Another GTD question.

Another GTD question.

[Alex Bochannek]

I have been following a lot of the discussions about how people use
Org-mode to implement GTD. I am currently using a system that isn't
that different from it, but I am trying to figure out the "best" way
to use Org-mode for GTD.

Let us say that your primary use for your GTD org file is the lists,
which is what most people have reported they do. For the purpose of
this discussion I am leaving out the whole tickler file concept.

Looking at Charles Cave's example Web page, he uses a mix of grouping
projects and contexts in outlines. I am in a similar situation where
my first instinct would be to group all my work next actions under
specific headlines (e.g., hardware, software, etc.) since I work in
different areas.

In a way, GTD is somewhat fuzzy about how to address this. It's not
really a context since while I am at work, I can work on any number of
tasks in the different areas. The context talks about what it is I
need, where I need to be, or a type of activity. Maybe I overlooked
something, but it seems like another dimension to the problem.

How does this relate to Org-mode? In Org-mode I see different
approaches to how one could structure these lists:

- Outlines
- Tags
- ToDo keywords
- Categories (files)

They all have their pros and cons and I wonder if we could
collectively come up with a good usage model. Let me propose a couple
of items and that could be a starting point for discussion. I am
purposely trying to pick examples that should cover the common
categories listed in the book. I am also "borrowing" from Charles
Cave's page:

o Call the bank about personal statements
o Go to post office to buy stamps
o Wait for SysAdmin to finish server install
o Hang new pictures at home
o Discuss new development process with boss
o Read through vendor proposal
o Server Install project at work
o Install software upgrade on laptop
o Learn more about font-lock in Org-mode

The examples are fictional (well, except the last one), but I think
they are all plausible.

I am really looking forward to hear what people who are already using
GTD with Org-mode suggest and hope that those how are not familiar
with GTD may still find the discussion interesting.

Alex.

Re: Another GTD question.

[Charles Cave]

Hi Alex,

 > I am currently using a system that isn't
> that different from it, but I am trying to figure out the "best" way
> to use Org-mode for GTD.

I don't think there is any "best" way to use GTD system or org-mode
for that matter.  One of the core principles of GTD is creating
lists of next actions organised by context. A context is some
restraint that only allows the action to be done in a particular
place, time, or with particular resources.


> my first instinct would be to group all my work next actions under
> specific headlines (e.g., hardware, software, etc.) since I work in
> different areas.

I think that is a good idea because when you review your "system"...
the org-mode file you can use these headlines if hardware, software,
etc as a checklist for the question "Are there any outstanding actions
on this equipment?".

 >  It's not
> really a context since while I am at work, I can work on any number of
> tasks in the different areas. The context talks about what it is I
> need, where I need to be, or a type of activity. Maybe I overlooked
> something, but it seems like another dimension to the problem.

Maybe WORK is a sufficient context for your office related
activities. I don't think you need to break it down any further than
that.

> 
> How does this relate to Org-mode? In Org-mode I see different
> approaches to how one could structure these lists:
> 
> - Outlines
> - Tags
> - ToDo keywords
> - Categories (files)

The lists can be structured anyway you want, but the strength of 
org-mode is scanning your whole file to build up a list of lines
matching a tag.  Organising the actions under specific category tags
is extra (unnecessary) work since org-mode does it for you.
Keeping the actions in context with other items may make more sense.


Let me add my context tags to yourlist


> o Call the bank about personal statements   :PHONE:
> o Go to post office to buy stamps  :ERRANDS:
> o Wait for SysAdmin to finish server install :WAITING:
> o Hang new pictures at home :HOME:
> o Discuss new development process with boss  :BOSS:
     (Create a tag for items (agenda) to discuss with your boss
> o Read through vendor proposal  :READING:
    (Can this reading be done at home or on your train trip to
      work? Reading is one of those activities that can be done almost
      anywhere)
> o Server Install project at work    :OFFICE:
> o Install software upgrade on laptop :OFFICE:
> o Learn more about font-lock in Org-mode :OFFICE:

So when you are at the office, you display the tags for OFFICE.

If you set up a meeting with your boss, you search for :BOSS so you
get the most out of your meeting with the boss.

When you are reviewing what you are waiting for, use the WAITING tag.

During your lunch break, you search for ERRANDS.

> I am really looking forward to hear what people who are already using
> GTD with Org-mode suggest and hope that those how are not familiar
> with GTD may still find the discussion interesting.

I'm still refining my system with org-mode so I am keen to hear from
others on this.

Charles

Re: Another GTD question.

