|
Bingo
cheers
Chris Maunder
|
|
|
|
|
For most postings without an ongoing discussion should not be necroed weeks, months, or even years after the last contribution: anyone intereseted in the issue probably has moved on, and likely the original question is no longer of interest in the context of modern tools, techniques and hardware capabilites. Would it be possible to lock such threads after X weeks of inactivity?
That could reduce clutter and the number of traps well-meaning devs could be caught in, trying to answer an outdated question...
GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)
|
|
|
|
|
There might still be some genuine answers that really bring more value to that question, and those would get lost.
I mean...
I have seen questions, that got an answered but was not accepted. Then a spammer posted something, we nuked it, a good answer was added with a new (/newer/better) alternative way of solving the issue.
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.
|
|
|
|
|
In most cases, if someone is looking for help, any actual help is of little use when posted two weeks late. And for homework it may already be too late two days later. If someone has a good idea on an old question, it's probably better to post that solution as tip/trick and refer to the Q/A question rather than adding a new solution to an old question.
GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)
|
|
|
|
|
But how are you going to police that? Most of the people who do this have no interest in writing articles, they just trawl the internet looking for questions that they can answer.
|
|
|
|
|
Not articles, just tips.
That said, I'm not sure how useful the tip section really is: it's not like you always search the tip section when you have a very specific problem and are looking for a very specific solution. I probably wouldn't ...
Maybe we should make a survey on how often people actually read the tip section, and how often they deliberately search it for a particular topic.
GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)
|
|
|
|
|
Stefan_Lang wrote: Not articles, just tips. Well it's exactly the same issue; how would you get them to do it, even if theyknew that was an option? And many of these "solutions" are only relevant to a single question and probably not valid as tips. Which is why we post Solutions rather than Tips ourselves.
I think the Tip section is as good as the Articles section, full of lots of useful stuff, if you just take the trouble to go there and look - something I have done many times. But, as SWMBO famously said, "why bark if you haven't got a dog?".
|
|
|
|
|
Stefan_Lang wrote: Maybe we should make a survey on how often people actually read the tip section Usefull has nothing to do with tip or article. It is related to relevance / accuracy of the information.
Stefan_Lang wrote: nd how often they deliberately search it for a particular topic. I don't search in the sections, I search for a topic in the whole site and have a look to the preview of the results list, going to the concrete items when I think it might have something useful. It doesn't matter if it is an article, a tip, a Q&A or a forum message. Matching / related information is always welcome, no matter in which format.
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.
|
|
|
|
|
Stefan_Lang wrote: In most cases, if someone is looking for help, any actual help is of little use when posted two weeks late. For that person yes, but there are over 14 million registered members + unknown not registered that look for information here. So, yes. What is late for someone, can still be helpful and up-to-date for another one.
Stefan_Lang wrote: And for homework it may already be too late two days later. If it is clearly homework, and the guy doesn't bother to try, there is less to no chance he will get an answer, it doesn't matter if that day, two days or a month later
Stefan_Lang wrote: If someone has a good idea on an old question, it's probably better to post that solution as tip/trick and refer to the Q/A question rather than adding a new solution to an old question. Both, yes and no. There are things that are so much specific to an issue, that they are better placed in the Q&A where the situation is described and the context is given for a concrete use case.
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.
|
|
|
|
|
Just because the original poster may not need the answer anymore doesn't mean others who stumble across the question won't.
cheers
Chris Maunder
|
|
|
|
|
Ok, looks like at least three people disagree with me on the potential usefulness of such postings, and no one so far chimed in to support my case. I am not convinced, but at least we had a good discussion.
Does that mean we should supply answers even to old questions in case we have something valuable to add?
GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)
|
|
|
|
|
Stefan_Lang wrote: Does that mean we should supply answers even to old questions in case we have something valuable to add? I would say... yes
We even got a heads up in the S&A Forum by CP Staff to allow this and not to report valuable answers no matter how old was the post.
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.
|
|
|
|
|
But it would be nice if, when you try to add an answer to an old question, you'd get a popup stating "This is an old question, are you sure you want to add an answer to it?"
|
|
|
|
|
Yes. I know of at least one forum that puts a reminder prominently on top of the edit box for entering a post in case the thread is older than two months. Something along the line of:
"This thread is older than two months. Be sure that you are really adding value before posting."
It's obvious because it shifts the edit box a little downwards, and it's nonintrusive. Also you'd notice it before spending time writing up your post, therefore I think it's better than a popup when trying to send a finished posting.
GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)
|
|
|
|
|
|
That's a good possibility too
@Chris-Maunder or @matthew-dennis the point already has the flag "rejected", but the debate continued... I think it is worth to consider this option
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.
|
|
|
|
|
Next question: what do you use to calculate that it's "two months old"?
- The date the question was first posted;
- The date the question was last edited;
- The date of the last answer;
- Something else?
I don't think that date of the last answer would work - that would suffer from the same problem as the revision date we currently have. Spam answers drag the question back to the top of the list and then get kicked into oblivion, but still count as the "last answer".
Personally I think the date of the last edit to the question would be the best bet. But it's still not foolproof: what if a spam answer drags the question back into the active list, and a well-meaning editor edits the question to fix the formatting, not noticing the original date? Would we need to add the warning to the edit box as well?
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Richard Deeming wrote: Spam answers drag the question back to the top of the list and then get kicked into oblivion, but still count as the "last answer". For that the "This question has got X answers so far" or "last answer posted on XXXX-XX-XX" could help, when the info remains after a spam nuke
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 would rely on people reading the message.
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Yes... that's true.
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.
|
|
|
|
|
Richard Deeming wrote: Personally I think the date of the last edit to the question would be the best bet.
Yes.
Richard Deeming wrote: But it's still not foolproof
Well, there is no such thing. But this is close enough.
|
|
|
|
|
Indeed. To quote Douglas Adams:
Quote: “A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools.”
GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)
|
|
|
|
|
A mistake which can easily be rectified by spending five minutes perusing QA.
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
1. I keep seeing old Q/A questions back on the front page. When I check, I find it's old, and nothing has been added to it. Except that under the original question it says something like 'update 18 hours ago' - only there is no version history, indicating there is no update at all!
Example: How Do I Replace A Word In A Text File In C++?[^] posted years ago, 'updated' 18 hours ago, but no new version, no new comment or solution.
Similar issue: A Q/A question on the front page that claims it's been 'updated' 2 days ago, but when you go investigate, the question and all solutions and comments are from 2012. There has been one update to the question in 2018 - but that is not '2 days ago'! See complexity analysis of this code[^]
Since I also keep seeing people necroing old questions by posting comments and solutions, I wonder if that 'updated' date refers to such a change that may have been deleted afterwards? If so, can the web code be changed to make sure the update time stamp gets reset as well?
GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)
modified 7-Nov-19 2:58am.
|
|
|
|
|
It's because "old questions" are a fertile ground for spammers to post on, in the hope that we won't notice.
The spam gets rejected, but the question has had a "response" which pushes it back up the list.
I'd agree that it looks wrong ... but it's difficult to see what can be done without adding a fair number of special cases to the codebase - and I'm never that fond of those in my code!
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
AntiTwitter: @DalekDave is now a follower!
|
|
|
|