i do know there are many analysis who sacrifice much of their time, talents, and computing power on these games, and i would like to pay those analysis their due respects.

the problem that i personally had this game, which is why i'm posting here, is that the analysis is spread out in 3 places. there's some in the forum, there's some in the sticky, and apparently there's a tree off site somewhere with more analysis.

i'm not sure the point of maintaining 3 places. does each place have a special purpose? because it seems like it just makes it harder to find stuff.

a way to manage me in the UMAN game would be for the team to maintain a list of "CPU volunteers", similar to forum hosts, who can be assigned to forum hosts. maybe like 10 per host or something. then each forum host could divy out his analysis needs.

If you do get an engine, there are lots of us willing to help mentor you on how to analyze effectively. In addition, please don't lose sight of the fact that one of the objectives is for you to learn more about chess. Don't let the machine take over your enthusiasm.

Here are some responses for you:

<the problem that i personally had this game, which is why i'm posting here, is that the analysis is spread out in 3 places. there's some in the forum, there's some in the sticky, and apparently there's a tree off site somewhere with more analysis.

i'm not sure the point of maintaining 3 places. does each place have a special purpose? because it seems like it just makes it harder to find stuff.

i think it would much more effective for me personally if there were a single resource for posting and reviewing analysis.>

I can see how a newcomer would find all of this to be confusing. Ideally, the main page, forums (which you are calling the sticky), and analysis tree all work together to provide an infrastructure to help the team. The forums are intended to be where specific moves or variations can be discussed in detail. Unfortunately, not all team members use the forum system. The analysis tree has limitations but in certain respects offers advantages that the forums do not. It is conceivable that eventually the tree will replace the forums. If you have not checked out the tree, go to the sticky, click on "Analysis Tree" and follow the directions at <jepflast>'s forum. Also read the post from <YouRang> a few days ago on my forum which explains how the forums and the tree can complement each other.

<here's an idea for the management.

a way to manage me in the UMAN game would be for the team to maintain a list of "CPU volunteers", similar to forum hosts, who can be assigned to forum hosts. maybe like 10 per host or something. then each forum host could divy out his analysis needs.>

At the beginning of the current game, I attempted to recruit a total of 24 "forum assistants" which would have a role similar to what you described. I was not successful. Very few team members are interested in such duty.

1. People who have been around forever
2. People who have been here for awhile
3. People who are brand new to the game

I think we need to make new people our VIP's for a month until they feel comfortable. My only suggestion is to consider the following...

If move A is ahead of move B by 50 votes, are you really "helping" move A by being the 51st marginal voter? Or are you just making it harder for the hard-working analysts working on move B?

<"We are all total amateurs here. We are here to win, learn, and have fun. > ... Not necessarily in this order all the time ; )

<You have chosen to be part of something really special.">

This is my opinion on virtually everything at this site.

Thanks for asking. Sorry to be slow and brief.

I continue to shoot for 4-6 forums at any given time. Occasionally, the situation demands more than six, but I never have fewer than four.>

I think this is still the correct approach. For myriad reasons, the forum system has not been properly used by analysts in this game, but I see this not as a reason to reject or minimize the forum system. Rather, I see it as a failing of the analysts that must be addressed in future games.

<How helpful has the Analysis Tree been to the team? How can it be improved?>

I think the Analysis Tree is fantastic, but I also think that it has not yet realised its full potential. There was a period where I was collecting every stray piece of analysis posted and entering it into the tree, and I think at these times the tree gave an extremely useful overview of the team's analysis. This process of collecting and sorting analysis is the very same process I formerly went through in constructing summaries, and I see the tree as a natural evolution of the team's data storage. However, with summaries, I used to have the benefit of only needing to produce one in particularly sticky situations - every say 3 moves. Maintaining the tree on the other hand is a constantly ongoing task, which requires several hours per day. Again, things would work better if analysts would take the time to post their analysis to the tree, but - with a few notable exceptions - this never seems to happen. To be honest, I feared that the tree might die once I stopped contributing to it, but others seem to have picked up some of the slack, so maintaining the tree does seem to be a role that can be naturally distributed between team members. The only problem is that if nobody is formally assigned to the role then some pieces of analysis may simply slip through the cracks.

