Subscribe to murtworld by Email



Life in files

Monday, September 05, 2005

For various reasons, I'm a little hesitant to go the way of "life in one big ass text file," although I can appreciate the benefits. I have decided, however, to try text files in general, along with some other (de facto) standard document types: PDF, Excel, HTML.

I've dabbled with this before, but never with the clear vision and overall gumption to make it work as I have now. For so many years I've wrestled with various information managers; perhaps initially the clunkiness of file systems led me down the path of searching for something better. The promise of a single, integrated store for all of my information was overwhelmingly tempting. What I didn't fully grasp at the time, however, is that the single-file/dedicated application approach comes with its own clunkiness and a lot of downsides.

Using info management software does have a lot of benefits. There are a lot of functions available that are not immediately available with plain text. But if you end up switching apps a lot as I have in the past decade, the benefits of each app are negated by the fact that all your info isn't in one place or format anymore - and usually there is no way to search through all the different apps in a reasonable fashion. While there are tools that will search the contents of any file regardless of type, it's generally a slow affair when you have thousands upon thousands of proprietary (and usually binary) files to search. And what if you can't install the software for some reason in the future? I have some text files from as far back as 1990 or so that are still instantly accessible, but I'm sure I have some info in some trial info manager from last year that I can't get to now. With text, your data is transparent and can be opened and manipulated with just about anything.

The recent release of Google Desktop v2 beta prompted me to try the plain text approach again. A nice global search provides the glue that can be missing if you just try to deal with files on their own. And since I live in text editors all day, Google Desktop is really just a fallback mechanism, since I can easily search through plain text files with a number of different methods. Add to this fact the lightweight nature of text files and the ability to easily interface with a plain text data store via scripting languages and I have to wonder why I held out so long for the unfulfilled promise of the info manager application approach.

Plus, as clunky as the Windows file system may be, it does provide a built-in tree hierarchy - the directory hierarchy. I'm now just using standard Windows directories for the hierarchy I'm used to creating in info management/PIM apps. I have a pretty firm personal hierarchy/ontology at this point, and combined with a real file manager like Directory Opus, the tree-based approach works just fine with normal files. I'd suggest doing this in something like c:\data instead of c:\documents and settings\chris\my documents - what an icky path.

On top of the text file approach for most lists and notes, I'm using Excel spreadsheets for things like budgets and other lists where I want to do math (yes, I know that sounds stupidly obvious but it's amazing how the "information manager" mindset can lead you to avoid files and even other applications like the plague) and the Scrapbook extension for Firefox for web clippings (which, sadly, I can't eliminate from my life). I've set the Scrapbook data directory to be c:\data\scrapbook, so this keeps everything in one directory that's easy to back up/transfer to other machines/etc.

We'll see how this all plays out, but I've got a feeling it could be for the long haul. I've wasted too much time in my life shoveling data into applications that might be cool but end up just being constricting. I've wasted too much time switching back and forth between a few of my favorite info managers. I've always loved HTML/CSS for its standards- (too bad certain browser makers can't see the light) and text-based goodness; now I'm just applying that to the rest of my stuff. I could go on and on about this but I'm breezing through it - my Google AdSense monitor keeps me honest by reminding me that blogging is not how I make a living.

Posted by murt at 11:14 PM


Anonymous Anonymous said...
You may want to try Brainstorm. It's a text-based note-taking system that has links... pretty interesting approach. I think it can import or export opml as well.
10/21/2005 1:22 PM  

Anonymous chrisw said...
Assuming GoogleDesktop can scan zipped file archives, you'd be able to aearch through those budgets etc too by instead doing them in, which saves in the compact OASIS OpenDocument XML format. (It can read/write .XLS, .DOC, .PPT,, .HTML, WPS, etc etc too.)
Hope this gives you some return - thank you for your blogs :)
8/23/2006 7:53 PM  

Blogger murt said...
I will probably move to OpenOffice as I upgrade computers, as I am tired of paying for Office.

Also, I am a fan of BrainStorm and use it often.
8/29/2006 2:52 PM  

Post a Comment

Links to this post:

Create a Link  |  Home