Index ← 3444 CFJ 3445 3446 → text
=============================  CFJ 3445  =============================

      A violation of R2446 has occurred this week.


Caller:                     G.                                

Judge:                      omd            
Judgement:                  TRUE            



Called by G.:               28 Apr 2015
Assigned to omd:            17 May 2015
Judged TRUE by omd:         23 May 2015 


Caller's Arguments:

The Herald did not publish a newspaper during the week of Apr 20-26.
This is a violation of R2446 that was technically triggered on the
cusp between weeks.  So the issue at hand is the "this week" part.

Rule 2426 (Cards) reads in part:
The Referee SHALL issue at least one Card per week, unless no
violations occurred that week.

Since the violation occurred "between" weeks, not "within" a week,
the question is whether the violation (or any weekly deadline) can be
attributed to a specific week for the above clause.  This is
relevant if someone tries to use the Referee's SHALL clause above to
deputize for the office - it's the only actual requirement of the
Referee office that can be deputized for.


Caller's Evidence:

Rule 2446/0 (Power=1)
The Agoran Newspaper

      The Agoran Newspaper is a weekly publication describing
      noteworthy recent events that occured in the public fora in the
      past 7 days. The Herald SHALL write and publish The Agoran
      Newspaper each week.


Judge's Arguments:

In general, if you have a requirement to do something in a particular
week, considering violations of that requirement to occur during that
same week would be seriously problematic in connection with R2426.  If
no rule violations occur for two weeks, but someone doesn't publish
their weekly report the second week, then under this interpretation
R2426 would require the Referee to issue a Card in the infinitesimal
interval of time at the end of the week; we should be predisposed to
interpretations that don't create impossible requirements.  Two
counterpoints that come to mind are that (a) R2426 is already
problematic without this interpretation, since the situation is just
as impossible for the Referee if a violation occurs during, say, the
last minute of the week (but such violations are probably
significantly rarer than missing weekly duties); and (b) the history
of weekly requirements is far, far older than R2426, so we should be
reluctant to give the former a fundamental-seeming property based on
the quirks of the latter (but I cannot find any past cases where the
issue was relevant).

Still, this interpretation just fundamentally seems icky to me.
Another weird case would be if violating the rule somehow triggered
the report to be published automatically, say under some partnerships
law, causing a paradox; this is more of a stretch since it doesn't
come up in current rules, but it doesn't seem unreasonable to want
automatic actions to work at least somewhat sanely should they be

Okay, so the next possibility is that it occurs in a 'witching hour'
between weeks, as the caller suggests.  I fundamentally dislike this
idea because it adds complexity to our treatment of time without any
explicit support from rule text: we already need to deal with 'right
before the week ends' and 'right after the week starts' (and if we
actually cared to be all Magic: The Gathering and get the ordering of
such actions well-defined, those two buckets would be quite sufficient
to place just about any interesting effects in); we don't need a third
time period in between the first two, which makes "current week" no
longer a well-defined query.

For this case in particular, making it impossible to deputise for the
Referee, despite violations occurring, is also clearly undesirable.

That leaves the possibility that the violation occurs right at the
start of the next week.  I can't think of any particular downside to
this except that it could possibly create ambiguous ordering with
other start-of-week effects, but precise ordering of violations is
unlikely to be important anytime soon.  So I'm going to go for this