[Christopher Kuettner]

>> to use Org-mode for GTD.
; self-documenting org-mode gtd-outline

* @today
** TODO [#A] next action items I have to do today (hard landscape)
** TODO [#B] actions I want to do today (to plan ahead gives my day some 
flow, since I work at home alone
** TODO [#B] I do: org-agenda-include-diary t
** TODO [#B] agenda is set to org-agenda-include-all-todo t
** TODO [#B] so I can work nicely off the daily *Agenda* while at 
computer (*Agenda* = hard landscape)
* @waiting for
- a list
- with waiting-for items
* @agenda
- another plain list
   - sometimes I have more than one
   - topic to share with a give persona
- actually this list is named @messaging because agenda is an org-mode 
reserved word
* @todo
  1. a sorted list of all my other next actions for @computer,
  2. @phone, @office, @desk contexts
  3. I sort roughly in the order I want to tackle them
  4. I try to get them all done 'til the next weekly review
  5. again: since working from home there are not that much 
interruptions, so the whole gtdish @whatyoucandonow has to emulate some 
workday-structure too.
  6. when @today is empty I work off this list
* @projects
  1. sorted list of projects inventory
  2. I try to have only the projects her, I am working on this week or 
the next
  3. project support materials are in directories in my home-dir
* someday
  ** already commited but not this week
  DROPLINE: <some date to speed me up>
* maybe
  ** whatever I have in this bucket comes
  ** sorted by importance because it is so much
* Special Areas Of Focus
** a structure of my
*** working environment
**** and it's subsets
**** like for example 'office management
**** or 'mac maintenance
** as a checklist to
*** gain overview while higher-level thinking is enabled
*** or to spin of new projects during weekly review
* daily major activities
I have a table here with things I want to do daily/weekly on a regular 
basis. Normally I print this out and mark the fields 'done after 
completion thorough the week.
* incoming
Some lines of text that tend to get appended her by some scripts or my 
hands from the device driver's CLI environment. (I run MAC OS X as 
device driver for my Emacs OS)

Re: Another GTD question.

[Alex Bochannek]

Charles, Uwe, Christopher,

thank you very much for your surprisingly diverse replies! It seems
everybody is coming up with a slightly different model and I am
starting to think that using a computer rather than the folders David
Allen proposes maybe gives you *too much* flexibility ;-)

Charles unintentionally gave a great example of what I have discovered
bothers me a bit about the different lists in GTD:

Charles Cave <charles_cave@optusnet.com.au> writes:

>> o Call the bank about personal statements   :PHONE:
>> o Go to post office to buy stamps  :ERRANDS:
>> o Wait for SysAdmin to finish server install :WAITING:
>> o Hang new pictures at home :HOME:
>> o Discuss new development process with boss  :BOSS:
>     (Create a tag for items (agenda) to discuss with your boss
>> o Read through vendor proposal  :READING:
>    (Can this reading be done at home or on your train trip to
>      work? Reading is one of those activities that can be done almost
>      anywhere)
>> o Server Install project at work    :OFFICE:
>> o Install software upgrade on laptop :OFFICE:
>> o Learn more about font-lock in Org-mode :OFFICE:
>
> So when you are at the office, you display the tags for OFFICE.
>
> If you set up a meeting with your boss, you search for :BOSS so you
> get the most out of your meeting with the boss.
>
> When you are reviewing what you are waiting for, use the WAITING tag.
>
> During your lunch break, you search for ERRANDS.

The items tagged PHONE, ERRANDS, OFFICE are pretty much obvious,
although the "Server Install" should probably be on a PROJECT list of
some sort. The READING, BOSS, and WAITING are trickier I
think. The tags are using the list names where most Next Action
sublists are essentially contexts. The Agenda and Reading lists aren't
quite the same though since Agenda is missing the information whose
agenda it is and reading isn't really a context, but an activity,
which is context independent. WAITING (and PROJECT) is even worse in a
way since once I have dozens of those, how do I find out who I am
waiting for, for example?

I have to say that I liked Christopher's example. No messing around
with tags and integration of diary and agenda. I suspect though that
Christopher is spending most of his day in front of a computer and not
a whole lot of time in meetings, for example, where next actions come
up. Is that suspicion correct? Also, do you include personal items in
the lists or is this pretty much just for work tasks?

The way Uwe uses different files as categories and tags as meta info
is something I do as well. Here's the structure I have come up with
and that I would be happy to have critiqued.

I am using my existing Notes.org file now with a #+CATEGORY: Work, a
new Lists.org fiel with a couple of different categories, one of which
is Home, and a Projects.org file for Work with just projects.

Since I keep meeting notes in Notes.org, I have #+TAGS set up to
include contexts as well as the meta info for different technical
areas. The contexts I use are:

#+TAGS: CALL(c) COMPUTER(o) ERRANDS(e) { @WORK(w) @HOME(h) } AGENDA(a) READING(r) 

I decided to make the general lists categories TODO types:

#+TYP_TODO: NEXT WAITING SOMEDAY PROJECT DONE

I am not sure yet that I really happy with this, but at least I can
show my TODO list now and have items listed as:

  Work: NEXT Do software thing :COMPUTER:Software:
  Work: WAITING Delivery of software :Software:
  Home: SOMEDAY Books to Read
  Home: NEXT: Buy Stamps :ERRANDS:

Or something along those lines. I am not really sure what to do with
the meta tags and whether I should attach them to WAITING items, for
example. There is no way to limit a view by tag or search string in
the TODO view, is there? That would make them a lot more useful to me.

Alex.

Re: Another GTD question.

