Tagging in FSIM

Theres’ a nice chat going on the Scrivener boards about File System Information Managers. I wrote a long reply to a post and thought I’d share parts here.

Metadata and tagging

A lot of my work flow is text based (it was once rtfd, then rtf, and if you go way back, doc and whatever AmiPro file extensions were) These days I’m using a lot of in-text hash tags in the text file for sub-characterizations; kind of like MMD tagging conventions. They used to scare me, in-text tags. I ended up with a bunch of them after exporting data out of Journler (it put a ‘Tag: xxxx’ line in the header of exported files) and I found them annoying because editing them was a process of open file> edit> save file one at a time. Somewhere along the line I learned about MassReplaceIt and my fear of tagging in text files disappeared (TextWrangler works well too). I think amberV mentioned it years ago in a post, but I had to figure it out on my own.

So, if I need to (which I’ll say is rare) I’ll add #LitCrit #TNStsg or some other CamelCase thing to the first few lines. The are easy to change, easy to delete, not quite so easy to add if there is no other existing tag, unless you’re a grep wizard, which I’m not.

For non-text files I’ve used Tagger and TagList. They are OpenMeta tagging tools, they are well designed, under active development, and free. It’s kind of a poor man’s Leap or Yep. And when DropBox went from 7.x to 8.x forum releases they were syching OpenMeta nicely. But I wish I felt better about OpenMeta, there is always the sinking feeling that Apple will pull the rug out one day and all the tags we’ve put on files in that ‘reserved space’ will be washed away. So my needs are well served with Tagger, but I worry from an architectural standpoint. (My needs tend to be like sticking an OpenMeta tag ‘#ReadNext’ on a bunch of pdfs that are filed away in their home folder. Or marking some Pages or MSWord drafts as ‘#ReviseThisThur’ — very temporary workflow oriented tags.)

Through blind faith, and an amberV post, I’ve been relying on full text search a lot. This so goes against my nature of ‘putting things in their place so they can be found’ but you know, it really works. Yes, the results list come up with some oddities, but it’s almost always a short enough list to get to the file I want. And you find relationships you would never be able to ‘tag’ your way into. To druid’s point this is where the black magic of DevonThink really is great. Their semantic recognition searching finds relationships you might not have anticipated, in this area I find DT to be without peer until you get to institutional strength corporate applications. With lots of data and a faulty memory this can be valuable.

DT will also let you index against an independent data set, which satisfies many of my concerns about having data in an app (EagleFiler does also). Only the organizational information you apply against your data is at risk. If I were handed a huge pile of unstructured data, that I didn’t create, that I had to make sense of, and then work with, I’d probably reach for DT as my primary tool.

I find that full text search gets me almost as good a result on my own files, but my data sets are not of the size of most researchers. I write fiction. I make stuff up. To feed my deterministic monkey I do make up lists of file names and save them in text files, occasionally. Searching on a file name with spotlight once it’s named as per amberV’s system gets you ‘just the one file’ every time.


I’ve actually been thinking a lot about Scrivener’s place in my work flow. I once used Scrivener not only as my writing tool but also as a long term repository, like a binder in which big collections of data lived. It’s just such a great tool to work in. But more and more I use it as a writing tool only. I dump a bunch of notes in from collected text files, web pages and the like, trolling through Notational Velocity or just the file system to find things, I synthesize, expand, write, whatever magic has to be done, and then get the finished work the heck out of the tool and into a more archival format. For me that’s probably a text file or a pdf. Again, amberV talked about this in an ancient post on these boards, it just took me a while to catch on.

I think of things in Scrivner as transitory, as part of a project, which will someday be finished, the output of which will go someplace else, even if it takes years.


I will have to say that the more time that’s passed, the more I love my file system info manager. I’ve corrected early mistakes, like not making the file naming conventions abstract enough – too much specificity and the thing breaks down under its own weight. And I don’t worry about application functionality as much, because my data is always secure. I’m using apps against data, not housing my data in an app.

This entry was posted in Productivity and tagged , , . Bookmark the permalink. Trackbacks are closed, but you can post a comment.


  1. Christian Tietze
    Posted September 9, 10 at 1:50 am | Permalink

    can you please expand more on:

    “DT will also let you index against an independent data set, which satisfies many of my concerns about having data in an app (EagleFiler does also).”

    It didn’t become quite clear, so I guess you use tags to imitate amberV’s “Notebooks” (a Boswell feature) — I still have to find a way of imitating collections of notes, not only assigning them but actually being able to see everything which belongs together in something like a folder.

  2. Posted September 9, 10 at 7:27 am | Permalink

    This might help you with data-in-DT vs data-indexed-by-DT…


    But, using DEVONThink’s own words

    “DEVONthink was designed as a document repository. Import your documents, and open them in an external editor if necessary using Data > Open With, or by using the contextual menu. Use indexing only when necessary”

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>