Forums suggestion: SOLVED folders

[bertaboy]

One of the great things about Gentoo's support forums is that so many problems are solved. That is why I'm suggesting that for each of the main support forums, that each has a sub-folder for solved problems. My reasoning is simple: Rather than wade through many pages of 0 or 1 post threads about the same problem, I believe it would be more efficient if we could search through only the threads marked SOLVED and in a folder that denoted that. In order to help the admins, I also think (although it might not be the most secure method) that the starter of the thread should have the power to move the thread to the SOLVED folder.

Suggestions?

[Ian]

Cool idea, but I'm not sure it's possible to do that without a major code rewrite.

I managed a few phpBB forums for my old boy scout troop a while back, and while you can do a lot of stuff with phpBB, I'm pretty sure you must be a mod to move stuff around, even if you wrote it. If it's doable, I'm all for it, but I have a feeling it'll be too much, unless perhaps someone wrote it as a phpBB plugin (but even those were rather tedious to install and maintain, a few years ago at least).

[Maedhros]

Unfortunately, phpBB 2.0.x doesn't support subforums, so we wouldn't be able to implement this until phpBB Olympus is released. Besides, we already have the Duplicate Threads forum so that you shouldn't have to wade through lots of the same threads without finding a solution.

Moved from Gentoo Chat.

[Frodg]

One issue we have already is that very few people actually go back to their thread and change the title to [SOLVED]...

While your suggestion is OK, we still will have the issue of users not indicating that an issue is solved either in the title or by moving it to the SOLVED threads folder..

More education (and a willingness by everyone using the forums) needed.

[rhill]

and then there's users like me that don't agree with marking threads as solved in the first place.

[rouben]

I think it's a neat idea, but overkill... Simply marking the threads as [SOLVED] or [STILL NOT SOLVED] is sufficient, IMO.

[bertaboy]

I think it's a neat idea, but overkill... Simply marking the threads as [SOLVED] or [STILL NOT SOLVED] is sufficient, IMO.

I guess the gentoo philosophy of "choices" here is either working or not, I'm not sure.... Or maybe it doesn't apply to this topic. Oh well, it was just an idea.

[jdgill0]

I think it's a neat idea, but overkill... Simply marking the threads as [SOLVED] or [STILL NOT SOLVED] is sufficient, IMO.

Would it be appropriate to add something about marking threads as solved to forum Guidelines? Maybe it would help facilitate getting forum users to mark their threads once they are satisfied they have a solution to their problem?

[penguinlnx]

and then there's users like me that don't agree with marking threads as solved in the first place.

you are technically right in suggesting that no problem is ever really 'solved'. but to me that is not the purpose of a '[solved]' thread. Obviously you could have one big button marked 'solved' which points to the User Manual. But this brings up my main point: Solved threads neatly fill up the void of special cases and problems not easily or adequately covered by any manual. It is sometimes better to have categories and chapters rather than one giant 'phonebook'.

[masseya]

I'm against using [solved] tags in a subject line for two main reasons, although I was originally for it.

1) As mentioned previously, no topic is ever really [solved] in the sense that there's always potential permutations of the same problem.

2) There's no reason for them from the standpoint that each topic should already only have one thread. Thus, if you have a problem with thing XYZ in Gentoo and you find a thread about thing XYZ it is automatically relevant to you because it's either solved or in the process of being solved. If it's 'solved' for you, great. Enjoy the fix. If it's not, then at least you'll know if they have all the latest details you have on the issue and you can add that you're also experiencing issues with the topic if you're willing to help debug it.

Also, eventually we're going to start marking this type of topic a duplicate post.

It pisses me off ...
MESSAGE TAGS
Proposal In response to subject search
"Fixed problems" forum

[rhill]

if you're looking at [solved] threads from the perspective of someone scanning a forum looking to provide help who doesn't want to waste time reading threads where the problem has already been fixed.

- how many times does a solved thread end in someone else going 'that didn't work for me? any help? anyone? hello? is this thing on?'? a lot of people skip over threads marked solved. i usually do. =P
- the best way to learn how to solve problems is by knowing how existing ones were fixed.

if you're looking at [solved] threads from the perspective of someone searching for a fix.

- problems very often have multiple possible causes. what worked for someone might not work for you, which leads to the type of post in my first argument.
- the best way to fix something is by understanding why it's broken. the threads that aren't solved might provide just as much important info as those that are.

then again i've also been known to say "you haven't truely known a people until you've read their bugzilla." so i may just be a crazy old man.

[nixnut]

1) As mentioned previously, no topic is ever really [solved] in the sense that there's always potential permutations of the same problem.

Imho opinion a thread should ideally only deal with one problem of one person. The alternative is massive threads like the grub sticky in Installing Gentoo, a ton of information, but wading through dozens of pages is a drama. Threads should be short and on topic, one topic.

2) There's no reason for them from the standpoint that each topic should already only have one thread. Thus, if you have a problem with thing XYZ in Gentoo and you find a thread about thing XYZ it is automatically relevant to you because it's either solved or in the process of being solved. If it's 'solved' for you, great. Enjoy the fix. If it's not, then at least you'll know if they have all the latest details you have on the issue and you can add that you're also experiencing issues with the topic if you're willing to help debug it.