[Carsten Dominik]

On Sep 30, 2006, at 7:25, Alex Bochannek wrote:
>
>   Work: NEXT Do software thing :COMPUTER:Software:
>   Work: WAITING Delivery of software :Software:
>   Home: SOMEDAY Books to Read
>   Home: NEXT: Buy Stamps :ERRANDS:
>
> Or something along those lines. I am not really sure what to do with
> the meta tags and whether I should attach them to WAITING items, for
> example. There is no way to limit a view by tag or search string in
> the TODO view, is there? That would make them a lot more useful to me.

Can you formulate an example search you would like to do?

- Carsten



--
Carsten Dominik
Sterrenkundig Instituut "Anton Pannekoek"
Universiteit van Amsterdam
Kruislaan 403
NL-1098SJ Amsterdam
phone: +31 20 525 7477

Re: Another GTD question.

[Piotr Zielinski]

Hi,

Org-mode is a major part of my GTD scheme, however, there are some
tasks for which I think it's not ideal.  One example are lists of
"lightweight" items; items that are not critical and nothing seriously
bad would happen if they are forgotten.  Interesting webpages to read
or papers to read/print are good examples.


FORGETTABLE LISTS

The following requirements describe the problem:

1. Adding new items must be as easy as possible, preferably just a
   single click of the browser button "Mark this page for future
   reading".  This is the most important requirement, which is not
   currently met by org-mode.

2. The list works like a stack: adding is possible only on the top.
   This keeps the items on the list in the approximate order of
   decreasing relevance to my current interests.

3. Only the say 10 most recent items are of any relevance.  If any
   earlier items haven't been acted upon, they are not relevant
   enough: I have at least 10 more interesting things to do.  I don't
   want to see such items; if they become relevant in the future I can
   always add them again.


MY LISTS

1. "To Read" for webpages/papers I want to read
2. "To Print" for pdfs to print when I'm in the office
3. "To Use" interesting internet services to try
4. "To Buy" for gadgets I might be interested in buying
5. "To Supermemo" for concepts to memorize with supermemo [1]
6. "Books to Read"
7. "Movies to Watch"


IMPLEMENTATION

Currently, I use del.icio.us to maintain them.  I have a menu in the
Toolbar menu that contains bookmarklets like "Add To Read" with the
following URL (one line)

javascript: function loadScript(scriptURL) { var scriptElem =
document.createElement('iframe'); scriptElem.setAttribute('src',
scriptURL); document.body.appendChild(scriptElem);}
loadScript('https://api.del.icio.us/v1/posts/add?tags=to-read&url='+encodeURIComponent(location.href)+'&description='+encodeURIComponent(document.title))

and also Live Bookmarks that point to the RSS of the relevant lists,
and display the first 10 or so items on each list.


METALISTS

Here are some places I use for maintaining lists: org-mode files,
browser bookmarks, amazon basket, amazon wishlist, amazon recommended
books, watchthatpage, google alerts, delicious, movielens, citeulike.
Each of them is different, has its strengths and weaknesses, and it
would be very difficult to integrate them into a single system.  My
solution at the moment, is just to have a metalist: a list in my
org file containing URLs of all my lists.

Piotr

[1] http://www.supermemo.net.pl/index.net

Re: Another GTD question.

[Chris Lowis]

> Here are some places I use for maintaining lists: org-mode files,
> browser bookmarks, amazon basket, amazon wishlist, amazon recommended
> books, watchthatpage, google alerts, delicious, movielens, citeulike.
> Each of them is different, has its strengths and weaknesses, and it
> would be very difficult to integrate them into a single system.  My
> solution at the moment, is just to have a metalist: a list in my
> org file containing URLs of all my lists.

You might not find it too difficult to integrate all of these lists
using a bit of coding. Most of the services you mention probably have
APIs, and it's likely there are some libraries for the scripting
language of your choice out there.

I wrote a quick bit of Ruby code, with a little elisp function to call
it, to pull my bookmarks from ma.gnolia into an emacs buffer, for
example. I don't know Ruby too well, but with the help of a library I
found, this was a 15 minute job. Extend this to all your other
services, and it'd be fairly easy to have a "master-list" in Emacs.

Chris





-- 
Chris Lowis
http://www.chrislowis.co.uk/

Re: Another GTD question.

[Alex Bochannek]

Carsten Dominik <dominik@science.uva.nl> writes:

> On Sep 30, 2006, at 7:25, Alex Bochannek wrote:
>
>>
>>   Work: NEXT Do software thing :COMPUTER:Software:
>>   Work: WAITING Delivery of software :Software:
>>   Home: SOMEDAY Books to Read
>>   Home: NEXT: Buy Stamps :ERRANDS:
>>
>> Or something along those lines. I am not really sure what to do with
>> the meta tags and whether I should attach them to WAITING items, for
>> example. There is no way to limit a view by tag or search string in
>> the TODO view, is there? That would make them a lot more useful to me.
>
> Can you formulate an example search you would like to do?

For starters a 'N t' to limit the view by tag would be good.

Maybe limiting the view by regexp would be a nice addition. 's' maybe?

Alex.

Re: Another GTD question.

[Charles Cave]

Alex Bochannek wrote:

> Charles unintentionally gave a great example of what I have discovered
> bothers me a bit about the different lists in GTD:
>
> The items tagged PHONE, ERRANDS, OFFICE are pretty much obvious,
[as contexts for lists]

> although the "Server Install" should probably be on a PROJECT list of
> some sort. The READING, BOSS, and WAITING are trickier I
> think. The tags are using the list names where most Next Action
> sublists are essentially contexts.

Your observations are correct. I was pushing the use of org-mode
tags to cater for identifying contexts for lists as well as making
pure lists.

I think it is best to structure the org-mode file to keep the
agenda items for each person separate from the tags.
So, if I want to make a list of things to talk about with Andrew,
I will have a section for Andrew, and similarly for items to discuss
with Belinda.

* Agendas
** Andrew
*** My annual performance review
*** Discuss PROJECT X delivery date
** Belinda

Charles

Re: Another GTD question.

[Alex Bochannek]

Charles Cave <charles_cave@optusnet.com.au> writes:

> I think it is best to structure the org-mode file to keep the
> agenda items for each person separate from the tags.
> So, if I want to make a list of things to talk about with Andrew,
> I will have a section for Andrew, and similarly for items to discuss
> with Belinda.
>
> * Agendas
> ** Andrew
> *** My annual performance review
> *** Discuss PROJECT X delivery date
> ** Belinda

This is definitely something I am struggling with. I haven't been
using my AGENDA tag as much since I end up just sending people email
anyway and therefore put a COMPUTER tag onto it.

What I did discover is that putting the projects into the same file as
my notes instead of a separate agenda file is working out a lot
better. I now intersperse next actions into my notes only if they are
not project related, but still keep everything in one file. I think I
will try the same with agendas and make AGENDA a TODO type rather than
a tag.

Thanks for the suggestion!

Alex.

Re: Another GTD question.

[Uwe Jochum]

Since orgmode has the ability to collect the information that is
distributed on different files into one agenda view, I do it this way:

I have for each of my complex projects its own org-file. At the head of
that file I place my todo-list that is connected to the given project,
below that comes everything else: notes, outlines and drafts; if I have
to write more than one paper for that project, I link it to the project
master file (the file with the todo-list). That gives me the context I
need. I add this file to the list of agenda files, et voilà: the
todo-list of that specific file shows up in my daily/weekly agenda
view. I think this is what you intend with "categories" (files).

I use tags for meta-information (if I may say so): a tag "research" is
used for an item of my todo-list (of whatever org-file) that has to do
with research. Well... This way I can see all my research-related
actions or todos, regardless in which file this information is burried.

For simple todos that are not connected to complex projects but collect
things that have to be done somehow, sooner or later, I use a separate
task-file that collects all of this stuff.

I don't know if this is the "reine Lehre" (pure philosophy) of GTD, but
in my view it comes more or less close to that.

Best,

Uwe

Re: Another GTD question.

[Carsten Dominik]

I really like this discussion started by Alex.
It has triggered for me a lot of
thinking and clarity about how to use Org-mode for a GTD system.  High
time, because my current system basically is "do whatever the closest
person pointing a gun to your head is asking".  Has kept me alive, if
stressed :-).

Charles Cave's [article/tutorial] gives a great overview over the basic
structure of GTD and his views on implementing GTD with org-mode.
Below are a few thoughts on how GTD elements can be represented in
org-mode.

2 The (too?) many organizational elements of Org-mode
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

There are many ways to apply structure to your notes using Org-mode:

- Categories (i.e. files)
- Lists (outlines)
- TODO keywords
- TAGS

and many of these can be used interchangeably.  For example, if I am
collecting the things I have to discuss with my colleagues Peter and
Sarah, I could

- use lists

: * Agendas
: ** Peter
: *** item p1
: *** item p2
: ** Sarah
: *** item s1
: *** item s2

- use TODO keywords

: #+TYP_TODO: Peter Sarah
:
: * Project X
: ** Peter item p1
: ** Sarah item s1
: * Project Y
: ** Peter item p2
: ** Sarah item s2

- use TAGS

: * Project X
: ** item p1          :Peter:
: ** item s1          :Sarah:
: * Project Y
: ** item p2          :Peter:
: ** item s2          :Sarah:

The same is true for contexts like `@work', `@home', `@computer' etc
as they are being used in GTD.  You could make a list of things to be
handled at your computer at home, or you could use tags for contexts.

So what is the best way to approach these issues, what method should
be preferred and why?  I think his is the core of the present
discussion.

For me personally, the main advantage of Org-mode is that I can keep
information relating to a project together in one place.  This is best
for many reasons, for example
- things that belong together, stay together
- easy review if a project is stuck

So I would not generally make lists for a specific contexts or people.
Lists for a specific person are unlikely as well.  Most of the time I
would use either TAGS or TODO keywords, also because the search
functions for tags and TODO keywords are the most powerful ones in
Org-mode.

3 CATEGORIES for broad splitting of the GTD system
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I am using separate files for things I need to do for HOME
and for WORK.  At work I use one big file for most things, but the
biggest tasks/projects I split off and put them into a separate file.

4 TAGS versus TODO keywords versus Lists to implement GTD elements
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


4.1 Context
===========

    I think most of us agree that contexts (location and required tools
    for doing a task) is something best implemented with TAGS in the
    org-mode system.  Tags like @home, @work, @phone, @computer, @mall
    can be defined and easily applied to any tasks that need that
    particular context.

4.2 The GTD lists as *status* of a task
=======================================

    Another important part of GTD terminology are different lists that
    contain tasks, for example NEXT ACTION, WAITING, or SOMEDAY/MAYBE.
    In the original GTD terminology these are called lists.  When
    thinking about implementation in Org-mode, it occurred to me that
    these are better called /status of a task/ which is then used to
    make corresponding lists.  Here is what I mean by status:

    - TODO: this is something that needs to be done, no further
      specification if this can be done now or not.
    - NEXT: this is something that can be done /now/, you have
      everything you need to start doing it.  This is what /next
      actions/ are about.
    - WAITING: This item cannot be done now, because we are waiting for
      something.  Somebody else needs to act, some material needs to
      arrive, etc.
    - SOMEDAY: means that you have not decided that this needs to be
      done.

    How should we go about implementing this structure in org-mode?

    1. We could make physically separate lists for each task status.
       As I said, I don't like this idea and will not discuss it
       further.

    2. We can use TODO keywords to implement these different states.
       Each time the state changes, we switch to a different TODO
       keyword.  This is very easy from an Agenda view: `1 t', `2 t',
       `3 t' etc directly switch to the corresponding keyword.  In the
       buffer, try `Shift-left/right' with the cursor on the keyword.

    3. We can use TAGS to implement this structure.  So each TODO item
       would have an additional tag, identifying the state of the task.
       - Advantage: you keep the simple on/off of a TODO item.
       - Disadvantage: When you mark an entry DONE, the NEXT tag (or
         whichever the current status is will stick around and put this
         item into your NEXT ACTION lists.

    Which of these two possibilities you choose really depends on your
    personal taste.  Since version 4.52 of org-mode matching TODO
    keywords has become as easy as matching tasks, so also from the
    technical point of view there is no preference. I am personally
    inclined to try option (2) first.

4.3 Agendas
===========

    Charles and Pete have already discussed here about making agendas
    (things to discuss with a particular person or group) either lists
    or tags.  I agree with them that it is best to keep tasks in the
    project context and use tags to produce the relation to a person.
    However, org-mode also gives you flexibility here.  Lets say you
    have to discuss a number of things with a person that is not
    related to projects, but for example to their and your personality,
    interaction with other people etc.  So you might want to sit down
    to write an agenda for discussing with them.  In this case, simply
    /also/ tag this special list with the name of the person/group.
    The a tag search will later link you to scattered items as well as
    the specific list you have drawn up.  For example:

    : * Agendas
    : ** Peter                               :Peter:
    : *** Issue 1
    : ** Sarah                               :Sarah:
    : *** Issue 1

5 Configuration of Org-mode
~~~~~~~~~~~~~~~~~~~~~~~~~~~

5.1 Option 1: TODO items are simple two-state
=============================================

5.1.1 Configure by using in-file options
----------------------------------------

    : #+TAGS: { NEXT(n) WAITING(w) SOMEDAY(s) }
    : #+TAGS: Peter(P) Sarah(S)
    : #+TAGS: { @office(o) @home(h) @mall(m) }
    : #+TAGS: { @phone(p) @computer(c) }

5.2 Option 2: TODO types cover task lists
=========================================

5.2.1 Configure by using in-file options
----------------------------------------

    : #+TYP_TODO: TODO NEXT WAITING SOMEDAY DONE
    : #+TAGS: Peter(P) Sarah(S)
    : #+TAGS: { @office(o) @home(h) @mall(m) }
    : #+TAGS: { @phone(p) @computer(c) }

6 Creating the GTD lists
~~~~~~~~~~~~~~~~~~~~~~~~

With a system set up as described above, you can easily create all
those lists you need for GTD.  For the example below I am assuming
that we are using TODO keywords for the status of a task - if you are
using tags for this as will, just reformulate the search to match a
tag instead of a todo keyword.

- Next actions at home
   TAG search for "@home//NEXT"

- What actions am I waiting for that Sarah has to do?
   TAG search for "Sarah//WAITING"

- All items to discuss in a meeting with Sarah and Peter
   TAG search for "Sarah|Peter"

etc etc etc.

Re: Another GTD question.

[Piotr Zielinski]

Hi,

In relation to Carsten's email, I'd like to ask about possible
integration of headings and plain lists.  I remember that such
integration was difficult because of the implementation of
outline-mode, but I don't know the details.  In case this is possible,
here are a few reasons why I'd like it:

1. As opposed to headings, plain list items can consists of more than
   one line.  On the other hand, they cannot be assigned tags or
   marked TODO.  So, sometimes, one has to use one or the other.  This
   is especially problematic, if you would like a TODO item below a
   multi-line plain list item.  It would be great if the features of
   headings and plain lists could be, at least to some extent,
   combined.

2. Code duplication.  At the moment, AFAIK, you need to write separate
   code for headings and plain lists.  Also, some features exist for
   both but in different forms, for example, TODO/DONE for headings is
   essentially the same as [ ]/[X] for plain lists.  I don't mind
   having different representation for the same concept as long as
   their share the same handling code.

However, I feel that any attempt at integrating integrating headings
and plain lists would require a significant rewrite.  Carsten, could
you please comment on the main difficulties of such integration?

Thanks,
Piotr

Re: Another GTD question.

[Carsten Dominik]

Hi Piotr,

thanks for sticking my head into that dark hole again :-)

Yes, the issues of headlines and plain list items have been bugging me 
for a long time, and am I glad to share my thoughts here - if only to 
grab this chance to organize them again.  Maybe some kind of solution 
will present itself eventually.

First of all, there is one subtle but critical structural difference
between the two concepts of outline headlines and items.

A headline starts a new subsection in a tree.  The body below the 
headline
only ends at the next headline.  This is, in fact sometimes unfortunate,
because you can never go back to the level of text where you were.
For example:

    * level one
      some text
    ** item1
    ** item2
      And now I would like to continue with the thought started with 
"some text"
      but I cannot!

Items allow you to do just that

    * level one
      some text
      - item1
      - item2
      And here we continue after "some text"

I see this problem as the single most frustrating issue.  When I want to
record a TODO in the middle of some longish text, I ----- cannot reaaly
do so.  Therefore, allowing plain list items to become TODO entries is
something I *really* want.

I am not sure if *anyone* has actually noticed, but org-mode does not 
provide proper outlining support for plain list items.  If you set 
org-cycle-include-plain-lists, org-mode is cheating.  Items are 
temporarily (for the duration of an org-cycle command) treated as 
outline headings.  So in the example above, "And here we continue..." 
is actually treated as part of item2, betraying the true structure.

So one thing to be fixed is definitely this: proper cycle functionality 
for plain lists.  The reason why this is more difficult is that to find 
the end of an item, you cannot simply use a regular expression search 
for the next heading, you have to walk down line by line to find the 
next line with the proper indentation.  That is why the handling code 
for promotion etc is separate.

If you are looking for common handling code for things like promotion 
and demotion, you can go one level up and use function like 
org-shiftmetaleft, which do the proper dispatching.

About checklists and TODO items, I think that it makes sense to have 
both, one lightweight, and a more heavy one for use in agenda buffers.

But what would be nice is to have the possibility to make a plain list 
item a TODO entry.  I'd really like this.  Here are the problems with 
it:

- A lot of the code handling TODOs was written early when there were
   no plain lists.  That code often explicitly assumes that TODO is 
preceded
   by the beginning of a line and a few stars.  Several regular 
expressions
   that are used all over the place implicitly make this assumption.  
This is
   just bad programming, but there you are, org-mode has been growing 
instead of
   having been designed from top to bottom.

- To make TODO in plain list items fully useful, I'd have to be able to
   apply tags to them.  However, other than headlines, the first line of 
a
   plain list item does not have a defined end, it can be filled and
   wrapped - so where would a good place be, where should TAGS be stored?
   Any good proposals?

- Another issue would be:  If I have a deadline or a scheduled item,
   should it refer to plain list item of which it is a part, or should
   it (as it does now) always refer to the nearest headline?  I guess
   the former would make more sense, greately increasing the complexity
   when scanning files for the agenda.  This would slow down creation
   of the agenda - maybe a price we could pay, I don't know.

The most important obstacle is that I would need something like 3
consecutive days with nothing else on my mind to make the changes
without introducing too many new bugs.  These I currently don't have,
they are very difficult to find.

Anyway, if there is ever a version 5 of org-mode, it will have these 
features
and require intensive testing. :-)


>
> 1. As opposed to headings, plain list items can consists of more than
>   one line.  On the other hand, they cannot be assigned tags or
>   marked TODO.  So, sometimes, one has to use one or the other.  This
>   is especially problematic, if you would like a TODO item below a
>   multi-line plain list item.  It would be great if the features of
>   headings and plain lists could be, at least to some extent,
>   combined.
>
> 2. Code duplication.  At the moment, AFAIK, you need to write separate
>   code for headings and plain lists.  Also, some features exist for
>   both but in different forms, for example, TODO/DONE for headings is
>   essentially the same as [ ]/[X] for plain lists.  I don't mind
>   having different representation for the same concept as long as
>   their share the same handling code.
>
> However, I feel that any attempt at integrating integrating headings
> and plain lists would require a significant rewrite.  Carsten, could
> you please comment on the main difficulties of such integration?
>
> Thanks,
> Piotr
>
>

--
Carsten Dominik
Sterrenkundig Instituut "Anton Pannekoek"
Universiteit van Amsterdam
Kruislaan 403
NL-1098SJ Amsterdam
phone: +31 20 525 7477

Re: Another GTD question.

[Christopher Kuettner]

> - A lot of the code handling TODOs was written early when there were
>   no plain lists.  That code often explicitly assumes that TODO is preceded
>   by the beginning of a line and a few stars.  Several regular expressions
>   that are used all over the place implicitly make this assumption.  

In outline mode there is the possibility to replace the stars with an 
reg-expression.  That means you can replace the star as the 
headline-indicator.  Maybe you can take some code from outline-mode.


> - To make TODO in plain list items fully useful, I'd have to be able to
>   apply tags to them.  However, other than headlines, the first line of a
>   plain list item does not have a defined end, it can be filled and
>   wrapped - so where would a good place be, where should TAGS be stored?
>   Any good proposals?

maybe you can rise the importance of org-tags-column like in "if a ":" 
is here, than this is a tag.

Aside from that...

What is the basic design model for org-mode?  What is org supposed to 
be?  Where it is headed? I thought I  got an outliner with 
dates-capabilities.  No it's almost a full fledged publishing platform...

I think you did a terrific job so far.  Maybe you have to make some 
fundamental decisions here...

Regards,
Christopher

RE: Another GTD question.

[Eddward DeVilla]

On 10/21/06, Christopher Kuettner <ckuettner@gmail.com> wrote:
> Aside from that...
>
> What is the basic design model for org-mode?  What is org supposed to
> be?  Where it is headed? I thought I  got an outliner with
> dates-capabilities.  No it's almost a full fledged publishing platform...
>
> I think you did a terrific job so far.  Maybe you have to make some
> fundamental decisions here...

Those are some interesting questions.  I certainly can't answer any of
them, but I don't know one aspect of Org that I hope is maintained.
It's really flexible.  Kind of like perl.  It has a lot of little
nifty features that you can use to manage and organize information (to
tasks, or whatever) and use can use any subset you want.  You can
learn it incrementally.  (...as I have.  It sound popular but I
haven't even touched the publishing...)  None of the features really
require the use of any other feature except maybe agenda and agenda is
just a flexible interface for gathering the information marked and
managed by the other features.  (Dates, tags, Todo state...)  Yet all
of the features work well together.  There's more that one way to do
most things.

I don't really understand GTD, elisp or project management that I
would try to guess a good direction for Org-mode, but I do hope it is
able to maintain a design where you can pick and choose the features
and assemble them as they suit you instead of trying to impose a
framework or style.

And yes, Carsten and company of done an excellent job.  For all I've
pestered him and the list, I don't say that enough.  I just picked
org-mode because I was looking for a replacement an orphaned outliner
that I depended on.  Org turned out to be better in many ways and has
since surpassed it in all ways.  It's changed how I manage list,
projects, todo and information in general.  I'm actually excited to
see how it will grow.  You've done great.

Re: Another GTD question.

[Carsten Dominik]

On Oct 22, 2006, at 2:39, Eddward DeVilla wrote:

> ... but I don't know one aspect of Org that I hope is maintained.
> It's really flexible.  Kind of like perl.  It has a lot of little
> nifty features that you can use to manage and organize information (to
> tasks, or whatever) and use can use any subset you want.  You can
> learn it incrementally.  (...as I have.  It sound popular but I
> haven't even touched the publishing...)  None of the features really
> require the use of any other feature except maybe agenda and agenda is
> just a flexible interface for gathering the information marked and
> managed by the other features.  (Dates, tags, Todo state...)  Yet all
> of the features work well together.  There's more that one way to do
> most things.

I am smiling happily at this description, it reflects very much of
what I am trying to do with org-mode.

- Carsten

Re: Another GTD question.

[Xiao-Yong Jin]

Carsten Dominik <dominik@science.uva.nl> writes:

> On Oct 22, 2006, at 2:39, Eddward DeVilla wrote:
>
>> ... but I don't know one aspect of Org that I hope is maintained.
>> It's really flexible.  Kind of like perl.  It has a lot of little
>> nifty features that you can use to manage and organize information (to
>> tasks, or whatever) and use can use any subset you want.  You can
>> learn it incrementally.  (...as I have.  It sound popular but I
>> haven't even touched the publishing...)  None of the features really
>> require the use of any other feature except maybe agenda and agenda is
>> just a flexible interface for gathering the information marked and
>> managed by the other features.  (Dates, tags, Todo state...)  Yet all
>> of the features work well together.  There's more that one way to do
>> most things.
>
> I am smiling happily at this description, it reflects very much of
> what I am trying to do with org-mode.
>
> - Carsten

I'm wondering if you could absorb all the nifty features from
emacs-muse or planner, etc.  Especially various export formats and
more text markups?

Xiao-Yong
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---

Re: Another GTD question.

[Carsten Dominik]

On Oct 23, 2006, at 9:21, Xiao-Yong Jin wrote:

> I'm wondering if you could absorb all the nifty features from
> emacs-muse or planner, etc.  Especially various export formats and
> more text markups?

In this generality: no.

The focus of muse is publishing, and that is not the focus of org-mode. 
  So blindly taking over more features is nothing I'd like to do.  If 
you have specific things you'd find particularly useful, and if you 
take the time to describe exactly how they should work instead of 
expecting me to do that, you chances are much better :-)

- Carsten

Re: Another GTD question.

