Nick Dokos <ndo...@gmail.com> writes: D> Reuben Thomas <r...@sc3d.org> writes: > >> On 16 March 2015 at 16:52, Subhan Michael Tindall >> <subh...@familycareinc.org> wrote: >> >> You can use a custom capture template and have timestamps of various >> sorts inserted. >> >> For example, I have one datetree I use that inserts a date/time stamp >> using %T (%t gives only date, not time) >> >> See the documentation for capture (hit C-c C C to get into the >> customize interface then scroll down) >> >> My question was about the datetree entry headings of the form "2015-03-16 >> Monday". These aren't controlled by the template. I was interested to know >> why these >> headings look very much like timestamps (and effectively are timestamps, >> though at the top level they mention just a year and at the second level >> just a year and a >> month), but aren't actual timestamps. >> >> Eric Fraga said "I don't think it would make sense for the headlines in the >> date-tree to have time stamps"; but my question is not why they don't have >> time stamps, >> but why they ARE not time stamps (purely in the formal sense: the >> information they contain is already effectively a time stamp, as far as I >> can see). >> > > This is third-hand knowledge and guesswork on my part, but I think that > datetrees are used for things like journals: "that's what I did that > day". Datetrees just give you a hierarchical structure of nodes for > easy navigation: you can look at your journal and open and close nodes > at will, so you can navigate to the date of interest. The fact that the > third-level headings look like timestamps is purely coincidental. > > Timestamps are given to things that are going to appear in an agenda: > "that's what I have to do today, tomorrow or next week". They are > completely orthogonal to datetrees in that respect. > > The stuff that ends up in your journal is stuff that (mostly) did not > appear in the agenda: all the little things that you did that day, > probably unplanned (otherwise they would be in the agenda!) > > Not that the headings in a datetree couldn't be made into timestamps; > but that's not what people use datetrees for[fn:1]. The one thing that > would be facilitated if they *were* timestamps, would be clicking on one > and getting the day agenda for that long-gone day, so you could > reminisce about the other things that you did that day, that did not end > up in your journal. Maybe that's enough reason to make them > timestamps, but there are other (perhaps less convenient) ways > to do that. > > Of course, I may be suffering from a failure of imagination: you might > be using datetrees in a completely different way, one where having the > heading be a timestamp is a very good idea, but I can't think of any: > if you *have* something in mind, do tell.
I often feel the same thing, that datetrees should use timestamps(the 'quiet' kind). The main reason is that I often write a journal entry the day after, and it would be easier to manipulate the date like you do a time stamp. > >> I was hoping to discover the rationale for the design from a developer :) > > You'll have to ask Carsten about it: he invented datetrees I believe (as > well as most of org), but he does not frequent org circles much these > days. > > Footnotes: > > [fn:1] Remember however my caveat about third-hand knowledge and > guesswork: I don't use datetrees. > > Nick > > > -- Joakim Verona