Eric S Fraga <ucec...@ucl.ac.uk> writes: > On Fri, 08 Oct 2010 15:01:49 -0400, Bernt Hansen <be...@norang.ca> wrote: >> >> Eric S Fraga <ucec...@ucl.ac.uk> writes: >> >> > Recently, but I cannot say for how long, I have found that dates >> > entered, for instance using "j" in the standard agenda view, no longer >> > choose a time/day in the future but seem to default to the current >> > year. For instance, today, typing "j 2 feb RET" (with a real space >> > between 2 and feb) jumps me to 2010 February 2, not 2011. > > [...] > >> >> Hi Eric, >> >> This was recently changed in commit >> 03b178d (Do not prefer future when jumping to a date in the agenda, >> 2010-09-21) >> by Carsten after an offline discussion with me. >> >> The behaviour changed for the 'j' command in the agenda only but not for >> other date prompts. > > Ah, okay, so I am not totally losing it... ;-) > >> The justification for this was at the start of a new month you need to >> enter the year to go back to a date a week or two ago in the agenda >> which seemed inconvenient. >> >> Carsten noticed that I had set org-read-date-prefer-future to nil in >> http://doc.norang.ca/org-mode.html and questioned why that was >> necessary. After a short discussion he decided to change the default >> behaviour for the agenda j command only. >> >> Please comment on whether this change is good or bad. The docstring >> should be more clear about this change if we decide to keep it. >> >> Regards, >> Bernt > > Well, I must say that I prefer the old way as it is more likely (on a > simple probabilistic view considering the full twelve months of the > year) that I am going to want a future date if I refer to a month > before the current one. I can understand your justification for > earlier in a month but I typically simply use, say, -7 or -10 then (as > I use +7 or +10 say for days in the future). So, I guess my view is > that the change is more bad than good... At the very least, I would > like this to be configurable, if that is at all possible? If not, I > am sure I can adjust! > > By the way, I guess I could see an argument for a date alone being for > the current month, whether future or past, much as time can be > considered already to be for the current day, whether future or past, > if the variable is configured as I have it (time), but even then we > should have a configurable variable? > > Regardless, the docs definitely have to change!
Personally I'm okay with reverting this commit if it is problematic. I'll leave the final decision on that up to Carsten. -Bernt _______________________________________________ Emacs-orgmode mailing list Please use `Reply All' to send replies to the list. Emacs-orgmode@gnu.org http://lists.gnu.org/mailman/listinfo/emacs-orgmode