[Xiao-Yong Jin]

Carsten Dominik <dominik@science.uva.nl> writes:

> On Oct 23, 2006, at 9:21, Xiao-Yong Jin wrote:
>
>> I'm wondering if you could absorb all the nifty features from
>> emacs-muse or planner, etc.  Especially various export formats and
>> more text markups?
>
> In this generality: no.
>
> The focus of muse is publishing, and that is not the focus of
> org-mode. So blindly taking over more features is nothing I'd like to
> do.  If you have specific things you'd find particularly useful, and
> if you take the time to describe exactly how they should work instead
> of expecting me to do that, you chances are much better :-)

No, I can't think of any now.  It just works fantastic so far.

Xiao-Yong
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---

Re: Another GTD question.

[Eddward DeVilla]

On 10/23/06, Carsten Dominik <dominik@science.uva.nl> wrote:
>
> On Oct 22, 2006, at 2:39, Eddward DeVilla wrote:
>
> > ... but I don't know one aspect of Org that I hope is maintained.
> > It's really flexible.  Kind of like perl.  It has a lot of little

wow.  Can you tell I was sleep deprived.  I *do* know one aspect I
hope is maintained.  I thought I even proof read that.  I'm afraid to
go back and read the rest.

Re: Another GTD question.

[Pete Phillips]

    Chris> Aside from that...

    Chris> What is the basic design model for org-mode?  What is org
    Chris> supposed to be?  Where it is headed? I thought I got an
    Chris> outliner with dates-capabilities.  No it's almost a full
    Chris> fledged publishing platform...

Someone mentioned that org-mode is a bit like perl. I agree. Way back,
someone described perl as "the Swiss army chainsaw of UNIX
programming". Over the last 12 months, I think org mode has evolved into
something akin to the "Swiss army JCB of organisational software" (to
stretch a metaphor until it screams for mercy!).

The 3.14 manual (May 2005 ?) says:

    Org-mode is a mode for keeping notes, maintaining ToDo lists, and
    doing project planning with a fast and effective plain-text system.

    Org-mode develops organizational tasks around NOTES files that
    contain information about projects as plain text.  ...... Org-mode
    supports ToDo items, deadlines, time stamps, and scheduling. 

The latest manual (up to 85 pages from 42!) makes the same claim.

So I think the basic design model is clear. Yes, it has had a massive
amount of development over the last year - the tags functionality (which
has its own section in the manual) was the function that made org-mode
indispensable to me, and the other major addition was the publishing
function (which I have never used, but clearly there are some users
who have found this the indispensable part of org-mode). However, the
underlying goals of maintaining TODO lists with a fast plain-text system
remains unchanged.

    Chris> I think you did a terrific job so far.  

And like others, I feel I don't say this often enough. Carsten is a real
star - I assume he has a day job where he has to do work to earn his
keep :-) but the level of support is astounding. I have had a few major
issues in the last year where I have been sending files and emails back
and forth to Carsten during the day, and he has fixed the problem within
a few hours. 

    Chris> Maybe you have to make some fundamental decisions here...

I disagree with this. As far as I can see, the development has been very
much in accordance with the original design criteria. Yes, it is now
much more sophisticated, and it allows you to do the same thing in many
different ways. This to me is a benefit of org-mode - I keep learning
new things I can do.  For example, the link with diary/calendar mode is
awesome - and this week I learned that I can change the dates of
Deadlines etc in the Agenda buffer using the Shift-Cursor keys, which
has saved a whole lot of tabbing to the org-mode buffer to change each
date, then 'Ctrl-X B'ing back! Very slick.

I implement a complete GTD system using org-mode, which I use
intensively - it runs my working and home life. At present I have 198
Work projects (using David Allen's definition of a project), and around
100 Home projects which I am managing with org-mode plus the 43 physical
folders and project files. If I ever feel that it is going in the wrong
direction, I will make my feelings known, as I depend on it.  

However, at the moment, I do not think Carsten needs to make any
fundamental decisions, as development appears to be continuing on the
right track.

Pete

Re: Another GTD question.

[Jason F. McBrayer]

Carsten Dominik <dominik@science.uva.nl> writes:

[A summary of the main options for implementing GTD in Org-mode.]

I'd just like to follow this up by saying that Option 2 (TODO keywords
are TODO NEXT WAITING SOMEDAY DONE; contexts and people are tags) is
basically the practise I've evolved for doing GTD with Org-mode, as
well.

My main point of indecision right now is one file vs. many files.  At
home I've had one file, and at work I've been trying one file per
project.  Because of the way timekeeping works, however, it looks like
I'll either be moving all of my projects back into one file, or
writing a lisp function to do a time summary table for all files in
org-agenda-files.

It is looking to me currently like one file for projects and tasks is
the way to go, with other files being useful mainly for longer notes,
journal entries, and references...


-- 
+-----------------------------------------------------------+
| Jason F. McBrayer                    jmcbray@carcosa.net  |
| A flower falls, even though we love it; and a weed grows, |
| even though we do not love it.            -- Dogen        |