17

As my experience of Stack Overflow has increased I realize that there are a lot of people who generate good answers because many have broad experience in many topics (or languages). This indicates that that the grouping of closely related topics (in this case languages and platforms) can provide a stronger Q&A.

Without this eclectic approach Stack Overflow would never have made it out of the commital stage (though I'm guessing it never really went through one).

The "Artificial Intelligence" proposal made it to beta but was rejected for low usage, nearly a year ago. There are now proposals for computer vision, machine learning and several robotics sites which seem strong but might go the same way. These new site proposals are just smaller target audiences within the previous beta topic that failed.

I would propose an "Artificial Intelligence & Robotics" site that covers all and provides that broad experience base, however I'm not sure how to go about this without creating another dead duck. In combining all these proposals we could create a Stack Exchange site that is worthy enough of the title, one with a stack of knowledge, experience and topics within a broad field. The problem is I don't know how to pull these groups together into a united family using the existing Area51 procedures.

Is it possible to resurrect a beta and merge with related proposals?

I need help co-ordinating this from the top so that we can start to build a community instead of being caught in part of a propose-commital-beta loop of doom in ever decreasing denominations.

Your counsel and guidance is greatly appreciated.

UPDATE

I've taken Bill's fine suggestion and forged ahead ... proposal here ... unfortunately there appears to be a cross-over with http://stats.stackexchange.com

Improve this question
3
  • 2
    I definitely think it's time to give AI another chance on SE. ai-class.com has attracted a lot of interest to the topic recently. It has a huge following on reddit, Twitter, and even has its own Q&A site on aiqus.com If a new Area 51 proposal is created, you might be able to promote it on those sites to see if it would reach the critical mass needed to stick this time.
    – Bill the Lizard Mod
    Commented Nov 18, 2011 at 13:00
  • thanks Bill, I think we had 160,000 people on the course, I'm quite active on aiqus.com but I find this to be heavily orientated towards the course itself and not the content. I will gladly spread the word. I will also try to contact Peter and Sebastian directly to help get them advocate it. What do we do about the merging issue though?
    – Moog
    Commented Nov 18, 2011 at 13:35
  • Partly due to @mbq's answer, I don't think this will work.
    – Mateen Ulhaq
    Commented Nov 19, 2011 at 3:25

2 Answers 2

Reset to default
1

The problem with AI (which actually killed AI.SE) is that it is an ultra-sweet honeypot for various, let's say, specific individuals which have an ability to totally overwhelm the real discussion.

Moreover, there will be a significant overlap with stats SE (machine learning), electronics SE (hardware stuff), signal processing SE (vision, hearing) and soon-in-beta computational science SE (sub-optimization, the rest of computational intelligence) -- I have doubts that there might be too little left for a healthy site.

Summing up, I fully support the idea of AI on SE, but I think it will be more successful as an avid spirit living across several sites than a resurrected zombie.

Improve this answer
3
  • Thanks for your well-structured argument, there are several great points here which lead me to consider matters more thoroughly. Perhaps there is a need for a broader discussion on overlapping topics on Stack Exchange. I suppose university courses suffer the same overlap. I have found that Stack Overflow is a great example community with overlap as diverse in it's content as it is I would have thought that a similar encompassing site that focused on the application of a purpose with enthusiasm rather than subject might prove more fruitful than a specialist site, it's a tough one to call an answer on.
    – Moog
    Commented Nov 18, 2011 at 18:38
  • My interest in this comes from the robotics side of things, and so I don't have the benefit of knowing the history of the AI situation you allude to. The way I see it is that even if Machine Learning and Computer Vision go their own way (being topics with many diverse applications), AI and Robotics still have enough common ground to thrive together. As for Electronics.SE, they purposefully chose to diverge from robotics topics awhile ago.
    – Sean
    Commented Nov 19, 2011 at 20:16
  • @Sean, I agree, seems the only way to get robotics off the ground is by doing joining forces machine learning under this inclusive topic. Too many stack sites are too focused. Proposals continually fail because the audience is simply too small. We don't want to be homeless and forced to bounce around various sites and proposals forever.
    – Moog
    Commented Nov 20, 2011 at 1:43
0

Just so that you know - there is already a seemingly thriving q/a site called metaoptimize.

Improve this answer
2
  • 1
    Just so that you know - this is an OSQA site, not a Stack Exchange site.
    – Kevin Vermeer
    Commented Nov 22, 2011 at 20:56
  • 1
    I know that - but it does the same job, albeit without the stack exchange brand.
    – vishvAs vAsuki
    Commented Nov 25, 2011 at 17:03

You must log in to answer this question.

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