It seems a number of users here have opted to use the unofficial KGS client (which has stopped working after the recent KGS update). I think it would be interesting to get a feel for how many users are (were) actually using it in practice.

A self-selected survey like this will tell you only how many people who read this forum and who want to answer the question use one or the other. Not very useful.

_________________
My blog about Macs and more: Kirkville

More or less the default of most surveys on forum, no ?

You mean the material you find on internet forums is subjective? That's a shocker. The survey results, just like posts, don't need to be "useful". I find surveys entertaining, much like I find it interesting to see what other hobbyist Go players think on topic X or what's their opinion on topic Y is. I don't come here for scientifically relevant statistics.

As for the survey, I have always used the official client. I did look for the unofficial one a few years back, but couldn't find it quickly enough, and I had mixed feelings about using an unofficial version of a client for a service that isn't designed to support an unofficial client. (Bit odd, because on the other hand I don't mind the jailbreaking of phones or custom firmware for other mobile devices. I think the difference for me is that KGS is offered for free and is developed by one man in his already limited spare time, so I tend to respect his decisions. It's different if I buy a piece of hardware, pay with my money for it and own the actual device -- based on my individual set of morals, this entitles me to modify it in whatever way I chose to as long as it doesn't cause any harm.)

Yep.

_________________
My blog about Macs and more: Kirkville

Is there a better way? A few occur to me, but they would have to involve KGS itself, which would then lend tacit support to this client. This is true whether it prompts the user to answer a question, or invisibly detects it in the server code.

A survey on an the most widely read western go forum seems a pretty good way. The results should be interesting, and possibly even useful. If I was the developer of a service and matching client, and found that a (statistically) significant proportion were using an unofficial client, it would give me solid data that perhaps some of the features in the unofficial client should be prioritized instead of others. Having said that, I realize the KGS is largely a labour of love, so user preferences don't play as strong a role as in a more typical commercial venture. I didn't really do this to inform WMS, merely because I was interested.

To be fair, there are maybe a few dozen people who post here regularly. Compare that to the thousands of people who use KGS. Just sayin'...

_________________
My blog about Macs and more: Kirkville

I've only used cgoban, but from what I've read about the features of cgoban-h, I would have no objection if they were integrated into the official client.

BTW, If anyone wishes to make suggestions on improving cgoban, there is a relevant SL page here.

_________________
Patience, grasshopper.

I tried cgoban-h long time ago. The features were nice, but perhaps more relevant question than whether we would like some neat / fun / good feature X is how it would come to happen. Obviously there is no shortage of people who would like to have their suggestions implemented. I have faith that there would even be enthusiastic volunteer coders who would make those changes in open-source philosophy. KGS is the #1 source of Internet Go for the western world, be that it may, it is still wms's project and he alone can choose the direction it will take.

Has ever wms express himself against some of the changes cgoban-h made effective ? Has he arguments against, e.g., coloring game list according to wins and defeats, or the quick browsing of sgf files ?

Or is it just a lack of time ?

I have never heard of cgoban-h, otherwise I surely would use it.

I know that he has said that human-voice countdowns would make it harder for him to maintain the client in multiple languages.

He could just pick a single language and not worry about it. It is not hard to learn to count to ten in any language... except maybe in Qwghlmian.

Also, I wasn't even aware of alternative versions of KGS. ><

I was speaking about what cgoban-h implemented.