| Life In 19x19 http://prod.lifein19x19.com/ |
|
| Proposed solution to the escapers problem... http://prod.lifein19x19.com/viewtopic.php?f=24&t=4266 |
Page 1 of 8 |
| Author: | Suji [ Sun Jul 17, 2011 1:01 am ] |
| Post subject: | Proposed solution to the escapers problem... |
So, several years ago, I moved from Yahoo! games to the FICS (Free Internet Chess Server). Escaping is a problem there as well, however, they have a solution to the issue that I find ingenious. Simply put they have a setting, called noescape, that if both players have it set, then a disconnect is a loss for the disconnecting party. I think most players would like this idea. |
|
| Author: | Toge [ Sun Jul 17, 2011 1:13 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
Proposed & working solution. I challenge anyone to come up with argument how this is unfair to anybody. |
|
| Author: | RobertJasiek [ Sun Jul 17, 2011 1:16 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
I have made suggestions to oppose escaping since 1995. Server administrators have not been interested. It is a problem of administrative decisions - not of the existence of escapers. First you need to convince the admins - only then it becomes useful to implement existing solutions. |
|
| Author: | BobC [ Sun Jul 17, 2011 1:29 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
That's quite clever. Players then have choice. |
|
| Author: | tj86430 [ Sun Jul 17, 2011 1:30 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
Did I understand this correctly: - I never escape, so I set the flag - I play against someone who also has the flag set - I have a connection problem with my ISP / hardware problem with my computer and I lose How is this better than just marking disconnect as a loss without any flags? |
|
| Author: | Toge [ Sun Jul 17, 2011 1:51 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
tj86430 wrote: Did I understand this correctly: - I never escape, so I set the flag - I play against someone who also has the flag set - I have a connection problem with my ISP / hardware problem with my computer and I lose How is this better than just marking disconnect as a loss without any flags? - Some people may have connection problems occasionally, so they may want to use the old system without flag. |
|
| Author: | Li Kao [ Sun Jul 17, 2011 1:52 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
And how would it interact with automatch? |
|
| Author: | Toge [ Sun Jul 17, 2011 1:53 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
Li Kao wrote: And how would it interact with automatch? - Match players with same flag setting. |
|
| Author: | tj86430 [ Sun Jul 17, 2011 2:00 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
Toge wrote: tj86430 wrote: Did I understand this correctly: - I never escape, so I set the flag - I play against someone who also has the flag set - I have a connection problem with my ISP / hardware problem with my computer and I lose How is this better than just marking disconnect as a loss without any flags? - Some people may have connection problems occasionally, so they may want to use the old system without flag. I thought someone would say that. However - I'd say that anyone (or, at least the vast majority of internet users) may have problems occasionally. How many of us has a computer that never fails and at least two lines with instantaneous switch over? - Also almost everyone has a system with fairly rare failures. Thus anyone not setting the flag would be viewed as intentional escaper, and almost everyone would only play people with flag set. How does this work in practice in the chess community? |
|
| Author: | BobC [ Sun Jul 17, 2011 3:11 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
I really believe poor connections is a red herring. If you go to any online poker house and open up and watch ten or so tables of play money (not real money) and watch 100 (ten on each table) players over the course of 20 minutes you will see that disconnect is a very rare event. Poker had its boom in 2004 when a majority of European players were on dialup. When I organise a Go game in real life I arrange a time, a place and (even though I am at work) and I make sure the game is uninterrupted and enjoyable for my opponents. It is only polite to do so. If my opponent is willing to invest his time then I have to make some commitment to organisation. If there is the remotest possibility that I will be disturbed I make sure my opponents know in advance. There is choice here. "Faulty" connection players can move to turn based servers (I complete 1.3 games a day on OGS so game rate isn't an issue). They can also opt for free games. The choice would be enhanced with a "flag" as described above. IMO it is only fair that those who may be stopped playing by their mother/wife/dog/children/choice of isp declare this in advance. |
|
| Author: | tj86430 [ Sun Jul 17, 2011 4:08 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
My point was not that bad connection etc is a frequent occurrence (although it may happen to practically anyone, regardless of "choice of isp"). My point was that IMNSHO the flagging system is as good as making disconnect equal loss without any flags. |
|
| Author: | Li Kao [ Sun Jul 17, 2011 4:25 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
So you'd split the player base, increase automatch times, add several options,... for an imaginary gain? No thanks. I get more unfinished games than with the current system. Since any game where either player has a disconnect can't be finished if that flag is enabled. Even the tygem escaper system is better that *that*. |
