Sweet.

@zultar

I hope you don't think I'm trying to sabotage GR integration. Even though I'm the senior admin, I'm not able to have divine communication with Kestas. I've already asked him for many things related to mod features and he's done most of them for me. I don't want to keep asking him for things unless I really believe they are necessary.

But, if we can find someone willing to code and people are OK with a cum GR that only includes WTA full/no press standard board, you'd have my support, for whatever that's worth.

So, CODER needed to make this happen: GR, WTA full press classic board, and WTA gunboat classic board in first iteration!!!!!

Perhaps you should start a new thread? I don't think many people know what's being discussed here.

I don't think there is a problem with the GR's as it is besides the fact that they are a bit delayed. While some people are very vocal about only wanting WTA full press included I think that the majority of people, myself included, prefer them in their current state. Also, I find the monthly rating enjoyable because it lets me so who is active, who's been moving up in the rankings, who's going down...etc. Simply adding a rating to profiles would take away from that unless a link to a spreadsheet style viewing was also available.

@jmo

That's a good point, I like seeing the whole spreadsheet as well. It would be rather easy to add that to the site, though. In fact, that would be the easiest way to integrate it to being with.

Do you not find the ranking less meaningful by having so many different games being considered?

@jmo - How, exactly, did you arrive at the conclussion that the "majority of people ... prefer them in their current state"?

I have a question for a coder who understand how difficult this would be to implement to the site. Would it be easier to make a number of GRs (ie WTA full press, WTA gunboat, WTA live, WTA Med... etc.) or to just give an option when setting up a game for it to be ghost rated or not. Just as there are some people that are very good at gunboat, there are some people who prefer PPSC, the med variant, the world variant, etc.

Though I disagree with jmo that the majority of people have 0 problem with GR as it is now, I think limiting it to just WTA classic games is a detriment to the overall enjoyment gained from GR. Just my 2 cents, take it for what it's worth :D

Draug asks a good question. Let's get feedback.

Prefers current method: JMO, Indybroughton. Cast your votes!

I prefer current method because I like to be able to compare with historical. And because I play a reasonable amount of PPSC and World games and would no longer be "relevant" in the GR system.

Ha. A polling feature would be nice for this forum. Coders? ; )

Just because I'm curious...

Current Method: 2
Added to WebDip: 1

Prefers current method: JMO, Indybroughton
Wants GR added to WebDip: EmperorMaximus,

@Draug, simple, if the rankings in their current state bothered people then they would/should have mentioned it, if they don't care enough to come and start threads and/or comment on existing threads about potential changes then I doubt they care enough to want it changed. I'm assuming we agree that the people who have commented on this thread do not represent a majority of the players on the site. I'm surprised I'm having a conversation with someone who has supposedly quit the site twice now.

"Do you not find the ranking less meaningful by having so many different games being considered?"
Well there are a lot of players who do not play WTA full press games all that often, I don't think gunboaters like PE and Lando should be excluded from a multiple skillful wins just because they play a slight variant of the game. I would rather keep global and ancient games counting for less, as they do now, since I still think it requires skill to win or draw in those variants as long as you are playing against decent opponents (which is measured by the gr equations).

@abge, "That's a good point, I like seeing the whole spreadsheet as well. It would be rather easy to add that to the site, though. In fact, that would be the easiest way to integrate it to being with."

I would not mind seeing GR integrated into the site if it was in spreadsheet form, as long as all the variants were counted.

@jmo - You said they *prefer* the current method. Apathy is *not* an affirmative vote for the status quo. They may not care if it changes or stays the same, but that doesn't mean they *prefer* (your word).the status quo. They may *prefer* to change it but be apathetic enough to not want to bother posting one way or the other.

@Indy - World Games may as well be irrelevant now. They are heavily reduced in their influence on GR as are ancient med games.

Hey, welcome back, Draugnar.
Mumbles "every time he comes back, he derails my thread...." :)

My thoughts:
- The statistics should be limited to the member profile page and not be added to our nametags.
- GR should still be calculated every month. Changing GR is an intensive process that would require too much too often.
- Categories (as many have said) should be: All, Classic, and Gunboat. Even just these three creates nine numbers when you remember we have Peak, Current, and All-Time.
- Asking for TGM's code is incredibly rude and you shouldn't do it. If he responds here and offers it, great. Otherwise you'll have to learn how it is calculated (from his extensive explanation) and code the beginnings yourself.

Folks,
The most important thing at this point is to get a CODER(s). Let's get one or two people committed to that first before we spend two weeks discussing what could ultimately end up being an academic issue.

For you zultar...

I'm all for including GR on the site and like the Idea of keeping both the traditional computation as it resides now and adding category based GR. And putting on profile vs. spreadsheet... These aren't mutually exclusive options. A GR comparative page could be built with sortable columns that list over all GR/position, and each of the categories and you could then sort on whichever column you wanted by clicking the header. I write these types of pages for client's all the time.

Finally, no, I don't have time right now to code anything for the site.

hmmm... combinations.

WTA vs PPSC = 2
white vs grey vs no press = 3
classic vs med vs world = 3
live vs non-live = 2

Did I miss any?

I don't see why it would be terribly difficult to track stats on all of that in real time... you could see how your GR is all lumped together, or filtering out individual parameters, or just looking at a very specific category (WTA classic full press nonlive e.g.).

No more of this monthly updating bullshit :P

All we need right now is a coder and a FIRST iteration. Then we can add to the codes and whatnots in the next one. It's doubtful that we can get something that would be perfect the first time.

Actually, you have 2x3x3x2 possible game types (i.e. 36 possible combinations) giving you potentially 2 + 3 + 3 + 2 + 6 + 6 + 4 + 9 + 6 + 6 + 18 + 12 + 18 + 36 + 1 total possible GR combinations. that's 132 possible GR scores... I think we'd want to break it down to a fewer categories. Maybe Live, non-Live WTA White Classic (for the WTA purists), ditto but PPSC (for the other purists), non-Live Gunboat, All Gunboat, All Classic, All World, and All Ancient Med.

And @YJ - Tyhe problem is that GR has to be computed based on each breakdown seperately. It's not a percentage thing you can just run on the fly because it's based on the GR for that given state of *all* the players in that game when the game ends.

Are we discussing how many different GR ratings there should be again? :)
I say: (1) WTA, classic, full press
(2) other games

I just realised a big issue with implementing directly into the system.
Not everyone that uses Kestas's open source software implements GR (we might be the only ones?)
I really don't think fragmenting the software into "public for everyone" and "webDiplomacy.net specific" is a good idea at all.

That's easily handled with a config file that allows it to be turned off (code reads the config and the library routines check the setting before doing anything with GR). That's what config files are for. But if the calculations are built into the site, it might be something other sites would like to use.

Yes, Draug, I am aware of all of this. I just don't see why it would be difficult to set up automatically - yes even for every combination. The coding is pretty trivial.

Welcome back Fag-Naur!

@YJ,
I don't know if the coding and data handling/storage would be an issue, but that is a bit much information for most to process. It's overdoing it to have every combination displayed or even calculated. I don't think there is more information (not data but usable and understandable information) to be had when you look at someone's specific subscore like WTA live world variant or PPSC ancient map gunboat. For most of us, we have a set of games we like to play and won't have many games in the other variants, so even the data calculated there is more anecdotal than informative.
If you want to play with people of comparable skills, it's probably more important to look at their overall score, WTA vs PPSC, live vs day length phase, and press vs gunboat. I think those four factors are more indicative of someone's skill levels than their performance in any single combination.