Print

[cgc0202]

Hi,

First, I must commend the Coppermine core group for creating a very good software and also all those who have develop mods/plug ins to make the product achieve even more useful applications for everyone.  The browser-based installation (and instructions) were well done also so that I have not encountered any problem with installing the software -- within a few minutes.

It is also the case that there are some users who are too lazy to read and would take the easy way -- and take the attitude: "I have encountered this problem, solve it for me."  What was more aggravating is when they do not even provide enough information.

The aforementioned reason might be the reason why many veterans who have been spending their precious time helping others could be quite terse (although some I should say even condescending) when they see a post that they knew have been answered before.

"Read the manual!"  "Search the forum!", etc. were the cryptic responses.

But, is it really that simple?  I like to think that many users do take their time to read and search.  Take this for example, I have search using:

"Alphabetical user gallery browsing" or variations, and the  response:

"Your query was not specific enough. Try using larger words, or less common phrases." 

This of course is not true because that is in fact the title of a thread:

http://forum.coppermine-gallery.net/index.php?topic=25556.msg117575#msg117575

The reason for that is because the search functions of forum boards or many open source softwares are rather primitive compared to Google.   I have been in both sides of the teach-learn situation.  There is something to be said when to provide information and when to tell someone to read more and search more.  Moreover, I will not second-guess the position of the moderators and veterans of the forum who decide to stick to certain attitude.

Clearly though, there must be some alternatives. I was confronted with this issue when I was trying to learn CPMFetch, and it gave me an idea.  [I must state here though that Vuud has been very good in trying to explain his Mod, and is the reason why I have progressed much faster in trying to understand his Mod.]

It is a very popular Mod, and there are several threads with one now 42 pages  long.  Anyway, in this particular issue, an Italian encountered some problem, and was told that his question has already been answered -- he should read the posts.  He explained that he did not understand English well, and it would take him more than a week just to read the whole thing.  There is also no assurance that he would indeed understand the solution, if he is not that tech-savvie.

Fortunately, one of the posters was kind enough to share the specific posts (burried somewhere in page 30 or therabouts). With Vuud providing a  helping hand, the issue was resolved eventually.  Reading the dated  exchange itself saved me from having to go through the entire  40 pages.  It solved the error I encountered (even before I posted the issue) but led to another error that turned out to be  an old "error" also.  But, as it turned out the old "error" I encountered required a different solution altogether. Thanks to the patience of Vuud, we were able to resolve the issue, eventually.

The reason why I related this is to demonstrate that it is not practical (almost sadistic) to tell someone that it has been resolved without even providing any clue where to look (or just to say, it is in the Manual).  The fact is, most threads are dead end -- and would be a great waste of time to pore through them.

Together, we could help each other.  We could create some sort of links repository of specific parts of forum threads that lead to solutions to specific issues that are frequently encountered.  This could be prepared in the form of "Frequently Asked Questions" but more detailed -- preferably one that is indexed using easily recognized terms or code words.  Those who may be inclined could prepare detailed tutorials, just like Vuud has done with his Mod.

Since I am myself going to benefit from such a repository of useful information, I would be interested to help.  I am even willing to set up wiki page in my own website for this purpose, if there are others who would volunteer to help.

Just an idea.  Thanks.

cgc0202

[Paver]

I agree you could try to consolidate the information on the forum with a links repository, but I'm not sure how easy it would be to condense the information appropriately so that the actual problem is easily searched on such a repository and in a timely way so you can keep up with the flood of posts.

I also agree that the search function could use some improvements.  Besides the message you noted (which I agree should be investigated and addressed if possible), I would suggest adding a "search by tag" field so you can search only "Solved" topics and leave out the "Invalid" or "Closed" tags.  This would easily cut down on the results if supporters use these tags appropriately (and as far as I have seen they do now - although I'm personally a little iffy on the "closed" tag for questions whose answers require a mod/plugin that has not been written by anyone yet as exemplified here).

The main reason for a terse "read the manual" or "search the forum" response is that supporters want to support all the posts as a first step, and if the supporter knows the answer is in the manual or has been mentioned in a thread before, the terse response is an accurate one.  If another supporter has enough time or specific memory info to add more details, he/she will.  The supporters would not be able to handle the current flood of posts without giving such terse - and I contend accurate - answers where appropriate.