I disagree. First of all if I search the forum, I'll look at topics marked [solved] first, because they have a higher chance of containing a solution for my problem. In other words, I spend less time resolving my problem (no matter how educational you may think reading all topics may be )
Second, when I'm answering posts I know I can ignore topics marked [solved] resulting in more time to help where help is actually needed instead of where it is no longer needed.

[masseya]

Imho opinion a thread should ideally only deal with one problem of one person. The alternative is massive threads like the grub sticky in Installing Gentoo, a ton of information, but wading through dozens of pages is a drama. Threads should be short and on topic, one topic.

Actually, the alternative is hundreds of threads posted by different people with the same problem. This is why the duplicate threads forum was created. The question of where to draw the line between one problem and another cannot be as simple as labeling them different so long as they are posted by different people. Neither can it be as simple as making the scope of "what is a problem" larger by saying, for example, that a problem getting aspell to do spell checking in KMail is the same as getting your filtering rules to work again since they are both problems with KMail. If you have found a thread on a particular problem that has become a monster to read, please report it and we'll take care of it either by splitting up the thread or by moderating it differently such as summarizing things in the first post like in the Grub error collection sticky.

I disagree. First of all if I search the forum, I'll look at topics marked [solved] first, because they have a higher chance of containing a solution for my problem. In other words, I spend less time resolving my problem (no matter how educational you may think reading all topics may be )
Second, when I'm answering posts I know I can ignore topics marked [solved] resulting in more time to help where help is actually needed instead of where it is no longer needed.

If done correctly, there should be only one thread that actually deals with your problem, solved or not. Thus, when you find that thread, you'll actually be able to know if there's a solution or not without having to search further.

Furthermore, thread subjects will always be a problem, regardless of [solved] tags. If you have two threads with the subject "[solved] aspell path issues", which one is the solution to your aspell pathing problem? Also, is that decision any easier if they are both labeled "apsell path issues"? In fact, they can actually be entirely misleading. Let's assume for the moment that there are two threads one with the [solved] tag and one without. If the one with the tag doesn't address your problem and the second one does, then you have actually wasted time by going with the one that had the tag.

Another problem with thread subjects is that I don't believe they are actually used by the phpbb search engine in any form. Thus, if there were two "aspell path issues" threads and one had a [solved] tag added to it, you wouldn't be able to differentiate them by inputting "solved aspell path issues" in the search engine. If the words didn't all appear in the thread, then you would actually get results that would misleadingly suggest that there was no thread that actually solved aspell path issues. I'm sure you'll agree that this is a real kicker. It results from phpbb not really being the best solution to a technical support forum. What's really needed is a phpbb mod that would allow for threads to have a status and improved search capacity. This would almost be a sort of reverse bugzilla where instead of posting bugs to be tracked, solved and documented by the dev team people are posting bugs that would be tracked by the original poster, sovled by the community, and ultimately become a part of the technical support search engine. This is a kind of discussion for another time though.

As to your second point, there are still a great many things that people post which are actually already solved on the forums. This means that by choosing not to read threads with that tag, you're likely going to come across another posters question which has been solved already and not know that it's been solved, which would hinder your ability to help that person. Perhaps this hinderance is greater or lesser than the hinderance that you would find when you read a thread hoping to help someone only to find at the end that it's already been solved. Who knows?

The bottom line is that people are lazy and adding [solved] tags takes more work. This work would have to come from somewhere, either from the original poster, who likely will have never read this post or care about it at all, or from a moderator, who would likely be driven insane by the extra effort required, even if we didn't try to make the forums backwards compatible with this. People are always going to post dupes out of lazy or non-existant searching. People are always going to choose not to post a "yes, that did it - thanks!" message out of laziness or simply out of not caring. People are always going to have uninteresting thread subjects, even with a tag system implemented, out of either laziness or not really understanding what their problem is. [Example: KMail at some point stopped using the settings for a spell checker set in the KDE control center. This could lead a poster to have a thread with the subject "kmail can't find Aspell path" or "control center spell check path not working" which would both apply to the same problem but seem vastly different from the subjects.]

As a result of this bottom line issue, I think the best reason to not make [solved] tags a part of official policy is that we still can't even keep up with the massive amounts of duplicate threads. (which btw, was an issue that had far, far more clear cut pros and cons than this one) I think the policy for subject headings outlined in the guidelines is fine as it is:

Choose a good subject - Do not make your subject say just "Help gentoo newbie," nobody wants to hear this. Describe your problem briefly in your subject, then you can describe your problem in greater depth in the body of your article. Repeat the subject in the body if it will make things clearer. Avoid "subject says all" etc. A better subject might have been "X Windows crashes whenever I run program foo."

This seems to be the closest to what a common user would do using basic common sense regardless of whether or not they have read the guidelines. We have 30120 views for the Guidelines topic and 84726 registered users at the time of this writing...and I've 'viewed' the Guidelines at the very least 100 times. However, if you want to join the grassroots campaign to make people have people add solved tags, please feel free. Adding it to your signature (as you have done) does get the word out.

