Pick some system then. Kaya doesn't really have a 'system'. It honestly feels unfinished.
The only pro that you keep bringing up is that no time is wasted. You get to save 20 or 30 seconds over other systems (to play a game that takes an hour or two).
There are so many cons that have been voiced in this thread. Real cons. For example the abuse issue of this system, you want to fix it when it becomes a problem. Why wait? Why wait until the server population gets large, and it becomes a problem, and you have to spend a week coding a solution? While everyone vacates the server because it gets difficult to find a proper game? Its kind of crazy to stonewall like this, I seriously don't understand.

It is your server, do what you want. It is just frustrating because earlier on you pounded your chest about adding Fischer time because its what the PEOPLE wanted, man. Now it seems like what the people want means squat-all. You are being dismissive with pretty bad reasoning in my opinion.

I don't understand either. Your arguments don't seem reasonable to me, let alone valid.

- Changing our opinion about who we want to play happens all the time. I will give you an example that applies to my personal experience: when I am waiting for a game, I don't mind playing anyone with 2 or 3 ranks of me with proper handicap. I also don't mind playing anyone 5 or 6 stones weaker, but I would rather not give more than 4 handicap stones as I don't feel confident playing handicap games and I believe that it is a teaching game anyway. Given the lack of games in Kaya right now, most games must be seen as teaching games - is this possible in Kaya? No, it's not. And it's not really selecting who I want to play, it's more tweaking the handicap settings based on the particular opponent.

- I can't follow your argument. Obviously whoever creates the game offer is the one who is expected to wait longer and, therefore, is the one who might not be ready to play at this exact moment.

- If you don't want to read the new timesettings, maybe you should just refuse the proposal.

Unfortunately, I believe this must be added:
Some of your arguments clearly indicate that you are not willing to accept change and just made up your mind about the issue. You are not proposing alternatives that fix the enumerated problems and dismiss them indicating some really minor trade-off that a negotiation system could create. That's too bad. I rest my participation in this topic.

It does not serve the community or Kaya to get riled up in a feature conversation. It is also silly to want to make a stronger argument by appealing to attack our intents or our way to work.

Kaya has the most broad and extensive user feedback system than any other server out there by 100 miles. Over 40 user requested features were implemented. Those are tracked, commented and done. Between the tweaki, email, chat and partially this forum, saying that we dont listen to our users is disregarding for the respectful every-day work we do.

I welcome this conversation on Kaya or the other official Kaya channels, where its much harder to get riled up and its a lot more practical when it comes to describing desires.

Keep tuned for what is coming next! There is likely going to be a mid release this weekend with a new feature provided by aleski

Regards.

_________________
Founder of Kaya.gs

It might be useful to compare the online situation to the real life one. Let's say I'm at a go club. I come because I want to play some go. In particular, I want to have a good game. What makes good game is my business, and it depends on my mood whether I want to play a stronger or weaker player, with or without handicap, fast or slow. I've been to the club before, and I can see that there are some people with whom I'd rather play than others. If one of those people looks like they want to play, either because they are sitting at an empty board, or seem just to be casually observing another game, I might approach them and ask if they would like to play. If they don't want to play, they make an excuse and basta. If they do want a game with me, they will respond affirmatively, we will sit down at a go board and negotiations will begin over handicap stones and time settings if any. When this is decided, we will wish each other a good game and begin. The same can take place in reverse, with someone approaching me.

To me, this is a good way of starting a game.

If we take a go server as being something similar to a go club, here is how I would compare my experience on kaya to the real world. My goal is to point out problems that I think should be addressed.

I open the website because I want to play go. I look around the room to see if anyone is looking for a game - i.e., if there are any open games with someone with whom I wish to play. If this is the case, the experience is still fairly similar. My next step would be to approach the person I want to play. This however proves to be awkward. I can either click play, which seems akin to wordlessly and obnoxiously sitting down at someone's table, or I can try first to start a conversation, either in the room or per pm. This also feels awkward, and to make matters worse, often goes unnoticed for minutes on end.

More often than not however, there are no game offers up. People are just hanging around, apparently not wanting to play go. This is surely a problem - though to what extent it has to do with how games are started, I don't know.

In any case, the option most like approaching someone for a game in a club is to send a pm. Whereas irl, this method usually yields a game quickly, on the net, it feels slow and cumbersome. On Kaya, it is particularly bothersome as pms tend often to go unnoticed. The other way would be to sit down at and empty table in front of a board myself. There are two ways of doing so on Kaya. One is to click "start a new game," the other is to shout in a room.

Looking first at shouting in a room, this option allows me to see who responds and to begin negotiations. This is quite like the club situation, with the caveat that rooms can be noisy and such requests can go unheard. Also, this way of starting negotiations is cumbersome.

The last option, clicking "start a new game," is also like sitting at an empty table in front of a board, however with the addition that I specify the game conditions. While this can work fine, several problems could - and do - arise. If nobody sits down right away at my club, I might get up and order a coffee, and upon returning to the table would not expect someone to have sat down and placed a stone (unless it was a tournament with a scheduled game - but that's another matter). Quite frankly, if someone did so, I would consider them rude. Games starting without both players present is a problem. Second, if someone sits down with whom I for whatever reason do not want to play, my only option on Kaya is to prematurely end a started game. This is also rude. Not being able to politely reject a challenge is a problem.