<Can we revive the idea of having a Summarizing Committee?>

My thoughts are that this role should be amalgamated with maintenance of the Analysis Tree - in fact that's essentially what my role evolved into. Unfortunately, to both maintain the analysis tree and condense that data into succinct summaries requires a completely prohibitive amount of time for a single individual. The work simply has to be distributed between a committee.

At the opening of the Summarizing Committee forum, I posted some guidelines to writing summaries in a group environment: whatthefat chessforum With the development of the Analysis Tree, much of this process has now been conveniently automated, and all data that is stored in the tree is immediately accessible to the other members of the committee.

yes and I have been afraid to check the tree since you have taken a break. I too think the tree is nowhere close to full potential. The tree can handle a very large team and contributing body.

I plan to do some analysis next game, since UMAN will not be as boring, and draw-seeking as GMAN2 has been.

I'm glad to hear the analysis tree has stood through a difficult dry season. The lack of forum use cannot be explained by an increase in analysis-tree use, since it has declined also, and so have vote sums.

A 3-0 team naturally was euphoric and when a draw or loss seemed on the horizon, naturally the euphoria vanished for some members. Rookfile, I'm convinced, won the first GMAN game for us. He was quite influential and seemed able to effortlessly sway votes to his move.

Also the realization that the GMT game was probably a draw after 36...Kf6 coupled with a tough switch to 48 hour voting seems to better explain the sentiment around here. I did not feel like contributing to an undefeated chess team, but now that our team has a draw, I'm ready to analyze lines.

I have an idea. Rather than have the <RV> page and that becomes our de facto fall back, I wonder if it might be useful to have a page where people with various engines post their engines proposed move.

So the page might say:

Move after 1.e4:

Randomvisitor Rybka 3 (28ply) 1...c5 2.Nf3 Nc6 (.18)

Zanshin Fritz 11 (22 ply) 1...e5 2. Nf3 Nf6 (.22)

Waitika Shredder (21 ply) 1. e6 etc.

That way there may not be such a tendency to rely on one computer source.

Last year, I suggested we organize all our work into a single google-doc and abandon the forum system.

Kutztown and I provide each other a necessary balance for the team. Yesterday, on the main page, some naive people still are asking about why GMT hasn't responded back to our uber-bloated list of 60 questions.

I have only asked 2 of my ideas to be contemplated, both involving making forum assistants do one hour of work per assignment.

The right balancing point for our team's ever changing composition is somewhere between Kutztown's and sentriclecub's management philosophies.

I trust kutztown to figure out where on that continuum to settle on, for the upcoming UMAN game.

Kutz and I go way back, he introduced me to the team and helped sell "guess-the-move" which i burnt myself out after one week when I upgraded it to "guess the variation", most exhausting job ever, and its effectively what kutztown does every week when he prepares forum assignments from an unknown number of candidate moves.

I have tremendous respect for his hard work. He seems to never burn out, where I am still burnt out from many months ago. I know he is more important to the team than I am, but I also have to ruffle his feathers sometimes to make him shift a little bit towards efficiency and avoiding redundancy.

Thank you to everyone who took part in this discussion. I will now further discuss a number of the items.

<1. The "bloat" issue>

A number of team members commented on this. This seems to be a two-part complaint: a) there are too many roles, some of which are assigned a permanent forum, and b) if a volunteer loses interest and does not keep up with the role (s)he volunteered for, the link to that person's forum remains on the sticky and at my forum - instead it should be quickly removed if it is no longer active.

I think one of these criticisms is fair and one is not.

I make no apologies for sanctioning the number of roles that I have. In every case, the role was the idea of the person who took the job or someone else's idea (in some cases mine). If someone comes up with an idea that will hopefully provide benefit to the team, and someone volunteers to fulfill the role, and the idea is not in any way contrary to cg's philosophy or how the team operates, I will always be happy to help organizationally. I do not want to stifle creativity just to keep the list of permanent forums more streamlined.

Now, many of these initiatives fizzle out, usually because the person who agreed to do it loses interest. I am forced to agree that when this happens, the links should be quickly removed from the sticky and from my forum. Not to do this gives the appearance of sloppiness. I plead guilty to this charge and I will try to do better. To be honest, I barely have time to do my own job for the team and don't have time to check up on everyone else. However, that's no excuse. If I don't have time, I will recruit help.