[nixnut]

Imho opinion a thread should ideally only deal with one problem of one person. The alternative is massive threads like the grub sticky in Installing Gentoo, a ton of information, but wading through dozens of pages is a drama. Threads should be short and on topic, one topic.

Actually, the alternative is hundreds of threads posted by different people with the same problem.

Which is imho opinion better than long threads and slow respons to questions. Let's face it, only one thread for one problem with a single clear solution is forum utopia. We don't have it now and we won't ever have it. The purpose of the support forums is in my opinion to provide support, preferably prompt and to the point. Currently the most workable way is to just answer support requests. Searching for dupes and reporting them is tedious. First you have to find them, then assess them (are they really duplicates?) then refer the original poster to the dupe and finally report the new thread as a dupe. Of course the orginal poster should have done that search himself, but that does not always happen, and even if he does search he is not always succesful.

This is why the duplicate threads forum was created. The question of where to draw the line between one problem and another cannot be as simple as labeling them different so long as they are posted by different people. Neither can it be as simple as making the scope of "what is a problem" larger by saying, for example, that a problem getting aspell to do spell checking in KMail is the same as getting your filtering rules to work again since they are both problems with KMail.

Exactly. So a lot support requests warrant their own thread.

Furthermore, thread subjects will always be a problem, regardless of [solved] tags. If you have two threads with the subject "[solved] aspell path issues", which one is the solution to your aspell pathing problem? Also, is that decision any easier if they are both labeled "apsell path issues"? In fact, they can actually be entirely misleading. Let's assume for the moment that there are two threads one with the [solved] tag and one without. If the one with the tag doesn't address your problem and the second one does, then you have actually wasted time by going with the one that had the tag.

The point is I do a search on terms relevant to my problem and a host of possibly related threads pop up. I still maintain that threads marked [solved] have a higher probability containing a solution for me than threads not marked as solved. That some of the threads marked solved may not be relevant goes for all threads, marked or not. So I scan the as solved marked threads first, and the others in the case the first scan yielded no solution to my problem.

Another problem with thread subjects is that I don't believe they are actually used by the phpbb search engine in any form. Thus, if there were two "aspell path issues" threads and one had a [solved] tag added to it, you wouldn't be able to differentiate them by inputting "solved aspell path issues" in the search engine.

Ah, you think I want [solved] in the topic title because you assume I use solved as a search term. I don't actually. I merely like search results having [solved] in the title. Purely a visual signal to help me decide which search results to look at first. Even if I did use solved as search term, I would still want that visual marker.

If the words didn't all appear in the thread, then you would actually get results that would misleadingly suggest that there was no thread that actually solved aspell path issues.

You lost me here. If I search for terms and there are no search results, I should try a different search with other terms. Is this still about using solved as search term? Didn't you just mention that topic titles aren't used for search? So, it's not relevant then is it...?

I'm sure you'll agree that this is a real kicker. It results from phpbb not really being the best solution to a technical support forum. What's really needed is a phpbb mod that would allow for threads to have a status and improved search capacity. This would almost be a sort of reverse bugzilla where instead of posting bugs to be tracked, solved and documented by the dev team people are posting bugs that would be tracked by the original poster, sovled by the community, and ultimately become a part of the technical support search engine. This is a kind of discussion for another time though.

Hmm yeah, I've been thinking about a indexing system that operates on an optimised periodic dump of the forums database. Seperate from the actual forums. Thread status however is just a more and systematic approach to the suggestion of marking topics. A lot more work for moderators too if you want to take such functionality seriously. But as you said another discussion for another time.

As to your second point, there are still a great many things that people post which are actually already solved on the forums. This means that by choosing not to read threads with that tag, you're likely going to come across another posters question which has been solved already and not know that it's been solved, which would hinder your ability to help that person. Perhaps this hinderance is greater or lesser than the hinderance that you would find when you read a thread hoping to help someone only to find at the end that it's already been solved. Who knows?

You again assume that I'll only read topics marked [solved]. That's not the case. I would look at marked threads first, then move on to the others. Besides that, the kind of threads that are answered many times already are often about things I don't need to look up anyway, but can answer directly from the wetware knowledge base.

The bottom line is that people are lazy and adding [solved] tags takes more work. This work would have to come from somewhere, either from the original poster, who likely will have never read this post or care about it at all, or from a moderator, who would likely be driven insane by the extra effort required

All effort on the forums in voluntary, so enforcing anything regarding marking threads is a non-issue, so workload exists only in the mind of whoever wants to accept work. Think it's too much work, don't do it, nobody is forcing you. Never mind going insane over that imaginary workload.

, even if we didn't try to make the forums backwards compatible with this.

Backward compatible?? What do you mean with that?

People are always going to post dupes out of lazy or non-existant searching. People are always going to choose not to post a "yes, that did it - thanks!" message out of laziness or simply out of not caring.

Often they won't, but some will. And I've seen in IG that more will, once you ask them (via signatures for example). I certainly agree that utopia won't happen, but I do believe things can be improved.

As a result of this bottom line issue, I think the best reason to not make [solved] tags a part of official policy is that we still can't even keep up with the massive amounts of duplicate threads.

That's a kind of nonsense argument imho. The two things are hardly related. You can have both a policy to request posters to mark their topics as solved when they feel their problem is resolved and encourage posters to search first to prevent dupes and ask to report dupes. I think that works complementary, not contrary.

However, if you want to join the grassroots campaign to make people have people add solved tags, please feel free. Adding it to your signature (as you have done) does get the word out.

Oh, I will. But some official encouragement for marking solved threads would benefit the forums in my opinion. Even just encouraging signatures like mine would.

[jdgill0]

I think having threads marked as "solved" could be beneficial. However, there are two sides to this coin, and I think both sides are correct. It really comes down to how people use the forums and what the expected behavior of the forums is to be. The problem presented by the pro-solved side seems to be that it makes it easier finding a solution and aids people that try to provide help. I can't argue against that, as I have read a number of threads only to find at the end it was solved for which I could have avoided the waste of time if the thread had been marked as solved in the first place. At the same time having threads marked as solved I think facilitates more duplicate threads.

Ideally there would be no duplicate threads presenting the same solution, but instead having only one thread with a particular solution. With fewer threads, there is less to search through and it's easier on people trying to help. Nothing new there, instead I really think the problem lies more with people not taking the time to do a proper search of the forums.

Perhaps a better idea would be to somehow enforce searching the forums prior to being allowed to start a new post. Maybe you only get the newtopic button after having done a search, for which the subject line is automatically filled in with the keywords from the search -- maybe not exactly that way, but an idea none the less. Something like this would at least force people into a search and possibly avoid them making a duplicate post just because they were too lazy to do a search in the first place or facilitate them joining an existing thread that is still looking for a solution to a problem that matches theirs. The end result being fewer needless threads.

Perhaps when you hit the newtopic button you are presented with a webpage where you have to check off (by actually visiting) the various points of help: the forum faqs, the stickies in a particular forum, the gentoo documention webpages, etc. After having verified your problem can not be helped, then you get to proceed with starting a new topic. Again, just another idea.

It's whatever everyone thinks I suppose. I personally don't think the "solved" issue is at the heart of the problem with the forums -- instead having too many duplicate threads or having too many threads that could have been solved by first doing a little research of the various help tools which are available (i.e. faqs, docs, etc). Maybe I am way off base here and have presented some stupid ideas -- go ahead and say so, I don't care.

[masseya]

The purpose of the support forums is in my opinion to provide support, preferably prompt and to the point. Currently the most workable way is to just answer support requests. Searching for dupes and reporting them is tedious.

I would say that accurate support is more important than fast and short support. Also, if you abandon threshing out the duplicate threads because it's faster to just provide an answer to a question than to go through the process of marking it as a duplicate and moving it, then the forums will become choked to death by worthless repeat posts of the same problems. This is why we moderate duplicate threads and why we encourage grouping extrodinarily similar problems in one thread.

It encourages accuracy as well, particularly in the cases where something was broken, fixed, broken a slightly different way and fixed again. Having the whole history of the issue in the single thread promotes better overall understanding of the real problem and it's associated issues. If you're concerned that you can't find a simple quick and dirty solution with long threads, just start with the first two posts and the last two posts. If nothing in that range fixes your problem or looks at all promising, then move on. Four posts really isn't all that much to read.

I still maintain that threads marked [solved] have a higher probability containing a solution for me than threads not marked as solved.

In general, I agree. I just don't think this higher probablility is really all that much higher nor is it worth the effort needed to maintain and moderate a system like this manually. The debate between [solved] tags is very similar to the user ranks debate, imho. Both of them could be made slightly more accurate with a lot more manual effort or with a phpbb subsystem that was designed to track this sort of thing. The only difference is that you can proactively campaign for a grassroots solution to the [solved] tag debate.

If the words didn't all appear in the thread, then you would actually get results that would misleadingly suggest that there was no thread that actually solved aspell path issues.

You lost me here. If I search for terms and there are no search results, I should try a different search with other terms. Is this still about using solved as search term? Didn't you just mention that topic titles aren't used for search? So, it's not relevant then is it...?

I'll answer your questions first and then rephrase slightly. Yes, this is still about using [solved] as a search term because many forums users assume that thread subjects are searched by the search engine, which seems pretty logical to me in that they are supposed to be a short description of the thread's contents or theme. Yes, I did mention that just prior to discussing it, but not everyone knows it and it's not exactly intuitive to have a [solved] tag in thread titles when it won't actually allow a user to use [solved] as a search term and get the results that they would expect. Lastly, I think it's extrodinarily relevent because before you start browsing thread subjects to see which ones might pertain to your particular problem, you do a search with the search engine. I would hate to make this some kind of official policy and all of a sudden find that it made our already lackluster search engine even harder to use because of non-obvious flaws.

