter to "articles" (the blogs and tip/tricks add noise), and set the date to Last Month and you'll get 3 results and the one you want is at #2. Not bad I thought.
Chris : Maybe if you automatically moved newer/more-viewed articles to the top, then the results would be similar to Google.
I can't bookmark this article (something which I have done for a number of other articles). Also the Icon to bookmark isn't appearing (neither are the other two). A box and broken graphic place holder appears in each position (with the bookmark being different -- perhaps due to the number of times I have clicked-on it).
I have checked My Bookmarks but do not see this link show-up!
Assume you click on "View the # replies to this message" link:
The faint horizontal gray bar that separates messages appears below the title of the response (bold, left justified) and the name of the author of the response, and time stamp (default text weight) to the right
This, visually, gives the disturbing impression that the title/respondent/time-stamp content line is part of the post above it.
Clearly, the horizontal bar should be above the title/respondent/time-stamp line.
I think I may have mentioned before that when you click the "View All Threads," from a view where you have expanded the responses, you are returned to the top of the "collapsed" page, which means that you have essentially "lost track" of how far down you had scrolled in the page in collapsed view.
Is there a technical reason why expanding the responses to a post, and then restoring the "all threads" view, cannot return you to the "collapsed" page scrolled down to its former position ?
This aspect (returning you to the top) is particularly annoying when you view comments on a long article, and expand them, and then restore the "all thread" view.