<2. Too many items on the sticky>

I have already agreed that links to discontinued efforts should be eliminated. Additionally, however, it was suggested that some links should only appear at <Forum Central> and should not appear on the sticky.

I believe the sticky should be limited to one screen - a team member should not have to scroll down to see all the items on the sticky. As long as that guideline is observed, I believe it would be wrong to streamline the sticky to the point that some active permanent forums do not appear on the sticky.

<3. The Analysis Tree needs to be kept up to date.>

I strongly agree. The usefulness of the tree is highly correlated to how up to date it is. I am going to try hard to find someone who will agree to take this on. Of course, we need to continue encouraging analysts to post their findings to the tree.

<4. Relationship between the forums and the AT>

Thanks to <YouRang> for this very astute observation: <One advantage of the AT is that you can see the complete line there, whereas that line might be split up across multiple forums as a result of new forum assignments. Consequently, the analysis forums should be seen as the place transient development work, since it is more suited for conversational collaboration. The AT would be seen as the permanent repository of stored analysis, since it is logically structured for that purpose. One thing doesn't replace the other -- rather, they complement each other.>

<5. The forum system has fallen into disuse>

This cannot be disputed. Many possible reasons for the falloff in use of the analysis forums in the GMAN2 game have been discussed and I won't repeat them here. I believe the forums will come back well in the GMMU game. If they do not, we will discontinue them. It is simply too much work to plan the forums if they are not going to be used.

<6. Expectations for forum hosts>

It has been suggested that forum hosts should be expected, at minimum, to do certain things (such as provide minimum analysis for the assigned move or line).

I have mixed feelings about this. On one hand, I am sympathetic to this notion, and I would love to see every forum host do some minimum work for every new assignment. On the other hand, I am wary of forum host burnout. I think of the <Deep Breath> experience. He did not quit just because of bloat - he quit because he was burned out. He faithfully analyzed the lines assigned to his forum and he said it became like an unpaid job. At the beginning of the GMAN2 game, I attempted to deal with this in part by recruiting multiple assistants for every forum host. Unfortunately, I failed. I do not intend to "crack a whip" and try to force forum hosts to do what they may not always have time and help to accomplish. I would love to hear how the forum hosts feel about this issue.

<7. Team captains, Top Analysts Suggestions forum, etc.>

There were a number of well-intentioned suggestions that have the common element of creating a class system within the team, wherein a small group of better players are given special status. I believe that this philosophy is antithetical to what cg intends, and I personally disagree with granting special status or privileges to a small group. There are many players on this team that I would not last for 20 moves against them OTB, but part of the beauty of the current World Team experience is the democracy - the weakest player has a vote equal to the strongest player.

Some might think that the Summarizing Committee elevates certain players above the rest. However, I think as long as they stay neutral and objective, such a committee does not violate the democratic spirit of the team.

<8. <capafan>'s ideas>

<capafan> suggested a few things that I want to follow up with him about, namely a main line forum, a contrarian forum, and the creation of a model forum. I think the contrarian idea was tried in GMAN1 and failed, and I do not see how the main line and contrarian ideas fit with what we do now, but I hope to follow up and discuss these ideas with him further.

<9. <benjinathan> idea - a forum to list the top choices of various engines>

A true democratic voting system should reflect the wishes of the majority. Therefor our team rating should be comprised from the average ratings of individuals representing the team collective rating.

If the rating of the majority block has 1500 ELO how much higher would then be our rating without advocating or trying to convince the majority what is considered best according to our higher rated players.

Why then are we dead set against the advocating what our top analysts suggest? We should consider our self lucky because we have RV to follow.

As much as I like the Analysis Tree, I do agree with some of the points raised by people who don't like it so much.

The main problem is the anonymity of the contributors. Imagine how troublesome the forums would be if cg.com didn't identify who the poster was. You wouldn't know if a given post was from <kwid> or from <poisonpawns> unless the poster chose to identified him/her self, and even then you couldn't be sure he/she was being truthful!

But that's pretty much the situation with the AT.

Ideally, the AT system should identify the person who submitted each comment, and even the person who added each move. In fact, for moves, it might be good if multiple people could 'endorse' a move addition by attaching their name to it.