Proposal: Deploy a "support.drupal.org" site which would provide quality support-answers

Sure, we can set up a dev site for people who want to work on this. When they've got something to show, we can then decide how to go forward.

What are the features needed on this support site that our forums don't offer?

In that thread (which I sort of read as I went), I think I heard:
- Better searching
- Question/answer format

Is that it?

The killer app here is for voting to be applied to answers, a la Stack Overflow. Especially one which will create an automatic meritocracy.

Plus, autosorting a la Stack Overflow (mentioned by Sam Tressler).

For those of us not as familar with Stack Overflow, can you explain what the auto-sorting feature would do?

I'm all for this, but the other day I was in #drupal and drumm (Neil) had people testing out some feature enhancements on the forums. So it might be a good first step to figure out what work was already underway before this initiative gained interest.

On a side note, as for the tools themselves, we have ArrayShift installation profile which is pretty close to a decent StackOverflow clone in Drupal. i built StaticBuzz as a proof of concept last year (it was done one Saturday so yes it looks ugly, but it was just to see that it could be done).

I'm interested in helping with this, especially in the context of a re-usable profile like ArrayShift. Since, as @rfay point out, the technical part isn't that hard, it would be great to see someone with a lot of UX focus head up this effort to keep it focused on the user experience problems, which are hard.

What are the features needed on this support site that our forums don't offer?

In that thread (which I sort of read as I went), I think I heard:
- Better searching
- Question/answer format

Is that it?

I think badges, earned permissions, and/or some other user incentive system is pretty critical to keeping everyone very focused on the question/answer format. It's those incentives, not any technical barrier, that keep Stack Overflow from turning into a generic discussion forum.

For those of us not as familar with Stack Overflow, can you explain what the auto-sorting feature would do?

Answers with the most votes show up higher on the list.

subscribe

Remark: Josh's initial post was in response to http://lists.drupal.org/pipermail/development/2011-January/037275.html

I'm willing to help when this kicks off, both with design and build and custom modules as required

I think that it would interest this thread that Stack Exchange will create a site specifically for Drupal questions, which at the moment is in commitment status. In the moment enough users will have committed to the proposal, the site will pass to beta status (private beta, and then public beta), and then it will get an official status.
It may take some times before the site will be official (which means it will have its own domain), but this is what most probably will happen.

Regarding the stackoverflow issue, has anybody considered that "Drupal" is a trademark and domains like drupal.stackoverflow.com might require a license?

@killes: the site is at the moment at commitment phase; this mean that a number of users must commit to the proposal, and then the site will be launched.
The site will be initially hosted at drupal-answers.stackexchange.com, and then (when it passes the beta phase) to its own domain (which domain, it will be chosen only later).
Considering the purpose of the site, and the fact the domain name will not be drupal.stackexchange.com, maybe they fall in the case of automatically granted license.

Anyway, the domain name is not the issue here (I guess they could still launch the site even if they use a completely different domain name).
What is relevant here is if we really want to create a site like Stack Overflow, when there will be a site similar to Stack Overflow for Drupal questions. Actually, the site will be very close to Stack Overflow as both the sites will use the same software.

I am personally not against creating support.drupal.org. The first reason is that it will be an official site; if then we create a site very similar to Stack Overflow, it would be a show case for what Drupal can do.
The question is then: do we have the resources to make this project reality?

Another feature mentioned as good on StackOverflow by Adam B. Ross today on the email list:

"definitive access to questions and topics that go unanswered, or unanswered by good answers. This informs documentation needs very well. (In fact, good answers can be clean enough to copy-and-paste as documentation.)"

I was reading the thread on the list, let me share some thoughts:

• IMHO issues are the heart of the communication on drupal community. In fact we have a support request category, that try to document questions.
• I am not really sure that creating and independent source of information is going to help. But well, independent is not really true as I suppose that is going to be search-able from drupal.org, but I think that removing that category from issues(and moving all support requests to the new place(tm)) would be part of the process.
• The main facts that I see as the improvement of having such a stackoverflow clone are:
  • voting + visibility based on that voting
    Feedback based on the amount of users looking at the questions/answers would be definitely a good way to rate the quality.
  • relevancy of the answer based on the karma of the user answering
    We are not collecting karma on d.o AFAIK, and way karma can be calculated is almost a flame, so it would be difficult to define one but maybe it can be doable if we think in "support karma". In the other side, we know that an answer from the maintainer of the current asked module use to be more relevant, so that could be a good first start.

