In my work I tend to use Tinderbox for taking/organising notes on sets of articles rather than books. This might be 30-150 articles of 4–16 pages, so the overall volume can be comparable to a (large) book.

I use two key top-level containers, Sources and MyNotes. (Actually there is another to contain my writing on the topic being reviewed, but that's not essential to the note-taking part) There are prototypes set up for *Sources (including key attributes for bibliographic details, plus boolean attributes for Used (so I can check if I've not used any of the sources reviewed) and for DoNotCite, so I can exclude sources from the bibliography section of my writing), for *Notes (which has a comparable Used_Note boolean attribute. (There's also a rule to change the colour of used sources/notes so I can easily see which haven't been used.)

Because the articles tend to be fairly short I usually make the initial notes on each article within the Text for each source (i.e. there's one tbx note per source). After a while, maybe when I've got a better idea of the overall subject, I make shorter topic-specific notes (using the *Notes prototype), either putting in links back to the Source they come from or just recording and Author/Date reference to the Source. I keep all the *Notes at a single level so I can review them in the Map view, where I organise them spatially, using adornments and links. I think this is critical to getting the real value out of Tinderbox in this type of work.

The *Sources prototype includes key attributes for Theme and Topic (both sets) to help with organising and understanding. One way I've found very helpful for this is to make an agent that selects a particular Theme; set its CleanupAction attribute to "none", and then open it in Map view. Turning off the auto cleanup allows you to move the notes around at will and group, use adornments etc.

Having used keywords for themes & topics on the *Sources, I often find that when I come to write up or just to think about how new notes fit in, that the Find window is very helpful - you can set the drop-down to Theme or Topic as appropriate, type in the keyword and get just the notes matching those keywords (though it took a while to appreciate how picky the Find is for keywords within sets - you have to type the whole keyword, not just part of it, and it is case-sensitive even if you untick the box for this).

I also have a prototype *Questions in which I log things I need to find out or to do.

There are also agents set up in advance to collect *Sources added recently (i.e. in last 2 days), *Notes added recently (and also stuff I've written for output recently), which I find helpful for picking up things the next day or so.

I don't use a *Quotations prototype but achieve something similar with the "quote" badge.

Hope this isn't too far off-topic to be of interest.

I also use the Kindle reader on iPad and the techniques Ben Worthington described.

Most of what I read is not books.  PDFs from many sources.  For work and for research I do a lot of annotating, highlighting, and clipping.  "How do I get this into Tinderbox" is always in the back of my mind.  I store most of my source material in DEVONthink, and lately have synced this to the iPad, where I do my annotating and clipping in GoodReader or its relatives.  Now that iOS 4.2 is out, I'm enjoying the ability to toggle between apps, so I'm developing outlines and clippings in Carbon Fin, toggling back and forth to the PDF reader.  It's easy then to grab that notation file from Carbon Fin and import the OPML into Tinderbox.

Then the fun begins, because as I read I'm thinking of "how does this relate to that" and Tinderbox is the only place where I can fully work through the questions that the reading raises.

My answer to Mark's "how do you prepare" question is: I get a lot of funnels - on the desktop, on the iPad, on the Kindle - so that I can cull and direct the material I need for my research into Tinderbox, which is at the top level of a value pyramid.

I'm going to try and use Tinderbox as a book journal this year, as my blog will feature more mentions of books that I've read (not full blown reviews, though). I will probably organize it as follows, with values for each note:

Chronological (books read in January, February, etc.)
Genre of book (nonfiction, historical, suspense, mainstream, etc.)

I want to keep it very simple to start. Before I've kept this sort of information in notebooks. But it would be nice to have it in a digital form more easily shared.

I think those are excellent ideas, Mark. Thank you!

Thanks for all the comments.

One question: How do you coordinate the bibliographic data in Tinderbox with your bibliographic program? I use Bookends for that latter purpose. Specifically, it seems like there is a duplication of typing to enter into two programs.

Christopher,

I've done a very quick example file for you, you can hopefully find it here:  https://public.me.com/ben_worthington

I cannot post my actual reading notes as they're mixed in a file with lots of confidential stuff.  Hope this helps.

I dont think I did this in the example posted but for sorting just go into the rename dialogue box of any container and adjust the sort order at the bottom.

Ben

To inspect an agent, select the agent and choose RENAME the note menu.  Shortcut:  press Enter (fn-Return).

For example, Ben Worthington's agent for 2011 Books has the query
Code:

CompletedOn(2011) 

which means, "The attribute "CompletedOn" contains the substring 2011. This will work nicely in the UK but might not work in the US, where default settings give you two-digit years like "1/8/11".

Another way to write this might be

Code:

$CompletedOn>date("1/1/2011") & $Compeleted<date("11/2012") 

Yes another way to write this:

Code:

$CompletedOn.year=2011