[O] Represent *everything* in Org-mode

[Karl Voit]

Hi!

I need your thoughts and feedback on this idea:

I am thinking of letting student(s) implement a (Python[1]) script
that imports[2] all kinds of data sources to generate simple (and
reduced) Org-mode heading entries and links to the original
information in order to represent the users digital life as complete
as possible.

Imagine, you have got one (additional) «archive.org» (or
«mylife.org_archive»[8]) which contains lots of small entries that
represent many things you are doing on your computer:

  * emails you send and receive
  * tweets you write
  * weblog entries you write
  * usenet postings you send
  * files you are creating (with a datestamp in its filename)
  * bookmarks you save (in delicious?)
  * SVN/git commits you are committing
  * SMS you send and receive (via smartphone)
  * ... and much more 

With this system, you can visit any day in the past to see, what
happened in your (digital) life that time. You can reconstruct
pretty much anything you were thinking, working, ... that day.

If you happen to know MyLifeBits[3] from MS Research, the papers
from Gemmell et al or the book «Total Recall»[4] you already know
what I am writing about: researchers implemented a (MS Windows only)
system to capture your digital life even with digital cameras and
screenshots of your desktop.

With Org-mode and a bunch of «connectors» this should be a fairly
easy job to do. Nothing proprietary here, the amount of data is not
that much as with those binary information from MyLifeBits.

I am thinking about a central management tool that writes the
Org-mode file(s), lets you add tags to specific sources and correct
time zone deltas caused by timestamps of services out of sync with
the time zone you are living at.

Then there are those «connectors»: one will parse through my
maildir[5] to collect sent (and received?) emails in order to
generate something like:

* [[file:/my/maildir/the_email][Urgend: Server just died]]  :email:work:
    <2010-01-17 Tue 08:12>
    :PROPERTIES:
    :FROM: address@hidden
    :END:

Another «connector» parses my monthly backup of tweets[6] in order
to generate entries like:

* [[http://twitter.com/status/0815][I hate dying hardware]]  :tweet:
    <2010-01-17 Tue 08:15>

Parsing a source like «locate» I can filter out files I am putting
an ISO datestamp into and generate:

* [[file:/albums/2010-01-17T08:21_rat.jpg][The rat that ate the \
    server cable]]  :file:
    <2010-01-17 Tue 08:21>

With another «connector» I am parsing my weekly delicious[7] backup
and generate entries like following for all my bookmarks:

* [[http://killrats.com][How to kill rodents]]  :delicious:animals:
    <2010-01-17 Tue 09:35>

Without such a combined agenda view, you would possible never know
which different things you were «using» that day when a rat was the
source of a hardware downtime.

This is not a new idea but as far as I know, it was never
implemented that complete outside of MyLifeBits.


So: is there something similar out there? Probably using Org-mode
already?

And: what do *you* think of this idea?


I'd like to have a central tool that manages the connectors as
mentioned above and small and easy to implement connectors for each
data source.


  1. I know that you guys would like to see that in ELISP but here
     at my side is sadly no ELISP knowledge available :-(
  2. Currently, only one-side-import (and no two-side sync) is
     planned.
  3. http://en.wikipedia.org/wiki/MyLifeBitso
  4. http://totalrecallbook.com/ (I'll have to read it soon)
  5. http://en.wikipedia.org/wiki/Maildir
  6. I am using http://grabeeter.tugraz.at/
  7. http://delicious.com
  8. In order to keep daily agenda small/fast and only «Archive mode» complete
-- 
Karl Voit