Should the policy on not asking module recommendations be re-evaluated?

Question

In the page - https://drupal.stackexchange.com/help/on-topic

There is a policy on asking recommendations on modules

Questions asking for recommendations or suggestions on hosting, tools, modules, themes, distributions, books, tutorials, documentation, or other off-site resources

I am only questioning the part about 'modules, themes, distributions'

As a long time Drupal user/developer this is still one of the most common type of question that I ask before I start developing something custom.

Is there a module that does this. I would usually ask this after doing a bit of search for such a module. I have to say that a lot of modules are not quite good at getting in front of Google search results when questions are asked that should have led to them.

Now my question should really be useful for another developer who is going to be asking such a question and the answer should definitely be useful for me when I ask and should also prevent the creation of yet another unnecessary module that gets created / contributed back.

I am not sure if this stackexchange policy is actually written to benefit the community of actual users out there or comes from a wishful thinking that the software recommendations stackexchange site should become the be-all-and-end-all for software recommendations.

I find exactly 111 questions asking for recommendations for a Drupal module on that software recommendations site (Keywords "Drupal Module" search)

I find exactly 219 questions with drupal keyword.

Now if we look at how google handles the content on the site. Here is a question from that site - Which Drupal module allows donations combined with a field where donators can write a message?

This is the google search for the same - "https://www.google.com/search?q=Which+Drupal+module+allows+donations+combined+with+a+field+where+donators+can+write+a+message

The above question is not in the first 10 search results.

What is the point of throwing questions into a forum if there are not enough people who are going to come across the question and respond to it.

Also what is the point of responding to it if no new user trying to answer the question is ever going to come across the question and the answer, asked and answered earlier.

I think this policy should be re-evaluated from the perspective of whether it is actually helping people at all.

Answer 1

I am not sure if this stackexchange policy is actually written to benefit the community of actual users out there or comes from a wishful thinking that the software recommendations stackexchange site should become the be-all-and-end-all for software recommendations.

I think I'm right in saying that the policy was decided on by this community, not handed down from Stack Exchange (before my time so I can't say for certain). Either way, neither Stack Exchange nor this community have a desire/mandate/whatever to push recommendation questions to software recs with the view that that's where they should live.

Module/library/etc recommendation questions are disallowed on some sites, like this one, simply because they've shown time and time again to produce low to poor quality content. When we know there's an entire easily-identifiable category of question which does that, it only makes sense to put some barriers up, or our mission of high quality content is constantly at risk.

In my experience, these questions fall within two extremes:

1. Those like What module can I use to build a CRM?, which are hopelessly broad and not the sort of thing we're trying to address here (regardless of whether someone's looking for an OOtB solution or trying to build it from scratch)
2. A list of requirements so specific that it could only be feasibly useful to the OP, and not future visitors.

In my opinion, which again is based on my experience here (and Stack Overflow), anything in the middle of those two extremes is too ambiguous to get a decent response. New requirements will almost inevitably be identified through the life of the question, and once it does finally become answerable, it's in category 2 - only useful to the OP.

I have to say that a lot of modules are not quite good at getting in front of Google search results when questions are asked that should have led to them.

I'd agree with you there, I often find myself wishing there was an AI bot or something I could present some statements to and get a module recommendation (any students with a summer break coming up, there's a project for you!)

But I don't think that's relevant here. If a question is going to be on-topic on Drupal Answers, it has to be of a type that can inherently add value to the site. Historically, recommendation questions just haven't shown that they can do that.

Letting them back in would be a mistake in my opinion.

Answer 2

As a community of users and contributors, I believe that one of our common objectives would be to help the community of users at large. This would mean that we would want to be able to assist users, especially new users, in their journey of getting good at Drupal and getting good at using, building and ultimately contributing back to Drupal.

So let me try to put some of the points + some new here. I hope we agree on these.

1. We are trying to help the community of Drupal users and contributors
2. New users of Drupal would struggle to find contributed modules from the list of 20k+ modules that are out there that meets their need.
3. These modules do not do very well in Google searches because the module page are not necessarily search engine optimized with common phrases
4. Users will keep asking similar but variations of the same question but with different specifics
5. Even if the specifics are different and detailed, users will use short search phrases to find answers
6. When people land on a page that do not exactly answer their specifics, they might ask a variation of the question or comment on the specific question.
7. Even if a recommendation question does not give an answer to a very specific requirement of a viewer it would still give the viewer some options to start with or continue their further research.
8. New users will appreciate the help that they get from the user community.
9. The above point (I believe) would also inspire more people to contribute towards helping more future users.

Even when a question does not get a full answer or perfectly accurate answer, it providers new readers some information to start with, it provides new contributors some stub to add/edit comments/answers on. The questions and their answers can definitely be edited as more content gets added to it. Even if the question does not necessarily get one 'correct' answer, it would give some option for the user to start looking at. Over time, these questions would have different answers with different vote counts that will progressively give more information / insights to users into the future.

I understand that doing this would create some additional moderation workload because it would open up the possibility of lot of variations on the same question. However I believe that these questions can either be merged or be marked as duplicates to organize this.

If we look at the objective of a recommendation question not as the way to an 'ideal recommendation' for a 'uniquely identifiable set of requirements' but as a way to direct the user to a set of different options and variations on the set of that person's requirements then a requirement question would still make sense in stackoverflow. This would help them in their journey of finding an answer to the question they were trying answer.

I do not know of another place as useful as drupal.stackexchange.com where new users could get answers to seemingly easy questions for experienced Drupal developers. I do believe that we will be less useful for people if we are only interested in addressing programming / code related questions that are very specific. We will want these new users to get to that level where they start asking such questions.

I believe that the content on stackoverflow plays a very important role for new developers as they navigate the complex maze of Drupal. As a community of users and contributors I believe that we should try to expand on what we do to be of more help to more users and not necessarily limit ourselves.