close old thread with link possibility to new one

[Gabi1]

after observing several old threads that has been followed up very recently, i think they should be closed to new posts, mainly to keep integrity of each thread. i believe this should apply to all boards.

However, we should be able to create follow - up threads, with links or quotes to the older ones.

Connecting that with a better search engine, it should allow us to more efficiently look up information on the forum.

[Ray.R]

Hi Gabi,

I don't know if it's that simple...  I'll give you an example of a post I saw yesterday.

Yesterday, someone replied with an attachment (example solution) to a question from years ago (2005 I think??).

I was about to reply the usual "hey bud, this post is x-years old"..

However, I have done that same thing in the past... >> post an answer to an old post <<

Here's why..

You do a search.  You come up with a few hits.  You read them..  none have solutions.  You figure:  "oh... no-one has come up with a solution".

You develop a solution and you post your solution to the threads that were looking for the same thing. 

Why do this?  Well, when someone else looks for the same topic, they (hopefully) will find those posts and your solution.  That keeps the number of new threads down, otherwise, you may end up with many hits for the topic and you spend time looking for the right one.

I don't know how much better the search engine can be.  Maybe a spinoff of your idea is to put the recent solution in a new thread and go to the similar/same older threads and post a link to your new answer...  Again cluttering the forum(???)

I don't see a problem with posting new answers to an old post (or updates).  But asking questions to someone who posted years ago is ... well.. euh....  you know 😉  But then again, I occasionally look at stuff I posted 5 years ago because I remember that I had a solution for something I did back then..

🙂

[altenbach]

It is scary how often an old thread is really obsolete, because the problem has been fixed in newer versions of LabVIEW or a new feature allows a much simpler solutions than was possible years ago.

Nobody in his right mind would systematically go through all the old threads, but they do show up in specific searches. Remember, a search is always conducted before opening a new thread.

I have no problem if somebody posts in an old thread as long as it is still ontopic. It even gives us a chance to update the solution according to new LabVIEW features.

If the old thread was perfect, nobody would need to post in it, but just copy and apply the solution. 🙂 So a new post in an old thread could mean that the solution was not clear or could be improved.

Still, I would prefer a soft barrier that whenever somebody tries to post in a thread that is for example over 6 month old, it would give a warning. This could be via a popup dialog or simply with a red message above the answer box:

Warning: This thread is over 6 month old.

[RavensFan]

I find that when I do a search for a topic, the results are ranked by relevance (perhaps this is a default setting for the search?).  Something may be marked 100% and be at the top, other things have lower percentages that drop as you go down the page and further onto other pages.  More often than not, it seems the most highly ranked messages are several years old.  I can't think of any example searches I've done to demonstrate, but it seems like I get more stuff from 2004, 2005 as highly ranked than messages 2006 or 2007.  Also what complicates things is that 5 messages will be listed, but they are all messages to the same thread, so it is more work to sift through them to find a separate thread with different and perhaps more relevant information.

Perhaps others in their searches are coming across the older threads fairly often near the top, and reply to them either because they are the most relevant, they think they are relevant, or just that they don't realize how old the thread actually is.  Think of Google.  Usually what you are looking for isn't the top message, but may be lower on the page or even on page 2 or 3.  You find similar links to different web addresses cluttering up the results.  And how far along do you go on later pages until you give up.  I usually don't go any deeper than page 4 on Google.

[Dennis_Knutson]

I don't understand how the relevance is even determined. I just did a search for VISA in the LabVIEW board and the relevance ranged from 100% to 7%. A higher score seems to apply when the search item is in the message subject but there is wide variation there. What I find myself doing more often than not is that before I even starting looking at the posts, I click the Date/Time to get them sorted from newest to latest.

[Ray.R]

---

@Ravens Fan wrote:

Also what complicates things is that 5 messages will be listed, but they are all messages to the same thread, so it is more work to sift through them to find a separate thread with different and perhaps more relevant information.

---

Yeah... that's very annoying..  Especially when you do a search to help someone else..  😉

[altenbach]

Sometimes a 5year+ thread is reopened claiming to have the same problem (except the cryptic unspecified problem seems to be with a enum vs. a text ring control (quite different!) and the 5 year old solution by Greg is apparenty ignored ;)).

 http://forums.ni.com/ni/board/message?board.id=170&message.id=279915#M279915

[Gabi1]

combined with the advantages pointed by JLV, i like Altenbachs' solution to have a red warning for old threads.

but it seems the main problem is the search engine: most of recent threads and questions deals with same questions as old ones. therefore the search should identify more recent threads as somewhat more relevent.

also, it is useless to have all the posts in the same thread. if i open the thread it is very likely that i will pass trough its history.

the relevence should then try to mach the max nb of hits per threads rather than per post.

[Ray.R]

---

@Gabi1 wrote:

but it seems the main problem is the search engine: most of recent threads and questions deals with same questions as old ones. therefore the search should identify more recent threads as somewhat more relevent.  << That might be difficult.  Older answers may actually be more relevant.  Look at older answers from Dennis or myself on the topic of Serial Communication..

also, it is useless to have all the posts in the same thread. if i open the thread it is very likely that i will pass trough its history.

the relevence should then try to mach the max nb of hits per threads rather than per post.  << I'd be happy is the result pointed to only 1 instance of a particular thread.  Imagine seeing 10 pages of hits, but actually only different 10 threads...  😉  only mildly exaggerated..  🙂