Life In 19x19

Some of you seem to not realize or just ignore how KGS handles the escapers. As the escaper keeps escaping more games, he will lose earlier games by forfeit. After enough escapes, forfeits will become immediate. By resigning such escaped games, you are undermining KGS's strategy of dealing with escapers - kinda ironic.

kivi wrote:Some of you seem to not realize or just ignore how KGS handles the escapers. As the escaper keeps escaping more games, he will lose earlier games by forfeit. After enough escapes, forfeits will become immediate. By resigning such escaped games, you are undermining KGS's strategy of dealing with escapers - kinda ironic.

I might consider giving credit to the KGS handling if I could find a single forfeit in my game history.

hyperpape wrote:I would love to see an improvement to the system, but I've yet to see a real improvement among the suggestions.

And it might be an unfair system, but it's ridiculous to talk about it being unjust.

What if instead of it taking 10 previous escaped games in order to have the oldest game forfeited, it only took 2 or 3 games?

Does anyone really need a 10 game leeway in order to honestly get back and do their utmost to finish incomplete games?

cdybeijing wrote:

hyperpape wrote:I would love to see an improvement to the system, but I've yet to see a real improvement among the suggestions.

And it might be an unfair system, but it's ridiculous to talk about it being unjust.

What if instead of it taking 10 previous escaped games in order to have the oldest game forfeited, it only took 2 or 3 games?

Does anyone really need a 10 game leeway in order to honestly get back and do their utmost to finish incomplete games?

I'd prefer this. And in fact, I think it'd even be alright to move the number down to zero instead of 2 or 3. Then, give an option to have a "mutually agreed upon postponement" of the game.

But I think that making it 2 or 3 games would be an improvement, still.

Helel wrote:...

The 10 games limit provides a buffer that allows people to leave games unfinished when they take a strong dislike of their opponent. ...

This is the feeling I got from another thread on this topic; It appears that there's been an intentional effort to allow people to keep from finishing games against people they don't like.

I really do not agree with the philosophy behind this.

Kirby wrote:

Helel wrote:...

The 10 games limit provides a buffer that allows people to leave games unfinished when they take a strong dislike of their opponent. ...

This is the feeling I got from another thread on this topic; It appears that there's been an intentional effort to allow people to keep from finishing games against people they don't like.

I really do not agree with the philosophy behind this.

Have any KGS admins confirmed this or suggested it? I had sort of thought this thread was mostly speculation/post-hoc rationalization by people who didn't know the exact reasons. I had assumed the rule was intended to help people who had connection problems.

hyperpape wrote:...Have any KGS admins confirmed this or suggested it? I had sort of thought this thread was mostly speculation/post-hoc rationalization by people who didn't know the exact reasons. I had assumed the rule was intended to help people who had connection problems.

Mef is a KGS admin. Actually, he's a senior admin.

Mef wrote: This has been discussed time and time again (dating back to RGG), and in fact has even been answered in this thread. The principle of "No one should be forced to continue a game they have no desire to finish."

The principle of "No one should be forced to continue a game they have no desire to finish.", to me, seems to suggest that if you don't want to finish a game for whatever reason (eg. you don't like your opponent, you're late for your date, etc.), then you don't have the responsibility to finish the game you've started.

I do not agree with this philosophy, because I feel that, when you start a game, you should see it through to the end.

On the topic of connection problems, again, I think that having a buffer of 5 to 10 minutes for people to get back into the game would suffice to alleviate any connection issues that people have.

These days, I can't even think of a person that has, for example, a dial-up connection, so it's surprising to me that this is still a really big issue. In any case, giving a time buffer should suffice to let people get back online.

Again, though, from Mef's comment above, it seems to me that the philosophy behind the current policy does not only have to do with connection speed, but also with whether a person feels like finishing a game they have "no desire to finish".

I live in a small city in a house built in the 1930s. I have DSL, but get poor speed even for that, and have frequent interruptions. I probably could still finish all my games if there was a 5 minute timeout that reset on every disconnection. If I had 5 minutes total, I think I would lose games on time.

Note that KGS has a mobile client, and mobile networks are pretty bad. Also, there are places that have even worse broadband than the US.