Overall, I agree with your intentions.  But I don't think it would be possible to have a links repository accurate enough and timely enough to be useful.  Searching the forum directly should be the best way to do so.  I do think more FAQs could be added based on questions & answers on the forum and that the FAQ could be served through a searchable database in addition to the docs page.

edit: An alternative solution to address the specific issue you mentioned of a topic with a large post count is to have sub-boards for mods/plugins.  That way support issues would have separate threads in the appropriate sub-board for the specific mod.  CPG-Contrib has started such a forum to do so for plugins.  I don't know if they would consider adding the most popular mods as well.

[cgc0202]

Yes, I understand your side of the issue Paver.  I don't know how you guys even find the time to answer stuff.  I cannot even respond to my emails on time.  Sometimes, people could be so demanding.  Once I had a personal website that had a following from all over the world.  People asked stuff and expected me to answer promptly and accused me of being evasive if I decide not to answer specific questions.  For these reasons, I understand the cryptic and terse response as noted in my initial post. 

There is no perfect solution, the one I proposed and the ones you in turn proposed will all help.  Many of the ones you proposed are doable immediately.  Perhaps, the one I proposed is most difficult to implement and realize because it would require collaboration on all sides.  Realisticaly, my prior experience was that in general many people help only if they think it would benefit their current needs.  Even with this realization, it is still good to propose the idea -- others may participate.

There is a well-featured CMS program that I am trying to use right now, but it is very frustrating because  there is literally no instruction manual at all on key features that I found important, and not enough clear response from the forum to make them work.

cgc0202

[Paver]

Hmm... the example you gave for a search phrase not returning a topic with the exact title is troubling.  It does return the topic if you search for "alphabetical user gallery".  For some reason using the word "browsing" kills that topic as being a valid search result. 

I tried using Google to search the same topic, but it also did not return a result.  Google of course does not have the latest posts (yet), but it should have the topic you mentioned.  But in general it seems like Google is not indexing all the posts on the forum, for some reason.  For example, this search should return multiple topics from the forum but it returns none - only a page from the docs.

[Joachim Müller]

Re forum search: the error message you got didn't appear untill 6 months ago. Imo it's related to the fact that the index of the SMF-driven forum db tables got huge. If there are too many hits, the internal search considers a string trivial and asks you to be more specific. Not sure what to recommend though. I'm shying away from adding another mod to SMF to improve this - it makes upgrading so hard. Another option would be to drop the policy not to delete threads: someone would have to go through older threads and get rid of the invalid/double threads, which would reduce the index. Huge task...

Re FAQ: everybody is welcome to suggest further FAQ to be taken into account on the FAQ pages that already exist.

Re harsh replies: happens on many boards, see http://www.slash7.com/pages/vampires It usually helps if there are more supporters. Volunteers wanted!

[Abbas Ali]

How about starting a wiki and slowly dumping the old threads?

[cgc0202]

Thanks Paver, Gaugau and Abbas,

But in general it seems like Google is not indexing all the posts on the forum, for some reason.  For example, this search should return multiple topics from the forum but it returns none - only a page from the docs.

Yeah, I tested this error many times the past few months here, with  different phrases using different posts that I knew existed, just to make sure.  The one I included in the initial post was just an example.

I have not used the Google search when I did the search tests on this forum.  I assume that it  was the internal search associated with SMF???  I thought that would be more focused. I am aware though of a number of Open source developers stating that they have difficulty with developing a good search mod for their own software.  Others stated that they do not use certain searches from Open source softwares because they are not Bolean-based and usually are not as good as those of Google.  So, I am not sure that adding another mod would really solve the limitations of the search string found in SMF.

Re forum search: the error message you got didn't appear untill 6 months ago. Imo it's related to the fact that the index of the SMF-driven forum db tables got huge. If there are too many hits, the internal search considers a string trivial and asks you to be more specific. Not sure what to recommend though.

In my own experience, when I search with Google I try to place as many relevant terms to narrow the choices, i.e., to be more specific. 

I was getting the reverse here in the Coppermine Forum.  To get some hits in the Forum search,  I have to reduce the terms I place in the search string -- but then, usually most of  the search results were not at all relevant to what  I wanted to look for.  At least, this has been my general experience with Forum searches.  [It is for this reason that I am an advocate of alphabetical listings, i.e., indices in directory form.  If there are good enough key words in such an A-Z listing directory index, people can find what they are looking for more intuitively through association much better than a search -- especially if you do not know what you are searching for very specifically. Again, that is from my own experience.]

Re FAQ: everybody is welcome to suggest further FAQ to be taken into account on the FAQ pages that already exist.

Yes, FAQs would help  very much, if the FAQs were asking direct questions in a manner relevant to the what a person is looking for.  I had this situation here also that what I was looking for was indeed already "solved" and answered through an FAQ, as I found out later.  I encountered the FAQ by accident, and indirectly through another post thread here.  The post was relevant to what I wanted answered.  The FAQ where it was answered would be far from a question I am likely to ask, in regard my inquiry.

