Posts Tagged ‘VimOutliner’

My software stack revisited – Organizing

Monday, December 27th, 2010

Back in the original post, I don’t think I actually had this category at all. In fact, I do believe all the applications I am about to list in this post are additions to my stack.

Knowledge management

Knowledge is power. But only if you have that knowledge accessible and stored in a fashion which makes the retrieval of the relevant parts of it both simple and timely.

This is more true the more knowledge you accumulate.

For storing growing amounts of knowledge, I have come to find a wiki the best alternative. It does require some initial thinking, about how to structure the information being saved within it, but in my experience the benefits outweigh the work. Simply think of it as an investment.

Some of the knowledge I collect is pure information which no one but me should have access to, so I didn’t especially like the idea of putting the stuff into a web-based wiki. And that’s where desktop wikis comes into relevancy.

I tried out Zim, a GUI-centric desktop wiki. It works great, and I do recommend it. And it is as lightweight as they come. But… that was before I realized there is a Vim plugin: VimWiki.


For some reason I find it much easier to break down problems into smaller pieces if I use software (or hardware, pen and paper FTW), than if I do it in my head.

Whatever the reason for this is, I have found a friend in another Vim plugin: VimOutliner. One of the tasks that VimOutliner helped me split into smaller pieces, and organize? This series of posts.


Like with the knowledge management (and Zim), I have already managed to iterate through a solution, which consisted of the CLI-based cal, which just prints out a simple three month view (previous, current, next) and should be included by default in most if not all GNU/Linux distributions, when which is a script that parse a file filled with rule-based constraints, all detailing dates, or ranges of dates, when things happen.

when is an excellent piece of software, and if you can take the syntax (me, being a programmer, found it quite straightforward) but I began feeling a lack of putting it all in context easily when adding new dates (specifically if I already had appointments on a specific date, and what dates would possibly be better).

Which made me try out calcurse. Not to replace when, but to complement it. Sadly neither when, nor calcurse, saves date/appointment data in a format which the other recognize. In fact, both save in a format only they themselves understand. This made synchronizing between the two inefficient.

I said that when is excellent. I also said that I’d iterated through a solution, which means I am no longer using when. And I wouldn’t replace an excellent piece of software with a mediocre one, so whatever I replaced when with, would need to be at least as good as when, but without the drawbacks.

Enter bumby who came along and urged me to try remind which does the exact same thing as when, but with a different syntax. But remind has one big advantage. There is another program, wyrd, which is to remind what calcurse was to when, with one difference: wyrd doesn’t save appointments in a format of its own. It is just a wrapper around reminds appointment files.

All these files, whether from when, calcurse or remind, are of course all plaintext.


Finally, comes an address book, simply named abook. It is an ncurses-based lightweight little application for storing and viewing contact information about people.

Thunderbird provides the same thing, so some duplication of effort / breaking of DRY is occurring here, but I feel it justified as I could fetch relevant information from abook, and be on my way contacting the person, before Thunderbird has loaded and come into a usable state.

The next post will be focusing on the largest single change to have been implemented since the original post: the addition of a server.