Kirby wrote:Mef is a KGS admin. Actually, he's a senior admin.

Missed that, and I'm surprised.

hyperpape wrote:I live in a small city in a house built in the 1930s. I have DSL, but get poor speed even for that, and have frequent interruptions. I probably could still finish all my games if there was a 5 minute timeout that reset on every disconnection. If I had 5 minutes total, I think I would lose games on time.

Yeah, I'd prefer if it were 5 minutes on a per-disconnection basis. This gives people time, and still eliminates the escaper problem. They can't stay away from the game forever, so they cannot escape.

Kirby wrote:

hyperpape wrote:I live in a small city in a house built in the 1930s. I have DSL, but get poor speed even for that, and have frequent interruptions. I probably could still finish all my games if there was a 5 minute timeout that reset on every disconnection. If I had 5 minutes total, I think I would lose games on time.

Yeah, I'd prefer if it were 5 minutes on a per-disconnection basis. This gives people time, and still eliminates the escaper problem. They can't stay away from the game forever, so they cannot escape.

I agree.
And anything much longer than 5 minutes per disconnect becomes really unfair towards the still connected player. If the internet connection is really that bad, there is still the option to play with longer time settings.

Kirby wrote:

hyperpape wrote:I live in a small city in a house built in the 1930s. I have DSL, but get poor speed even for that, and have frequent interruptions. I probably could still finish all my games if there was a 5 minute timeout that reset on every disconnection. If I had 5 minutes total, I think I would lose games on time.

Yeah, I'd prefer if it were 5 minutes on a per-disconnection basis. This gives people time, and still eliminates the escaper problem. They can't stay away from the game forever, so they cannot escape.

I disconnect, you wait, I reconnect, I disconnect...

Sometimes I play blitz at lunchtime. 5 to 10 minutes becomes awkward.

hyperpape wrote:...

I disconnect, you wait, I reconnect, I disconnect...

Yes, but you cannot be gone for more than 5 to 10 minutes or you will lose the game. Today's escapers can leave the game and stay away for days.

Sometimes I play blitz at lunchtime. 5 to 10 minutes becomes awkward.

Is it more awkward than two days, for example?

If you cannot commit to finishing a game through, though, you could simply play a free game.

The point is, having a time limit enforces commitment to the game. You cannot leave just because you feel like it.

chipping in... with a different angle.

From what I gather KGS has a lenient escaper policy.

In online poker the responsibility for the internet connection is clearly with the players.
If you have a dodgy connection you will lose money. Typically you are permitted one minute to reconnect. This is done to help the flow of the game and to make sure there is no "work around" on a losing hand.

Internet poker was far more international with more diverse internet connections than Go ever was and there was a lot more at stake.

If a poker room adopted the a KGS type policy then there would be many ways to avoid losses which are a part of the game and would be unfair on "robust connection" players.

On Tygem you lose after about 5 minutes on disconnect. This is clean and understood by the players. If there is a genuine/rare disconnect a rematch costs nothing. If disconnection is a common problem for a player then possibly he is destroying the game enjoyment for his opponents and should consider a turn based server.

One thought -- I try not to treat any particular game of Go as spectacularly important. If something important related to family, work, or friends comes up while I'm playing a game on KGS, I would like to drop the game without a second thought. If all my opponents spoke English, I'm sure it would be fairly easy to say "Hey, I'm sorry, something's come up," and come back to the game later. In practice, though, most of my opponents seem to be Japanese. I like that the lenient policy on leaving one's games gives me the option to simply close the game and come back to it later, even without any conversation with the other player.

Kirby wrote:
Mef is a KGS admin. Actually, he's a senior admin.

Mef wrote: This has been discussed time and time again (dating back to RGG), and in fact has even been answered in this thread. The principle of "No one should be forced to continue a game they have no desire to finish."

I've mentioned before that I've given up on these discussions since nothing new is ever brought up, but I would just like to clarify on this since I'm being quoted -- When I say this I'm doing my best to paraphrase/recall the sentiment wms expressed when the new escaper policy was implemented (and likely when the old was was implemented too...wms if you read these threads please feel free to correct me if my memory is poor). This isn't a backwards rationalization of the system, it was one of the driving principles behind designing it.