http://area51.stackexchange.com/proposa ... ioaqR_AEA2

I have used Senseis Library many times.

The thing about stackexchange sites like stackoverflow.com is that they are simply more user-friendly than SL pages or forums for that matter.

On stackoverflow.com (which is about programming languages), all questions/answers are 1) sorted 2) tagged and 3) rated. These three systems in combination are remarkably useful. Using the tags and sorting (sorting works based on full-text searching) you filter the questions you see when browsing and search through old ones. The tags and sorting also ensure that people find similar questions when writing their own, so that a potentially already-answered question can be found.

The rating ensures that, if a question/answer was particularly highly rated by everyone interested in a topic, all the other 'late-comers' to the discussion will see it. And if a 'late-comer' adds a new answer, people will notice it, because the high rating will ensure that people see someone has added an answer.

Furthermore, whenever anyone adds an answer to an old question, people see it.

So, in my opinion, SE sites are far superior to forums or wikis when it comes to sharing knowledge, whereas for pure information or opinion, wikis and forums respectively are better.

On the other hand I don't find
unstructured forms of information, like forums or wikis, useful for getting real knowledge. They're better for either mixtures or pure distillations of opinion and information, respectively.

If you look at stackexchange.com, it is a perfect way to learn things,
because people who give good answers are rewarded points, and people
who give bad answers are subtracted points. So, your ranking in go
does not automatically determine the respect you get. People look at
your answers, and judge whether you answered the question.

So, if you want to ask something like, 'is this play an example of aji
keshi,' you will not be ignored. Everyone using the page will see your
question, how many replies it has received, and if other people have
already given a good answer.

There was a website like this before called Shidogo. Shidogo only failed because it failed to gather a user-base before introducing the SE-method of sharing knowledge. Hence very few people ended up learning about it. This time, the site will need to gather about 200 people before it can begin operating.

As the site gathers momentum, it will move into the planning stage, and they will assign us a URL. The planning stage is to ensure that more people know about the site this time. Shidogo failed because only like 20 people ended up hearing about it.

20 people sharing knowledge is not enough to answer questions about go. With 300-400 users, all posting diagrams from their own games, asking questions about the finer points of strategy, the new site and its wonderful design will be able to ensure that everyone can ask questions that will get good answers.

Good luck to you. I will participate and support the project, but I still have my doubts as to the efficacy of the solution as applied to Go. I hope to be proven wrong, though.

Bear in mind that, even as a programmer, I still don't find myself using Stack Overflow as often as I would have expected when I first joined. There's lots of good information there, but many times I need a discussion more than I need an answer. I find this to often be the case in Go as well.

So, I'll use the new site however I can, but I still expect most, if not all, of my knowledge sharing will happen through L19 and directly on KGS.

I'll throw in my support on Area 51 as well, but as a regular user of SO I'm not convinced that Go is a good match for the "1 question == 1 correct answer" format of stack exchange. I'd love to be wrong about this, however.

Edit: committed and tweeted

There seems to be open source clones of stackexchange. But I don't know how good they are compared to SE. Perhaps we should setup an instance of one of those instead of trying to push it through Area51. That way we could perhaps integrate it with the L19 forum(For example sharing accounts).
http://shapado.com/
http://www.osqa.net/

_________________
Sanity is for the weak.

That's an interesting idea. The Area51 proposal is taking a long time to get through the commitment stage.

Looks like we'll hit the end of the commitment phase soon. After that we'll get into the closed beta phase. If you want to participate in the closed beta you need to sign up before that happens:
Game of Go proposal

The next things to do after we're in beta are:

* Adding content
* Lobbying the SE coders to add go specific markup (SGF, board image,...)

_________________
Sanity is for the weak.

Your biggest problem will be the proper naming of things. On stack overflow, since it is about programming, proper names are almost always used by the questioner no matter how ignorant he is. He is using this language, and this OS, and he is trying to link that particular library.
With go on the other hand, many things do not have names, and those that have names can only be named by those who recognize them.
Beginner: "Why do I have problems with this?" ( draws diagram)
Dan player: "You have an empty triangle."

In programming, I usually know the names of things before I learn about them. In go, I know the names after I learn about them - if they have names at all.

_________________
Help make L19 more organized. Make an index: https://lifein19x19.com/viewtopic.php?f=14&t=5207

Next question:
Beginner: "Why do I have a problem with this?" (draws diagram)
Dan player: "You should have made an empty triangle here."

_________________
http://tchan001.wordpress.com
A blog on Asian go books, go sightings, and interesting tidbits
Go is such a beautiful game.

I had originally been committed to this site, simply for solidarity reasons. I pulled out of that commitment, though, because I am still not convinced that the Stack Exchange software is the correct tool. Don't get me wrong, SE is the perfect hammer ... but I don't think Go makes for a very good nail in this metaphor.

IMO, position search is crucial.

_________________
Occupy Babel!

True. Unfortunately it has to be fuzzy positional search.

When a beginner is asking the diagrammatical question of "why do I keep losing after I pincer here?", the search engine has to be able to lump all pincers together prior to being able to search. That question, and many others, involve general areas of the board, not particular coordinates. This can be some tricky programming.

_________________
Help make L19 more organized. Make an index: https://lifein19x19.com/viewtopic.php?f=14&t=5207

It might have to be fuzzy positional search in order to do everything you want it to do, but non-fuzzy positional search is still useful.

_________________
Occupy Babel!

Genuinely useful fuzzy positional search is about .05 steps away from a full-fledged non MC 9 dan pro bot.

_________________
That which can be destroyed by the truth should be.
--
My (sadly neglected, but not forgotten) project: http://dailyjoseki.com

I think the sensei's library positional search with wildcards is close enough for most things.

Sad news: http://area51.stackexchange.com/proposa ... ioaqR_AEA2. (I guess that is good news for senseis and L19 in a way).

_________________
Occupy Babel!

Worth noting that for those still interested in this style of community it still exists as a subsection of http://boardgames.stackexchange.com...probably for the best as I'm not sure go alone would support enough of this style activity (closest thing is probably BQM on SL)

the beta stats are staggering. For a 27 day old beta, you are supposed to get 1500 views/day ? Thats quite hard on any kind of site you make, if you dont go paying for advertising.

_________________
Founder of Kaya.gs

I think that's just where you are supposed to get to *eventually*. Besides, to launch the site at all you in theory need to already have 200 committed members.

Most of the sites at area51 that are close to leaving beta have been running for much longer than 27 days.