I’m still thinking about copywork.
The previous experiment (Lovecraft Copywork) was no good.
In fact, it sucks, and not just because the UI is rough.
Using it now feels more like a typing exercise, not like copywork.
For this case to work, I think a better approach would to to create beautiful letter/a4-sized pages of text with large font that can be printed and used as a reference for the writer to hand-copy, away from the workstation.
This is key. Hand-copy. Slow things down.
A book of these examples would be great. But what’s the difference between this and using a paperback of Lovecraft stories?
- Goal (copy vs read)
- Form factor (large vs small)
- Line spacing (large vs small)
- Font size (large vs small)
What else?
- Curated for the most interesting (public domain) stories.
- Stories from related authors (e.g. yellow king)
- We could trim a story to one or two key/fun/interesting/etc scenes.
- We could have a large margin to allow notes
- We could have blank page opposite each scene to allow copywork directly into the book
- Spiral bound so it sits flat.
- Number lines, and actual lines under the text.
- Progress indicators (percent) through large stories
- Divide the book by themes, and then stories/scenes
- Vocab notes before/after/footnotes/margin notes/etc.
- Color coding literary devices (if possible)
- Word counts by story/page/etc.
- Bookmark ribbon for keeping track where you’re up to
- etc.
Very cool. Some of this is do-able with print on demand, other stuff would require a specialty book printer.
Also, I’m pretty sure there’s a bug in the JS, as the prose does not make sense from one block to the next. Oh well.
I was thinking back to copywork for programmers.
We could do the same thing, in terms of the design of the book.
Large font, large form factor, one algorithm per page, etc.
Copywork requires work, and transcribing from print back to the terminal is a heavy-enough lift I think.
The MVP is a static site, then perhaps progress to beautiful PDFs, then a book.
If I can’t let go of this, I may need to do a pilot with a class of algorithms/data structures/etc.