_________________
Patience, grasshopper.

まさか俺様を無視するつもり！？

…… RZ……

I find the recent discussion rather confusing. If I make a game offer, of course it should be me who gets to choose if I want to play a particular opponent. My reasons for potentially changing my mind don't matter.

Dear Kaya.gs, I also find that your arguments are not really being real arguments at all. It sounds that your mind is made up already and you want to stick with your decision. The benefits you describe don't sound like they'd actually give me any benefits. I don't see any problems having steps A=>B=>A. Making it a must to use only steps A=>B feels like: As an idea it is something that might be cool, but if turns out not to be, it still must be implemented just because it was a cool idea in the beginning.

Again: My game, I choose who I play and when the game begins. As a matter of course. The opposite would be just silly.

I'm afraid this is the heart of the issue. Earlier when someone raised the suggestion to give people the optional choice of proper custom games - which should override all concerns about waiting time and whatnot - it was shot down with "I didn't like it at all".

I agree that Gabriels position is frustrating, but too be fair, If someone had made a suggestion in the suggestion section, and everyone had voted for it instead of being angry in this thread, he probably would have fixed the system.
Edit: I even did it myself
http://kaya.uservoice.com/forums/130479 ... tion-phase

https://kaya.uservoice.com/forums/13047 ... tion-phase

edit: I also found this one. It has 98 votes, more than every other requested feature except one: https://kaya.uservoice.com/forums/13047 ... tart-timer

_________________
Patience, grasshopper.

One comment about that suggestion. It is from before being able to pick handicap stones(as described in my first comment there).
It had about 70 vote points back then. It is a bit tainted.

_________________
Founder of Kaya.gs

It seems that this topic has been discussed in sufficient depth already, but anyway:

While I like the new approach for being simpler and innovative I am unclear about the valid use case that I do not want to play again with my last opponent. Will there be a flag 'Exclude previous opponent' in the game setup dialog?
This does not feel like an elegant solution.

When developing software you want to go for improvements. Sometimes, after you have the new solution in place, you realize that the 'old way of doing things' has actually been better. At least in my company (software vendor) that has happened multiple times. I have the feeling that this will be one of those instances.

_________________
Stay out of my territory! (W. White, aka Heisenberg)

Many games are set up such that there is a certain, defined ranked/ladder mode. Press a button, the game does some skillmatching player searching magic and you're in the game. Then custom games where you set up a lobby, can invite people, people can just hop into the room, chat while in lobby and set the game settings in that lobby. Has been working pretty damn well elsewhere.

anyone else smell oncoming feature creep?

It's almost concerning to see how open gabriel has been with integrating stuff the users want, but when he makes one decision on his own the community instantly turns on him and tells him off. I believe Gabriel has thought this through thoroughly and though we might not completely agree, it prevents some other negative things that can occur when you have negotiation (for example, always be declined because of your rank)

Gabriel has been open minded, so please return by doing the same thing.

on topic: good job guys!

_________________
admin of the ASR league and KGS admin

I think the arguments for and against true automatch and a negotiation phase have been explored sufficiently.

Kaya.gs: Your server, your design decisions.

If it were my server, I'd want license to be autocratic and I'd pray that my users follow along, at least for long enough to see if my decisions work, or not. Innovation is impossible, otherwise.

If he has thought it through extensively, he should share with us the counter arguments to what we have stated.
people can already be declined because of their rank, so it isn't that.



so... the shouldn't listen to feedback? innovation is certainly not impossible if you have to listen to you customer base. the reason people are complaining is because it isn't working.

A random check just done now - 28 players online, six games being played (i.e. 12/28 or 42% of online players playing a game). At the same time, KGS has 1648 players online, with 449 games (54% of online players playing a game). This does not strike me as clear evidence of "not working" system. However, 28 players is a tiny sample, so even if the figure was only one game, it would still not constitute proof of the position that the system is not working. It will be easier to judge when there is a far larger group of online players.

Perhaps. That statement was based on the fact that usually there are at most 3 games going on.
I just counted- 29 people on line, and 4 games, which is also better than usual based on my experiences
edit: sorry 3 games 1 review

This reminds me of this comic:



give Kaya some more time, it could get to 100%

_________________
Tactics yes, Tact no...

I think there's no evidence that Gabriel isn't listening. He's just not agreeing. Just because he doesn't change his mind as soon as the mob starts to grumble doesn't mean that he won't fix it if he comes to believe that it's not working. At the moment, he likes the way it is, despite the vocal majority here, but that doesn't mean he doesn't hear what people are saying or that he will never change his mind.

Also, people are complaining not because it's not working, they're complaining because they don't like the wayit works.

_________________
Patience, grasshopper.

1. I never said he wasn't listening, someone seemed to think the current advice/listening to advice conversation wasn't working, which was what I was objecting too.
2. it isn't working (at least not as well as KGS). there is no how about it, this system is showing worse results than KGS in any reasonable measure.