CVS History for TIP #185

VersionAuthorDateQuick Comparison
1.1dkf2004-04-12 04:10:24 -0700 View this version
Log Message
  Created for John H. Harris <JHHarris at valley dot net>
1.2dkf2004-04-12 06:19:38 -0700 View this version
Log Message
  Reference number fix and additional keyword
1.3dkf2004-04-12 09:53:35 -0700 View this version
Log Message
  Typo
1.4WebEdit2006-05-30 20:44:38 -0700 View this version
Log Message
  Revision through web from Zubair <sheik_zubairahmed at yahoo dot com>
1.5dgp2006-05-30 21:40:15 -0700 View this version
Log Message
  A web edit mangled this TIP badly.  Reverted to previous version.
1.6dkf2006-12-05 01:50:52 -0800 View this version
Log Message
  Feature set of 8.5 is now closed to new TIPs without prior agreement of the
  TCT; TIPs targetting 8.5 now redirected to 8.6
1.7dkf2008-06-10 05:36:04 -0700 View this version
Log Message
  CFV by DKF
  Vote closes [clock format 1213704000]
1.8dkf2008-06-17 06:05:14 -0700 View this version
Log Message
  Rejected 0/6/0
  
  For: none
  Against: DKF, JE, MS, KBK, JN, JH
  Present: none
  
  Rejection Comments:
  
  By DKF:
  > > The TIP as proposed runs a big fat wrecking ball through Tcl's value
  > > system by adding a lot of extra magic, causing major disruption to the
  > > way many things work in Tcl at the moment; implementing the TIP to
  > > high-enough quality would be very hard work (the NULL handling of every
  > > single Tcl command would need checking, and a significant number of
  > > extra tests would need to be written, to say nothing of the impact on
  > > extensions) and would probably slow things down[*] as well. Since only a
  > > single use-case needs this sort of thing, it would be far easier for
  > > something targeted at that to be done (e.g. as part of TIP#308 - TDBC).
  
  By JE:
  > > In addition to all the reasons DKF mentioned why
  > > TIP#185 is a Bad Idea, it would also do major damage
  > > to the semantics at the scripting level.  Do not want.
  
  By KBK:
  > > TIP 185 is an extraordinarily bad idea, because its implications
  > > extend to every Tcl command that accepts arguments, creating new
  > > opportunities for errors everywhere. The language abundantly
  > > supports other means for applications to deal with "missing data"
  > > in the correct, application-dependent manner.
  
  By JH:
  > > Verdict may be out on the concept as a whole, but no on this particular
  > > tip.

Powered by Tcl
TIP AutoGenerator - written by Donal K. Fellows