If the words don't appear in the thread (i.e. not in the subject line), then you would get results that are misleading from the standpoint of the search terms entered. In the past we have had people send us feedback/questions phrased something like, "I searched for the thread called 'aspell pathing issues in KMail' using the search string 'aspell pathing issues kmail' with the box to reguire all search terms checked and it didn't say there was a thread, but I remember reading it and I eventually found it without the help of the search engine. Why didn't my search find it?" Well, when we go in and look at the thread we see something like this:

Code: Select all

+------------------------------------------------------------------------------+
| Subject: aspell pathing issues in KMail                                      |
+------------------------------------------------------------------------------+
| Message Body: I was trying to spell check an email recently and it           |
|   couldn't seem to find the spell checker in my path environment             |
|   variable.  Any thoughts?                                                   |
+------------------------------------------------------------------------------+

As I'm sure you can tell, the search terms don't appear in the message body, which is the only part that's searched by the search engine. Thus, this will never be returned if you search only for those words. A good example from the forums is performing a search for "Another fluxbox n00b" which is the title of this thread, which at the time of this writing only contains the word 'fluxbox' and would be given the same relevance on the search engine as any other thread that had that word. However, this thread and the one you are reading right now actually contain all three of those words and appear higher despite not being the more obvious result the searcher would have intended to find.

In short, I would find a grassroots campaign to get people to post a follow up saying "thanks that solved my problem" to the end of the thread more useful because it would actually get indexed by the first line of defense: the search engine.

Hmm yeah, I've been thinking about a indexing system that operates on an optimised periodic dump of the forums database. Seperate from the actual forums. Thread status however is just a more and systematic approach to the suggestion of marking topics. A lot more work for moderators too if you want to take such functionality seriously. But as you said another discussion for another time.

Yeah, a moderator can dream though, right?

You again assume that I'll only read topics marked [solved]. That's not the case. I would look at marked threads first, then move on to the others. Besides that, the kind of threads that are answered many times already are often about things I don't need to look up anyway, but can answer directly from the wetware knowledge base.

Actually, I'm not really as concerned about you as I am about people who are far newer to the forums. Any vetern poster who actually has an opinion on this subject or is even reading this post would likely be able to find what they need and more importantly is far less likely to throw their hands in the air and give up on Gentoo as a result of being frustrated with our policies. A much newer user finding out that they violated X, Y and Z official policies in their first thread along with the fact that they were likely already having some problem with Gentoo prior to coming here could turn out very differently. I'd really like to not require all that much of our users who are coming here simply to get answers. I don't mind folks on a grassroots level asking politely for people to use [solved] tags but making it official policy for them to self manage the database while getting answers to their questions becomes a little awkward, imho.

All effort on the forums in voluntary, so enforcing anything regarding marking threads is a non-issue, so workload exists only in the mind of whoever wants to accept work. Think it's too much work, don't do it, nobody is forcing you. Never mind going insane over that imaginary workload.

I whole heartedly disagree with this. There are a lot of moderators and admins that have put a lot of work into these forums and they do it because it's fun and they want to. If we were to enforce a new policy that was hideously annoying to all or most of them it would no longer become fun and we'd lose a lot of the daily work that people have come take for granted. There's nothing imaginary about the workload people take on as a voluntary responsibility. Not getting paid for something doesn't make it less meaningful. In fact, for many people that makes it more meaningful. People care deeply about the things they do with their own time, so I can very easily see it an overly annoying, mildly effective policy causing a lot of angst amongst the people who have to enforce it. Now, would all mods and admins be arm-twisted into personally being responsible for a new policy - no. And certainly it would be something we could add moderators to handle if it would provide a significant enough improvement. However, my point is don't be so eager to dismiss strain to a core group of volunteers as a "non-issue."

, even if we didn't try to make the forums backwards compatible with this.

Backward compatible?? What do you mean with that?

There are 318440 threads in the forums and 285.60 new ones made every day at the time of this writing. Would you want to be the one that sorts through these to add [solved] tags to the ones that get resolved? Heck, just think about keeping track of the future posts. That would mean logging on and checking 285 posts to see if they have proper solved tags when they get resovled which could be days, weeks, or never from the time they were posted. Anyhow, the other 300,000+ posts were what I was referencing with 'backwards compatible'. Doing that 5 days a week you'd still have to handle over a thousand posts a day to finish the job in under a year. Obviously this calls for an automated solution such as the one mentioned earlier that we could discuss in a different context.

Often they won't, but some will. And I've seen in IG that more will, once you ask them (via signatures for example). I certainly agree that utopia won't happen, but I do believe things can be improved.

Keep encouraging it. If enough people really like the practice, we'll find a way to help moderate it and make it official. I do the best I can to keep an open mind about things like this because they can change so quickly. Right now though, the interest just doesn't seem to be there. Plus, it really would help to have this kind of thing less manual and more built-in to phpbb.

You can have both a policy to request posters to mark their topics as solved when they feel their problem is resolved and encourage posters to search first to prevent dupes and ask to report dupes. I think that works complementary, not contrary.

The more bookkeeping type things we require of our users the easier it becomes for them to just blow it all off and get to the point of what they really want at that moment.

[nixnut]

The purpose of the support forums is in my opinion to provide support, preferably prompt and to the point. Currently the most workable way is to just answer support requests. Searching for dupes and reporting them is tedious.

I would say that accurate support is more important than fast and short support.

marketing/crm theory would disagree with you. Customers want to have attention asap. Giving that to them (i.e. someone willing to listen to them, take them seriously) is more important than the also important accuracy. Besides that, often you need to ask for more information first before you can discern the actual nature of the problem. Don't get me wrong, accurate answers are very important, but for new posters a prompt respons is often even a bit more important.

