i was at least trying to imply that i didn't want to change anything this round. breaking ties in this category would only matter for team averages, which are exponentially less likely than individual ties. i think rank order was probably easiest to implement on the fly. i was merely trying to suggest that making a note of who submitted in what order doesn't seem like much less work than making a note of who submitted when, assuming a program like excel does the work of figuring out durations (i'm also thinking in terms of excel 2007, which has -IFS functions that allow you to specify multiple criteria for average, sum and count in a single function -- it's handy for simplifying things).

Oh yeah, you are indeed correct. I had in essence been doing the same thing in my list which was to put down the one he was most likely to submit. Then when I copied my list here I forgot he did have other games that qualified.

In any case I agree that your reasoning is correct when you say it won't matter in regard to any bonus points as to which of his games are selected. Also for that matter whether his submission actully occured when you made your post or at the last possible moment doesn't make any difference either since Team E will be 1st and Team D will be 2nd in submission no matter what the order from here on out (opps, sorry didn't mean to let the cat out of the bag so to speak).

I'm glad that PlayJeff's game doesn't make a difference because I do believe that by current rule neilo/FB should probably randomly select one of the 4.


I agree this is a legitimate argument but not necessarily everyone agrees. We do need to resolve this in the time between rounds, just that now is not that instant.

In some sense the 4 year 11200 game is a no brainer choice since all other choices are either longer games and/or lower scoring games, so who in their right mind would choose a game other than that. What we have done in the past is to try and intuit the players *intent*. When there is but a single game it is clear they only intent the player could have had was to submit that game. One could reasonably argue that between the following games there really is only one logical choice.

4 year 11200
5 year 9600
4 year 8800
5 year 7400

On that basis I could support the argument that PlayJeff's intent would obviously be to submit the 4 year 11200 game.

Where I would have a problem is if there was a 1 year 4000 game in there. That case is much more difficult to argue because it takes what should be a negative for the team (a player's non-submission) and turns it into a potential positive (the choice between either a potential speed bonus or potential score bonus).

Again I think you made the same argument for captains choice in the AltMeta MVL thread and IIRC the bulk of the opinion was agaist this. Still I have no issue with opening up this and deciding this anew because the previous discussion was in the context of the AltMeta and this subtly different, however we do need to wait until the round is over.

We should get a ruling from Neilo about which of PlayJeff's 4 games should be counted. As I said I do see only one logical choice and it won't matter anyway, but neilo should either approve this choice or make a random selection himself. Certainly if it were only random selection we are asking for FB could provide that but the ruling to accept the "obvious" choice should come directly from neilo.

Of course PlayJeff could spare us all this consernation by merely showing up and submitting. Essentially all the discussion about "auto-submit" is trying to find a balance between divining the player's intent versus giving the team an undeserved advantage.

I forget. I did read through some of the stuff there but frankly didn't have the patience to find the significant info. What I did read seemed to indicate that captains selection really hung on the crux of that issue of being able to tweak the speed/score choice at the last possible moment and thereby seemed to take something that should be bad (i.e. a non-submission) and turn it into a benefit for the team. Clearly this is tempered by the desire to not be in the business of punishing folks for things they possibly can't control like a teammate having a real life event.

As I said I think we should rethink this aspect as it realtes to current rules as opposed to as it relates to AltMeta because they are slightly different discussions. Also now that folks have seen how it really plays out in a round perhaps more people will be interested in participating in the discussion. Anyway the fact that it has come up means we should explicitly deal with this in the rule thread between rounds.


I don't think both methods result in the same answer. In rank order the extremes are highly exaggerated. It's easily possible that the 1st submission could happen in two weeks and the 20th submission in 4 weeks. The ratio in this case is 2 to 1 in a time base system however rank order makes being 1st 20 times more valuable than being 20th not twice. Also there is the point about there being no difference in beating someone's submission by a second or beating them by a week which is the case in rank order. I agree that conceptually rank is easier than time but if all the data is going into excel anyway it's not that much more difficult. Really the data collection is somewhat less because with rank order you need to list the post number and then put the posts into rank order whereas time is simply the timestamp of the post. I do agree that you end with more characters to record it but using ctrl-c from the previous post to get January 11, 2008 17:31:14 was pretty easy.

As an example I just copied the timestamp of the OP followed by the timestamp of Neilo's post and then the timestamp of my post.

I then summed my timestamp and neilos timestamp then divided by two for the average and subtracted the start time and placed the result at the bottom of the column and changed the format of that cell to a number with two digits of accuracy. The result is the following.

12/5/2007 18:39
1/11/2008 17:31
1/11/2008 18:48
36.98

This says that the average time of our two posts was 36.98 days. Elsewhere I converted each timestamp into relative days from the start and saw that neilo's timestamp is 36.95 days from the OP and mine is 37.01 so the average make total sense. In a rank order system any result with two people will always come up with the same result of 1.5. Clearly 2 samples is an extreme example but any system that results in the same answer every time for two samples regardless of when the submissions occur is probably not a good choice of a system to implement the comparison of 5 sets of 4 samples.

Hi!
I have a big problem with my game. I can not post it to the metaverse. Server keeps rejecting it because "application serial number does not match the number at metaverse server".

I've tried updating the serial, updating the game, re-registered the game with the serial I've just received from StarDock, deleted the sig.bin, updated serial through SDC... Nothing worked. On the metaverse accound under registrations I see the serial for my DA and DL, but both are the same, despite I've recived from SD two different ones.

Suggestions please?

I can zip and e-mail my savegame to a trusted person to check what's wrong or to post it for me.

BR, Iztok