Conversation
|
It looks as if the intended behavior is for the test to remain as is, but that |
|
If |
|
Looking carefully, there are clearly some issues with the declared languages in this section. |
|
FWIW, the tutorial consists of notes from the 2015 Redex workshop.
The lecture notes supported live demos and worked at the time. The
problem sets were derived from working solutions.
I guess when we migrated these notes into our docs, things got shuffled
in ways that leaves copy-and-paste readers perplexed.
What you need to state first is a guideline for what you want to achieve,
what kind of reader you want to support.
Do you want readers to chase the definitions need to explore a particular
point in the notes? Text books take this approach, partly because the
chasing can serve a pedagogic role.
Do you want easily flowing test? Do you want to support it with margin
notes that link to complete files?
Do you want text interrupted by ever-growing programs? People already
find scribble’s format heavy because so little content fits into large
vertical space.
Don’t just fix local problems randomly.
|
|
The tutorial gives the impression that run can run it at any point, up to the point one is reading. To me, that is useful. For now, I will just keep sending "random" pull requests that "fix" things I run into. |
3 participants
Language
Lambda/nis defined and used in a test, but the =a(lpha) function is defined in terms of languageSDand uses thesdmetafunction that is also defined onSD.