A comment on another post asked, “how’s it going with PlanText?”
And the short answer is, not so well.
Not that the application is not wonderful. I find PlainText to be the best of the iPad/iPhone note taking and writing applications available. It’s better than Elements (which I find aesthetically unattractive) iAWriter (which I find functionally constrictive) and SimpleNote (which is architecturally off when it comes to its support for a text file based system).
My problem is how PlainText, the file system, and desktop editors like Notational Velocity interact. It’s caused me to rethink bits of my system, if only on the margins.
PlainText relies on folders for categorization. I think this is very smart. Folders are easy, inherent in the file system, and durable. They have the added benefit of managing sync volumes since by definition they break sets of files up into smaller groups.
But in a single folder system, like the one that is required for Notational Velocity, this categorization by folder just doesn’t work. With NV, or any other single folder based info management system, categorization relies on full text search and various in-text tags. I uses WikiCase (aka CamelCase) tags in my entries prepended with a # – against which a search with spotlight of other full text search tools usually brings up the right subset of items.
PlainText does not yet have a search function. It probably will soon. WriteRoom and TaskPaper both do, and it works great in those apps. But without search now, tags like #TNSModerism are useless. Until search comes you have to use folders, and when you do, NV is neutered.
Just as a reminder, I name all my files like this…
101113-N2-The Sunny Day is Better Than None
…where… “101113” is the date “-N2-” is a category code (that in this instance means ‘Notes that I created’ – N1 would be someone else’s text) “The Sunny Day is Better Than None” is a unique identifier (for Books the UI is “Author – Book Name”)
If it were an unfinished draft it would be …
N2-The Sunny Day is Better Than None
… since files not dated are not committed to the InfoBase and they are not dated until I’m finished with them. In a single folder system this works nicely since undated items float to the top of a file name sort.
On my desktop I can search against these files with Spotlight or NV, using the hash tags, or text, or names. But this organizational system system falls apart on mobile devices.
Which really isn’t that big a deal, I guess. Over the past few weeks I’ve found that when I’m using the iPad I actually only work on a small subset of my files anyway, and they tend to be the most current ones. I’ve also noticed that my mobile use is dominated by adding new entries rather then editing old ones. It’s rare that I need to go deep in the InfoBase on my iPad. If I really do I can use the file title search in DropBox as a stop gap.
So what I’m doing now while waiting for search to show up is breaking the InfoBase up into short term relative folders, based on the primary tags I’m using, all for the sake of PlainText. And I’m not using NV.
Once you modify the default font for QuickLook to something legible (see the MacPilot app for a solution to this), a list of folders, QuickLook and good old WriteRoom works as a passable substitute for NV.
This does all raise a marketing opportunity for some programer. There is a real need for a multi-folder, drive-by text editor, just like NV, but able to switch folders, editing as it goes. Some of the NV forks out there are getting close, in particular the one at Elastic Threads. But still, this is a function that many writers would embrace, text (or rtf) files in a folder, and an editor to work on each one sequentially.