|
| Author: | HermanHiddema [ Sun Jul 17, 2011 4:26 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
Toge wrote: Proposed & working solution. I challenge anyone to come up with argument how this is unfair to anybody. Anyone who does not set the flag will be looked at as suspect, and people will refuse to play them. |
|
| Author: | Harleqin [ Sun Jul 17, 2011 4:28 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
Here in germany, most ISPs do a regular 12h- or 24h-disconnect. Also, some people do have a shaky connection. Making those people second class citizens is not an option for me. KGS has solved the escaper problem:
The real escaper problem is the hysteria about every disconnect. |
|
| Author: | RobertJasiek [ Sun Jul 17, 2011 4:44 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
Harleqin wrote: KGS has solved the escaper problem Nonsense. Escaping is more than disconnecting. Worse are things like repeated Undos and bad dead stone markings. Also the KGS "solution" is pretty bad; if you don't know exactly how it works, the escaper opponent still gets away with it. |
|
| Author: | Dusk Eagle [ Sun Jul 17, 2011 6:40 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
Harleqin wrote: ...most ISPs do a regular 12h- or 24h-disconnect What? That would infuriate me if I got disconnected in the middle of a game (of Go or otherwise) due to my ISP being unable or unwilling to provide 24/7 service. I'd even have to plan my games around when I knew that disconnect was coming. I think I just gained a bit of appreciation for Canadian telecoms, which is something I haven't had gained in a long time. |
|
| Author: | Li Kao [ Sun Jul 17, 2011 6:59 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
Dusk Eagle wrote: Harleqin wrote: ...most ISPs do a regular 12h- or 24h-disconnect What? That would infuriate me if I got disconnected in the middle of a game (of Go or otherwise) due to my ISP being unable or unwilling to provide 24/7 service. I'd even have to plan my games around when I knew that disconnect was coming. I think I just gained a bit of appreciation for Canadian telecoms, which is something I haven't had gained in a long time. It's not as bad as it sounds. This only happens if you try keep a continuous internet connection for over 24h. So if you shut down your computer at night it won't happen. And many routers have a feature that tells them to disconnect once a day at a fixed time(say 4AM) which avoids the disconnect for the rest of the day. I don't think I have ever hit the 24h disconnect. It's intention is to prevent you from hosting a full server at home because by giving you a new IP address every 24h. WLAN is a *much* bigger problem in practice. Especially since KGS doesn't deal gracefully with packetloss. |
|
| Author: | karaklis [ Sun Jul 17, 2011 8:19 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
A slight improvement suggestion for the "non-escapers" option: The one who has either escaped or disconnected is granted a grace period of five minutes to get connected again and resume the match. With the "resume" button that KGS has (in contrast to IGS) this should be an easy task. As for the automatch, this should always be in the "non-escapers" mode just as it is now, or better: with the five-minutes-return grace period. |
|
| Author: | danielm [ Sun Jul 17, 2011 9:31 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
Harleqin wrote: KGS has solved the escaper problem:
The real escaper problem is the hysteria about every disconnect. It's not solved, because:
Whether that is a smart thing to do is another matter, but this is still "officially sanctioned abuse", and thus this can't be classified as a complete solution. I have never seen a complete solution to this problem that doesn't involve some degree of manual administration, and the KGS system isn't it either. Whether its drawbacks are preferable to other systems is mainly a matter of preference. For the record, I do prefer the current system to the "noescape" flag system on FICS, which is either ineffective or going to split up the already not-that-huge pool of players. Even if it works due to sufficient players on both sides of the flag, you still have to choose between a rock (dealing with escapers) and a hard place (having games decided by unintentional disconnect of either player every once in a while). FICS' best feature regarding this issue IMO is not the flag, but the adjournment system with the option of manual adjudication (which also helps to weed out regular abusers). |
|
| Author: | lovely [ Sun Jul 17, 2011 10:47 am ] |
| Post subject: | Re: Proposed solution to the escapers problem... |
I don't really understand the general uproar over escapers. If someone escapes from a game I'm playing I get satisfaction in knowing that I smashed him so hard that he couldn't stand playing longer. I disconnect every once in a while because I'm using a neighbor's Wi-Fi; when their system goes down, I go down. That's why I put a note in my profile: "Note: sometimes I disconnect because I'm using a neighbor's Wi-Fi. If I do, please wait a few minutes until I return. Usually I return within seconds." I have no control over when/if it disconnects. I really like Tygem's system in the five minutes to reconnect policy. It works well and I've been able to continue playing games a few seconds after I disconnected without any problems. |
|
| Page 1 of 8 | All times are UTC - 8 hours [ DST ] |
| Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |
|