Should playtime for the Comp be increased?
Al
Should playtime for each game be increased by at least 1 hour
to 3 hours. That way longer games could be included.
A lot of games can be intricate and unsolvable in 2 hours and I
feel that the player should have at least 1 more hour of playtime.
Its very possible that it may take the extra time to really give
a good judgement of the game being played especially since
it's a Comp version.
Feedback, flames welcomed.
Benjamin Caplan
The point of the two-hour rule is to encourage people to enter who
aren't up to making a longer game; by tradition, longer games should be
entered in the Spring Thing (or, in the southern hemisphere, Fall
Fooferall).
There is a sort of gap, though; moderate-sized games (say, three to
eight hours) don't really feel quite right in either.
Conrad
I'd be more interested in seeing a qualifying round added on. That
way, mental midgets could flame the qualifying-round judges rather
than new authors.
Conrad.
Al
Explain what you mean by a qualifying round.
I took a look at the scores and am still stumped by these
so-called "Std. deviations" and "averages" that are used to
determine
the winners when several games had identical scores.
There were games that had identical number
of votes and that's all that should count.
Byzantine Perspective had 106 votes yet came in 9th ! !
14 more than Rover.
Snowquest and Duel in the Snow had 101 votes each yet
Snowquest loses to a game with 9 less votes?
Gleaming the Verb had 102 votes and came in 21st.
Byzantine Perspective should have come in first based on number of
votes
Gleaming the Verb 2nd, and a
and second voting round should have determined the 3rd and 4th places
between Snowquest and Duel in the Snow.
this is like saying that even though Obama got more votes but because
of the
"average" and "std deviation" McCain became president.
If anything there should be a re-vote on the top 3 places when two or
more of
the games entered have the same number of votes IMHO.
Ron Newcomb
I mentally adjust for my playing skill -- or lack thereof. 2 hours to
a veteran IFer is, like, 4 for me. It's not cheating in favor of
longish comp games, since they're all treated to my "relative 2-hour"
rule. :)
Yoon Ha Lee
I have to say that 2 hours hits the sweet spot for me: it's long
enough for me to get *some* sense of a game (1 hour would be too
short, for instance), while giving me an escape hatch. If the time
increased even to three hours, I'm not sure I'd play comp games for
judging, because of the extra potential time commitment. I don't have
the brain cells for multiple longer games, and I'd rather play more
games than fewer longer ones.
YHL
Sarah
first last year.
http://ifcomp.org/comp08/results.html
Benjamin Caplan
> Byzantine Perspective had 106 votes yet came in 9th ! !
> 14 more than Rover.
>
> Snowquest and Duel in the Snow had 101 votes each yet
> Snowquest loses to a game with 9 less votes?
>
> Gleaming the Verb had 102 votes and came in 21st.
>
> Byzantine Perspective should have come in first based on number of
> votes
> Gleaming the Verb 2nd, and a
> and second voting round should have determined the 3rd and 4th places
> between Snowquest and Duel in the Snow.
>
> this is like saying that even though Obama got more votes but because
> of the
> "average" and "std deviation" McCain became president.
Are you a troll?
You're conflating two very different implications of "number of votes"
that come from completely different voting systems.
In a democratic election, a candidate "getting a vote" means that
someone voted _for_ that candidate -- cast a vote saying _that_ the
candidate should win. There is only one type of vote, the "yes" vote. A
voter that dislikes a candidate will cast no vote for that candidate.
In this system, it makes sense for the candidate with the most votes to win.
In the IFComp, a game "getting a vote" means that someone gave their
opinion on _whether_ that game was good -- cast a vote saying _whether_
the candidate should win. There are several types of votes
(specifically, numbers from one (zero?) to ten). A voter that dislikes a
game will cast a vote assigning a low score to that game.
In this system, it makes no sense for the game with the most votes to
win. A game that gets 100 judges each voting 1 point (a very bad score)
should certainly not win over a game that gets 50 judges each voting 10
points (a very good score). The first game has 100 judges agree that
it's a terrible game, while the second game has 50 judges agree that
it's an excellent game.
A game receiving lots of votes in the IFComp simply means that lots of
people felt like saying what they thought about it. What they thought
about it might be either good or bad.
Conrad
You find an "average," Al, by adding together the number of scores (a
more accurate word than "vote) and dividing what you get by the number
of scores you added up.
So, the average of 4, 5, and 6 = (4+5+6) / 3 = 15 / 3 = 5. In that
case the 4 and 6 "cancel out" to 5.
You can verify yourself that the average of 1, 5, and 9 is also 5.
A "standard deviation" measures how spread out the data are. (4, 5,
6) has a much smaller standard deviation than (1, 5, 9). So, games
with low standard deviations are games that judges agreed more about
scoring, and games with high standard deviations are games that judges
disagreed about.
--I can't tell you the math for standard dev. off the top of my head,
but I'm sure you can find it on Wikipedia.
Conrad.
Conrad
What I mean by a qualifying round is that games must *earn* the right
to be in the Comp, by passing a little pre-competition judging round.
They don't let everyone into the Olympics who show up, and I think too
many people become resentful of roughly-made games. It's bad for new
authors, it's bad for the community, and it's bad for the emotionally
incontinent reviewers.