Also, if you abandon threshing out the duplicate threads because it's faster to just provide an answer to a question than to go through the process of marking it as a duplicate and moving it, then the forums will become choked to death by worthless repeat posts of the same problems. This is why we moderate duplicate threads and why we encourage grouping extrodinarily similar problems in one thread.

Good point, but is it working? Not very well, is it. And as mentioned earlier, it's often complicated by first having to ask for more information before you can even properly assess if the problem is a dupe and point out the thread it's a dupe of.

It encourages accuracy as well, particularly in the cases where something was broken, fixed, broken a slightly different way and fixed again. Having the whole history of the issue in the single thread promotes better overall understanding of the real problem and it's associated issues.

I disagree. That leads to unwieldy threads. Ok, one solution would be to edit such threads into a more readable form, compiled into a sort of Gentoo Troubleshooting Guide and refer to the relevant part of that document.

If you're concerned that you can't find a simple quick and dirty solution with long threads, just start with the first two posts and the last two posts. If nothing in that range fixes your problem or looks at all promising, then move on. Four posts really isn't all that much to read.

mmm, I'll try that for a bit. I doubt that I find that satisfactory, but only one way to find out.

Lastly, I think it's extrodinarily relevent because before you start browsing thread subjects to see which ones might pertain to your particular problem, you do a search with the search engine. I would hate to make this some kind of official policy and all of a sudden find that it made our already lackluster search engine even harder to use because of non-obvious flaws.

Ok, I get your point. It's about the expectations of newer users concerning the search feature. I agree that if we would like to have the benefits of having threads marked solved, they should outweigh the disadvantages. Perhaps some expectation management (i.e. user education) might help in this regard. Getting this across to the new user then becomes the next problem

In short, I would find a grassroots campaign to get people to post a follow up saying "thanks that solved my problem" to the end of the thread more useful because it would actually get indexed by the first line of defense: the search engine.

Yes, for searching on solved, not for the visual cue. Having topic titles searched would be even better though. Hmmm, maybe I should dig into the forum code again, look for something workable (for a 10GB forum that is).

Any vetern poster who actually has an opinion on this subject or is even reading this post would likely be able to find what they need and more importantly is far less likely to throw their hands in the air and give up on Gentoo as a result of being frustrated with our policies. A much newer user finding out that they violated X, Y and Z official policies in their first thread along with the fact that they were likely already having some problem with Gentoo prior to coming here could turn out very differently. I'd really like to not require all that much of our users who are coming here simply to get answers. I don't mind folks on a grassroots level asking politely for people to use [solved] tags but making it official policy for them to self manage the database while getting answers to their questions becomes a little awkward, imho.

Oh, absolutely. I guess my idea of such a policy is different from yours. In my view the policy would be to encourage users to mark threads solved, not force them. That would never work.

All effort on the forums in voluntary, so enforcing anything regarding marking threads is a non-issue, so workload exists only in the mind of whoever wants to accept work. Think it's too much work, don't do it, nobody is forcing you. Never mind going insane over that imaginary workload.

I whole heartedly disagree with this. There are a lot of moderators and admins that have put a lot of work into these forums and they do it because it's fun and they want to. If we were to enforce a new policy that was hideously annoying to all or most of them it would no longer become fun and we'd lose a lot of the daily work that people have come take for granted. There's nothing imaginary about the workload people take on as a voluntary responsibility. Not getting paid for something doesn't make it less meaningful. In fact, for many people that makes it more meaningful. People care deeply about the things they do with their own time, so I can very easily see it an overly annoying, mildly effective policy causing a lot of angst amongst the people who have to enforce it. Now, would all mods and admins be arm-twisted into personally being responsible for a new policy - no. And certainly it would be something we could add moderators to handle if it would provide a significant enough improvement. However, my point is don't be so eager to dismiss strain to a core group of volunteers as a "non-issue."

mmm, I guess I reaaally missed getting my point across here. I certainly don't mean to imply that moderators and admins are not taking their responsibilities seriously. Nor do I mean to imply they don't experience a workload. My apologies if I offended anyone. But, well, even moderators should keep a perspective on things. They are not forced to moderate, it is a choice. And thus any workload they are (thankfully) willing to accept, is only a workload for as much they wish to accept. At any moment they can say "sorry, no more right now". Well, at least that's how I see it. Any kind of work they do is a gift (and gladly recieved with my sincere thanks). And as I mentioned above, I would propose a policy of encouragement, not a policy of enforcement. So it would be a non-issue, nothing to enforce really. Simply requesting (not requiring ofcourse) the highly visible forum regulars to add a sig similar to mine could already make a big difference. Such a request by a moderator carries quite a bit of weight I think, at least in my mind.

Anyhow, the other 300,000+ posts were what I was referencing with 'backwards compatible'. Doing that 5 days a week you'd still have to handle over a thousand posts a day to finish the job in under a year. Obviously this calls for an automated solution such as the one mentioned earlier that we could discuss in a different context.

Ah, got it. Well, I would disregard the 300k "old" threads. For one processing them manually is unworkable, processing them automatically is not an option yet either. Apart from that, a lot of those threads will become obsolete sooner or later. So, not really worth the effort imho.

