BrianInGermany / Xzerpt

A framework for clustering research notes

Xzerpt: a framework for clustering research notes

Description:

Xzerpt uses idea clustering and source management to bring structure to your research notes and keep track of what ideas you found where.

Record your ideas in a online interface and then create printable Xzerpt cards you can cluster into stacks on your desk.

Xzerpt is hosted at xzerpt.com.

Usage

1. Data collection

• Conduct research and fill out Xzerpt templates:

  

  • The template contains the following fields:

    • Type: quote, paraphrase, or own opinion
    • Source: unique name or nickname you keep source details saved under (wherever)
    • Page: page number in source (0 for without pages i.e. website)
    • Topic: brief summary of idea
    • Statement: the idea itself

  • Source details should contain the name/nickname used in the csv and a way to find the source (URL, book title, ISBN, etc):

    Xzerpt
      https://github.com/BrianInGermany/Xzerpt
    brian
      https://github.com/BrianInGermany
    bible
      https://www.bible.com/

  • Pressing Cache Card will add the card to temporary storage. When you're done caching cards, click Download and Clear Cache to get your CSV file.

  • Caveat: The | symbol is the delimiter and using one in the template will break the CSV creation. Substitute them with some other character.

2. Card generation

• Click the Create Cards from CSV button on the homepage to open the card creator.

• Drag and drop Xzerpt CSVs into the field, and press Create Cards:

  

• Sample output:

  

• Print the website using the browser to get your cards

  • Print to A4-size PDF and then print the PDF two-to-a-page for A6 index cards.
  • On A4 size without headers, the cards print two-to-a-page

3. Card clustering

• Create a coordinate system on your desk with 1, 2, 3, 4 on one axis and A, B, C, D, E on the other. Make it big enough to put one idea card in each cell.

• Cluster similar Xzerpt cards together. Once two or three cards are in one stack, name the stack. Write down its name on this chart (pdf link):

  

• Rename stacks as often as you like! The point is to have an appropriate name at the end. You may also want to merge or split stacks as you go.

• When you are finished, you have stacks that represent the low-level structure of your information, like this:

  

4. Stack clustering

• Once you have a set of named stacks on your table, label each stack with a post-it according to your names in the chart.
• Now we're going to cluster the clusters to create a structure for our research!
• Repeat step 3. with a blank chart, this time clustering stacks instead of individual cards and naming these superstacks instead of the stacks.

5. Sequentialization

• For each superstack, put its stacks in a logical order. The resulting sequence represents the high-level structure of the information you collected.
• Optional: For each stack, put its cards in a logical order. This is your low-level structure.
• You should now have one big heap of cards containing ordered superstacks which contain ordered stacks which contain individual cards.
• This is your table of contents!

6. Composition

• The final step of the framework is to weave your superstacks, stacks and cards into a new composition.
• Use ordered superstacks and stacks to write table of contents.
• Put the heap of ordered stacks and superstacks on your desk, and integrate the cards' contents one by one into your paper, presentation or speech.

Background: Excerpting

The system behind Xzerpt is a type of academic note-taking called excerpting (German: Exzerpieren)

• Taking Notes from Research Reading, University of Toronto
• How do I excerpt?, Uni Kassel
• Ein Exzerpt schreiben, scribbr.de
• Taking Notes, gatech.edu
• How to Take Good Notes