So, although I personally like the open famework of the Comp, I think
Comps with a qualifying round would be better all around.
Conrad.
Vivienne Dunstan
> I have to say that 2 hours hits the sweet spot for me: it's long
> enough for me to get *some* sense of a game (1 hour would be too
> short, for instance), while giving me an escape hatch. If the time
> increased even to three hours, I'm not sure I'd play comp games for
> judging, because of the extra potential time commitment. I don't have
> the brain cells for multiple longer games, and I'd rather play more
> games than fewer longer ones.
Me too. I find 2 hours long enough for judging, and practical in terms
of getting through all the games. Any longer and I'd be far less likely
to try. Also if I really like a game I can always play it for longer,
though I judge based on my impressions from the initial 2 hours.
Viv
Robb Sherwin
> I took a look at the scores and am still stumped by these
> so-called "Std. deviations" and "averages" that are used to
> determine
> the winners when several games had identical scores.
>
> There were games that had identical number
> of votes and that's all that should count.
>
> Byzantine Perspective had 106 votes yet came in 9th ! !
> 14 more than Rover.
>
> Snowquest and Duel in the Snow had 101 votes each yet
> Snowquest loses to a game with 9 less votes?
>
> Gleaming the Verb had 102 votes and came in 21st.
>
> Byzantine Perspective should have come in first based on number of
> votes
> Gleaming the Verb 2nd, and a
> and second voting round should have determined the 3rd and 4th places
> between Snowquest and Duel in the Snow.
>
> this is like saying that even though Obama got more votes but because
> of the
> "average" and "std deviation" McCain became president.
>
> If anything there should be a re-vote on the top 3 places when two or
> more of
> the games entered have the same number of votes IMHO.
hahahahhahahahahhahhahahahahahahahahHAhah
Al
I think what you're trying to do is compare IF to professional sports
where
you have quarterfinals, semi-finals and then finals.
The Comp is not something that has a season and then the top games
go into the above "finals" rounds.
I think that is what you are trying to get at.
Mike Roberts
> On Nov 17, 9:48 pm, Al <radic...@evcohs.com> wrote:
> > Explain what you mean by a qualifying round.
>
> What I mean by a qualifying round is that games must *earn* the right
> to be in the Comp, by passing a little pre-competition judging round.
>
> Comps with a qualifying round would be better all around.
Adding a qualifying round doesn't have to take anything away from the
openness. Simply make both rounds open to all judges - everyone would
decide for themselves which round (or rounds) they want to participate in.
You could also make it more of an advisory round than a qualifying round:
don't eliminate any games, but just post a set of initial scores, which the
rest of the judges are free to use or ignore.
I'd suggest a very coarse scale for the first round, maybe just pass/fail.
If there's a two-round structure, the second round will act as an amplifier
for the first, so I don't think you want to let the smaller pool of
first-round judges actually pick winners. Pass/fail wouldn't tell you which
games are best so much as which are worst. That's *usually* easier for
people to agree on, so amplifying that sort of minority opinion would
probably be more acceptable.
You know, now that judges are free to post reviews during the judging
period, this could all be done unofficially, without any new Comp rules.
Someone could set up an unofficial first-round web site that collects votes
and posts the results at some specified time, say two weeks into the judging
period. Judges who want a more filtered experience could wait until the
results are posted to start their judging. I *think* this would all be
acceptable under the current rules. If someone sets this up as an
experiment, and it works, then it might make sense to talk about
incorporating it into the official Comp rules.
Mike Roberts
mjr underscore at hotmail dot com
Jim Aikin
>
> I'd suggest a very coarse scale for the first round, maybe just pass/fail.
I like this idea a lot -- I think it would focus more nuanced attention
on the better games and, as a side benefit, allow slightly longer games
to play well in the Comp, since judges would be inclined to cut a bit of
slack for longish games that did well in the preliminary.
But I would suggest something a tiny bit more fine-grained than
pass/fail. Something like 0/1/2 (0 being "probably not worth your time,"
2 being "wow, check this out!", and 1 being "dunno -- maybe").
--JA
Benjamin Caplan
> 1 being "dunno -- maybe").
Or, not-quite-equivalently, "interesting, not sure if it works or not --
worth thinking about carefully before judging."
-- that is to say, Golden Banana of Discord candidates.
Aaron A. Reed
Al, what's the problem you're trying to solve?
It seems to me that longer games have other venues for expression
(like the Spring Thing, or even *gasp* a release outside a
competition).
I think the time limit makes all the Comp games of a relatively
similar scope... it kind of wouldn't be fair to have something like
"The King of Shreds and Patches" competing for the same award as
"9:05."
--Aaron
Conrad
I feel there's a big difference between a qualifying round and a
concerted effort to blacklist games: namely, if there were a
qualifying round, the author is free to further refine his game and to
enter it into the next year's Comp. Whereas a blacklist only unfairly
prejudices judges: presumably you'd have a process by which you'd
report that "75% of blacklist judges consider this game not to belong
in the Comp."
But if judges were opt-in -- as they'd have to be; a handful of judges
couldn't review all the games quickly enough -- then they'd probably
only go to the list if they really disliked a game. So the process
would be biased against games.
In effect, a blacklist would just be an organized form of the spewing
we already have.
Conrad.