[Rd] sort() generic? [Re: Suspicious behaviour of sort on POSIXct ..]

[Dannenberg, Alex]

With all due respect, I don't think that the change to the sort function is "a good change".  New versions should be backward compatible whenever possible, i.e. unless there's a very compelling reason to force large amounts of old code to break.  In this case, the sort function could have had an additional "unclass" argument with a default value of FALSE.  That way, old code wouldn't break and new code could be written with unclass set to TRUE for speed.  I hope this will be the case for 2.3.4

_______________________________
Alexander Dannenberg
Geode Capital Management, LLC
53 State Street . Boston . MA . 02109
Tel 617.392.8123  Cell 617.372.6805
alex.dannenberg at geodecapital.com
This e-mail, and any attachments hereto, are intended for us...{{dropped}}