Network blog just had a post about "guessing game" type questions and issues with them.

Should we take this as a hint and consider getting rid of theme and plugin recommendation questions?

This came up several times and I think current points to consider are:

  • some of recommendation questions might be good, but most of them are not.

  • we've been handling them from the start and it is explicitly in scope per current FAQ.

  • it might be easy to decide to get rid of them, but it will be hard to get people to understand how are those not fitting the scope of WordPress site.

share
add comment

7 Answers 7

I think the vast majority of the problem questions are because they are to localized, for example, "I need a plugin for a dentist reservation system".

But I also think some questions about plugins are useful so we need some guidelines.

It's funny because I once asked a plugin question, and it has over 3k views and is somewhat helpful.

share
2  
There may be some terminology hair-splitting involved here, but in the question you asked, while you did specifically ask for a Plugin recommendation, what you were really asking for (and what the accepted answer provided) was an implementation method to solve a specific issue. I see such a question as perfectly valid. –  Chip Bennett Apr 12 '12 at 15:59
    
That is true, I guess a good line would be, does it create an actual discussion instead of just a link, which kaiser addressed above. –  Wyck Apr 12 '12 at 16:02
add comment

We have to define some very strict and clear criteria for on topic questions.

Close it, if:

  • a search engine can answer the question,
  • it is about the usage of a plugin or a theme,
  • the theme or plugin should just copy some (more or less) famous web site.

These questions are boring. They add just noise most of the time. Especially the theme recommendation questions are usually good close candidates.

share
    
+1. I'm pretty certain I know the question that prompted this answer; it involved usage support for a specific Plugin. I struggle to see the value to the WPSE community to have such a question on the site, as well as the benefit to the end user to ask such a question at WPSE rather than asking the Plugin developer directly. –  Chip Bennett Apr 11 '12 at 14:29
    
So, would you consider this question as falling under usage of the Plugin? I'm leaning that way, but wanted some feedback before commenting as such in the question. –  Chip Bennett Apr 12 '12 at 12:42
    
@ChipBennett It seems to need handwritten code, and it may help other readers too. I'd say it is on topic. –  toscho Apr 12 '12 at 17:27
    
I can imagine good questions about a class of plugins (such as Photo Galleries) where the user lists plugins they've tried, lays out the cross-cutting issues with the ones they've tried, and asks for recommendations for a plugin that avoids said problem. Then the issue is not really about a specific plugin but more about a frequent problem with a class of plugins. –  mrwweb May 2 '12 at 18:50
    
@mrwweb Can you name one example? –  toscho May 2 '12 at 18:59
    
I had this question in mind: wordpress.stackexchange.com/questions/48747/… –  mrwweb May 2 '12 at 19:05
add comment

I trust the plugin recommendations of people here on WPSE because they tend to come from a developer's perspective, looking both at functionality and the various costs of implementing a plugin vs rolling your own function. I can't think of anywhere else I'd find this breadth of knowledge all in one place. I hope recommendations can stay.

share
add comment

"Guessing game questions don’t help others"

Plugin recommendations do help others - they are searchable (think: "What plugin can I use for X situation" - that is super searchable for future Google searchers in a similar situation).

We cannot fix the problem of out-of-date plugins by making rules that discourage the referencing of plugins. We need to treat the problem (out-of-date plugins) not the symptom (links to said plugins).

I do agree (with Chip) that plugin recommendations that are too localized or are not constructive should be avoided. Coming up with that decision algorithm is difficult though. Plugin recommendations should stay.

share
    
» We need to treat the problem (out-of-date plugins) not the symptom (links to said plugins).« If the problem would be solved and those plugins removed, what then? We'd have hundreds of answers with abandoned links. How would you go and then fix that? –  kaiser Mar 5 '12 at 0:43
add comment

I agree that the majority of theme or plugin recommendation questions fall in the category of bad questions, But since a few are actually good and even if too localized to a user requirements, generate some nice answers of combining a few plugins or customizing a plugin to answer the requirements.

I myself wrote a few plugins based on these questions and I have to admit that when i need a plugin for something i come here and search for it.

As for plugins being outdated, the WordPress team said something about removing plugins that haven't been updated in the last year or so, don't know when that actually happen but its a turn in the right direction.

I really think that people wont understand how are those not fitting the scope of WordPress site and for that alone they should stay in.

share
2  
Old plugins in repository now are not shown in search, but they are still there. –  Rarst Feb 29 '12 at 23:15
add comment

We explicitly made such questions in-scope; however, finding such a Theme/Plugin recommendation question that isn't either too localized (based on unique set of user requirements) or else not constructive (tending to solicit polling, opinions, extended discussion, etc.) is extremely difficult.

Personally, I think they're not a good fit for WPSE, unless we can better-define what it is about such questions that would cause them to be useful to the purpose of WPSE.

share
    
I second @Chip - apart from relatively often being too localized or non constructive, this type of question tends to create a huge amount of dupes or very similar questions. –  Johannes Pille May 10 '12 at 13:33
add comment

YES!

WPSE point of view

The majority of plugins is outdated.

Plugin dev point of view

You get no profit (unless you take non-calm, pushing and bold users and call them "worth it"), so most plugins stop their dev cycle, stay in the repo 1) and get slowely outdated 2).

1) Official repo has no clean-up crew, so they simply stay.

2) Some get outdated fast (Ex. Image Uploader > Plupload).

Plus

Users don't take a look at the date/version, the Q was asked and they simply download (the yet - 8 years later - available) plugin and use it.

Problem

Then we have to clean up the mess and have a few new Qs:

Typical Q/A

  1. »I got a blank page! Help!« »How do I activate debug?«

  2. »I got dozens of errors! Help!« »Turn off all plugins, switch to blablabla…«

  3. »My Plugin isn't working« »You can fix it with this or that" »Where do I place this code?" »Well… sigh … you're better off searching for a new plugin.«

  4. »I need a Plugin for something incredibly kool that can do [insert task that you never imagined]" »Try this or that

  5. BACK TO N°1… *)

*) Which means we just double the users that a) backtrace it and try to fix the longely outdated plugin, which brings up more annoying Qs and b) more users that have problems when the new plugin recommendation is outdated in some month. The snake bites her tail…

share
add comment

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .