Developing a Malkovich Web App

[tapwater]

The Q/A of shidogo.com will end. A few days ago, I emailed the guy in charge of all Stack Exchange websites, pleading to let shidogo continue as-is, and describing how much activity it has received in a mere few days. Sadly, this was his response:

Unfortunately, we longer offer any type of services for Hosted Stack Exchange sites. Sites are scheduled to expire sometimes after the date listed on the site.

I want to thank all those who went out of their way to try and keep shidogo alive; I deeply appreciate your efforts. But it seems that shidogo.com will no longer have Q/A in a matter of days. However, that does not detract from the ultimate goal of shidogo.com, which is to organize online Go learning community so that Go players can learn more and improve his/her game.

In my call for help post, Marcus mentioned and led me to discover Malkovich games. After browsing through a single game, I became very excited and immediately started thinking of different ways to turn this into a standalone, streamlined web application.

The concept of a Malkovich game is very exciting. It is always enlightening to discover what high-level players are thinking during their matches, and compare their thought process to your own. Comments and questions from spectators increase the level of learning even greater, all while keeping the game interesting by hiding each opposing player's thought comments from the other.

I believe Malkovich games will be the next big step in the online Go learning community, and for that I wish to take a part in its development.

Of course, I cannot do it alone. I'm sure there are others among you who are just as enthusiastic about a Malkovich web app, which is why I posted to this forum. If you have any interest in helping this happen, from concept development to code development, join the discussion!

----

First step: we'll need an explicit outline of the rules of a Malkovich game. I understood the general idea while browsing a game thread, but there are other little things that confused me, such as triggers. Maybe we can make a wikipedia page for Malkovich games?

[Dusk Eagle]

What would a Malkovich web app even do?

[tapwater]

The Malkovich web app I'm thinking of would be a streamlined way to play and comment on Malkovich games:

• Each move will have all player and spectator information about it in an organized, easy to navigate layout.
• The data will be stored on the server, but exportable as a download.
• Spectators will be able to comment on a past move and have that comment grouped with that move (as opposed to forums, where it might appear after a more recent move).
• The players actually playing the game will have a much easier time creating and organizing thoughts, diagrams, and triggers, as well as responding to spectator questions that are aimed directly at them.

These are just some of the goals I have. As I've mentioned before, we'll need a seriously good ui to pull it off, and that fact only makes the project even more interesting.

Oh, and everything will be open source! Hurray!

[Marcus]

tapwater touched on a good number of my own thoughts.

I've been toying with design ideas for this for a while, mostly because I need to get some design and coding practice in. It seems like an interesting problem set to work with, and has a decent design space to work in.

[Joaz Banbeck]

...

First step: we'll need an explicit outline of the rules of a Malkovich game. I understood the general idea while browsing a game thread, but there are other little things that confused me, such as triggers. Maybe we can make a wikipedia page for Malkovich games?

There is really one explicit rule - beyond the rules of the game of go itself - and that is that all posters respect the hide tags. Players should not read between hide tage unless an associated label specifically invites them to do so, and other posters who repeat something from inside hide tags should keep it hidden.

Triggers came later. I think that Araban - then writing under the name of Sol.ch - started them in the second Malkovich game on GD between him and me.

[CarlJung]

IIRC http://egoban.org/ allws you to comment individual moves in any order you like. It has been a while I looked at it and the site has been redesigned since I last visited it. I believe this would already cater to 95% of Malkovich needs if you are unhappy with the forum format.

[Marcus]

It's not really about being unhappy with the forum format (at least for me it isn't). I need to flex my design and coding muscles, as I am out of practice. This just seems like an interesting way to do so, for me.

The reality is, I think I could design a really neat web application around the Malkovich Game concept, and it doesn't matter (to me) if it ends up being used or not.

Now, tapwater probably has different reasons for interest, and I'm assuming, from his previous attempt at a Go website built on the Stack Exchange engine, that he would like his efforts to have some degree of success in terms of usage. I'm willing to help him out in any way I can, and if it works out, great! If not, I'm not so worried.

[tapwater]

IIRC http://egoban.org/ allws you to comment individual moves in any order you like. It has been a while I looked at it and the site has been redesigned since I last visited it. I believe this would already cater to 95% of Malkovich needs if you are unhappy with the forum format.

Yes, like Marcus said, it's not about being unhappy with what we have. For me, it's about making something people find better, easier, and more useful. When comparing two similar services, it's all about execution. If I'm trying to build a useful app and no one wants to use it, something's wrong with it, and I need to improve it.

[calm]

I am willing to add to it. Unfortunately at this point I think my main contribution
is to learn Javascript

I think the main functionality for such a web app has already been provided.
It's called a "wave" ( aka Google wave ).

The two main things I think you want to add to waves are:

a goban widget
a comment scoping widget

Once those two contributions are added then I think that the rest of the work is
really just playing test games and figuring out how to glue things together.

[tapwater]

I think the main functionality for such a web app has already been provided.
It's called a "wave" ( aka Google wave ).

I appreciate the suggestion, but I'd like for malkovich app to be its own open source entity. This way if someone wants to improve malkovich, they'll easily be able look at the source code, write and submit a patch for everyone to enjoy.

I am willing to add to it. Unfortunately at this point I think my main contribution
is to learn Javascript

Great! Javascript is such interesting, madly misunderstood language. There's never been a better time to learn javascript, especially since it is now being used everywhere, including on the serverside (!). I would be very willing to help you learn

Speaking of javascript, the malkovich app will be built on CouchApp, which is essentially pure javascript. I've been working on getting the app to a manageable start. I'll probably post a new thread once a minimal codebase is ready to be presented.

[rubin427]

For those players that simply cannot trust their opponents, there may be some value to:

1) Showing a list of all users that have opened this a particular hidden tag. (implys that guests can not view hidden tags?)

2) Agreeing, at the start of the game to keep all player's hidden comments unreadable to everyone until the end of the match - and also explicitly disallowing kibitzing until the end of the match. (i.e. game is private until players agree on the result.)

If I was going to play a complete stranger in a Malkovich game, either one of these options would help me rest a little easier. After all, playing a Malkovich game seems like a huge investment of time and energy.

[tapwater]

1) Showing a list of all users that have opened this a particular hidden tag. (implys that guests can not view hidden tags?)

I was planning to simply not show spectator comments for the players of the game, but your idea is a great extension to that; you must log in to view comments. The logged-in requirement makes it much easier for a player to avoid accidentally seeing spectators' or his/her opponent's comments. Sure, if a player really wanted to cheat he could create another account and view the comments, but that requires extra effort, and I don't think malkovich players would want to cheat anyway.

2) Agreeing, at the start of the game to keep all player's hidden comments unreadable to everyone until the end of the match - and also explicitly disallowing kibitzing until the end of the match. (i.e. game is private until players agree on the result.)

Ideally the hiding will all be automated, so a logged in player will not even have the option to view a "hidden" comment; the comment will simply not appear, unless that comment is explicitly directed towards that player. For these directed comments, we could even create an option where the comments only appear to the player a few moves after they are made.

If I was going to play a complete stranger in a Malkovich game, either one of these options would help me rest a little easier. After all, playing a Malkovich game seems like a huge investment of time and energy.

Hopefully the matches will be friendly, since the players should be playing to teach those who are watching, rather than the players' own benefits

[sumiyaka]

If I was going to play a complete stranger in a Malkovich game, either one of these options would help me rest a little easier. After all, playing a Malkovich game seems like a huge investment of time and energy.

Hopefully the matches will be friendly, since the players should be playing to teach those who are watching, rather than the players' own benefits

Yeah, I've never worried that my opponent could "cheat" at the Malkovich game. I guess we could be a special case because we don't have any observers commenting at the moment, and I feel anything I can see he can see too... and vice-versa.

In any event, this is simply a shameless ploy to get people to take a look at that game.