The more bookkeeping type things we require of our users the easier it becomes for them to just blow it all off and get to the point of what they really want at that moment.

Right, so it boils down to relation management, meaning communication, once more. Not enforcing, but requesting users to behave like we want them to and explain why that is good for both them and the other forum users. Still, easier said then done, I know.

[masseya]

marketing/crm theory would disagree with you. Customers want to have attention asap. Giving that to them (i.e. someone willing to listen to them, take them seriously) is more important than the also important accuracy. Besides that, often you need to ask for more information first before you can discern the actual nature of the problem. Don't get me wrong, accurate answers are very important, but for new posters a prompt respons is often even a bit more important.

This is a good point, but I think too many of Gentoo's users are somewhat technologically savy and uninterested in being 'handled' in any way. If we had a large number of folks coming to Gentoo with very little computer experience period, then I think this would be more important. (Sidenote: It would also be easier because extremely new people to computers usually are asking questions that are much easier for the average Gentoo forums user to answer with both accuracy and speed.)

Ok, one solution would be to edit such threads into a more readable form, compiled into a sort of Gentoo Troubleshooting Guide and refer to the relevant part of that document.

That was the original intention of the FAQ Forum, for better or worse.

Ok, I get your point. It's about the expectations of newer users concerning the search feature. I agree that if we would like to have the benefits of having threads marked solved, they should outweigh the disadvantages. Perhaps some expectation management (i.e. user education) might help in this regard. Getting this across to the new user then becomes the next problem

Right. And what a fun problem that is.

Yes, for searching on solved, not for the visual cue. Having topic titles searched would be even better though. Hmmm, maybe I should dig into the forum code again, look for something workable (for a 10GB forum that is).

Another nice visual cue would be a relevancy rating returned with each search. Even something as simple as giving the number of search terms that were found in the thread would be helpful imho. Basically, all these issues can be summarized in the multitude of threads about how bad the search engine is. It's really the main problem you're having here as well in my opinion. Being able to perform a search and get the results you want or need shouldn't require anything more than a good search engine. (i.e. We shouldn't have to put [solved] tags in thread titles manually.)

But, well, even moderators should keep a perspective on things. They are not forced to moderate, it is a choice. And thus any workload they are (thankfully) willing to accept, is only a workload for as much they wish to accept. At any moment they can say "sorry, no more right now". Well, at least that's how I see it. Any kind of work they do is a gift (and gladly recieved with my sincere thanks). And as I mentioned above, I would propose a policy of encouragement, not a policy of enforcement. So it would be a non-issue, nothing to enforce really.

To some extent this is true, and we have had people walk away or take long breaks (myself included) after becoming frustrated with the various aspects of their voluntarily accepted positions. The problem I have is that I can't see a policy of encouragement working. It sounds too much like an oxymoron. "Optional policy" just doesn't compute for me. The best I can understand it is that it would be where I, as an admin, would be required to say you, as a user, can opt to do this [solved] tag thing. Personally, I don't think it provides nearly enough benefit to make such a request of a user, but if it becomes official policy, then I would be required to say that it's our policy to encourage the practice. Furthermore, it's hard for a mod/admin to put something like that in their signature without implying that it's an official policy, mostly because of the ambiguity of the voice of the signature. It makes a concerned reader ask, 'Is that statement coming from "them, the moderators" or "him, the fellow forums user"'?

Right, so it boils down to relation management, meaning communication, once more. Not enforcing, but requesting users to behave like we want them to and explain why that is good for both them and the other forum users. Still, easier said then done, I know.

Right, and I don't have a problem with anyone putting something in their signature to support this so long as it is not worded as if it were official policy or some other form of requirement. I think the best official solution to this would be a better internal search engine. Actually, that would be a good solution to a lot of things.

[nixnut]

If we had a large number of folks coming to Gentoo with very little computer experience period, then I think this would be more important. (Sidenote: It would also be easier because extremely new people to computers usually are asking questions that are much easier for the average Gentoo forums user to answer with both accuracy and speed.)

Well, that's more or less my experience in IG, more and more users new to linux, never mind gentoo, with sometimes the barest grasp of concepts that are becoming (or have become) second nature to us. And a lot of these questions are easy to answer, easier to answer than to find the original "dupe" in fact, a point made previously.

Ok, one solution would be to edit such threads into a more readable form, compiled into a sort of Gentoo Troubleshooting Guide and refer to the relevant part of that document.

That was the original intention of the FAQ Forum, for better or worse.

I realise that. I guess I'm a bit too focused on the install since install related questions are what I spend most of attention on. So I was thinking of a document that follows the structure and steps of the installation handbook closely. But, that could be done in the FAQ forum too, of course, but the handbook is quite readably in links for example, while the forums are a lot harder to read in a text based browser.

Another nice visual cue would be a relevancy rating returned with each search. Even something as simple as giving the number of search terms that were found in the thread would be helpful imho. Basically, all these issues can be summarized in the multitude of threads about how bad the search engine is.

Good points, but with a manual [solved] marking, you can assume that there's high probability that the thread contains a solution to the problem it deals with. With a calculated relevancy rating, that probability is lower. But I agree that such a rating would be useful.