So, IMHO the best way to introduce the stackoverflow clone is to drupalize it with our way to deal with issues. I mean, use support request category as the base for that site. Actually I think that support.drupal.org should be a "view" of the issue queue that supports the stackoverflow way of support. And that means that we should be able to vote only on that category of issues. About the karma, maybe we could only take in account "module maintainers" as a pre-defined good feedback, and let the rest to be qualified based on the voting.

I am not sure about all technical details to make that possible, but sound like the best idea to me if we want to make subsite.

@marvil07, I agree that project-specific questions are already handled well by issue queues, and yours is the first suggestion I've seen that this might change with a Q&A site. I don't see a lot of project-specific questions on Stack Overflow's Drupal tag. I see more "how do I do this?" questions, where the answers often point to specific projects. I don't use the forums at all (no one has ever suggested them to me), but I gather from what others have said that this would have much more overlap with the forums than with issue queues. That overlap is something we'd need to be careful about, but assuming there is some scope that would make sense for a Q&A site, the exact scope is probably a question better addressed in the context of a working prototype.

As nearly everyone who has expressed interest in helping with this has also mentioned interest in ArrayShift, I suggest we mark this as "postponed" until ArrayShift has a recommended release and we all focus on getting ArrayShift to a recommended release as a first step.

And actually, for Drupal Core, I don't think we use the issue queue for support much.

subscribe

sub

sub

Now that I've had a few months to think about this, the more I'd like to see these features on Drupal.org forums and issues. We do share infrastructure, but having all the data in one place is much more convenient for things like showing related post/issues and including reputation on user profiles. There are certainly ways around this, but it adds more things (that would be nice) to do. If we don't leverage the rest of Drupal.org, we might as well send support to StackExchange, because they are going to be better at their core business.

Another related link is the g.d.o group: Support Infrastructure.

I would like to see how I can help with this.

I also joined the Support Infrastructure group.

sud

Do we intent (in the long run) to deprecate/merge the forums and/or support requests in favor of support.drupal.org? I'm asking this because despite the fact that forums, issue queues and Q&A sections have different(iating) purposes they somehow overlap with each other. Taking a step back and looking at the general idea, they simply seem like different ways to provide the same thing: support. They only differ in the way they do it.

I myself stick to the issue queues and hardly ever comment on the documentation section and I never even touch the forums. There is no specific, conscious reason behind this choice of mine, but if I were to explain it, I'd point out these two facts/problems that are of the most concern to me:

1. There is no way in the forums or the documentation comments to tell in a glance:

• whether the issue discussed has been addressed or not
• in which version of the module or core
• which was the final decision/solution (best answer)

In the issue queues we have metadata attached to each specific problem that help in these areas. Plus we have the filtering based on these metadata when we search for a specific issue - not a simple string-based search.

2. If some of us chose the forums, others stick to the issue queues and the rest of us like the documentation comments, don't we this way have a "scattered force" in these three different places when we could instead all focus in a single place in order to help make it as perfect as it can be?

These thoughts lead me to the conclusion that if we want a single, (almost) perfect place where users turn for support, then this place needs to combine as many features from our current support ways and at the same time address as many shortcomings they currently have.

I am wishing to help in this...

We are working on cleaning up Drupal.org-related issue queues. Many issues have been open for years with no resolution because there were no volunteers interested in taking on the task or staff time available to complete the work.

We are going through old issues and marking them as Closed (won't fix). If you feel your issue has been closed in error, please do comment on the issue and let us know. If we know something is critical, we may reconsider the closure of a feature request or bug report.

Thank you for your understanding while we work to clean up the Drupal.org queues.