Re harsh replies: happens on many boards, see http://www.slash7.com/pages/vampires It usually helps if there are more supporters. Volunteers wanted!

That is the reason why I posted this thread Gaugau.  Frankly, I get easily stymied in a very terse situation. I would be willing to help if I know what I am talking about, but I do not have background in programming -- even the simplest php scripts that contain more than one statement confounds me.  But  I see this in every tech forum  I have participated in: the newbies or non-tech savvie people like me asking the same questions over and over again, because what they were looking for -- at least those who would take the time to look -- was buried  usually in the avalanche of the forum threads .  That is the inherent limitation of forums.

While I am not be tech-savvie, what I am quite familiar with at least in tech forums like this, is that I understand your frustration (repeating yourself to newbies and non-tech savvie people).  I also understand quite fully well, the side of people who are asking questions and are trying desperately to find solutions to their questions, but do not know where to begin looking for stuff.

How about starting a wiki and slowly dumping the old threads?

That was also my suggestion in the original post Abbas, in regard to your suggestion about wiki.  The solution is not something that could be done by a single magic bullet and not by a single person -- everyone must help if this is to have some resolution. In that sense, the collaborative environment of a wiki helps.

Wiki does help, if prepared properly -- incorporating  the FAQs suggested by GauGau and the other suggestions by Paver, like including forum topics that have been resolved.  As Paver pointed out, the term resolved would be those that answered the original question.  Those marked solved because they would be included in the future CPGv1.5 are not truly resolved -- but that information is as important too.  It would imply -- do not expect any solution any time soon.

I am willing to volunteer help shape such a wiki -- from the perspective of the user and in a more free-wheeling and "relax" environment.  I have some ideas on how to go about this but it will take me a novel to explain it.  So, what I will do is prepare a wiki pilot in a subdomain of my website.  I will integrate some of the ideas presented here so far, as a start.  Everyone may participate -- with various degrees of permissions -- to suggest or auto-submit other suggestions, answers or edit, depending on their permissions level.  Also, the wiki I have in mind will not be anonymous wiki (like wikipedia) to minimize wiki-related vandalism.

If many will participate,  there is no reason why such a wiki should not work, at least in alleviating the issues presented here.  If it does help, in any way, once the wiki pilot matures then then the core group of Coppermine may opt to import all the stuff relevant to CPQ to create a more specific wiki page for CPG.

What do you think?

cgc0202

[Joachim Müller]

Starting a wiki has been discussed many times on the dev-only board. All who suggested starting it failed to implement one, because it lies in the nature of a wiki as collaboration form: edit-wars as a result is only one of the things that you have to be prepared for. Imo, starting a wiki has become a fashion for everything and everyone. To me, it's a buzzword - the technology will only work for certain communities and topics. In the nineties, everybody needed a guestbook. Look who still uses one. Nowadays, everybody needs a wiki, preferably with AJAX technologies (yet another buzzword). Imo, devs should do what they do best: develop apps. When an app like Coppermine becomes more and more popular, it's only natural that people who don't fulfill the minimum requirements (in terms of personal skills and being not tech-savy enough) to install, run and tweak it will still pop up on the support mechanism (wether it's a maling list, forum, FAQ list, wiki, helpdesk, whatever) and clutter the channels there.

For me as a project manager, it's not a question if there should be a wiki or not: if there are people who are both capable (in terms of skills) and willing to set it up and maintain it, that's fine. It's just fine as well if people contribute in another way (i.e. by writing additional docs, creating tutorials). I'll continue to work on this project for as long as it's fun for me and I'll stop working on it isn't any longer. But: my workload for the project is already high enough, so I'm not ready to look into setting up a wiki.

The speed of development of cpg1.5.x has decreased, yet there are still a lot of possible features to implement that are on the to-do list. If devs spend even more time on user support, they can spend less time on development - that's a fact.

So basically, I'm willing to "sacrifice" some non-tech-savy users who might leave in frustration because they're not capable to use the existing support mechanisms in favor of developing new versions for the benefit of the whole community. For me, Coppermine is still a project aimed at geeks (people who know how to run their website, who know at least HTML well enough to apply custom modifications). People who just barely know their way around on their webhost's control panel and expect to find a "one-click install" app are welcome to try Coppermine out, but if they fail to get it running and are reluctant to learn something during the process of getting their issue solved, I'm not worried too much.

Joachim