The problem I have is that I can't see a policy of encouragement working. It sounds too much like an oxymoron. "Optional policy" just doesn't compute for me. The best I can understand it is that it would be where I, as an admin, would be required to say you, as a user, can opt to do this [solved] tag thing. Personally, I don't think it provides nearly enough benefit to make such a request of a user, but if it becomes official policy, then I would be required to say that it's our policy to encourage the practice.

Ok, well, that's something to decide amongst yourselves, admins and moderators. If you like the practice, you can promote the idea. Like with the "Adopt an Unanswered Post" thing. Is that official policy or just an admin/moderator with a good idea? I haven't been asked by a moderator to add that link to my sig, so it wouldn't be policy then, right? Anyway, the result is that I see quite a bit of users (and at least one admin) with a sig that links to unanswered posts. It is that kind of actions I'm thinking of. As I said before, ideas of and suggestions from admins and moderators carry a lot of weight.

Furthermore, it's hard for a mod/admin to put something like that in their signature without implying that it's an official policy, mostly because of the ambiguity of the voice of the signature. It makes a concerned reader ask, 'Is that statement coming from "them, the moderators" or "him, the fellow forums user"'?

I see your point. I actually wasn't thinking of moderators having such a sig. That would be entirely up to the individual moderator. Though a sig grass root movement would have more impact if some moderators took it up as well. But if there were consensus about accepting a [solved] encouragement policy, the conflict of personal versus official opinion would no longer exist.

I think the best official solution to this would be a better internal search engine. Actually, that would be a good solution to a lot of things.

Agreed on the search engine improvement.

Concerning the other aspects, I agree you have some valid points, but I think those are mostly outweighed by the benefits as I see them or side issues (related, but less relevant). But one can only agree to disagree on personal opinions when disagreeing on them

[Tamsco]

I think people should just add a [Solved] in their title, it's just as much work for an admin to add that to the title as it is to move the thread.

I do however think solved/unsolved should be a thread attribute (like sticky) so that people should be able to search through solved posts only.

Mike

Edit: Moreover the problem is that one person asks a question then someone else will come in and say "Hey I have a question like that" If the thread starter's question get's answered, does it get moved to Solved if the next guys question is still outstanding? Does it get marked solved? Or do we just say 1 question 1 thread.

[hq4ever]

This for several reason's :

1. It's forces the poster to EDIT his original post, a topic recently emerged on this forum.
2. It causes the poster (some times) to delete a word or two from his initial [Subject] line. those possibly removing meaningful key words that could assist the next person in finding the topic posted.
3. It's a waste of time, unnecessary inconvenience, and a redundant task typing each and every time [SOLVED].

What I offer is simple, yet effective:

Instead of demanding the user to edit his messages, allow the original poster to click a button, that would in turn execute a SQL, that would mark the post in the forum database as solved.
For the user, it a mere "1 Click" action, on something that looks like Click here to set your problem as SOLVED

Users visiting the forum will know that the message has been marked as solved by a small Green "v" next to the post title.
Where to the forum members not using X (I assume there's quite a few of those in here ) the alternative text for the V image file will be "[SOL]" so that for them, knowing the status won't be a problem either.

EDIT: those, the post's will look like this:
normal : http://img217.echo.cx/img217/5181/gentoopost1ti.jpg
solved: http://img214.echo.cx/img214/8523/gentoosolved2xj.jpg
*V image taken from : http://w3.rz-berlin.mpg.de/~wm/wm3a.html

****

If the solved posts would be marked in the forum db, it would be a very useful yet quite easy feature to implement the idea posted here : Forums suggestion: SOLVED folders and even simpler: Allow the users to search only SOLVED topics.

****

I don't have much knowledge of the inner working of phpbb but I do know that it's a plug in enabled forum. For this I think that implementing a "1 Click" button + Adding 1 more column to the thread's table in the db, should not be a problem.
At least I hope so (:.

Thank you.

--
Cheers,
Maxim Vexler (hq4ever).

Do u GNU ?

[pjp]

SOLVED shouldn't be used period. A thread that has been solved for one person, may not solve it for another. It creates a mess.

Merged previous.

[hq4ever]

Why not ?

After all [SOLVED] doesn't lock the thread from posting.

Besides on most cases a problem being solved for one person is exactly the same problem anoter is trying to find a solution for, this as I see it is the main reason why [SOLVED] was intruduced in the first place.

Apart from all that, if a user searches for a problem, find's one with the title [SOLVED] but the solution there is not solving his problem, what stopping him from opening a new thread ?

[pjp]

Why not ?

Read my entire post, not just the first sentence.

And also, when people post "solved" enough, it'll be added to the stopwords list as useless, and not be searchable.

[hq4ever]

Read my entire post

Please read mine.

"when people post "solved" enough, it'll be added to the stopwords list as useless, and not be searchable."

Exactly where my suggustion fits, don't make people post "solved" allow them to define in the search page to search only with in "solved" topics.
Besides, "solved" saves the time of people that come to help other users, a person seeing "V" (as I suggested) or "solved" or any other indicator that the post has been answered will not need to spend time reading the post only to find out that the poster already been helped.

Save [Solved] !